[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]
	England and Wales High Court (Technology and Construction Court) Decisions
You are here: BAILII >> Databases >> England and Wales High Court (Technology and Construction Court) Decisions >> Yorkshire Water Services Ltd. v Taylor Woodrow Construction Northern Ltd. [2004] EWHC 1660 (TCC) (08 July 2004) URL: http://www.bailii.org/ew/cases/EWHC/TCC/2004/1660.html Cite as: [2004] EWHC 1660 (TCC)

[New search] [Printable RTF version] [Help]

IN THE HIGH COURT OF JUSTICE
QUEENS BENCH DIVISION
TECHNOLOGY AND CONSTRUCTION COURT

		Royal Courts of Justice Strand, London, WC2A 2LL
		8th July 2004

 


Mr Justice Forbes:

1. General Introduction. This case is concerned with certain works of improvement ("the works of improvement") carried out by the Defendant ("Taylor Woodrow") to the sewage treatment works at Knostrop ("Knostrop") that is owned and operated by the Claimant ("Yorkshire Water"). At all material times, Yorkshire Water's Project Manager for the works of improvement was Earth-Tech Engineering Ltd ("Earth-Tech")
2. Knostrop receives and treats wastewater from a large catchment area that includes the city of Leeds. After treatment at Knostrop, the treated wastewater (the effluent) is discharged directly into the River Aire.
3. Taylor Woodrow is a company that carries on business as a building contractor. On 28^th August 1997, Taylor Woodrow submitted a written tender for the works of improvement. By a letter dated 26^th November 1997, Yorkshire Water informed Taylor Woodrow that its tender had been accepted in the sum of £12,458,537.19 and that the letter should be taken to constitute an order for Taylor Woodrow to commence the works of improvement with effect from the contract commencement date.
4. Yorkshire Water's engagement of Taylor Woodrow to carry out the works of improvement was under the terms of a design and build contract ("the Main Contract") which incorporated the IChemE Red Book 3^rd Edition General Conditions of Contract ("the Red Book") and provided for a contract commencement date of 8^th December 1997.
5. The Main Contract also incorporated a Specification that required that the 6 tank Sequencing Batch Reactor ("the SBR": as to which, see below), which was to be designed, supplied and installed by Taylor Woodrow as part of the works of improvement at Knostrop, should comply with certain stated performance requirements and/or to perform within certain specified parameters.
6. Stated broadly, it is Yorkshire Water's case that, at all material times since its commission in April 1999, the SBR was unable and thus failed to operate to the required capacity and/or to produce effluent of the required quality: see paragraph 5 of Yorkshire Water's Re-Re-Re-Amended Particulars of Claim ("the Particulars of Claim").
7. The First Part 20 Defendant ("Biwater") was the process sub-contractor appointed by Taylor Woodrow in connection with the works of improvement. At an earlier stage in the overall proceedings, there was an issue between Biwater and Taylor Woodrow as to whether a sub-contract had ever been concluded between them.
8. That particular question was the subject of a preliminary issue tried by me in February 2003. In the event, I decided that a sub-contract had been agreed between Taylor Woodrow and Biwater and that it was in the so-called "Yellow Book" form of sub-contract. Strictly speaking, the sub-contract had not actually been in the "Yellow Book" form, although it was convenient to refer to it as such. However, for the avoidance of doubt, the expression "the Yellow Book form of sub-contract" is a reference to the "Biwater Model Form of Conditions of Sub-Contract for Process Plants, December 1997, Rev. O", provided to Taylor Woodrow by Biwater on 21^st April 1999 with amendments: see my judgment in this Preliminary Issue dated 16^th May 2003 ("the Sub-contract Preliminary Issue") passim.
9. The trial of this action commenced on 22^nd May 2003. At that stage, Taylor Woodrow and Biwater were separately represented and delivered separate openings. After the openings, the hearing was adjourned until 9^th June. By the time it resumed, Taylor Woodrow and Biwater had reached a settlement between themselves, the terms of which were set out in a formal written agreement ("the settlement agreement") to which a schedule was attached.
10. By Clause 1 of the settlement agreement, Biwater admitted its liability to indemnify Taylor Woodrow pursuant to the terms of the sub-contract and agreed not to seek permission to appeal my decision in the Sub-contract Preliminary Issue, whilst making it clear that it was not intended to create any wider liability than that which arose under the sub-contract that had been held to exist between Taylor Woodrow and Biwater.
11. By Clause 3.1 of the settlement agreement, Biwater agreed to continue to conduct the action on behalf of Taylor Woodrow. That obligation extended to pursuing the counterclaims that Taylor Woodrow had maintained against Yorkshire Water. Furthermore, by Clause 2.1 Taylor Woodrow admitted the Biwater counterclaim against Taylor Woodrow, but Biwater in turn agreed to release Taylor Woodrow in relation to any part of that counterclaim which was not established against Yorkshire Water.
12. I do not understand Yorkshire Water to take any point as to the nature and effect of the Taylor Woodrow/Biwater settlement agreement. However, there are some outstanding issues between Taylor Woodrow/Biwater and the Second Part 20 Defendant ("Elga") with regard to the settlement agreement. As a result, it has been agreed that any necessary further argument/submissions by those particular parties on that aspect of the matter should be left over until all other issues in the case have been decided (Transcript, Day 35/29).
13. Elga was the sub-contractor, appointed by Biwater under the terms of a written agreement dated 31^st March 1998, for (inter alia) the design, manufacture and supply of the SBR. Elga is a UK associate company of U.S. Filter Inc, an American company that owns all relevant design rights in the type of SBR with which these proceedings are concerned (see paragraphs 151-152 below). US Filter Inc was sometimes known as Jet Tech.
14. At this stage, it is convenient to note that, as already indicated, Taylor Woodrow itself has made a counterclaim against Yorkshire Water that, broadly stated, consists of the following two elements: (i) various claims for outstanding amounts said to be due from Yorkshire Water to Taylor Woodrow and (ii) a claim that passes on Biwater's claim against Taylor Woodrow (also brought in these proceedings) for an extension of time together with various claims for additional and outstanding amounts said to be due from Taylor Woodrow to Biwater. Finally, Elga has a counterclaim against Biwater for £101,894.55, which largely consists of the agreed outstanding balance of its subcontract price, less retention. I will turn to consider these various additional claims at an appropriate stage after I have dealt with my findings and conclusions with regard to the principal claim in the main action, i.e. Yorkshire Water's claim against Taylor Woodrow.
15. The Main Contract Regime in Outline. At this stage in my judgment and before referring to the relevant provisions of the Main Contract, it is convenient to quote and adopt as correct part of the written outline submissions that were prepared for the purposes of the Main Contact Preliminary Issues by Mr Elliott QC on behalf of Elga, in which he gave the following general (and, I believe, to the extent quoted, essentially uncontroversial) outline of the contractual regime of the Main Contract in this case:

"1. The contract between (Yorkshire Water) … and (Taylor Woodrow) … was the Model Form of Conditions of Contract for Process Plant (Lump Sum) published by the Institution of Chemical Engineers. It is known as the …"Red Book". Certain amendments were made to the Conditions but these did not change the essential form and structure of the Red Book.

2. The Red Book is not a building contract although it has some similarities to a building contract. Certainly, physical construction work is required to be completed by the Contractor within a time frame. However the main object of the contract is the creation of a process plant. … At the core of this case is the Sequencing Batch Reactor. The process system was to deal with sewage through biological action. …

3. The Red Book places emphasis on testing. The Construction Completion Report has to state that the Plant is in such a physical condition that pre-operation take-over tests can be carried out safely. Thereafter Taking Over of the Plant by the Purchaser and final Acceptance by the Purchaser depend on assessment of performance as demonstrated by testing. By reference to the individual clauses, after physical construction works are completed (Clause 33) take over tests and procedures are carried out to gauge whether or not the Contractor is entitled to hand over the Plant; Clause 34 and Schedule 6. After Take-Over, the Purchaser has possession of the Plant and he carries out performance tests to check how well the Plant is operating; Clause 35 and Schedule 7. Following successful Performance Tests an Acceptance Certificate is issued (Clause 37.2).

4. Clause 36 of the Red Book provides for defects in the Plant to be made good either before or after Take-Over. If the defect is made good before Take-Over the remedial work is subject to the take-over tests. If it is made good after Take-Over the Project Manager may require the Contractor to repeat any appropriate take-over tests. (36.3)

5. The Red Book envisages that a Purchaser may wish to take over and accept a Plant even though there has not been 100% compliance with either take-over tests or performance tests: see Clauses 34.9 and 37.9. Furthermore, under Clause 35.10 the Purchaser is obliged to accept the Plant even though there has been a failure (within specified tolerances) to pass performance tests. The Purchaser accepts the Plant together with an entitlement to liquidated damages.

6. Once the Take-Over tests have been passed the Purchaser must take possession of the Plant. It is his obligation to have the performance tests carried out (Clauses 35.3 and 35.4). The Plant is now the Purchaser's to run (Clause 34.2). Take Over having taken place, the Contractor has limited rights of access; for example, to supervise performance tests (Clause 35.3) and then to make adjustments (Clause 35.6); or to make good any Defects notified under Clause 36.1. The fundamental effect of Take-Over is to transfer possession and control of the Plant (and risk in it) from Contractor to Purchaser.

7. Liquidated damages are not leviable under this contract for late completion of the construction phase. However they are leviable for late completion of take-over tests (see Clause 15 and Schedules 5 and 9) - and therefore late completion of construction will indirectly result in liquidated damages. If the Plant does not pass a take-over test, it has to be repeated (34.5). Parts of the Plant can be excluded from Take-Over. Also Take-Over can take place with minor items still to be completed.

8. Once Take-Over has been achieved there is a period for performance testing of 365 days. …

9. Liability for defects is covered by Clause 36. The remedy stipulated in the clause is that the defects are to be put right. The Contractor has to do this (36.1). If the defect is his fault, he has to bear his own cost. If it is not, he gets paid for the work (36.2). If he fails to make the defect good the Purchaser can do so and the Contractor has to reimburse him his reasonable Cost (36.7). Any disputes as to the quantum of the Cost are determined by an Expert, not by the Court (36.7). …

10. If a Plant does not pass the take-over tests or if take-over tests are not carried out, either Take-Over does not take place and liquidated damages under Clause 15 and Schedule 9 are leviable or the procedure under clause 34.9 is followed. This permits Take-Over to take place despite a failure to carry out or pass a take over test so long as the Contractor agrees. In that event the Certificate may either exclude parts of the Plant from taking over or may specify the take over-tests which the plant failed to pass."

16. Mr Elliott also submitted uncontroversially that the Red Book sets out procedures that limit either party's ability to refer matters to the Court and suggested that the Main Contract contains its own system of relief and remedies that the parties have to follow. Mr Elliott submitted that in this context, Clause 44.2 of the General Conditions of Contract is central to the decisions that fall to be considered in this case. So far as material, Clause 44.2 provides as follows:

"Except in the case of termination of the Contract under the terms of Clause 41 (Contractor's Default), the liability of the Contractor to the Purchaser for any breach of contract shall be limited to the expenses, charges, damages and reimbursements expressly provided in the Contract. … "

17. It was Mr Elliott's submission that the effect of Clause 44.2 is that a claim for general damages for breach of contract and/or negligence is not permissible, except in circumstances where the Contract has been determined under Clause 41 in the event of a Contractor's default. Mr Elliott accepted that a defect in the Plant might amount to a "material breach", entitling the Employer to determine the Contract under Clause 41.2 (c), but he emphasised that Yorkshire Water had chosen not to go down that particular route in this case and submitted that Yorkshire Water was, thus, still contractually bound by the restrictions contained in Clause 44.2.
18. It was therefore Mr Elliot's submission that Yorkshire Water had simply overlooked or ignored the effect of Clause 44.2 of the Contract when it originally commenced proceedings in August 2001 and claimed damages for breach of contract and/or negligence (see paragraph 427 below and the terms of the original un-amended Claim Form). He pointed out that it was only once the point was taken in the pleadings served by Taylor Woodrow, Biwater and Elga that Yorkshire Water had amended the Claim Form and its Particulars of Claim to delete the claim for damages for breach of contract and to substitute for it the present contractual claim for "Payment of all sums due under the terms of the Contract… ": see paragraph 450 below, the Amended Claim Form and the terms of paragraph 1 of the Amended Prayer.
19. Mr Elliott suggested that, having realised its mistake, Yorkshire Water was now, in reality, trying to pursue a claim for general damages for breach of contract under the guise of a claim to recover sums alleged to be due under various provisions of the Main Contract: i.e. Sub-Clause 34.8, Sub-Clause 35.10/Schedule 10, Sub-Clause 36.7 and Sub-Clause 36.11. He submitted that this necessary change of tack had resulted in Yorkshire Water having "to shoehorn" a claim for general damages into contractual provisions that were intended to form part of a carefully structured and systematic contractual procedure for specific financial relief and remedies.
20. Mr Elliott submitted that the financial claims now advanced by Yorkshire Water are ones that have inevitably given rise to significant problems of construction of the Contract, because the present claims necessarily involve a consideration of (i) the rights and remedies provided by the contractual terms in question and (ii) the conditions required for those rights and remedies to arise. As Mr Elliott pointed out (correctly, in my view), that is a very different process from having to prove the amount of damages to which Yorkshire Water would be entitled as compensation for such foreseeable damage as has been caused by the various alleged breaches of contract. On behalf of Taylor Woodrow/Biwater, Mr Streatfeild-James QC supported and adopted all the foregoing submissions by Mr Elliott.
21. The Relevant Contractual Provisions of the Main Contract In the paragraphs that follow, I set out the various relevant contractual provisions by reference to some of the various documents that go to make up the Main Contract, namely: (i) The Agreement, (ii) The Special Conditions, (iii) The General Conditions, (iv) The Schedules and (v) The Specification.
22. The Agreement: So far as material, the written Agreement between Yorkshire Water and Taylor Woodrow, provided as follows:-

"THIS AGREEMENT is made …

BETWEEN Taylor Woodrow Construction Northern Ltd …

(hereinafter called "the Contractor") of the one part and Yorkshire Water Services Limited (hereinafter called the "Purchaser") of the other part.

WHEREAS

1. The Purchaser wishes to have process plant to be known as

Knostrop … Improvement Works constructed at Knostrop …, Leeds and to this end wishes the Contractor to complete the design of and to execute and complete the Works (as hereinafter defined).

2. The Contractor is able to complete such design of and to execute and complete the Works for the consideration and upon the terms hereafter appearing.

NOW THIS AGREEMENT provides as follows:

1. The following documents only and their annexes, if any, shall together constitute the Contract between the Purchaser and the Contractor and the term "the Contract" shall in all such documents be construed accordingly.

(a) this Form of Agreement

(b) the Special Conditions

(c) the General Conditions of Contract

(d) the following Schedules

Schedule 1 Description of Works

Schedule 2 Supply by Purchaser

Schedule 3 Documentation for Approval

Schedule 4 Final Documentation and Manuals

Schedule 5 Times and Stages of Completion

Schedule 6 Take Over Procedures

Schedule 7 Performance Tests and Procedures

Schedule 8 Terms of Payment

Schedule 9 Liquidated Damages for Delay

Schedule 10 Liquidated Damages for Failure to Pass Performance Tests

Schedule 11 Rates and Charges

Schedule 12 Training of Purchaser's Staff by Contractor

Schedule 13 Quality Assurance System

Schedule 14 Limitations on Subcontracting

(e) the Specification and Documentation (if any) listed therein or annexed thereto

(f) Form of Tender and Appendix Part 1 and Part 2

(g) Acceptance Letter and Taylor Woodrow's letter dated 18/5/98

(h) Agreement for the Preferred Contractors

For the purpose of identification the said Special Conditions, Schedules and Specification are bound together with this Form of Agreement and have been signed on behalf of the Purchaser and the Contractor.

2. The Contract as hereinbefore defined constitutes a full statement of the contractual rights and liabilities of the Purchaser and the Contractor in relation to the Works and no negotiations between them nor any document agreed or signed by them prior to the date of the Contract in relation to the Works shall hereafter be of any contractual effect.

…

6. In case of conflict between any of the documents accompanying this agreement, the order of precedence shall be as follows:

The Agreement;

The Form of Tender, Acceptance Letter and Taylor Woodrow's letter dated 18/5/98

The Special Conditions;

The Schedules;

The General Conditions of Contract of the Institution of Chemical Engineers for Process Plant Lump Sum Contracts;

The Specification and Documentation (if any) listed therein or annexed thereto.

The Agreement for Preferred Contractors.

7. For the purposes of Sub-Clauses 7.2, 8.3 and 28.2 of the General Conditions the date of the Contractor's tender shall be the 28 August 1997

8. The date for the commencement of the Works as defined in Clause 1 of the General Conditions shall be 8 December 1997 …"

23. The Special Conditions. So far as material the Special Conditions provided as follows:

"The Conditions of Contract shall be Clauses 1 to 46 of the Model Form of Conditions of Contract for Process Plants, suitable for Lump Sum Contracts in the United Kingdom, (1995 Edition) published by the Institution of Chemical Engineers and hereafter referred to as the General Conditions of Contract and are added to and amended by The Special Conditions: -

THE SPECIAL CONDITIONS

The General Conditions of Contract shall be read subject to the following additions and, amendments:-

Definition of Terms Substitute the following definitions:

…

"Contract" means the General Conditions of Contract, the Special Conditions, the Schedules, the Specification and Documentation, The Preferred Contractor Agreement, the form of Tender and Appendices, Acceptance Letter and the Agreement…

Delete the definition of "Cost" and replace with the following: -

"Cost", "costs, losses and claims" and "the amount to be added to or deducted from the Contract Price" or the valuation of new prices within the Conditions of Contract, other than prices that can be valued from the Schedule of Prices or based on the Schedule of Prices shall be valued in accordance with the Schedule of Cost Components included in Schedule 11 of this Contract. The use of such Schedule of Cost Components provides a reasonable profit and no additional profit will be added under the various clauses"

…

Add sub-clause

"36.11 "In addition to his obligations under Sub-clause 36.1, the Contractor shall be responsible for making good or for the cost of making good any defects to which this Sub-clause applies and which may be discovered in any section, part or portion of the Plant during the period of twelve years (or, if less, the design life specified or agreed) after whichever, in relation to that section, part or portion, is the later of the following dates:-

(a) the date of the relevant Taking-over Certificate; and

(b) the date such section, part or portion is brought into use by the Purchaser for its intended purpose.

The defects to which this Sub-clause applies are defects of the kind referred to in Sub-clause 36.1(a):-

(aa) which would not have been discovered upon a reasonable examination by the Project Manager at the date of the relevant Taking-over Certificate; and

(bb) which are notified by the Purchaser to the Contractor as soon as reasonably practicable after their discovery for the purpose of allowing the Contractor to inspect the defects discovered; and

(cc) which do not arise from a failure by the Purchaser to ensure that the Plant or any section, part or portion of the Plant has been operated and maintained in accordance with the Contractor's instructions and good engineering practice at all times." …

24. The General Conditions. So far as material, The General Conditions of Contract provided (inter alia) as follows:-

"1. Definition of Terms

Unless the context otherwise requires, the following expressions shall have the meanings hereby assigned to them. …

"Contract" [See the Special Conditions, quoted above] …

"Cost" [See the Special Conditions, quoted above] …

"Defect" and the "Defects Liability Period" have the meaning assigned to them in Sub-clause 36.1. …

2. Interpretation

…

2.2 The Contract documents shall be construed as mutually explicable. However, in the event of conflict between any of the documents comprising the Contract, the order of precedence shall be as follows: … [see Clause 6 of the Written Agreement, quoted above] …

3. Contractor's responsibilities

3.1 The Contractor shall carry out and complete the Works in accordance with the provisions of the Contract in consideration of the payment by the Purchaser of the Contract Price.

3.2 Subject to the express provisions of the Contract, all work carried out by the Contractor under the Contract shall be executed with sound (design) workmanship and materials, safely and in accordance with good engineering practice and applicable regulations and shall be to the reasonable satisfaction of the Project Manager. …

…

33. Completion of construction

33.1 As soon as the Plant, or any appropriate part thereof, is in the opinion of the Contractor substantially complete and ready for inspection, the Contractor shall so notify the Project Manager by means of a Construction Completion Report. The Contractor shall propose a programme for such inspection (including the carrying - out of any tests required at that time by the Specification) commencing not sooner than seven days nor later than fourteen days after the date of the notice, unless the Project Manager otherwise agrees in writing.

33.2 Upon the satisfactory completion of any such inspection the Contractor and the Project Manager shall sign the Construction Completion Report, with an endorsement stating that the Contractor has demonstrated to the Project Manager that the Plant or part thereof is substantially complete and in a condition such that any procedures needed before the Plant is put into operation may safely be carried out. Such Construction Completion Report may include a note of any minor items requiring completion before the issue of a Take-Over Certificate, as provided in Clause 34 (Taking Over).

33.3 If the Project Manager is not satisfied that the Plant or part thereof referred to in Sub-clause 33.1 is substantially complete, he may endorse the Construction Completion Report form accordingly, stating in what way the Plant or part thereof is not in accordance with the Specification. The Contractor shall then complete the Plant or part thereof as necessary and shall repeat the procedure described in Sub-clause 33.1. …

34. Taking over

34.1 If the Contract provides for the Plant to be taken over by specified sections, then the provisions of this clause shall apply as if the references therein to the Plant were references to any specified section thereof.

34.2 As soon as the construction of any part of the Plant has been demonstrated to be complete in accordance with the provisions of Sub-Clause 33.2 and the Plant is, in the opinion of the Contractor, ready for the conduct of any of the procedures specified in Schedule 6 (Take-Over Procedures), which may include take-over tests, the Contractor shall so notify the Project Manager and shall specify a time not sooner than seven days and not later than fourteen days after the date of the notice when the Contractor intends to begin to conduct such procedures. If the Project Manager requires the Contractor to carry out any tests or take-over procedures which are not included in Schedule 6, such requirements shall be treated as a Variation. …

34.5 If the Plant fails to pass any take-over test, then the Contractor, after making such adjustments as he considers necessary, shall repeat such test in the presence of the Project Manager at a time agreed between the Contractor and the Project Manager, or notified at least fourteen days in advance by the Contractor to the Project Manager. If the Project Manager is of the opinion that any such adjustments made by the Contractor make it desirable to do so, he may require the Contractor to repeat any other of the take-over tests which has already been successfully carried out, and the Contractor shall do so. Save as aforesaid the Contractor shall not be bound to repeat for the purposes of this clause any take-over test that has already been carried out. …

34.7 As soon as any minor items referred to in Sub-clause 33.2 have been completed and all the procedures specified in Schedule 6 have been successfully carried out (subject to the provisions of Sub-clause 34:10) including any which affect the operability or safety of the Plant, the Project Manager shall issue to the Purchaser and to the Contractor copies of a Take-Over Certificate stating that the Contractor has satisfied the requirements of the Specification and Schedule 6, whereupon the Plant, apart from any parts that are excluded from the taking over by the terms of the Certificate, shall be at the risk of the Purchaser who shall take possession thereof. The Purchaser shall thereafter be responsible for the care, safety, operation, servicing and maintenance of the Plant, and if Clause 35 (Performance Tests) applies, shall start up and operate the Plant, and prepare for and carry out the performance tests.

Upon the issue by the Project Manager of a Take-Over Certificate the Purchaser shall forthwith sign such Certificate, and shall deliver it to the Contractor. The Take-Over Certificate may include a list of minor items still to be completed by the Contractor, and a list of take-over procedures omitted by operation of Sub-clause 34.10.

34.8 As soon as shall be reasonably possible after the issue of the Take-Over Certificate the Contractor shall correct, repair, adjust or compete to the satisfaction of the Project Manager all items noted on the Take-Over Certificate as not being complete at the date of the said Certificate.1f the Contractor shall fail so to do the Project Manager may arrange for such work to be carried out by others, the cost thereof to be to the account of the Contractor.

34.9 Notwithstanding the failure of the Plant to pass any take-over test or if certain procedures have not been carried out, the Project Manager may with the Contractor's consent issue a Take-Over Certificate in respect of the Plant. Such a Certificate may either exclude from taking over such parts of the Plant as are specified therein, being parts for which the specified procedures have not been carried out, or may specify the take-over tests which the Plant has failed to pass. In either case the Contractor shall remain liable until the end of the Defects Liability Period, should the Project Manager so require, to undertake any omitted procedures or to repeat any tests which the Plant has not passed and in the former case the Contractor shall become entitled to a separate Take-Over Certificate in respect of the excluded part as soon as the applicable take-over procedures have been completed satisfactorily. …

35. Performance tests

35.1 If under the Contract the Contractor makes specific guarantees in respect of the performance of the Plant, verifiable by performance tests, the provisions of this clause shall apply. Otherwise this clause shall form no part of the Contract.

35.2 The performance tests to be carried out on the Plant shall be those specified in Schedule 7 (Performance Tests and Procedures) or, if not so specified, as may be agreed between the Contractor and the Project Manager before any part of the Plant is taken over. The performance tests shall be carried out as soon as is reasonably practicable after the Plant, or the specified section of the Plant to which such tests relate, has been taken over by the Purchaser.

35.3 Performance tests shall be carried out by suitably trained and experienced employees of the Purchaser under the supervision of the Contractor and in accordance with the manuals provided by the Contractor pursuant to Clause 20 (Documentation) and such other instructions as the Contractor may give in the course of carrying out such tests. If such instructions conflict in any way with the said manuals, the Contractor shall issue them in writing in the form of an amendment to the manuals. The performance tests shall be carried out as far as practicable under the conditions, if any, detailed in the Contract.

35.4 The Purchaser shall give the Contractor at least fourteen days notice of his readiness to carry out the performance tests, including a proposal for the time at which the tests should commence. The Contractor shall acknowledge, at least seven days before the time proposed by the Purchaser, that he will be prepared to supervise the tests.

35.5 Every performance test shall be carried out to completion (as may be specified in Schedule 7) unless either the Purchaser or the Project Manager or the Contractor shall order it to be stopped because its continuance would be unsafe or cause damage to property.

35.6 If the Plant or any specified section thereof fails to pass any performance test or if any performance test is stopped before its completion such test shall, subject to Sub-Clause 35.9, be repeated as soon as practicable thereafter. Meanwhile the Purchaser shall have the right to operate the Plant under the general scrutiny of the Contractor. The Purchaser shall permit the Contractor to make adjustments and modifications to any part of the Plant before the repetition of any performance test and shall, if the Contractor reasonably requires, shut down any part of the Plant for such purpose and restart it after completion of the adjustments and modifications, which shall be made by the Contractor with all reasonable speed. The timing of any such shutdown shall be agreed by the Project Manager taking into account the Purchaser's statutory obligations. The Contractor shall, if so required, by the Project Manager, submit to the Project Manager for his approval details of the adjustments or modifications which he proposes to make. The Contractor shall make such adjustments and modifications which he proposes to make. The Contractor shall make such adjustments and modifications at his own cost, subject to the provisions of Sub-clause 35.12, unless he can show that the need for them was caused by the Purchaser or by any other contractor employed by the Purchaser, in which case the Contractor shall be paid by the Purchaser the Cost of making such adjustments and modifications plus a reasonable profit thereon.

35.7 The results of the performance tests shall be compiled and evaluated jointly by the Purchaser or the Project Manager and by the Contractor. Feed and product rates and the consumption of utilities shall be averaged over the specified period for the relevant test, unless Schedule 7 otherwise provides. Any necessary adjustments to take account of actual operating conditions and measuring tolerances shall be made in accordance with the provisions of the said Schedule. Any effect on the results of such performance tests which can reasonably be shown to be due to the prior use of the Plant by the Purchaser shall be taken into account in assessing such results.

35.8 Subject to the following sub-clauses, the operation of the Plant by the Purchaser shall, as far as the Purchaser's statutory obligations permit, be subordinated to the provisions of this clause until the Plant has passed all the performance tests.

35.9 If the Contactor fails to demonstrate that the Plant has passed all its performance tests within a period stated in the Appendix to the Form of Tender [365 days] from the date of the relevant Take-Over Certificate, providing no unreasonable delay has been caused by the Purchaser in commencing or carrying out his obligations under Sub-clauses 35.2 and 35.3 above, the Purchaser shall thereafter be free to operate the Plant as he sees fit.

35.10 If the Plant does not pass any performance test within a period stated in the Appendix to the Form of Tender [365 days] from the date of the relevant Take-Over Certificate, the Contractor shall pay liquidated damages to the Purchaser in accordance with Schedule 10 (Liquidated Damages for Failure to Pass Performance Tests), provided that the results of the performance tests are within any limits specified in the said Schedule. Upon payment of such sum the Contractor shall become entitled to the issue of an Acceptance Certificate in respect of the Plant stating that the applicable liquidated damages have been paid in respect of the shortfall in performance. If the results of any performance tests are outside any limits specified in Schedule10 then the Purchaser may either:

(a) accept the Plant subject to such reasonable reduction in the Contract Price as may be agreed between the Purchaser and the Contractor or, in default of agreement, as determined by an Expert in accordance with Clause 45 (Reference to an Expert); or

(b) reject the Plant and proceed in accordance with Clause 41 (Contractor's Default).

35.11 If the Plant or any specified section thereof fails to pass any performance test and the Contractor in consequence proposes to make any adjustment or modification thereto, the Project Manager may notify the Contractor that the Purchaser wishes to defer such work until a time convenient to the Purchaser, in which case the period stated in the Appendix to the Form of Tender [365 days] specified in Sub-clause 35.10 shall be extended by the length of time of the deferment. In such event the Contractor shall become entitled to the issue of an Acceptance Certificate as if the Plant or specified section thereof had passed the performance test in question, but the Contractor shall remain liable to carry out the adjustment or modification and to satisfy the performance test within a reasonable time of being notified to do so by the Project Manager. If however the Project Manager fails to give any such notice within three hundred and sixty-five days of the date of the last Take-Over Certificate, the Contractor shall be relieved of any such obligation and the Plant or specified section thereof shall be deemed to have passed such performance test.

35.12 If the Contractor incurs additional cost as a result of any delay by the Purchaser in permitting access to the Plant by the Contractor, either to investigate the causes of failure to pass a performance test, or to carry out any adjustments or modifications, then the Contractor shall be entitled to be paid his reasonable Cost incurred by reason of such delay as an addition to the Contract Price.

36. Liability for Defects

36.1 If at any time before the Plant or any section thereof is taken over pursuant to Clause 34 (Taking Over) or within the defects liability period, which in relation to the Plant or any section thereof is a period of three hundred and sixty five days, after whichever is the later of the following dates:-

(i) the date of the relevant Taking-over Certificate; and

(ii) the date such section is brought into use by the Purchaser for its intended purpose (but not so as to cause such period of three hundred and sixty five days to extend beyond two years after the date of the relevant Taking-over Certificate),

the Project Manager shall:

(a) decide that any work done or Materials supplied by the Contractor or any Subcontractor is or are defective or not in accordance with the Contract (normal wear and tear excepted) or that the Plant or any portion thereof is defective or does not fulfil the requirements of the Contract (any such matter being herein called a "Defect"); and

(b) as soon as reasonably practicable give to the Contractor notice in writing of such decision giving particulars of the alleged Defects; and

(c) so far as may be necessary place the Plant at the Contractor's disposal;

then the Contractor shall as soon as reasonably practicable make good the Defects so specified subject to the Purchaser's affording the Contractor the necessary access and, where appropriate, permitting any defective Materials to be removed. The Contractor shall, if so required by the Project Manager, submit his proposals for making good any Defect to the Project Manager for his approval.

36.2 If any Defect is attributable to any breach of the Contract committed by the Contractor the Contractor shall bear his own Cost of making good the Defect subject to the provisions of Sub-clause 36.10. In the case of any other defect made good by the Contractor the work done by the Contractor shall be treated as if it were a Variation ordered by the Project Manager and shall be valued accordingly.

36.3 If a Defect is made good after the issue of a Take-Over Certificate the Project Manager may require the Contractor to repeat any appropriate take-over test following the making good of any such Defect for the purpose of establishing that the Defect has indeed been made good. …

36.5 If the Purchaser reasonably requires that any Defect notified to the Contractor under Sub-clause 36.1 be made good urgently and the Contractor is unable to comply or refuses to make good any such Defect within a reasonable time, then the Purchaser may, without prejudice to any other remedies or relief available to him under the Contract, proceed to do the work in such a manner as the Project Manager may approve.

36.6 If the Contractor shall neglect or refuse to make good as soon as reasonably practicable any Defect for which he is responsible under Sub-clause 36.2 then the Purchaser may, without prejudice to any other remedies or relief available to him under the Contract, proceed to do the work, provided that the Purchaser gives at least fourteen days notice of his intention in writing.

36.7 If the Purchaser has made good a Defect in pursuance of either Sub-clauses 36.5 or 36.6 then the Contractor shall reimburse the Purchaser his reasonable Cost of so doing. Any dispute as to quantum shall be referred to an Expert for resolution in accordance with Clause 45 (Reference to an Expert). …

36.11 [inserted by Special Conditions: see paragraph 23 above]

37. Acceptance

…

37.6 All uncompleted work and all unsatisfied or uncompleted performance tests shall be conducted as soon as practicable after the issue of an Acceptance Certificate pursuant to Sub-clause 37.3. 1f matters beyond the Contractor's control continue to prevent the carrying out of the performance tests throughout the remainder of the relevant Defects Liability Period, then, unless it is otherwise established that the Plant, or any specified section thereof, was throughout the said Defects Liability Period incapable of passing any or all of such performance tests by reason of some default of the Contractor, such tests shall at the end of the Defects Liability Period be deemed to have been satisfied. …

38 Final Certificate

38.1 Subject to the provisions of Sub-clauses 38.2 and 38.3, as soon as the Defects Liability Period for the Plant has expired or the Contractor has made good all Defects that have within such a period appeared in the Plant, or any specified section thereof, whichever is the later, the Project Manager shall issue both to the Purchaser and to the Contractor a certificate (herein called a 'Final Certificate') stating that the Works or the specified section of the Plant, as the case may be, is finally complete as from that day. …

41. Contractor's default

41.1 If the Contractor goes into liquidation (other than a voluntary liquidation for the purposes of reconstruction or amalgamation) or has an administration order made against him or carries on his business or any part of it under an administrator or receiver or manager for the benefit of his creditors or any of them, then without prejudice to any other rights or remedies which the Purchaser may possess the Purchaser may forthwith determine the employment of the Contractor under the Contract.

41.2 If the Contractor shall be in default in any one or more of the following respects, that is to say:

(a) without reasonable cause wholly suspends or abandons the carrying out of the Works before completion thereof; or

(b) fails to proceed regularly and diligently with the Works; or

(c) commits any other material breach of the Contract;

then, without prejudice to any other rights or remedies which the Purchaser may possess, the Project Manager may give the Contractor written notice specifying such default and if the Contractor shall fail to commence and diligently pursue the rectification of such default within a period of fourteen days after receipt of such notice, or shall at any time thereafter repeat such default, then the Purchaser may forthwith determine the employment of the Contractor under the Contract.

41.3 In the event of the employment of the Contractor under the Contract being determined under Sub-clauses 41.1 or 41.2 above, the following shall be included within the respective rights and duties of the parties:

(a) the Purchaser may employ and pay other persons to carry on and complete the Works and he and they may enter upon the Plant and use the Contractor's Equipment, Materials and any other things whatsoever brought to the Site by the Contractor or which have become the property of the Purchaser pursuant to Sub-clause 25.1, and the Purchaser shall not be liable to the Contractor for any fair wear or tear or accidental damage that may occur to such Contractor's Equipment, Materials or other things;

(b) the Contractor shall, when so required by the Purchaser, forthwith deliver to the Purchaser all Restricted Information together with all Documentation and technical information prepared by the Contractor as referred to in Clause 20 (Documentation);

(c) the Contractor shall, when so required by the Purchaser, assign to the Purchaser all its rights under any subcontracts.

41.4 All reasonable Cost incurred by the Purchaser in the circumstances described in Sub-clause 41.3 (a) shall be recoverable from the Contractor as damages. …

44. Limitation of Contractor's liability

44.1 The Contractor shall not be liable to the Purchaser by way of indemnity or by reason of any breach of the Contract for:

(a) any loss of production or of any contract that may be suffered by the Purchaser; or

(b) any wastage, loss or contamination during its use in the Plant of any process consumable which shall be deemed to include feedstocks, chemicals, biochemicals, catalysts and utilities; or

(c) any loss or damage arising from any design or information which the Purchaser has specifically instructed the Contractor to use;

except to the extent that recoveries in respect thereof are obtained under insurance effected pursuant to Clause 32 (Insurance).

44.2 Except in the case of termination of the Contract under the terms of Clause 41 (Contractor's Default), the liability of the Contractor to the Purchaser for any breach of contract shall be limited to the expenses, charges, damages and reimbursements expressly provided in the Contract. Nothing in the Contract shall in any way be interpreted as affecting or limiting any liability which the Contractor, may have under the Consumer Protection Act 1987 or in respect of personal injury or death caused by the negligence of the Contractor (as defined in Section 1 of the Unfair Contract Terms Act 1977). …"

25. The Schedules. So far as material, the Schedules provided as follows:-

"SCHEDULE 6

TAKE-OVER PROCEDURES

6.1 GENERAL

6.1.1 This Schedule describes the Take Over Tests to be carried out under the Contract. The tests (summarised in Table 1) include general tests detailed in paragraphs 6.1.1 to 6.1.9.2 inclusive and particular tests in paragraphs 6.2 to 6.6 inclusive.

6.1.2 The purpose of these tests is to demonstrate as far as practicable that plant and equipment are reliable in their operation and that their output and performance meets the stipulated criteria.

6.1.3 The Take-Over Procedures are written on the basis that all main items of Plant, as specified in Schedule 1, Section 1.3.2 'a' -'g' under the Contract, are tested independently of one another.

6.1.4 The Take Over Test for any Section of the Plant shall not commence until:

i) A Certificate of Construction Completion for the Section has been issued.

(ii) Three draft copies of the operating and maintenance manuals and 'As-Built' Drawings and draft information for the production of the Safety file for the Section have been forwarded to the Project Manager.

(iii) Full and detailed method statements and programme for the Section Take-over Procedures have been approved by the Project Manager. …

6.1.6 Reliability Test

A 14 continuous day reliability test will be carried out on the Sections of Plant to demonstrate reliable operational performance under automatic control. …

The Section will be deemed to have passed its reliability test if the plant and equipment operates without failure/breakdown during the 14 day test period. …

SCHEDULE 7

PERFORMANCE TESTS AND PROCEDURES

7.1 General

7.1 This schedule describes the Performance Tests as defined in Clause 35 of the Conditions of Contract to be carried out on the following Plant supplied and installed under this Contract. Refer to Table 7.1.

7.1.2 The Performance Test period shall commence from the date of the issue of the Take-Over Certificate for each Section of the Plant.

7.1.3 At the time of Testing, the flows and loads arriving at the sewage treatment works may be below the design figures. In order to demonstrate that the sidestream SBR Plant will achieve the specified effluent quality whilst treating design flows and loads it may be necessary to simulate the design flows and loads. The Contractor is required to submit for approval by the Project Manager his own proposals for such simulated tests.

7 .1.4 Each Section of the Plant will be tested over the range of flow and load variations that normally occur at the existing works.

7.1.5 Each Section of the Plant will be tested over a series of four 28 day test periods, to reflect the seasons, over a maximum period of 52 weeks as directed by the Project Manager.

7.1.6 The Performance Tests shall demonstrate compliance with the following requirements.

The Contractor shall achieve the guaranteed levels of operational performance provided at the time of tender, as indicated in the Schedule 7 Forms (Performance requirements) as detailed:

Guarantee TPI – Effluent Quality

Guarantee TP2 – Flow Capacity …

Guarantee TP9 – Reliability …

Guarantee TP11 – Sludge Production and Sludge quality…

7.1.7 If the Performance Tests demonstrate that all the determinands specified in the guarantees have achieved the specified compliance then the plant will be deemed to have passed the Performance Tests and an Acceptance Certificate will be issued.

7.2 Effluent Quality (Guarantee Form TPI refers)

7.2.1 The purpose of these tests is to demonstrate that the SBR Plant is capable of producing effluent of the specified quality over the full range of flows and loads and that the whole plant will operate reliably.

During these tests the Contractor shall demonstrate that the SBR Plant is capable of meeting all the criteria specified.

Flows through the SBR Plant during the test are to be set by the Project Manager in co-operation with the Purchaser's staff. Operation and effluent quality, during the flow tests, shall be monitored by both parties. The Contractor shall ensure that the quality of the effluent produced during the tests meets specified requirements and does not cause failure of the existing consent of the existing works.

The Contractor shall be responsible for all costs and expenses arising from failure to meet effluent standards during these tests. …

7.2.4 Sampling and Testing

The SBR Plant shall be tested over four 28 day test periods which will take place as set by the Project Manager during the 52 week period following the issue of the Take-Over certificate. The plant will be tested against its ability to meet the specified effluent quality at all times and all rates up to the maximum design values given in the specification.

During each test period daily 24 hour time based composite samples will be taken from the designated sampling points to check that the plant meets the UWWTD requirements for composite sampling. Also during each period, spot (grab) samples shall be taken once a day on a random basis from the agreed sampling points to check that the plant meets the NEP consent standards. …

7.3 Flow Capacity (Guarantee TP2 refers)

Tests will be conducted to confirm that each item of plant is hydraulically capable of passing the guaranteed flow capacity. Over pumping may be necessary to build up the required flow. …

7.10 Reliability (Guarantee TP9 refers)

7.10.1 The reliability of each item of plant installed under the Contract shall be demonstrated during the performance tests. Compliance with the requirement for reliability shall be on the basis of 100% plant availability over all the respective 28 day test periods. Where standby plant is available, failure of duty plant and automatic change of status of standby plant to duty plant and use of such plant to cover the failure will not constitute duty plant unavailability. However, a repeat failure of any plant will be deemed a failure. In the event of downtime the testing period will be extended. …

SCHEDULE 7 FORMS - PERFORMANCE REQUIREMENTS

The Contractor shall be required to achieve the following levels of operational performance to demonstrate that the plant fulfils the requirements of Contract.

The Contractor shall guarantee the following figures for performance of the plant when treating flows and loads not exceeding those given in specification.

Guarantee TPI -Effluent Quality

a) …

b) Sidestream SBR Plant

i As determined by composite samples of sidestream plant effluent assessed using the Environment Agency look-up table.



ii As determined by spot samples of Sidestream plant effluent assessed using the Environment Agency look-up table.



Guarantee TP2 – Flow Capacity

a) Low Level Works

i) Fine Screens

Flow to Full Treatment 238380m3/d

ii) 2nd Lift Pumps

Flow per pump 950 1/s

iii) …

b) High Level Works

i) Fine Screens

Flow to Full Treatment 230720m3/d

ii) Sidestream SBR Plant

Max Flow from High Level Primary Tanks 76138 m3/d

Max Flow from ASP Final Tanks 59805 m3/d

Total Max Flow to Sidestream Plant 135942 m3/d

Plus Return Liquors 1340 m3/d

…

SCHEDULE 10

LIQUIDATED DAMAGES FOR FAILURE TO PASS PERFORMANCE TESTS.

10.1 Guarantee TP1 -Effluent Quality

Liquidated damages do not apply to Guarantee TP1 -Effluent Quality, but if the Contractor is unable to meet the requirements of Schedule 7, Performance Tests, for guarantee TP1 for the Sidestream SBR plant then the Contractor shall pay to the Purchaser the actual costs incurred by the Purchaser not only for his own staff, experts, consultants, contractors and all necessary equipment and materials to achieve the standards of the Performance Tests, but also for the additional cost of running the plant and loss of use.

10.2 Guarantee TP2 -Flow Capacity

Liquidated damages do not apply to Guarantee TP2 -Flow Capacity, but if the Contractor is unable to meet the requirements of Schedule 7, Performance Tests, for guarantee TP2 then the Contractor shall pay to the Purchaser the actual costs incurred by the Purchaser not only for his own staff, experts, consultants, contractors and all necessary equipment and materials to achieve the standards of the Performance Test, but also for the additional cost of running the plant and loss of use. …

10.9 Guarantee TP9 – Reliability

Liquidated damages do not apply to Guarantee TP9 – Reliability but if the Contractor is unable to meet the requirements of Schedule 7, Performance Tests, for TP9 for plant reliability then the Contractor shall pay to the Purchaser the actual costs incurred by the Purchaser not only for his own staff, experts, consultants, contractors and all necessary equipment and materials to achieve the standards of the Performance Test, but also for the additional cost of running the plant and loss of use. …

10.11 Guarantee TP11 – Sludge Production and Sludge Quality

(a) Sludge Production

If the sludge produced by the … SBR … during the performance test detailed in Schedule 7 exceeds the sludge production figure guaranteed by the Contractor, the Contractor shall pay liquidated damages to the Purchaser. These liquidated damages shall be calculated as follows:

Liquidated damages = (D - (1.05 x P)) x unit cost x 365 x 6.71

Where D = average measured daily sludge production and P = guaranteed daily sludge production. The unit cost will be that for an alternative disposal route that would be paid by Yorkshire Water at the completion of the Performance Test. 6.71 is the discount factor.

(b) Sludge Quality

Liquidated damages do not apply to Guarantee TP11-Sludge Quality, but if the Contractor is unable to meet the requirements of Schedule 7, Performance Tests for the sludge solids content figure guaranteed by the Contractor, then the Contractor shall pay to the Purchaser the actual costs incurred by the Purchaser not only for his own staff, experts, consultants, contractors and all necessary equipment and materials to achieve the standards of the Performance Test but also for the additional cost of running the plant and loss of use. …

10.15 Limit of Liquidated Damages for Failure of Performance Tests

The Contractor shall pay liquidated damages for failure of Performance Tests to the Purchaser up to a limit of 10% of the Contract Price, with respect to guarantees TP3, TP5, TP8, and TPl1 (a), TP12 (a) only. Guarantees TP1, TP2, TP9, TPIO, TPl1 (b), TP12 (b), E1, and E2 are absolute guarantees which must be achieved and are not subject to liquidated damages."

26. The Specification: So far as material, the Specification provided as follows:-

"PARTICULAR SPECIFICATION

0.1 GENERAL

…

0.1.13 GENERAL REQUIREMENTS

The major element of the Project is the upgrading of the existing treatment works and provision of additional treatment capacity to ensure discharge quality of the final effluent meets the consent criteria set by the Urban Waste Water Treatment Directive and the National Environment Programme as given in Clause 0.2.3

…

0.2 PROCESS REQUIREMENTS AND INFORMATION

0.2.1 WORKS FLOWS

	Low Level	High Level	Units	Units	Units
Units	Current	2005 Future	Current	2005 Future	2005 Future
Population	219,529	217,331	324,913	321,055	hd
DWF	102000	124333	97000	101967	m3/d
FFT	211253	238384	205000	230715	m3/d
Formula A	428000	456000	560000	563970	m3/d

0.2.2 WORKING LOADING

a) Current Situation

The works average crude and settled loads for 1994 were as follows:

	Low level	High Level	High Level	High Level	High Level	High Level
High Level	BOD	COD	AMM	BOD	COD	AMM
Crude (tonnes/day)	20.78	57.24	2	26.93	82.30	2.49
Settled (tonnes/day)	18.74	48.72	2	12.57	32.68	2.53

b) Future Situation

For design purposes the following settled loads (tonnes/day) to the works are anticipated.

	Low level	High Level	High Level	High Level	High Level	High Level
High Level	BOD	COD	AMM	BOD	COD	AMM
Peak	24.25	66.34	3.36	17.55	46.68	3.34
Ave	20.45	56.04	3.26	16.45	43.52	3.20

The peak figures relate to 80%ile loads

0.2.3 PERFORMANCE CRITERIA

Standards to be Achieved

The standard required to meet UWWTD and NEP Regulations are as follows: -

1. Analytical Standards

NEP

Performance against these standards is measured by spot sampling using the EA look-up table.

	BOD	NH3	SS
(mg/l)	30	8	50
Upper Tier (mg/l)	64	30

UWWTD

Performance against these standards is measured by using composite samples using the EA look-up table.

	BOD	COD
(mg/l) or (% removal)	25(70)	125(75)
Upper Tier (mg/l)	50	250
Upper Tier for UWWTD only apply if % removal not achieved	Upper Tier for UWWTD only apply if % removal not achieved	Upper Tier for UWWTD only apply if % removal not achieved

…

0.2.6 DESIGN REQUIREMENTS

1. …

2. The Contractors designs shall cater for all likely variations in flow and loads to the works.

3. The plant proposed by the Contractor shall be based on modern but well proven technology entirely suitable for the purpose and the prevailing conditions. The plant shall be fully automated requiring minimal operator intervention and economic to operate. The layout selected shall facilitate safe and convenient access for operation and maintenance requirements. …

0.2.6.3 SIDESTREAM SBR PLANT

The Sidestream plant shall comprise a 6 basin SBR Plant and shall treat 49% of the flow and load from the Low Level ASP and 33% of the High Level primary tank effluent flow and load as specified in 0.2.10. The plant shall be capable of treating the design flow and load to the effluent standard specified in 0.2.3 with one basin out of service.

A new flow splitter chamber shall be constructed downstream of the Low Level ASP final settling tanks. At this chamber 51% of the final tank effluent shall be split off to be blended with the proportion of untreated primary tank effluent, and 49% shall be passed to the new SBR pumping station.

A new flow splitter chamber shall be constructed at the High Level works where 33% of the High Level primary tank effluent shall be split off and passed to The new SBR Pumping Station. The remainder of the flow shall pass to the existing High Level filters. The facility shall also be available to pass a fixed flow to the SBR pumping station upto the maximum design flow of 76 tcmd, irrespective of the flow being passed to the High Level filters.

The new SBR pumping station shall comprise 3 No variable speed submersible pumps operating on a 2 duty/1 standby basis which shall discharge to the SBR inlet channel.

The SBR plant shall provide a minimum of 28 hours retention period at DWF. The SBR will be the US filter Jet Tech design complete with Jet Aerators and manifolds, circulating pumps, blowers decanting units and sludge wasting piping, pumps and valves.

The effluent from the SBR plant shall connect into the outfall pipework downstream of the existing humus tanks where it will blend with the humus tank effluent. The facility shall be available to recirculate the SBR effluent by gravity onto The High Level filter beds over a range of flows from zero to the maximum design flow from the High Level works (ie 76 tcmd). An automatic recirculation system shall be provided so that at times of Low flow, a minimum hydraulic loading rate of 0.75m3/m3/d to the filters is achieved subject to a minimum total flow from the ASP plant and High Level primary tanks of 102.4 TCMD.

0.2.6.4 SIDESTREAM SBR PLANT SLUDGE TREATMENT

Any surplus sludge produced by the SBR shall be thickened to between 3% and 9% DS with an average of 6% DS and pumped to the existing sludge storage tanks numbers 5 and 6, with four actuated valves for automatic sludge tank pocket selection.

The sludge shall be thickened in 2 No belt thickeners operating on a duty/standby basis. The belt thickeners shall be sufficiently sized to provide a maximum operating period of 16 hours per day

Polyelectrolyte solution shall be dosed into the sludge flow to condition the sludge prior to thickening. The polyelectrolyte shall be dosed via 2 No dosing pumps operating on a duty/standby, basis. The make-up system shall use the Big-Bag system.

All sludge liquors from the thickening process shall be discharged to the inlet of the SBR plant. The plant shall be capable of treating the additional flow and load from the sludge liquors.

0.2.7 …

0.2.10 DESIGN PARAMETERS

The Contractor shall ensure that all design parameters specified below are not exceeded under normal operating conditions

Temperature range 7(C to 17(C

i) Existing ASP (Low Level)

Max Flow :122 tcmd

DWF :63 tcmd

Peak BOD Load (settled) :12.42 tonnes/d

Ave BOD Load (settled) :10.47 tonnes/d

Peak AMM Load :1.72 tonnes/d

Ave AMM Load :1.67 tonnes/d

RAS rate :65 tcmd

Design MLSS :3000mg/l

Max 3200mg/l:

ii) Existing Low Level Filters & Humus Tanks

51% from Low Level ASP

Max Flow :62.2 tcmd

DWF :32.5 tcmd

49% from Low Level primary tanks

Max Flow :116.3 tcmd

DWF :60.7 tcmd

Total Flow & Load

Max Flow :178.6 tcmd

DWF :93 tcmd

Peak BOD Load (settled) :13.7 tonnes/d

Ave BOD Load (settled) :11.5 tonnes/d

Peak AMM Load :2.34 tonnes/d

Ave AMM Load :2.29 tonnes/d

iii) Existing High Level Filters & Humus Tanks

Max Flow :154.6 tcmd

DWF :68.3 tcmd

Peak BOD Load (settled) :11.76 tonnes/d

Ave BOD Load (settled) :11.02 tonnes/d

Peak AMM Load :2.24 tonnes/d

AV AMM Load :2.16 tonnes/d

iv) Sidestream SBR Plant (Excluding return Liquors)

Max Flow :136 tcmd

DWF :65 tcmd

Peak BOD Load (settled) :7.58 tonnes/d

Ave BOD Load (settled) :6.89 tonnes/d

Peak AMM Load :1.77 tonnes/d

AV AMM Load :1.73 tonnes/d …"

27. The Alleged Breaches of Contract. Yorkshire Water alleges various breaches of contract against Taylor Woodrow, as pleaded in paragraphs 76 to 81 of the Particulars of Claim. However, it is only necessary to quote paragraphs 76, 78 and 79, as follows:-

"The Breaches of Contract

76. In breach of the terms of the Contract, the Plant as designed, constructed and installed by TW (Taylor Woodrow) does not comply with the performance requirements stated in the tender documents and/or the Schedules to the Special Conditions and/or the Specification. In particular:

76.1 The Plant has not been designed for average flow and/or is deficient in volumetric treatment capacity;

76.2 The Plant as designed and constructed is incapable of treating the maximum design flow(FFT) of 136 tcmd to the required standards; and/or

76.2A The Plant is incapable of treating 60 tcmd from the Low Level ASP; i.e. 49% of the maximum flow; and/or

76.3 The Plant does not have sufficient capacity to treat the design loads and flows to the required standards; and/or

76.4 The Plant is incapable of operating to the required standards with one tank out of service."

…

78. Further or alternatively, and in breach of the terms of the Contract, Taylor Woodrow failed to carry out all its work with sound design, workmanship and materials, safely and in accordance with good engineering practice to the reasonable satisfaction of the Project Manager in that:

78.1 The Plant as designed, constructed and installed by Taylor Woodrow does not comply with the performance requirements stated in the tender documents and/or the Schedules to the Special Conditions and/or the Specification. Yorkshire Water repeats paragraph 76 above.

78.2 The Plant is not suitable for the purposes for which it was designed and/or constructed and/or installed. …

78.3 The Plant does not have sufficient selector capacity (which results in poor sludge settling characteristics).

78.4 There is a lack of capacity in the sludge surplussing system and sludge thickening plant.

79. Further or alternatively, and in breach of the terms of the Contract, Taylor Woodrow failed to propose or provide plant which was based on modern but well proven technology which would be entirely suitable for purpose and prevailing conditions and/or to provide an SBR which would be to a US Filter Jet Tech design complete with jet aerators and manifolds, circulating pumps and valves (paragraph 0.2.6.3 of the Specification) in that:

79.1 Contrary to standard US Filter Jet Tech design, there are no influent distribution manifolds nor is there any provision of selector/contact zones within each tank; and/or

79.2 The inlet has been installed so that it is positioned at the same end of the tank as the outlet, thereby increasing the risk of solids in the effluent during a fill decant; and/or

79.3 The valves on the recirculation pipework leak, allowing liquor migration through adjoining pipework between pairs of tanks, resulting in erratic changes in MLSS levels.

79.4 The sludge collection manifold which was installed at Knostrop in place of an IDSC was much shorter and had fewer apertures. As a result, hydraulic forces cause sludge to enter the removal pipework more rapidly than the sludge can flow into the vicinity of the outlet, a void forms in the sludge layer and is immediately filled with dilute sludge from the region directly above the outlet. Eventually, a volume (sic) in the shape of an inverted cone is formed which makes a direct connection (known as a "rat hole") into the region of dilute sludge. As a result, the concentrated sludge layer then remains intact and only dilute sludge is removed (a process known as "rat-holing").

79.5 There have been persistent problems with the blowers, namely excessive consumption and/or leakage of oil, tripping out under load and excessive wear to filters. A number of attempts were made between about mid 2001 and April 2003 to resolve the problems by the supplier (a firm then known as Hick Hargreaves and since taken over by Edwards BOC). It is, as yet, uncertain whether the most recent repairs have been successful.

79.6 In the course of recent repairs carried out to one of the recirculation pumps, it was discovered that the intake pipework to the recirculation pump was filled with a back thixotropic type gel."

28. However, it is very important to note that, in paragraphs 82 and 83 of the Particulars of Claim, it is made clear (as pointed out by Mr Elliott: see above) that Yorkshire Water's entire claim for appropriate financial relief arising out of the alleged breaches of contract is founded solely upon its entitlement to financial relief and/or remedies under certain terms of the Main Contract and not upon a claim for general damages for breach of contract, as follows:

"82. By virtue of Clause 44.2 of the General Conditions of Contract the liability of TW to YWS (Yorkshire Water) for any of the aforesaid breaches of contract is limited to the expenses, charges, damages and reimbursements expressly provided by the Contract.

83. Accordingly and as appears further below, YWS brings this claim for the necessary remedial works (both already carried out and to be carried out) under Clause 34.8 and/or Clause 36.7 and/or Clause 36.11 and/or Clause 35.10 and/or Schedule 10. In particular, YWS claims to be entitled to recover the cost of the necessary remedial works on the grounds that:

…

83.2 Clause 36.7 provides that TW is liable to pay YWS the reasonable Cost of making good the Defect notified by the Project Manager under clause 36.1 namely

83.2.1 the Plant not achieving the final effluent quality as required by the contract; and/or

83.2.2 the Plant having inadequate capacity for the treatment of design f1ows and loads: and/or

83.2.3 the Plant not having sufficient selector capability: and/or

83.2.4 the Plant not being able to operate without excessive recourse to fill/decant; and/or

83.2.5 the Plant not achieving nitrification to the required standard; and/or

83.3 Clause 36.11 allows YWS to claim the Cost of making good any defects which:

(a) would not have been discovered by the Projects Manager at the date of the relevant Take-Over Certificate; and

(b) were notified by YWS to TW as soon as reasonably practicable after their discovery; and

(c) do not arise from a failure by YWS to ensure that the Plant or any section, part or portion of it has been operated in accordance with the Contractor's instructions and good engineering practice at all times.

83.4 Schedule 10, whether as a standard alone provision or in conjunction of Clause 35.10, which provides that TW shall pay YWS the actual costs incurred by YWS not only for its own staff, experts, consultants, contractors and all necessary equipment and materials to achieve the standards of the performance test, but also for the additional costs of running the Plant and loss of use."

29. In addition to its claim for financial relief under paragraph 83 of the Particulars of Claim, Yorkshire Water claims £237,587.50 liquidated damages for the delayed completion of the take-over tests (paragraph 90A of the Particulars of Claim) and in subparagraph (ii) of the prayer Yorkshire Water seeks a declaration that (A1/53): "(a) there was a Defect in the Plant; and (b) Yorkshire Water was and/or is entitled to make good the Defect; and (c) in relation to each of those items of work which Yorkshire Water proposes to carry out in order to make good the Defect, Taylor Woodrow is liable to pay the reasonable cost thereof once each such item of work has been completed." So far as concerns its claim for financial relief under paragraph 83, Yorkshire Water claims a total of £12,010,640 for the actual and prospective cost of the necessary remedial works and related/ancillary costs: see paragraphs 89 to 90 of the Particulars of Claim and Yorkshire Water's Revised Quantum Schedule. In paragraphs 90A to 95 of the Particulars of Claim, Yorkshire Water's total financial claim is pleaded as follows:

"The Claim

90A. As set out in paragraph 53A above, YWS is entitled and claims to recover liquidated damages in respect of the delayed completion of the Take-Over tests on the SBR Plant at a rate of £15,478 per week of delay in the total sum of £237,587.30

…

92. Further … YWS is entitled and claims to recover the costs of each and/or all the aforesaid remedial works pursuant to Clause 36.7 on the grounds that they represent the reasonable Cost of making good the Defect, namely:

92.1 the Plant not achieving the final effluent quality as required by the Contract; and/or

92.2 the Plant having inadequate capacity for the treatment of design flows and loads; and/or

92.3 the Plant not having sufficient selector capability; and/or

92.4 the Plant not being able to operate without excessive recourse to fill/decant; and/or

92.5 the Plant not achieving nitrification to the required standard.

93. Further or alternatively, if (contrary to YWS' primary case) it is a pre-condition to any recovery under Clause 36.7 that YWS should have actually carried out the works and actually incurred the cost of doing so, YWS claims to be entitled to:

93.1 the cost already incurred in making good the Defect (i.e. including the Phase 1 works and any works carried out by the date of trial); and

93.2 a declaration that:

(a) there was a Defect in the Plant, namely;

(1) the Plant not achieving the final effluent quality as required by the Contract; and/or

(2) the Plant having inadequate capacity for the treatment of design flows and loads; and/or

(3) the Plant not having sufficient selector capability; and/or

(4) the Plant not being able to operate without excessive recourse to fill/ decant; and/or

(5) the Plant not achieving nitrification to the required standard; and

(b) YWS was and/or is entitled to make good the Defect; and

(c) In relation to each of those items of work which YWS proposes to carry out in order to make good the Defect, TW is liable to pay the reasonable Cost thereof once each such item of work has been completed.

94. Further or alternatively, YWS is entitled and claims to recover the costs of each and/or all the aforesaid remedial works pursuant to 36.11 on the grounds that those costs represent the cost of making good defects which would not have been discovered upon a reasonable examination by the Project Manager at the date of the Take-Over Certificate.

95. Further or alternatively, YWS is entitled and claims to recover the cost of each and/or all the aforesaid remedial works pursuant to 35.10 and/or Schedule 10 of the Contract on the grounds that those costs represent the actual costs incurred by YWS not only for its own staff, experts, consultants, contractors and all necessary equipment and materials to achieve the standards of the performance tests, but also for the additional costs of running the Plant and loss of use."

30. The Main Contract Preliminary Issues. At this stage and because of their continued relevance to the issues that I have to decide in these proceedings, for convenience I propose to set out the various answers that I gave in the Main Contract Preliminary Issues. The detailed reasons for those answers are to be found in the judgment that I handed down on 2^nd May 2003.
31. The First Main Contract Preliminary Issue. Can Yorkshire Water have an entitlement to any recovery under either Clause 35.10 or Schedule 10 of the Contract if either no Performance Tests were carried out under Clause 35 of the Contract or no results were obtained from such tests?
32. Insofar as Sub-Clause 35.10 makes provision for the Purchaser's entitlement to Liquidated Damages under Schedule 10 ("the liquidated damages regime"), it does so on the basis of the taking and evaluation of the relevant performance tests (i.e. those tests that come within the "liquidated damages regime": see below). In my view, the expression "…does not pass…" must be construed accordingly. Furthermore, it is important to bear in mind that the contractual obligation to carry out the performance tests rests firmly on the Purchaser. If the words "does not pass" were to be given the wider meaning for which Yorkshire Water contended, this could produce the bizarre result of a Purchaser being able to claim liquidated damages from a Contractor, in circumstances where the fact that the Plant had not been tested during the relevant period was due entirely to the Purchaser's own breach of the obligation to carry out the necessary performance tests.
33. Accordingly, my answer to the first part of this question is that Yorkshire Water does not have an entitlement to recover liquidated damages under sub-clause 35.10 if no performance tests were carried out under Clause 35. It is common ground that, at the date these proceedings were amended and reissued (i.e. 18^th October 2002), no performance tests had been carried out (see below). It therefore follows that Yorkshire Water does not have any relevant right of recovery under sub-clause 35.10.
34. Paragraph 6 of the written Agreement (see above) expressly provides that Schedule 10 is to rank in precedence over the General Conditions (including Clause 35) and the express words of Schedule 10 make it clear that liquidated damages do not apply to the absolute guarantees, including Guarantees TP1, TP2, TP9 and TP11(b).
35. It is important to note that no amendment was made to the terms of sub-clause 35.10, which is expressly limited to making provision for an entitlement to "liquidated damages": (see the terms of Clause 35.10, quoted above). In my view, the express wording of sub-clause 35.10 makes it clear that it applies only to those performance tests that come within the liquidated damages regime, i.e. it applies only to those performance tests in respect of which there is provision for "liquidated damages" under Schedule 10 if the performance test in question is taken and failed (i.e. not passed). The express words of Schedule 10 also make it clear that the liquidated damages regime does not apply to the absolute guarantees (including TP1, TP2, TP9 and TP11(b)), which therefore do not come within the provisions of sub-clause 35.10. However, if I am wrong about that and sub-clause 35.10 does apply in the circumstances of this case, I am satisfied that the outcome of these proceedings remains wholly unaffected, for reasons that will become apparent later in this judgment.
36. Although Schedule 10 is headed "Liquidated Damages for failure to Pass Performance Tests", it is clear that the Schedule also makes express provision for certain specified forms of financial relief and/or remedies in respect of the "absolute" guarantees that have been expressly excluded from the liquidated damages regime (i.e. Guarantees TP1, TP2, TP9, TP10, TP11 (b), TP12 (b), E1 and E2: see paragraph 10.15).
37. In my view, Yorkshire Water's Schedule 10 entitlement to recover "the actual costs incurred" etc. in respect of TP1, TP2, TP9 and TP11(b) will arise if "the Contractor is unable to meet the requirements of Schedule 7, Performance Tests …" within the relevant timescale. It is not necessary for the performance tests to have actually been carried out and failed in order for a Purchaser to be able to prove the inability of a Contractor to meet the necessary requirements of the Schedule 7 performance tests in respect of any of the absolute guarantees within the relevant timescale.
38. Accordingly, for the foregoing reasons, my answer to the second part of the first question is that Yorkshire Water can have an entitlement to recovery under Schedule 10 in respect of the absolute guarantees, including TP1, TP2, TP3 and TP11(b), even though no performance tests were carried out under Clause 35 within the relevant time scale. Of course, in order to establish that entitlement, it will be necessary for Yorkshire Water to prove, by some means other than by showing that the Plant failed the relevant performance tests, that Taylor Woodrow was "unable to meet the requirements of Schedule 7" in respect of the relevant absolute guarantees within the relevant time scale. In the event, this proved to be the central issue in these proceedings.
39. The Second Main Contract Preliminary Issue Can Yorkshire Water have any entitlement under Clause 35.10 and/or Schedule 10 to recover costs as pleaded in paragraph 95 of the Yorkshire Water Claim if the said costs have not been incurred? I did not give an answer to this particular question because I came the conclusion that it was more satisfactory to defer my answer until the matter could be fully considered at trial. Accordingly, I will turn to consider this question in the next main part of this judgment (see paragraphs 54-67 below).
40. The Third Main Contract Preliminary Issue Can Yorkshire Water have an entitlement under Clause 35.10 and paragraphs 10.1 or 10.2 or 10.9 or 10.11(b) or 10.12 of Schedule 10 to recover costs as pleaded in paragraph 95 of the Yorkshire Water claim if Yorkshire Water do not achieve the standards of the Performance Tests? Although I was inclined to give a qualified "Yes", because I accepted Yorkshire Water's submission that the expression "to achieve" should be construed as meaning "in order to achieve", I came to the conclusion that the better course was to defer my final decision until I had heard the evidence. Accordingly, as with the second question, I will turn to consider this question in the next main part of this judgment (see paragraphs 68-77 below).
41. The Fourth Main Contract Preliminary Issue Can Yorkshire Water have an entitlement to be reimbursed the costs of making good a Defect under Clause 36.7 as claimed in paragraph 92 or 93.1 of the Yorkshire Water Claim if the pleaded necessary remedial works have not been carried out? It is clear that the Purchaser's contractual entitlement under sub-clause 36.7 is one of reimbursement by the Contractor of the reasonable cost of making good a relevant defect. The words "has made good" are conclusive. In my view, the words "… if the Purchaser has made good a Defect … the Contractor shall reimburse the Purchaser …" (my emphasis) make it clear that the obligation of the Contractor, to reimburse the Purchaser's reasonable cost of making good the defect under this particular sub-clause, is conditional upon the Purchaser having made good the defect in question. It is only once a defect has been made good that the obligation to reimburse the reasonable cost arises. On that basis, my answer is "No."
42. The Fifth Main Contract Preliminary Issue Can Yorkshire Water have an entitlement to be reimbursed the costs of making good a Defect under Clause 36.7 as claimed in paragraphs 92 or 93.1 of the Yorkshire Water Claim if the pleaded necessary remedial works have not been approved in advance by the Project Manager? The wording of sub-clause 36.5 does not compel the conclusion that the Project Manager's approval must be obtained in advance of carrying out urgently required remedial works. In my view, the words "the Purchaser may … proceed to do the work in such a manner as the Project Manager may approve" have a meaning sufficiently wide to encompass such an approval, whether given in advance, during or after completion of the works in question. Given that the sub-clause is dealing with urgently required works, that breadth of meaning makes perfectly good sense. For those reasons, my answer to this preliminary issue is therefore "Yes".
43. The Sixth Main Contract Preliminary Issue Can Yorkshire Water have an entitlement to be reimbursed the costs of making good a Defect under Clause 36.7 as claimed in paragraph 92 or 93.1 of the Yorkshire Water Claim if the Defect has not been made good? My answer is the same as for the Fourth Preliminary Issue and for the same reasons (see paragraph 41 above).
44. The Seventh Main Contract Preliminary Issue On the assumption that the facts in paragraphs 57 and 57A of the Yorkshire Water Claim are correct, can Yorkshire Water have an entitlement to recover costs from Taylor Woodrow pursuant to Clause 34.8 of the General Conditions of Contract as claimed in paragraph 91? Clause 34 deals with the following two main categories: (i) parts of the Plant that have been taken over and are the subject of the Take-Over Certificate and (ii) parts of the Plant that have been excluded from take-over by the terms of the Take-Over Certificate: see sub-clause 34.7. I also agree that within the first of those categories there can be a sub-category, namely minor items that have still to be completed (see the final sentence of sub-clause 34.7).
45. In the present case, Attachment B to the Take-Over Certificate dated 25^th February 2000 comprised the sub-category of minor items that were still to be completed and Attachment A listed those items of the Plant that had been excluded from take-over altogether.
46. The words "…not being complete …" mean exactly what they say, i.e. they refer only to those items that are expressly noted on the Certificate as incomplete. In this particular case, those words refer to the items listed in Attachment B as "Outstanding Minor Items". In my opinion, the words plainly do not refer to those parts of the Plant that have been expressly noted as excluded from take-over by the terms of the Take-Over Certificate, as envisaged by the first sentence of sub-clause 34.7 (a function performed by Attachment A in the present case).
47. Accordingly, the answer to this question is "No".
48. The Eighth, Ninth and Tenth Preliminary Issues (8) Can Yorkshire Water have an entitlement to recover costs of work of correction, repair, adjustment or completion pursuant to Clause 34.8 if the said work has not been carried out? (9) Can Yorkshire Water have an entitlement to recover costs of work of correction, repair, adjustment or completion pursuant to Clause 34.8 if the said cost has not been incurred? and (10) Can Yorkshire Water have an entitlement to recover costs of work of correction, repair, adjustment or completion pursuant to Clause 34.8 if the items noted on the Take-Over Certificate as not being complete have not been corrected, repaired, adjusted or completed to the satisfaction of the Project Manager? Since the word "cost" in sub-clause 34.8 is not emboldened it is not subject to the restricted contractual definition given to the word "Cost" in the Special Conditions (see above). The words "…to the satisfaction of the Project Manager…" refer to the standard to which the Contractor should have carried out the work in question. The expression "…such work…" serves to identify the relevant work, which the Project Manager is then empowered to make arrangements for others to carry out, following the Contractor's default. The expression "…such work…" does not refer back to those words that serve only to state the standard of the work that should have been carried out by the Contractor in the first place (i.e. it does not refer back and/or embrace the words "…to the satisfaction of the Project Manager…").
49. Accordingly, the Purchaser's entitlement to recover the cost of the work to be carried out by others, pursuant to the final sentence of this sub-clause, is not expressed to be conditional upon the Project Manager's satisfaction having been achieved in respect of the relevant work (which I understand to be the point of question 10).
50. This latter conclusion is not surprising, in my view, because the final sentence of sub-clause 34.8 is expressed in sufficiently wide terms to embrace both the actual and/or prospective cost of the relevant works. Accordingly, my answer to each of these questions is "Yes".
51. The Eleventh and Twelfth Preliminary Issues: Clause 36.11 (11) Can Yorkshire Water have an entitlement to recover the costs of making good defects as claimed in paragraph 94 of the Yorkshire Water Claim if the said costs have not been incurred? and (12) Can Yorkshire Water have an entitlement to recover the costs of making good defects as claimed in paragraph 94 of the Yorkshire Water Claim if the defects have not been made good? The relevant words of this particular sub-clause are sufficiently widely expressed to embrace both the actual and/or prospective cost of making good such a defect. Accordingly, my answer to each of these questions is "Yes". In the event, however, no recoverable costs under this particular sub-clause have been proved in these proceedings (see paragraphs 88-91 below).
52. Issues of Construction: Main Contract. Although my answers to the Main Contract Preliminary Issues dealt with many of the construction issues in relation to the Main Contract, a number of construction issues remain outstanding, the principal ones being as follows:

(i) Main Contract Preliminary Issues 2 and 3, with regard to which I deferred giving my answers ("Preliminary Issues 2 and 3");

(ii) Yorkshire Water's continued claim to be entitled to recover financial relief under sub-clause 36.7, despite my answers to Preliminary Issues 4 and 6 ("the sub-clause 36.7 point");

(iii) the proper construction of the provision "The plant shall be capable of treating the design flow and load to the effluent standard specified … with one basin out of service": see paragraph 0.2.6.3 of the Specification ("the one tank out of service point");

(iv) whether Yorkshire Water was contractually entitled to an SBR that incorporated an "Influent Distribution and Sludge Collection Manifold" (an "IDSC"): see, in particular, the expression "The SBR will be the US filter Jet Tech design complete with … manifolds …" in paragraph 0.2.6.3 of the Specification ("the IDSC point").

53. I therefore now turn to deal with each of these outstanding issues with regard to the proper construction of the Main Contract.
54. (1) Preliminary Issues 2 and 3. (a) Preliminary Issue 2: In essence, the Second Main Contract Preliminary Issue is concerned with Yorkshire Water's contractual entitlement to recover various future and/or prospective/anticipated costs pursuant to the provisions of clause 35.10 and/or Schedule 10. However, as stated and for the reasons given in my answer to the First Main Contract Preliminary Issue (see paragraphs 32 to 38 above), I have come to the conclusion that, in the circumstances of this case, Yorkshire Water have no relevant right of recovery under sub-clause 35.10. Accordingly, it is only Schedule 10 that falls to be considered for the purposes of this preliminary issue.
55. Both Mr Streatfeild-James and Mr Elliott emphasised that Yorkshire Water's entitlement under each of the relevant Schedule 10 "absolute guarantees" (i.e. TP1, TP2, TP9 and TP11(b)) was to be paid "the actual costs incurred" by Yorkshire Water in respect of certain defined matters. It was their submission that the natural meaning of the quoted words was unambiguous, that the words clearly refer to costs that Yorkshire Water has actually incurred and that they do not refer to sums that Yorkshire Water has yet to spend. It was submitted that to allow Yorkshire Water to recover, under any of the Schedule 10 absolute guarantees, costs that it has not yet incurred would be to violate the natural meaning of the language used and, in particular, to ignore the words "actual" and "incurred" and the use of the past tense in the case of the latter expression.
56. It was Mr Elliott's submission that this construction of the relevant words was entirely consistent with and supported by the context in which they were used. He referred to the terms of Schedule 10 and submitted that the costs recoverable in respect of the Schedule 10 "absolute" guarantees are clearly distinguished from the regular liquidated damages recoverable under the same Schedule in respect of the other guarantees, because the former are determinable by the costs that have actually been incurred by the Purchaser, whereas the latter are fixed or predicted in advance: see, for example the formulae for calculation of liquidated damages in respect of TP3 – Power Consumption, TP5 – Polyelectrolyte Usage and TP8 – Water Consumption.
57. Mr Elliott submitted that, although there was the foregoing clear distinction between the two types of financial recovery to which the Purchaser was entitled under Schedule 10, the scheme of the Schedule showed that they shared a feature in common namely that, before recovery is permitted, both recoverable costs and liquidated damages are fixed and ascertained. In the case of the absolute guarantees, this is achieved by limiting the recoverable costs to such costs as the Purchaser has actually incurred. In the case of the other guarantees, calculation of the liquidated damages depends on ascertaining the relevant unit costs for use in the applicable formula: see paragraph 41 of Mr Elliott's written closing submissions.
58. Mr Elliott suggested that this common feature may explain why the costs recoverable in respect of the Schedule 10 absolute guarantees are included in a Schedule entitled "Liquidated Damages for Failure to Pass Performance Tests". By the time the Purchaser can claim his costs from the Contractor in respect of an absolute guarantee, he will do so for a sum certain, namely the "actual costs incurred". The claim will thus be for a liquidated contractual debt and, as such, consistent with the heading to the Schedule and "the liquidated damages regime" in respect of the other Schedule 10 guarantees. In my view, there is much force in that submission.
59. In the event, the principal submissions on behalf of Yorkshire Water, with regard to the proper construction of the expression "actual costs incurred" in Schedule 10, were made in writing by Mr Sears QC and Ms Livesey in a document dated 20^th November 2003, submitted after completion of final speeches and in accordance with the directions I had given on 6^th November in order to deal with a post-speeches citation of authority by Mr Sears on this issue. Detailed written submissions, dealing with this late citation of authority, were also submitted in accordance with my directions by Mr Streatfeild-James and Mr Elliott on behalf of Taylor Woodrow/Biwater and Elga respectively.
60. In broad terms, Mr Sears' submission was to the effect that the appropriate approach to the proper construction of the material terms of Schedule 10 is that which was stated by Sir Thomas Bingham M.R. and Steyn L.J. in the course of their judgments in Arbuthnott v. Fagan (unreported), 30 July 1993; Court of Appeal (Civil Division), in passages that were quoted and applied by Mance J. in Charter Reinsurance Co. Ltd v. Fagan (1997) AC 313 at pages 325/326, as follows:

"Courts will never construe words in a vacuum. To a greater or lesser extent, depending on the subject matter, they will wish to be informed of what may variously be described as the context, the background, the factual matrix or the mischief. To seek to construe any instrument in ignorance or disregard of the circumstances which give rise to it or the situation in which it is expected to take effect is in my view pedantic, sterile and productive of error. But that is not to say that an initial judgment of what an instrument was or should reasonably have been intended to achieve should be permitted to override the clear language of the instrument, since what an author says is usually the surest guide to what he means. To my mind construction is a composite exercise, neither uncompromisingly literal nor unswervingly purposive: the instrument must speak for itself, but it must do so in situ and not be transported to the laboratory for microscopic analysis." (per Bingham M.R.)

"I readily accept Mr Eder's submission that the starting point of the process of interpretation must be the language of the contract. But Mr Eder went further and said that, if the meaning of the words is clear, as he submitted it is, the purpose of the contractual provisions cannot be allowed to influence the court's interpretation. That involves the process of interpretation in the fashion of a black-letter man. The argument assumes that interpretation is a purely linguistic or semantic process until an ambiguity is revealed. That is wrong. Dictionaries never solve concrete problems of construction. The meaning of words cannot be ascertained divorced from their context. And part of the contextual scene is the purpose of the provision. In the field of statutory interpretation the speeches of the House of Lords in Attorney-General v. Prince Ernest of Hanover … showed that the purpose of a statute, or part of a statute, is something to be taken into account in ascertaining the ordinary meaning of words in the statute: …it is important to bear in mind that the purpose of the statute is a permissible aid at all stages in the process of interpretation. In this respect a similar approach is applicable to the interpretation of a contractual text. That is why in Reardon Smith Line Ltd v. Yngvar Hansen-Tangen …Lord Wilberforce, speaking for the majority of their Lordships, made plain that in construing a commercial contract it is always right that the court should take into account the purpose of a contract and that presupposes an appreciation of the contextual scene of the contract." (per Steyn L.J.)

61. It is to be observed that the importance of context to the meaning of words was also emphasised by Lord Hoffman, in the course of his speech in Charter Reinsurance (supra), when he said this (see page 391):

"I think that in some cases the notion of words having a natural meaning is not a very helpful one. Because the meaning of words is so sensitive to syntax and context, the natural meaning of words in one sentence may be quite unnatural in another. Thus a statement that words have a particular natural meaning may mean no more that that in many contexts they will have that meaning. In other contexts their meaning will be different but no less natural."

62. In Charter Reinsurance the key question was whether the words "actually paid", as used in three contracts of reinsurance, meant that the liability of reinsurers under each of the relevant policies was limited to the sum in respect of which the reinsured (i.e. Charter Reinsurance) had discharged its liabilities in respect of the risks that it had insured. It was the reinsurers' case that this was the natural meaning of the words used and that there was nothing in the context which required the words to be given any different meaning. However, that argument was rejected and it was held that the words did not impose a temporal precondition to recovery, but were there "for the purpose of measurement". In the conclusion to his judgment, Mance J. put the matter in this way (see page 350):

"Adopting an approach to construction which is neither uncompromisingly literal nor unswervingly purposive, I consider that the words of the ultimate net loss clause in all three contracts were intended to ensure that reinsurance recoveries were at each stage and ultimately measured in all conceivable circumstances by reference to the precise net amount of the commitment involved in the settlements achieved between the reinsured and the original assured; and that they did not have an additional purpose of introducing a temporal precondition to recovery in the form of disbursement or other satisfaction of any such amount. … Putting the matter at its very lowest, the words of the ultimate net loss clause were not and are not in my view clearly designed for this purpose, nor are the words of the contract taken as a whole. On the contrary, in my judgment, they were designed, and are to be read as designed, for the different purpose of measurement."

In the course of his speech in Charter Reinsurance, Lord Mustill expressed his conclusion in the following terms:

"…I am now satisfied that the purpose of "the sum actually paid" in clause 2(c) is not to impose an additional condition precedent in relation to the disbursement of funds, but to emphasise that it is the ultimate outcome of the net loss calculation which determines the final liability of the syndicates under the policy. In this context, "actually" means "in the event when finally ascertained," and "paid" means "exposed to liability as a result of the loss insured under clause 1." These are far from the ordinary meanings of the words, and they may be far from the meanings which they would have had in other policies, and particularly in first-tier policies of reinsurance. But we are called upon to interpret them in a very specialised form of reinsurance, and I am now satisfied that, as Mance J. expressed it in his judgment at first instance, the words in question did not have the purpose of introducing a temporal precondition to recovery in the form of disbursement or other satisfaction of the precise net commitment between Charter and its reinsured, but were there "for the purpose of measurement"."

63. Accordingly, Mr Sears submitted that, in construing the meaning of the words "actual costs incurred", as used in the Schedule 10 absolute guarantees, it is important not to adopt a purely linguistic or semantic approach but to consider them in their context, part of which is the purpose of the provisions in question. It was his submission that the words are used in a section of the contract that is entirely concerned with the measure of the Purchaser's loss and/or the measure of recovery that the Purchaser is entitled to make in the event that the Contractor is unable to achieve the relevant performance requirements. He submitted that when, as here, the words are used in a context that is concerned with the measurement of loss and/or level of recovery, they ought not to be construed so as to impose a condition precedent to entitlement by reference to the actual disbursement of funds by the Purchaser, but should be construed so as to emphasise that it is the ultimate outcome of the remedial works that determines the final liability of the Contractor.
64. Mr Sears therefore submitted that, in the context in which the words "actual costs incurred" were used in this contract, the word "actual" should be construed to mean "in the event when finally ascertained" and the word "incurred" should be construed to mean "exposed to liability as a result of the need to achieve the standards of the Performance Tests": see the analogous construction of the words "the sum actually paid" by Lord Mustill in Charter Reinsurance (supra).
65. Like Mance J. in Charter Reinsurance, I have adopted an approach to construction that is "neither uncompromisingly literal nor unswervingly purposive" (see the quotation from his judgment in paragraph 62 above) and have come to the conclusion that Mr Streatfeild-James' and Mr Elliott's submissions as to the proper construction of the words "actual costs incurred" are correct. In my opinion, in order to recover its relevant costs pursuant to the provisions of the Schedule 10 "absolute guarantees", it is necessary for Yorkshire Water to have expended those costs (i.e. the costs must be actual and they must have been incurred by Yorkshire Water). The absolute guarantees therefore do not extend to recovery of future or anticipated costs, i.e. costs that Yorkshire Water has yet to spend.
66. As it seems to me, that is not only the natural meaning of the words, but a purposive approach also clearly shows that to be the correct construction of the relevant words, when considered in their context: i.e. the context being Schedule 10, a Schedule that makes a number of detailed provisions for the contractual recovery of costs and/or damages by the Purchaser, pursuant to the various Contractor's performance "guarantees". The general scheme or purpose of Schedule 10 is to make appropriate provision for the Contractor's contractual liability and the Purchaser's contractual entitlement under each "guarantee" to be in respect of a liquidated sum, calculated by reference to the costs actually incurred (in the case of recoverable costs under the "absolute" guarantees") or by use of the relevant unit costs in the applicable formula for calculating liquidated damages (in the case of the other "guarantees"), and not involving any estimate or prediction of future costs: see paragraphs 56 to 58 above.
67. Accordingly, my answer to the Second Main Contract Preliminary Issue is "No." However, if I am wrong about that and the expression "actual costs incurred" should be construed as submitted by Yorkshire Water, the outcome of these proceedings still remains the same for reasons that will become apparent later.
68. (b) Preliminary Issue 3: For the reasons already given, it is only Schedule 10 that falls to be considered in relation to the question posed by this Main Contract Preliminary Issue (see above). Accordingly, the issue raised by the question is whether, under the relevant provisions of Schedule 10, Yorkshire Water is entitled to recover the costs claimed in paragraph 95 of the Particulars of Claim, if Yorkshire Water has not yet achieved the standards of the performance tests. The relevant words of Schedule 10 are as follows:

"… if the Contractor is unable to meet the requirements of Schedule 7, Performance Tests [for the absolute guarantee in question]…then the Contractor shall pay to the Purchaser the actual costs incurred by the Purchaser …to achieve the standards of the Performance Tests .."

69. Mr Streatfeild-James and Mr Elliott each submitted that the words of Schedule 10 are unambiguous and clear. They submitted that, applying the ordinary and natural meaning of the words, it is only after the works have been carried out and the costs incurred that it can be properly judged whether they were carried out and/or incurred to achieve the performance standards in question. It was submitted that this approach is consistent with the approach under Clause 36, which does not depend on the subjective intention of the party claiming to have made good a defect, but on whether the defect has actually been made good.
70. It was Mr Streatfeild-James' submission that the remedial works undertaken by Yorkshire Water (and the costs incurred to date) have not been incurred to achieve the standards of the performance tests. In particular, it was argued that, on Yorkshire Water's own case, the Yorkshire Water remedial scheme has not resulted in the SBR being capable of producing effluent of the specified quality over the full range of flows and loads specified in the Main Contract (as is required by paragraph 7.2.1 of Schedule 7: see above), since its maximum capacity is limited to 76 tcmd (i.e. the originally specified proportion of effluent from the high-level primary filters). This, it was submitted, is fatal to any claim brought under Schedule 10, since the wording of the Schedule is clear and limits Yorkshire Water's contractual entitlement to recovery of "the actual costs incurred … to achieve the standards of performance tests" (emphasis added).
71. For his part, Mr Elliott submitted that if Yorkshire Water's remedial scheme fails to produce a plant that is capable of meeting the standard of the performance tests, it would not be accurate to describe the money spent on the remedial works as costs incurred "to achieve the standards of the performance tests". Rather, the claim will be one for the recovery of costs incurred in not achieving the standards of the performance tests. It was Mr Elliott's submission that, although these costs might (in certain circumstances) be recoverable as damages at Common Law, it would be wrong and contrary to the Scheme of the Red Book to compel the Contractor to bear those costs pursuant to Schedule 10. Similarly, if the Contractor were to claim such costs (i.e. the cost of unsuccessful remedial works) as a variation under sub-clause 36.2, it would be wrong to expect the Purchaser to bear them.
72. Mr Elliott submitted that the standard wording of the Red Book qualifies what could prove to be the prohibitively expensive consequences of failure of performance for the Contractor by providing liquidated damages and/or abatement provisions as in sub-clause 35.10. He accepted that the terms of Schedule 10 removed that protection in the case of the "absolute" guarantees. He submitted, however, that Schedule 10 provides for a different kind of limitation of liability in the case of the "absolute" guarantees, entirely in keeping with the general scheme of the Red Book.
73. It was Mr Elliott's submission that the terms of the "absolute" guarantees are such that the Contractor is not to be held liable for the potentially unlimited and very expensive failures (or simple inability) by the Purchaser to achieve the same absolute standards that the Contractor is required to meet. He submitted that it is not enough for the Purchaser to say that a genuine or even reasonable attempt has been unable to achieve the requisite standards. He contended that, in order to recover the relevant costs of remedial works pursuant to the Schedule "absolute" guarantees, the Purchaser has to prove that the relevant remedial works have been successful in achieving the performance standards in question. Mr Elliott maintained that success in achieving those standards by the Purchaser must be proved and can only be proved by the taking and passing of those performance tests.
74. However, Mr Slater submitted that the meaning of the relevant words is not limited in the manner suggested by Mr Streatfeild-James and Mr Elliott. He stressed that there is nothing in the wording of any of the "absolute" guarantees that can be said to impose a condition precedent for recovery of Yorkshire Water's remedial costs that the relevant performance tests actually be taken and passed after completion of the remedial works in question. I am satisfied that, on a literal as well as purposive construction of the relevant words, Mr Slater is correct in that submission.
75. As indicated in my original answer (see paragraph 40 above), I accept Mr Slater's submission that the words "to achieve" should be construed to mean "in order to achieve". In addition and adopting the same approach to construction as before (see paragraph 65 above), I also accept Mr Slater's further submission that, if the remedial costs in question are incurred as the result of appropriate steps being taken by the Purchaser towards achieving the relevant performance standards, then they are costs incurred in order to achieve the standards of the relevant performance tests, even though they have been incurred and recovery is sought at a stage when it may not yet be possible to meet all the performance standards (e.g. with regard to full flows and loads).
76. In my judgment and putting to one side the question of how such matters can be proved, as such they constitute costs that come within the widely expressed and natural meaning of the material words of Schedule 10 and I reject Mr Streatfeild-James' and Mr Elliott's submissions to the contrary effect. On that basis and provided that the other requirements of the Schedule 10 "absolute" guarantees are satisfied, I am satisfied that remedial costs incurred by Yorkshire Water can be recovered under the terms of the applicable "absolute" guarantee, even though no full and complete performance test has yet been carried out. It is on that basis that (as was my inclination before) my qualified answer to the Third Main Preliminary Issue is "Yes".
77. Before turning to the next contractual issue, it is worth noting that it is Yorkshire Water's case that the results of a "performance test" taken in February/March 2003 (as to which, see paragraph 452 below) show that, as a result of the remedial works carried out so far, the modified SBR was able to achieve the relevant performance standards with regard to its treatment of influent that had been (by then) limited to a specified proportion of the effluent from the high-level primary tanks, up to a maximum of 76 tcmd: see paragraph 452 below and Schedule 7 and Guarantees TP1 and TP 2 (b) (ii) in paragraph 25 above. It was therefore Mr Slater's submission that the costs incurred by Yorkshire Water in carrying out those remedial works were properly recoverable pursuant to the provisions of Schedule 10 as construed in my foregoing answers to the Main Contract Preliminary Issues 1 to 3 inclusive.
78. (2) The Sub-Clause 36.7 point. Mr Slater submitted that Yorkshire Water was entitled to recover its costs of remedial works already carried out pursuant to sub-clause 36.7 of the General Conditions of Contract, notwithstanding my existing answer to the Fourth Main Contract Preliminary Issue. In that answer I made it clear that the Contractor's obligation to reimburse the Purchaser's reasonable costs of making good a defect under sub-clause 36.7 is conditional upon the Purchaser having made good the defect in question (see paragraph 41 above).
79. Mr Slater acknowledged that Yorkshire Water has always put its case on the basis that the relevant defect (which, for the purposes of this part of his submissions, he accepted had been expressed both originally and subsequently as a single defect and to which I shall hereafter refer as "the overall defect") was the failure or inability of the Plant to achieve the requisite final effluent quality: see Earth-Tech's letter dated 14^th March 2000 (C64/339: and see paragraph 270 below) and paragraph 34 of Yorkshire Water's closing submissions. Having regard to the terms of that letter and to paragraphs 58 and 60 of the Particulars of Claim (A1/31), Mr Slater was clearly right to make that concession. In the event, for the reasons that I give later in this judgment, I am satisfied that the SBR was capable of achieving the requisite final effluent quality (i.e. it was not defective as alleged by Yorkshire Water). However, Mr Slater went on to submit that Yorkshire Water has also always made it clear that there were a number of reasons why the Plant did not achieve final effluent quality and/or that there were a number of items of work required to be done in order to achieve the requisite quality: see Earth-Tech's letter dated 7^th April 2000 (paragraph 290 below) and see also paragraphs 59 and 83.2 of the Particulars of Claim (A1/30 and 38).
80. It was therefore Mr Slater's contention that each of the items listed in paragraph 83.2 of the Particulars of Claim contributed to the overall defect and that each was a "sub-species of (the overall) defect" (see paragraph 38 of Yorkshire Water's closing submissions). Mr Slater submitted that, although the single overall defect was the Plant's failure to produce effluent of the required quality, the overall defect was itself the product of a number of individual defects or deficiencies in the design and construction of the Plant (hereafter called "the individual deficiencies"), each of which required to be remedied in order to improve the overall performance of the Plant and achieve the necessary contractual standards. He submitted that this was precisely what Yorkshire Water had endeavoured to do by carrying out the remedial works in question. Mr Slater submitted that neither the wording nor my answer to the Fourth Main Contract Preliminary Issue prevented recovery of the costs of remedial works under sub-clause 36.7, even if the works had not achieved a complete remedy, provided the works in question had gone some way towards making good the overall defect by dealing with some of the individual deficiencies. Stated shortly, it was Mr Slater's submission that sub-clause 36.7, when properly construed, permits recovery of costs incurred in carrying out remedial works that have the effect of partly making good the overall defect: see paragraph 41 of Yorkshire Water's closing submissions. Mr Slater stressed that sub-clause 36.7 does not state that the overall defect has to be made good completely. He therefore submitted that the relevant words of sub-clause 36.7 are susceptible to a construction that allows recovery of costs incurred in making good the overall defect in part: see paragraph 43 of Yorkshire Water's closing submissions.
81. In my view Mr Slater's foregoing submission is an ingenious attempt to circumvent my existing answer to the Fourth Main Contract Preliminary Issue. However, I agree with Mr Streatfeild-James and Mr Elliott that Mr Slater's submission, to the effect that sub-clause 36.7 can be so construed that Yorkshire Water are entitled to recover its costs incurred in making good a defect in part, cannot be reconciled with the answer that I gave when deciding that preliminary issue and in which I stated (inter alia):

"… the obligation of the Contractor, to reimburse the Purchaser's reasonable cost of making good the defect under this particular sub-clause, is conditional upon the Purchaser having made good the defect in question. It is only once a defect has been made good that the obligation to reimburse the reasonable cost arises". See paragraph 41 above.

Accordingly, for those reasons, I reject Mr Slater's foregoing primary submission with regard to the construction of sub-clause 36.7.

82. Mr Slater also maintained that, in any event, each of the matters listed in Earth-Tech's letter of 7^th April 2000 constituted a defect within the meaning of sub-clause 36.1 and that either the letter itself constituted sufficient notice under paragraph (b) of that sub-clause or, in the alternative, no or no further notice under sub-clause 36.1 was required to enable Yorkshire Water to recover its remedial costs pursuant to sub-clauses 36.6 and 36.7. In any event, for the reasons that I give later in this judgment, I am satisfied that (at all material times) the SBR was not defective as alleged and that the "remedial works" in question were not actually necessary.
83. Furthermore, if it is assumed that these further submissions by Mr Slater are correct and if it is further assumed that the SBR was defective as alleged, the contractual position with regard to Yorkshire Water's entitlement to be reimbursed its remedial costs under Sub-clause 36.7 would remain as before: i.e. it is conditional upon Yorkshire Water having made good the defect in question.
84. In my view, although the letter of the 7^th April 2000 itemises a number of alleged deficiencies in the Plant, in each case it is clear that it is the whole Plant which is said to be defective as a result of each such deficiency (i.e. the Plant does not meet the overall required performance standards because of the deficiency in question): see also the terms of paragraph 83.2 of the Particulars of Claim.
85. However, despite the remedial works that have been carried out, it is common ground that the SBR as "remedied" does not now meet the overall required performance standards: stated shortly, the SBR is now restricted to the specified proportion of effluent from the high-level primary settlement tanks and does not treat the specified proportion of the ASP effluent. In those circumstances, to the extent that any of the deficiencies alleged in the 7^th April letter might have been successfully addressed by any of the remedial works carried out to date, the deficiency in question has only been remedied in part because the SBR, as presently constituted, is unable to achieve fully the required contractual performance standards. On that basis, Yorkshire Water's remedial costs are still not recoverable under sub-clause 36.7, having regard to my existing answer to the Fourth Main Contract Preliminary Issue and my reasons for that answer. However, I accept that this conclusion is entirely academic, having regard to my conclusion (the reasons for which I explain later in this judgment) that times the SBR was not defective as alleged (i.e. it was capable of meeting the contractual performance requirements at all material times).
86. Accordingly, for those reasons and putting sub-clause 36.11 to one side for the moment (as to which, see paragraphs 88 to 91 below), I have come to the conclusion that Mr Streatfeild-James and Mr Elliott are correct in their submission that this case is almost entirely concerned with Yorkshire Water's contractual entitlement to recover its relevant remedial costs, pursuant to the material provisions of Schedule 10 of the Main Contract. When considered in conjunction with my conclusion on Preliminary Issue 2 (see paragraphs 54 to 67 above), this particular conclusion does, of course, have a significant effect on Yorkshire Water's entitlement to financial relief in these proceedings. In effect, the consequence is that, so far as concerns its claim for the costs of remedial works, Yorkshire Water's entitlement is limited to the cost of remedial works that have been carried out (i.e. its incurred costs, not its prospective costs).
87. It is important to stress that Yorkshire Water's claim in these proceedings is not a claim for general damages for breach of contract. As I have already explained, Yorkshire Water deleted its original claim for damages for breach of contract at a very early stage of these proceedings and did so for good reason: see paragraphs 16 to 20 and 28 above. Yorkshire Water's claim is confined to its entitlement to financial recovery and liquidated damages pursuant to various specified terms of the main contract. Accordingly, a claim for general damages has not been the subject of either evidence or submissions in these proceedings. It is therefore neither possible nor appropriate for me to decide or even speculate as to what the position would have been, either as to liability or quantum, if any of Yorkshire Water's many and various complaints about the SBR had been the subject of a claim for damages for breach of contract.
88. It is also convenient to deal with sub-clause 36.11 at this stage of my judgment. This particular sub-clause is concerned with financial recovery in respect of latent defects: see paragraph 75 above. My answers to Preliminary Issues 11 and 12 were to the effect that Yorkshire Water was entitled to recover its costs of remedying such defects, even if those costs had not yet been incurred or the defect in question not yet remedied: see paragraph 51 above.
89. On the pleadings, Yorkshire Water's claim under sub-clause 36.11 is limited to defects alleged in respect of the sludge handling system: see paragraph 101C(5) of Yorkshire Water's Re-re-re-amended Reply ("the Reply"). In the event, as Mr Streatfeild-James observed, this limited claim does not appear to have been pursued (although there was no formal indication to that effect at trial). However, Mr Slater did not deal with any specific sub-clause 36.11 claim in either his written opening or closing submissions and there is no obvious reference to any such claim in respect of any latent defect in the sludge handling system in Yorkshire Water's updated and amended Quantum Schedule ("the Quantum Schedule").
90. Although there is a claim for the cost of upsizing the filtrate drain in the sludge handling system (see Item 6 of the Quantum Schedule), I agree with Mr Streatfeild-James' submission that this does not constitute a claim for the cost of remedying a latent defect under sub-clause 36.11. This particular problem had clearly been identified at the time Earth-Tech issued the Take-over Certificate on 25^th February 2000, because it was one of the items of plant listed in Appendix A as excluded from the Take-over of Plant (item (vi)). The cost of remedying this particular defect is therefore not recoverable under sub-clause 36.11: see the terms of 36.11(aa), quoted in paragraph 23 above.
91. Accordingly, I am satisfied that Yorkshire Water is not entitled to make any financial recovery in these proceedings pursuant to the provisions of sub-clause 36.11.
92. (3) The One Tank out of Service Point. It is Yorkshire Water's case that the Main Contact required that the SBR should be capable of operation with one tank out of service, without any time limit and under all specified flow and load conditions (see paragraph 219 of Mr Slater's written opening submissions).
93. As Mr Streatfeild-James pointed out (Biwater's closing submissions, paragraph 603) the starting point is the relevant contractual obligation, which is set out in paragraph 0.2.6.3 of the Specification (see above), as follows:

"The sidestream plant shall comprise a 6 basin SBR Plant … The Plant shall be capable of treating the design flow and load to the effluent standard required in 0.2.3 with one basin out of service."

94. Mr Streatfeild-James emphasised that by paragraph 0.2.6.3 of the Specification the SBR is described as "a 6 basin SBR Plant". He suggested that, although the SBR was required to be capable of treating the design flows and loads on 5 tanks, it is clear from the wording and context of the paragraph in question that this was to be the exception rather than the rule. Mr Streatfeild-James submitted (correctly, in my view) that the phrase "with one basin out of service" suggests that the usual mode of operation would entail all 6 basins being in service. He also made the telling point that, unlike other items in the Specification (e.g. the submersible pumps, the belt thickeners or the polyelectrolyte dosing pumps), the 6^th tank/basin of the SBR is not described in the Specification (or elsewhere) as a "standby" unit. As to the latter point, Mr Streatfeild-James submitted that if what was required was a plant that would operate on 5 tanks, with one tank spare at all times, one would expect that to be spelled out clearly in the tender/contract documents, if for no other reason that that nobody would price the job on such a basis unless it was clearly spelled out.
95. It was therefore Mr Streatfeild-James' submission, supported by Mr Elliott, that the requirement that the SBR be capable of operating on 5 tanks was for the purpose of enabling necessary operations, such as the routine maintenance and/or repair of the SBR, to be carried out. They therefore submitted that paragraph 0.2.6.3 of the Specification should be construed accordingly (i.e. that the SBR should be capable of operating on 5 tanks for such periods as would be required, from time to time, to carry out such necessary operations as routine maintenance and repairs). It was also suggested that this construction entirely accords with the how the relevant obligation would normally be understood by experienced people in the wastewater industry (see, for example, paragraph 7.6.1 of the first report of Yorkshire Water's expert, Dr Brian Chambers ("Dr Chambers": E1/32), paragraphs 252-253 of the first report of Taylor Woodrow's expert, Dr David Johnstone ("Dr Johnstone: E2/53-54) and paragraph 8.5 of the second report of Biwater's expert, Dr Nigel Horan ("Dr Horan": E3A/tab 2/32). Furthermore, as Mr Streatfeild-James pointed out, that was how the Specification was understood at the time by the parties: see, for example, the evidence of Elga's Head of Projects, Mr Derek Workman ("Mr Workman"), at Transcript, Day 17/89, the evidence of Biwater's design process engineer, Mr William Dunn ("Mr Dunn"), at Transcript, Day 19/146 and 151-152 and C7/358 (see paragraph 177 below) and the evidence of Yorkshire Water's Waste Water Technical Support Manager, Mr Jeffrey Holmes ("Mr Holmes") at Transcript, Day 4/82-83, where he said this about a note that he had written at the time (C15/253), whilst emphasising that the plant should be capable of operating on 5 tanks for as long as was needed for such purposes:

"Q. …This is a note from you to Mr Sawyer, including some comments …

A. Yes.

Q. If you look under 3, it says: "More capacity. The Sidestream Plant Specification called for the plant to achieve the required consent with one process unit out of action to cover for breakdowns, etc." That was your understanding at that time, was it not, Mr Holmes?

A. No. It says "et cetera". So my understanding was … – my understanding was that (the) plant should be able to treat flows and loads on five basins, and that could cover for breakdowns or the "etc" could be any number of reasons to cover for maintenance operations. The plain fact is that it was my understanding and the maths show that we wanted to be able to operate fully on five basins only, full stop. There is no qualification in the tender document on that."

96. So far as concerns Elga, Mr Elliott also stressed that, in the course of correspondence with Biwater in 1997, USF/JetTech had made it clear and accepted that one tank could be taken out of service for only a limited time and that the SBR could not be expected to perform a 28 day test with only 5 tanks being operated. In its fax to Biwater dated 13^th August 1997, USF/JetTech stated as follows:

"A) Our SBR designs are based on the ability to take one basin out of action for short periods of time and still meet effluent consent standards. A number of cycle time distribution charts have already been sent to you to show how this can be achieved. However, they are not based on performing the 28-day test periods (as detailed in Schedule 7) with one tank out of action. We trust this is acceptable."

97. On the same day (i.e. 13^th August 1997), Biwater responded in the following terms:-

"Re your fax of this pm we respond in the order of asking:-

A) OK …"

98. The Biwater fax of 13^th August 1997 was subsequently incorporated into the Elga Sub-contract as a Contract document and (it is suggested) itself also incorporated the USF/JetTech fax by direct reference. If necessary, I will deal with the significance of these particular matters when I come to consider the proceedings as between Biwater and Elga. However it is worth noting that there is no suggestion that the views expressed by USF/JetTech (as set out in the 13^th August fax) were passed on to either Taylor Woodrow or Yorkshire Water.
99. For his part, Mr Slater submitted that the relevant obligation was unambiguous in its terms and that the specified 5-tank operation capability was clearly expressed to be without limitation of time. It was his contention that there was no basis for implying any temporal limitation into that obligation and that an examination of the relevant factual context clearly shows that the parties (i.e. Yorkshire Water, Taylor Woodrow and Biwater) did not intend that the obligation in question should be subject to any such limitation, particularly one that is so imprecise and vague, with possibilities ranging from a very short to a very lengthy period of time: see, for example, paragraph 607 of Biwater's closing submissions, in which a period of "around a week at a time" is suggested and Response 31.1 of Taylor Woodrow's Replies to Biwater's Request for Further Information (A3/Tab3/19), in which the suggestion is that a period of at least a year might be required for maintenance, as follows:

"…Biwater should have ensured that the SBR was capable of meeting the contract requirements with one tank out of service on an extended basis, namely the time required to service or maintain one tank out of six. In practice this could mean a 5 tank operation for a very long extended period because each of the 6 tanks could be taken out of service in succession. If each tank was out of service for 2 months, this would mean that the SBR had to be capable of operation on 5 tanks for at least a year."

100. Mr Slater referred to the terms of Yorkshire Water's original invitation to tender, in which it indicated that its preferred option was for a minimum of 4 process units capable of meeting final effluent quality at full design loading and at 75% availability (i.e. 3 units out of 4 achieving the final effluent quality): see B5A/262.
101. Taylor Woodrow's tender proposal in respect of the preferred option (which was for a 4-tank plant) was expressed (inter alia) as follows (C11/160):

"3 No units are capable of treatment to the consent standard and with all units operating the final effluent standard is likely to be a much higher quality than the consent requirement … which could compensate to the overall works discharge should there be any slight reduction in performance of the high or low level filter plants."

102. In its tender proposal in respect of its alternative option (which was for a 6-tank plant), Taylor Woodrow stated (C11/163):

"5 No units are capable of treatment to the consent standard and, with all units operating the final effluent standard from the sidestream plant is likely to be a much higher quality than the consent … which when one considers that the sidestream plant handles around one third of the flow, will probably ensure that the effluent discharge from the whole of the works is likely to be of a better quality. "

103. Mr Slater submitted that there is nothing in either of the passages quoted from Taylor Woodrow's tender proposals to suggest that there would be any restriction or temporal limitation on the ability of the Plant to operate with one tank out of service. On the contrary, it was Mr Slater's submission that in each of the quoted passages Taylor Woodrow was at pains to stress that, with all tanks in operation, the Plant would have additional treatment capability that could make an important contribution to improving the overall quality of the effluent discharged from the whole of the works.
104. Mr Slater submitted that, not only was there an obligation under paragraph 0.2.6.3 of the Specification, that the SBR be capable of treating the design flow and load to the requisite standards with one basin out of service, but that Taylor Woodrow was also contractually obliged to provide an SBR that was capable of carrying out the performance tests successfully with only 5 basins in operation: see, in particular the opening words to the Schedule 7 Forms – Performance Requirements, quoted in paragraph 25 above: i.e. "The Contractor shall be required to achieve the following levels of operational performance to demonstrate that the plant fulfils the requirements of the Contract …".
105. It was Mr Slater's submission that the effect of the latter contractual provision was to oblige Taylor Woodrow to demonstrate in the performance tests compliance with all the requirements of the Contract, including (in particular) that the Plant should be "capable of treating the design flow and load to the effluent standard specified in 0.2.3 with one basin out of service …". On the other hand, Mr Streatfeild-James stressed that there was no express requirement in Schedule 7 for performance testing on 5 tanks and emphasised that it is not clear how a performance test on 5 tanks for the entire 28 days would have allowed Yorkshire Water to assess the ability of the SBR to meet the power or water consumption guarantees imposed under the contract. He also pointed out that clause 7.2.1 of Schedule 7 (see paragraph 25 above and B1/137) provides that the purpose of the effluent quality performance tests was to ensure that "the SBR Plant" was capable of producing effluent of the specified quality and made the powerful point that the expression "Plant" is described (and thus, in effect, defined) in clause 0.2.6.3 of the Specification as a "6 basin SBR Plant" (see paragraphs 26 and 93 above).
106. Again adopting an approach that is neither uncompromisingly literal nor unswervingly purposive (see paragraph 65 above), I have come to the conclusion that Mr Streatfeild-James' submissions are correct. I accept that the extent and/or duration of the period of time for which the SBR might be required to operate with one tank out of service, to enable necessary operations such as maintenance and repair to be carried out, is not specified. However, I cannot accept Mr Slater's submission that that clause 0.2.6.3 of the Specification means that Taylor Woodrow was obliged to provide a 6 basin plant that was capable of operating indefinitely with one basin out of service as if it were, to all intents, a 5 basin plant with one basin on standby. It is clear that the parties understood that the purpose of this provision in the Specification was to enable such necessary operations as maintenance and repairs to be carried out whilst the otherwise "6 basin plant" continued to operate. In my view, Mr Streatfeild-James's construction (see paragraph 95 above) is not only consistent with that proposition but it is also consistent with the general description of the Plant in question (i.e. "a 6 basin SBR Plant") and consistent with the appropriate performance testing of significant contractual guarantees (i.e. power and water consumption: see Schedule 10, Guarantees TP 3 and TP 8: B1/160-161). Furthermore, I am not persuaded that such a construction is in conflict with the passages in Taylor Woodrow's tender proposals to which Mr Slater drew attention. In my view, these did no more than state that the ability of the proposed Plant to operate with one basin out of service (whenever required to do so, but without specifying for how long) meant that it was likely to be able to improve the overall quality of the effluent from the Knostrop works when all tanks were actually in operation: see paragraphs 101 to 103 above. I do not accept that these assurances carried any implied suggestion that the plant would be capable of being operated indefinitely with one basin out of service.
107. Accordingly, in my judgment, the Main Contract did not require that the SBR should be capable of being operated indefinitely under all specified flow and load conditions with one tank out of service and I accept Mr Streatfeild-James' and Mr Elliott's submissions to that effect. However, if I am wrong about that and clause 0.2.6.3 should be construed as suggested Mr Slater, I am satisfied that the outcome of these proceedings will remain unaffected in any event, for reasons that will become apparent from the later stages of this judgment (see, in particular, paragraphs 529-530 below).
108. (4) The IDSC Point. It was Yorkshire Water's case that Taylor Woodrow was contractually obliged to design, supply and install an SBR that was equipped with the standard Jet Tech IDSC manifold.
109. Taylor Woodrow's tender submission included certain technical literature published by USF/JetTech concerning the "Omniflo" SBR, in which the "Influent Distribution/Sludge Collection Manifold" (i.e. the IDSC) is expressly described as one of the "key elements in the "Omniflo" system", the others being "… the Vari Cant Aeration System, the OMNIFLO Flow Control Piping System, the OMNIFLO Microprocessor Control System, one pump per reactor and one blower for every two reactors" (C11/10). Furthermore, the description of the Anoxic Fill phase of the Omniflo SBR cycle that is given on the same page of the literature is as follows:

"During Anoxic Fill influent is distributed throughout the settled sludge through the Influent Distribution Manifold. Pumps are not operated, no power is used, influent is not diluted by mixing, making biological nutrient removal much more reliable."

110. Mr Slater referred to the USF/JetTech technical literature and to paragraph 0.2.6.3 of the Specification, which provides that " …The SBR will be the US filter Jet Tech design complete with Jet Aerators and manifolds, circulating pumps, blowers decanting units and sludge wasting piping, pumps and valves". He submitted that the words of the Specification (in particular, the expression "… complete with … manifolds …"), when read in conjunction with the USF/JetTech technical literature, were sufficient "to import an obligation to supply the standard Jet Tech ID manifold", alternatively that there was an obligation to supply an equivalent form of influent feed arrangement which would be equally effective in achieving the same performance guarantees: see paragraphs 141 to 142 of Yorkshire Water's closing submissions.
111. However, as Mr Streatfeild-James pointed out, the evidence of Dr Andrew Cannon ("Dr Cannon", a process engineer with Elga) and Mr Kenneth Norcross ("Mr Norcross", formerly USF/JetTech's Vice President of Technology) was clearly to the effect that a significant number of satisfactory USF/JetTech SBRs were actually designed and built without an IDSC manifold. According to Mr Norcross, this was particularly the case with regard to SBRs that were designed to deal with industrial wastewater (see Transcript, Day 23/105). I accept their evidence as an accurate statement of the position. Although USF/JetTech's proprietary IDSC can be and is properly described as one of a number of "key elements" in the USF/JetTech Omniflo SBR, it is clearly not one that must necessarily be incorporated into every USF/JetTech designed SBR to ensure its successful operation.
112. In my view, the words of the Specification upon which Mr Slater principally relied, even when read in conjunction with the USF/JetTech technical literature, cannot be construed so as to "import" a contractual term that specifically required Taylor Woodrow to provide the USF/JetTech proprietary IDSC as a component part of the SBR that it was obliged to design and install under this design and build contract. The word themselves do not compel such a construction and the fact is that many successful USF/JetTech SBRs have been designed and built without IDSCs.
113. Furthermore, the Main Contract was a performance contract in which the design obligation (and thus the design decision whether to include an IDSC) rested entirely with Taylor Woodrow, who was also the party who "guaranteed" the performance of the SBR in a number of appropriate respects (see Schedule 10 of the Main Contract, paragraph 25 above).
114. So it was that Mr Holmes, like others in the Yorkshire Water team, clearly recognised and readily acknowledged that Yorkshire Water could not insist on the inclusion of an IDSC, when he described his concern at the design omission of the IDSC in his evidence, as follows: "We at YWS did know that, despite our reservations, we were not in a position to demand the inclusion of an IDSC, given that Taylor Woodrow/Biwater were the ones designing the plant" (D1/88 paragraph 33).
115. Accordingly, for those reasons, I am satisfied that the Main Contract was not subject to the specific contractual term for which Mr Slater contended on this aspect of the matter. Of course, this particular decision with regard to the alleged specific contractual term does not preclude the conclusion (if it be appropriate) that, in the particular circumstances of this case, the SBR as designed and supplied by Taylor Woodrow lacked adequate or sufficient selector capacity because (inter alia) it was designed and built without an IDSC or an equivalent satisfactory method of introducing the influent into the SBR. In the event, for the reasons given later in this judgment, not only am I satisfied that the SBR was not defective as alleged but it is my view that the SBR did not lack selector capacity as the result of the designed omission of the IDSC (see, in particular, paragraphs 481 to 488 below).
116. The Main Factual and Scientific/Technical Circumstances. I now turn to give an account of the main factual and scientific/technical circumstances relating to this difficult and complex case, as I find them to be on the evidence that I have heard, seen and read. Although this section of my judgment contains a fair amount of detail, I emphasise that my primary purpose is to set out only the main factual and scientific/technical findings that have led me to the various conclusions that I have reached. Even so, the result is a dauntingly long judgment. The task of evaluating all the evidence in this case has been an extremely challenging one. It has taken me a great deal of time and much anxious thought. I have done my best to read, absorb and remind myself of all the relevant evidence. I have been greatly assisted in my task by the well presented oral arguments put forward on behalf of the parties and, in particular, by the careful and detailed written closing submissions prepared by Mr Slater, Mr Streatfeild-James and Mr Elliott, all of which I have read and carefully considered. I have reached my various conclusions only after having given full consideration to all the points made by Counsel in their submissions, both oral and written. Where appropriate in the course of this judgment, I have identified those submissions that I have found particularly helpful or persuasive. Inevitably in a case of this magnitude and complexity, it is not possible to refer to every point made on behalf of the parties. I emphasise, however, that I have made every effort to take all submissions and all the evidence fully into account.
117. The improvement works were required by Yorkshire Water to enable Knostrop to comply with the revised effluent standards that were to be brought into effect by the National Environmental Programme on 1^st July 2000 and the Waste Water Treatment Directive on 31^st December 2000.
118. The pre-existing plant at Knostrop consisted of two essentially separate sewage treatment systems for dealing with the raw sewage received from the two main sewers that serve different parts of the overall catchment area. The two sewers are known as the "High-Level" and the "Low-Level" sewers and the separate treatment processes connected to each are referred to in the same way. The high-level system receives sewage mainly from the more domestic northern and eastern areas of the city. The low-level system receives sewage mainly from the city centre and the industrial areas to the south.
119. The pre-existing low-level plant consisted of primary sedimentation, followed by two separate forms of treatment operating in parallel, namely biological filtration and an Activated Sludge Plant ("the ASP"). The pre-existing high-level plant consisted of primary sedimentation, followed by biological filtration.
120. The central feature of the improvement works was a process system that was designed to effect appropriate biological treatment of a specified proportion of the wastewater passing through each of the high-level and low-level systems and diverted to the new "Sidestream" Plant for that purpose. The core component of the new Sidestream Plant was to be the SBR.
121. The Biological Treatment of Sewage. Broadly speaking, the biological treatment of sewage is based on the natural oxidation of organic matter and ammonia by an accumulation of suitable bacteria and other micro-organisms. These microbes consume the organic matter and nutrients in the sewage, thereby oxidising it and multiplying their own numbers. In effect, the biological treatment of sewage in a sewage works is an intensification of naturally occurring processes, but in a controlled and closed environment where the bacteria involved do not deplete the naturally limited amounts of dissolved oxygen in rivers and the like, which fish and other forms of natural aquatic life need in order to survive.
122. There are two principal processes by which the biological treatment of sewage is achieved; one is called "carbonaceous oxidation" (which deals with biodegradable organic matter) and the other is called "biological nitrification" or simply "nitrification" (which deals with ammonia).
123. Carbonaceous oxidation is achieved by the oxidation of organic matter by microbes to produce carbon dioxide and water. Nitrification is the process whereby certain bacteria (often called "nitrifiers") oxidise the ammonia in sewage to the oxidised forms of nitrogen (i.e. nitrates, which are nutrients that can, as it happens, promote aquatic plant life).
124. In order to meet the relevant consent standards (see paragraph 154 below), it was necessary for the Knostrop SBR to be capable of both carbonaceous oxidation and nitrification. A plant that carries out both treatment processes (i.e. carbonaceous oxidation and nitrification) can be loosely referred to as a "nutrient-removal plant", although that expression is more accurately applied to a treatment plant that is also designed to remove nitrates and phosphates (Transcript, Day 31/30).
125. Biological Oxygen Demand (BOD) is a measure of sewage strength and is the amount of oxygen, usually expressed in milligrams per litre, required by the biodegradable material in the sewage to achieve its biological oxidation. The expression "BOD" is also used as an indicator of the readily biodegradable material or "food" for the bacteria that is contained within the sewage and is often used as a convenient term by which to refer to the biodegradable material itself. In the United Kingdom and many other countries, the BOD of a particular sewage is established by measuring the amount of oxygen taken up during a laboratory test carried out over a period of five days at a temperature of 20C, in which the natural conditions in a river are simulated ("5-day BOD"). 5-day BOD is usually expressed as BOD5. As is normal practice in the UK wastewater industry, all the BOD values used in the works of improvement (e.g. in the Specification and in the design) were BOD5. Unless otherwise indicated, all references to BOD in this judgment are to BOD5.
126. In the UK, an equation invented/developed by Dr Chambers and Mr Arthur Boon ("the Boon and Chambers equation": as to "Mr Boon", see below) is usually used to calculate the actual 5-day BOD from the results obtained during the test. The BOD5 is usually expressed as milligrams of oxygen removed per litre of sewage or simply as mg/l. Thus, if the sewage in question is found to have a BOD5 of 200mg/l that would mean that every litre of the sewage that is discharged to the river would remove a total of 200mg. of oxygen from the river over a five-day period. However, not all the biodegradable material is necessarily oxidised in five days and there may well be a continuing (albeit, normally very small) oxygen demand after five days. So it is that some other countries (e.g. in Scandinavia) conduct the BOD test over seven days (Transcript, Day 16/136), which results in BOD values expressed as BOD7.
127. Chemical Oxygen Demand ("COD") is a more stringent measure of sewage strength. The COD test utilises sulphuric acid and a temperature of 120C to speed up the reaction and is thus a much more exacting and more rapid process than the 5-day BOD test. The COD test conditions are such that all the organic material in the sewage is converted to carbon dioxide, i.e. all readily biodegradable organic material together with all other organic material of a less or non-biodegradable nature. Consequently, the value for COD in a particular sewage will always be significantly higher than for the BOD. As the COD test measures the total organic fraction in the sewage and the BOD test measures only the organic fraction that is biodegradable, the COD:BOD ratio is a crude measure of the biodegradability of the sewage in question. Thus the COD:BOD ratio is one that is widely used in the industry as a rough indication of the treatability of the sewage in question. In broad terms, the higher the ratio of COD to BOD, the more difficult the sewage is to treat (see also paragraph 331 to 336 below).
128. It should be noted that the sewage or wastewater that is treated by such biological treatment processes is no longer in its original "raw" state. Immediately after entering the sewage works, the raw sewage is put through appropriate stages of screening and primary sedimentation that remove all stones, grit, large objects and settleable solids. These earlier stages remove up to 60% of the BOD and suspended solids that are present in the original raw sewage. Having gone through these primary stages, the sewage is known as "settled sewage". The settled sewage is then subject to further treatment by one or more of the various types of biological treatment process, before being finally discharged to the local watercourse as treated effluent.
129. There are a number of different forms of biological treatment of settled sewage (hereafter referred to as "sewage" or "wastewater") that are designed to remove both the remaining BOD and the ammonia prior to discharge. However, all such forms of treatment have the following features in common: (i) a means whereby the sewage is contained for enough time for the necessary biological activity to take place, (ii) a means of maintaining an adequate concentration of micro-organisms in contact with the sewage that is to be treated, (iii) a method of aeration that provides enough oxygen to meet the oxygen demand of the organic material and the ammonia in the sewage and (iv) after completion of the treatment process, a system or means whereby the micro-organisms are separated from the treated wastewater that is to be discharged to the watercourse as treated effluent.
130. Broadly speaking, all types of biological treatment can be split into two basic types, defined by how contact is maintained between the micro-organisms and the influent sewage (i.e. feature (ii) in the previous paragraph). The two basic types are (i) the "attached growth" process, of which the pre-existing biological filtration processes at Knostrop are examples and (ii) the "suspended growth" process, of which the pre-existing ASP at Knostrop and the new SBR are both examples.
131. The Attached Growth Process. In this type of biological treatment, a tank or basin is filled with an inert support material, such as coal, blast furnace slag, plastic modules or polystyrene. The sewage is passed slowly over this material and the micro-organisms in the sewage attach themselves to it, using an adhesive that they produce naturally. This adhesive not only attaches the bacteria that produce it to the support material, but it also entraps other bacteria, protozoa and viruses. After a short time, the support material becomes covered with a layer of living organisms, known as a "biofilm" or "humus", which is able to remove BOD and ammonia from the sewage by biological activity.
132. The sewage passing slowly over the biofilm is also provided with oxygen, either by simple contact with the atmosphere or by bubbling air through the sewage itself. The biofilm micro-organisms remove the BOD and ammonia as a food source and, as a result, new micro-organisms are produced in a continuous process of self-renewal by the biofilm. There is also a continuous parallel process whereby the older micro-organisms of the biofilm die, become detached from the support material and are discharged from this stage of the treatment process as part of the partially treated wastewater.
133. The partially treated wastewater is then passed to a sedimentation tank where the detached/dead biofilm is allowed to settle out as sludge. This type of sedimentation tank is usually called a "humus" tank, but is sometimes referred to as a "secondary sedimentation tank" or a "final clarifier". Once the sludge consisting of the detached/dead biofilm has settled to the bottom of the humus tank, the layer of clear wastewater that remains above the sludge can now be discharged to the watercourse as treated effluent. The sludge from the bottom of the humus tank is then collected and combined with the sludge produced at the primary sedimentation stage ("the primary sludge") for further treatment. There are many types of attached growth processes, of which trickling filters and biological aerated filters ("BAF") are two examples.
134. The Suspended Growth Process. In a suspended growth process, no inert support for the micro-organisms is used. Instead, the mass of micro-organisms (known as "the biomass") and the sewage are kept in intimate contact with each other in a tank that is continuously mixed and aerated ("the aeration tank"). When mixed with the sewage in the aeration tank, the biomass undertakes exactly the same biological reactions that are carried out by the biofilm in an attached growth process. The mixture of sewage and biomass is known as the "mixed liquor". The "Mixed Liquor Suspended Solids" (the "MLSS") is the concentration of the suspended solids in the mixed liquor and is usually expressed in milligrams per litre of mixed liquor (mg/l). The MLSS is an important design and operation variable in activated sludge processes. Thus high values of MLSS concentration are normally required for nitrification to proceed effectively (E3/79).
135. After the mixed liquor has been retained long enough for satisfactory oxidation of the BOD and ammonia by the biomass to take place, it is then passed to a separate sedimentation tank, where the biomass suspended in the mixed liquor settles to the bottom of the tank as sludge. The clear treated effluent is then decanted and discharged to the watercourse. It is typically the case with a suspended growth system that some of the settled sludge is recycled to the aeration tank ("the return sludge") with the remainder being removed (a procedure known as "wasting" the sludge) and combined with the primary sludge for further treatment. All suspended growth systems that involve the recycling of sludge back to the aeration tank are known generically as "activated sludge" systems.
136. Many sewage treatment plants use an "activated sludge" system for the biological treatment of wastewater. As already indicated in what I have just said about suspended growth systems in general, all conventional activated sludge systems (e.g. the ASP at Knostrop) have the following essential features in common: (i) an aeration tank, in which the influent wastewater is treated by being mixed with the biomass and supplied with oxygen, thus enabling the bacteria in the bio-mass to oxidise the organic matter and ammonia: the mixture of wastewater and biomass being known as mixed liquor; (ii) a separate settling tank, in which the biomass ("the sludge") is allowed to settle, thus separating it from the treated wastewater, which can then be decanted for further treatment or discharge; (iii) a system for recycling some of the sludge back to the aeration tank, or otherwise retaining it within the process and (iv) a sludge handling system for the removal of excess sludge (i.e. the process known as "sludge wastage"), so that the process can continue to operate under reasonably constant conditions (i.e. with stable MLSS levels). It can therefore be seen that the sludge is a bi-product of the treatment process itself. Ultimately the amount of the sludge will depend upon the BOD and TSS (Total Suspended Solids) loads that are being treated by the plant in question. The term "sludge yield" is used to describe the amount of sludge produced during the treatment process by reference to the amount of the BOD load treated. Thus the sludge yield is the amount of sludge (in kg) produced per kilogram of BOD treated. It is an important component in the sizing of the sludge handling system, since it allows a contractor to calculate how much sludge a sludge handling system will be expected to process.
137. The period of time for which the bacteria forming the sludge is allowed to reside in a suspended growth treatment plant (as opposed to being wasted/discharged) is called the sludge age. Another way of describing what is meant by the expression "sludge age" is that it is the number of days' growth of sludge that is retained as activated sludge. If the sludge age is long, then slow-growing species of bacteria can maintain themselves within a more diverse population of resident species. It is important to note that the bacteria which can convert ammonia into nitrates (i.e. the nitrifiers) comprise mainly species which grow slowly. Accordingly, for the plant in question to achieve and maintain nitrification, a sludge age that is long enough to establish a sufficient resident population of nitrifying bacteria will be required. Conversely, a sludge age that is too short will result in an insufficiency of nitrifiers and nitrification will not be achieved or will be lost.
138. Sludge Settlement, Bulking Sludge and Selection. Since all activated sludge plants include a settlement stage, it follows that the ability of the biomass to settle under the action of gravity is crucial to the satisfactory performance of the treatment process. There are two standard laboratory tests used routinely at wastewater treatment plants to measure activated sludge settling properties and commonly referred to by their initials, namely SVI (Sludge Volume Index) and SSVI (Stirred Sludge Volume Index). Both tests involve placing a sample of mixed liquor in a graduated cylinder and measuring the volume occupied by one gram of sludge after a 30-minute settling period. The greater the volume occupied by a gram of sludge, the poorer its settleability. The results obtained are therefore quoted as numerical values, high values being indicative of poor settleability. SVI tests are simple to perform, but SSVI tests (which can be corrected or uncorrected), while more time consuming, are considered to be a better indication of the true settling properties of the sludge in question. The operational difficulties resulting from poor sludge settleability are a common cause of effluent failures in activated sludge plants. Poor sludge settlement (also known as "bulking sludge") is nearly always associated with the presence of excessive numbers of microbes/bacteria that have a diffuse, filamentous structure. Although these filamentous microbes can perform effective treatment, they do not settle as rapidly as the more desirable "flocculating" species, which are the types of bacteria that naturally combine into particles (known as "flocs") that settle readily by gravity.
139. An activated sludge plant should therefore include appropriate process design features so as to promote good sludge settlement. Broadly speaking, these design features make appropriate provision for conditions that encourage the development of the readily settling flocculating species of bacteria and inhibit and/or prevent excessive growth of the filamentous species. Such design features are known as selectors.
140. However, microthrix parvicella is a filamentous species of microbe that causes poor sludge settlement and is not readily discouraged by selection. The significance of this will become apparent later in this judgment. At this stage, it is helpful to quote from the evidence of Elga's expert witness, Mr Dirk Hielke Eikelboom ("Mr Eikelboom"), in which he gives the following succinct description of this particular and (in the context of this case) important species of microbe (see paragraphs 39-41 of his first report):

"39. Microthrix parvicella is a filamentous species which is detrimental to good settlement. The process conditions in nutrient removal plants are extremely favourable to the growth of M. parvicella and the usual selector mechanisms for promoting floc forming species are not so effective at discouraging this species of filamentous bacteria. This is because it is able to take up food in the absence of oxygen. It feeds on only one type of organic compound and can adsorb this food to the surface of its cells. Floc forming species are unable to take up this specific food if oxygen is not available.

40. The sole food for Microthrix parvicella is Long Chain Fatty Acids. These LCFAs result from sewage (usually domestic sewage) containing fats and lipids. In the absence of oxygen these fats and lipids break down (hydrolyse) into LCFAs, by which the availability to M. Parvicella increases.

41. If LCFAs are available, but oxygen is not, the Microthrix parvicella will tend to predominate over floc forming species. It is therefore an important cause of bulking sludge in nutrient removal plants. Its population size shows a marked seasonal pattern, with its maximum population at the end of the winter."

141. Selectors are usually deliberately designed regions or zones of an activated sludge system where defined high concentrations of BOD can exist under particular operating conditions, producing what is known as a high floc load when the influent sewage is introduced to that zone. Creating a high floc load is one of the main ways by which the flocculating bacteria are encouraged to predominate. This is because a high floc loading ensures the maximum amount of food is available for the micro-organisms to "feast" on. It is achieved by the introduction of a "feast" of food (i.e. consisting of the biodegradable content of the sewage) to a relatively small proportion of the contents of the aeration tank (i.e. the bacteria forming the biomass/sludge) in a selector zone. Many flocculating species are able, if food is present in excess, to take up the "food" rapidly (i.e. to "feast" on it) and to store it inside their cells, leaving it to be metabolised during the later aeration stage. In this way the "food" ceases to be available to other species (particularly filamentous ones). The flocculating species can be further encouraged by an adequate supply of air to the selector zone (to allow absorption of the available food at the maximum rate) and elsewhere in the aeration tank (to allow the food to be metabolised).
142. However, in a completely mixed tank (i.e. one where there is no physically separate "selector zone", as was the case with the Knostrop SBR, as originally designed and installed: see below) the total amount of influent and the total biomass, when mixed, will result in a lower food concentration for the microbes and, thus, a lower Food to Micro-organisms ("F/M") ratio (see paragraph 143 below). Consequently, the usual floc promoting advantages of the full-scale "feast" conditions of a separate selector zone are somewhat diminished in such circumstances. Furthermore, the combination of a lower F/M ratio and a filling stage that is almost entirely anoxic (see paragraph 147 below) means that the process conditions in a typical SBR can be very favourable to the growth of microthrix parvicella (see also paragraphs 140 above and 145 to 147 below).
143. This localised high concentration of BOD in the selector zone (the so-called "feast" condition) is usually identified by a high F/M (Food to Micro-organism) ratio. The selector zone is then followed by a region of low BOD concentration, where little food is available (the "famine" condition, i.e. a low F/M ratio). It is in this zone that the main aeration takes place. This enables the bacteria to metabolise the absorbed food. Ideally, the flocculating microbes should break down all the absorbed food before they are returned to the selector zone. The existence of feast/famine conditions in an activated sludge plant is generally considered to be essential for encouraging the development and dominance of the flocculating species of microbes, thus ensuring good sludge settlement. This process of selection is called kinetic selection.
144. However, many flocculating species of bacteria can also use nitrates in place of dissolved oxygen in order to metabolise the food that they have absorbed. This permits what is called metabolic selection to take place in anoxic conditions, i.e. conditions in which oxygen is absent, but nitrates are present. This process, during which the nutrient nitrates are converted into nitrogen gas, is known as "denitrification".
145. In general terms, anoxic conditions can promote the flocculating species of bacteria and discourage filamentous species. However, the notable and important exception to the latter part of that generalisation is microthrix parvicella which, as I have already pointed out (see paragraph 140 above) is able to take up food in the absence of oxygen. Furthermore, in the absence of oxygen, the fats and lipids that are commonly found in domestic sewage rapidly hydrolyse into Long Chain Fatty Acids ("LCFAs"), which are the sole food source for microthrix parvicella. However, flocculating species of bacteria are not able to take up LCFAs as food if oxygen is not present. Consequently, if LCFAs are available in the sewage, but oxygen is not, then microthrix parvicella will tend to predominate over the flocculating species, to the general detriment of the settleability of the bio-mass. The population size of microthrix parvicella also shows a marked seasonal pattern, with its maximum population at the end of the winter. Furthermore, it is important to note that, as indicated below, most if not all of the fill stage in a typical SBR treatment process is carried out in anoxic conditions (see paragraph 147 below).
146. SBR Treatment Systems. SBR systems are a variant of the conventional activated sludge process. In an SBR system the treatment operations are all carried out sequentially in the same tank on discrete batches of wastewater according to a timed sequence (hence the name "Sequencing Batch Reactor"). The various treatment operations in an SBR are therefore separated temporally rather than physically, the latter being the usual arrangement in a conventional activated sludge plant (see paragraph 133 above). The treatment of a continuous inflow of wastewater is only possible in an SBR system if more than one tank is installed in parallel, so that the influent can always be received by a tank that is ready to start the treatment cycle. The total amount of treated effluent that is decanted and discharged from an SBR system each day must be equal to the total amount of wastewater that is admitted into the system. Put another way, the number of decant events per day, when multiplied by the volume of each decant, must be equal to the total influent flowrate for that day.
147. Typically, the SBR treatment process divides into 5 stages, collectively making up one treatment cycle. The start of each treatment cycle is (i) the filling stage, in which wastewater is introduced into the process tank containing the total biomass. Generally speaking, the tank is filled under primarily anoxic conditions. During this period the BOD within the influent is absorbed and stored by the microbes. However, the bacteria are not able to metabolise this "food" until aerobic conditions are created within the tank by the addition of dissolved oxygen. Aeration usually (but not always) commences late in the filling stage and continues into the next stage, which is (ii) the react or treatment stage. After completion of the filling, the contents of the tank are or continue to be aerated for a given period. The process of aeration adds the dissolved oxygen ("DO") that is necessary to enable the microbial population to metabolise the organic matter that it had previously absorbed.
148. During the metabolic process the organic matter is oxidised by the microbes. Aeration continues until the required degree of treatment has been achieved. The aeration is then switched off and (iii) the settling stage is initiated. In the absence of aeration, the suspension of activated sludge solids will gradually settle, leaving above it a layer of treated effluent. As with the conventional activated sludge plant, good settlement of the sludge is crucial to the satisfactory performance of the process. The layer of treated effluent is then removed during (iv) the decanting stage. This usually involves the physical movement of mechanical equipment (known as "decanters") through the layer of treated effluent towards the top surface of the settling sludge (known as the "sludge blanket"). After the decanting stage, there is usually (v) an "idling" phase, while the tank waits to receive the next batch of influent. The whole sequence then repeats itself, with the various other parallel tanks at different stages of the treatment cycle at any given moment.
149. In the Knostrop SBR sludge handling system, as designed and built by Taylor Woodrow/Biwater/Elga, the "wasted" sludge was removed from the bottom of each SBR basin into a sludge buffer tank via sludge collection manifolds located at the bottom of each basin. From there the sludge was fed to gravity belt thickeners for thickening. After thickening, a small amount of the wasted sludge was returned to the SBR basins (the "sludge return liquors"). The vast majority of the wasted sludge was then sent to the waste handling plant in another part of the Knostrop works for final disposal.
150. It is worth noting that, from time to time and in certain circumstances, an SBR basin may commence the filling stage of its next treatment cycle before completion of the decanting stage of its last cycle. Such an event is known as a "fill-decant".
151. There are various versions of the basic SBR process available from different suppliers/manufacturers. The proprietary features of a particular SBR system are normally concerned with the details such as the filling and decanting systems and the automatic system for controlling the timed sequence of the various treatment operations. In this case, the design details of the SBR were owned by an American company called U.S. Filter Inc as a result, so I understand, of its acquisition of another American company called Jet Tech Inc. ("Jet Tech") that had been responsible for the original development of the design. So it was that the SBR with which this case is concerned was marketed as the U.S.Filter/Jet Tech "Omniflo" Sequencing Batch Reactor. As already stated, Elga is and was at all material times, a UK associate company of U.S. Filter Inc (hereafter referred to as "USF/Jet Tech").
152. The Background to the USF/JetTech "Omniflo" SBR. The "Omniflo" SBR first originated when Jet Tech moved from the production of jet aerators into SBRs in the 1980's (Transcript, Day 23/84-85). Mr Norcross was one of the founders of Jet Tech and was responsible for developing the original design. He was also responsible for the development of an internal design manual and a computerised system of design in the form of an Excel spreadsheet. According to Mr Norcross' description, this enabled the designer of a particular Jet Tech SBR plant to enter relevant information onto the standard design spreadsheet, which then proceeded to carry out a series of design calculations and checks automatically (Transcript, Day 23/87). The software to control the Omniflo SBR was originally produced by 3 different companies, of whom ICI turned out to be the most reliable (Transcript, Day 23/88).
153. None of the underlying design features of the Omniflo system to which I have referred in the last paragraph were disclosed during these proceedings, apparently because of USF/JetTech's concerns in protecting its proprietary interests in the system.
154. The New Consent Standards. The new consent standards that were required to be met at Knostrop were as follows: (i) 50 mg suspended solids per litre of effluent; (ii) 30 mg BOD per litre; 8 mg ammonia per litre; (iv) 125 mg composite COD per litre and (v) 25 mg composite BOD per litre: see paragraph 20 of the Particulars of Claim.
155. Preparations for the Knostrop Project. In April 1997, Yorkshire Water first embarked upon an investigation of the possibility of using the USF/JetTech SBR. As part of that process, Yorkshire Water carried out a tour of USF/JetTech installations in the USA under the guidance of Mr Norcross. As a result of its investigation, Yorkshire Water decided to make use of USF/JetTech technology in a number of its projects in the UK.
156. Yorkshire Water's internal authorisation for the works of improvement to proceed was given in November 1996 by means of an appropriate Investment Authorisation (C2/250) and, as already indicated, Earth-Tech was duly appointed Project Manager. The original intention was to use a GPSX standard software model to design the works of improvement, using Yorkshire Water flows, loads and other relevant parameters. In May 1996, Ms. Nicola Henderson ("Ms. Henderson") joined Earth-Tech and was appointed the lead process engineer in connection with the Knostrop project in about December 1996. Ms Henderson remained in that position until June 1999, when she was succeeded by Dr Jo Hemphrey ("Dr Hemphrey").
157. Part of the work required to establish the relevant flows and loads had already been carried out by the time Ms Henderson joined the Knostrop project. Following her appointment, part of Ms Henderson's job was to draw together and compile the information as to the relevant flows and loads, information that was to form the basis of the contract with Taylor Woodrow in due course.
158. By January 1997, Yorkshire Water had decided upon its contractual strategy for the works of improvement and this was summarised in documentary form (C3/27). Stated broadly, it was Yorkshire Water's intention to invite tenders in respect of pre-designed extensions to Yorkshire Water's existing ASP, rather than leaving the design of the relevant plant to the potential contractor on the basis of a design envelope.
159. In relation to the relevant flows and loads, Yorkshire Water had investigated the apparent discrepancies between the figures for 1994 and those for 1996 (C3/119). As a result of those investigations, Yorkshire Water decided to disregard the figures for 1996, which were characterised as atypical. On that basis, the project team (led by Ms Henderson) proceeded to prepare load information sheets, using assumed COD:BOD ratios. As I have already stated, the COD:BOD ratio is one that is widely used in the industry as a rough indication of the treatability of the sewage in question.
160. After all the load information sheets had been worked up and approved, the Yorkshire Water project team (including Earth-Tech) then carried out a Process Risk Assessment Review in February 1997 (C5/110). This consisted of an account of the progress made since the original authorisation and a review of the information and the design that had been prepared. Somewhat surprisingly, the GPSX modelling design exercise had predicted that an additional 17 ASP lanes would be required, as opposed to the 2 additional lanes that had been contemplated in the authorisation. As a result, it was decided to change the overall contractual strategy and to adopt an "open" approach to a "Red Book" contract, with the contractor being allowed to assess the possible options (C5/115).
161. The Tender Period – February to August 1997. In February 1997, as two of Yorkshire Water's preferred contractors, Taylor Woodrow and Biwater were involved in various pre-tender meetings with Yorkshire Water at which Yorkshire Water's idea of an extension to the existing ASP was discussed. However, consistent with Yorkshire Water's February decision to change its contractual approach, Taylor Woodrow was later, in May 1997, provided with the tender documents for the proposed works of improvement and left to come up with its own design for the side-stream plant on the basis of the information contained in those tender documents. In addition to being invited to tender for Yorkshire Water's preferred option of a side-stream plant to treat the High Level flows (described in the tender documents as "the Compliant Scheme"), tenderers were encouraged to produce alternative options.
162. Tenderers were provided with copies of the GPSX report, which included data that had been prepared as part of the design originally intended by Yorkshire Water. Also included in the tender documents was the information as to loads that had been produced by Ms Henderson and her team. Although the specification provided only details of the maximum flow (136 tcmd) and the dry weather flow ("DWF": 65 tcmd), Earth-Tech informed Biwater during the tender period that the minimum flow to the plant would be 0.55 DWF (35.75 tcmd) and that the average flow would be 1.25 DWF (about 85 tcmd) (C8/200).
163. It appears that tenderers were not actually provided with a comprehensive process flow diagram for the whole of the existing works at Knostrop. However, as Mr Dunn observed in his evidence (D3/5/paragraph 2.5), this was an aspect of the matter that Taylor Woodrow/Biwater found out for themselves by "going to site and inspecting the works".
164. As I have already indicated (paragraph 95 above), Mr Dunn is a consulting process engineer employed by Biwater. He is and was at all material times very experienced in the treatment of wastewater, but had no practical "hands on" experience of SBR plants prior to February 1997 – his knowledge of that type of process being, at that time, confined to seminars and the like.
165. Mr Dunn worked on the Knostrop project at the design and development stage and contributed to the design of the improvement works. It was Mr Dunn's responsibility (inter alia) to examine the requirements of the Specification, to determine the viability of the Compliant Scheme and to develop a suitable alternative scheme.
166. Mr Dunn also advised on the suitability of processes and plant and equipment developed by other members of the Biwater team. He was involved in the development of the process submission that formed part of Taylor Woodrow's tender for the works of improvement. In particular, Mr Dunn was responsible for the conceptual design of the "Alternative Scheme" that was submitted to Yorkshire Water as part of Taylor Woodrow's tender for the works of improvement and which was, in due course, constructed and installed at Knostrop. This involved deciding on the overall layout of the works and the choice of additional treatment systems to be used.
167. As I have already stated, the SBR was one of the major elements of the works of improvement. Mr Dunn did not actually design the SBR, but he did provide the necessary information to Elga for use in its design of the SBR. I accept Mr Dunn's evidence that Yorkshire Water's tender documents and addenda were the origin of the information that Mr Dunn provided to Elga for that purpose.
168. At the beginning of June 1997, Mr Dunn had a telephone conversation with David Gibson of Balfour Maunsell ("Mr David Gibson"), who was then acting as Yorkshire Water's management consultants in connection with another and unrelated project at Wath-on-Dearn. Mr David Gibson asked if Mr Dunn was going to attend a seminar that was to be held by Earth-Tech about the CASS SBR (Earth-Tech's own proprietary design for an SBR system) that Earth-Tech was building for the Wath-on-Dearn project. In the course of the conversation, Mr David Gibson asked whether an SBR was being contemplated for the Knostrop project and it was this conversation that actually prompted Mr Dunn to consider the possibility of using an SBR for the works of improvement at Knostrop.
169. Mr Dunn remembered that Jet Tech had supplied aeration equipment in connection with other work carried out by Biwater for Yorkshire Water at Knostrop in 1996. Mr Dunn was also aware that Elga (a UK associate company of USF/JetTech) designed SBRs. On 3^rd June 1997, Mr Dunn telephoned Mr Scott Davis ("Mr Davis") of USF/JetTech and made a general inquiry as to the suitability of an SBR for the Knostrop project.
170. It should be noted that Elga actually had very limited knowledge of the USF/JetTech SBR process at this stage. As Dr Cannon explained, the process knowledge relating to this particular type of SBR plant was predominantly in the USA and Elga was heavily dependent on American (i.e. USF/JetTech) expertise (Transcript, Day 21/70-72). Thus, although the USF/JetTech SBR had by then an established and successful record in the USA, neither Elga nor Biwater had any actual experience of the USF/JetTech SBR and, as Mr Streatfeild-James observed (Biwater's closing submissions, paragraph 30), USF/JetTech itself "was zealous in the protection of its proprietary product".
171. On 4^th June 1997, Mr Dunn faxed an initial inquiry (C7/50) to Mr Davis (dated 5^th June), to which he attached a hand-drawn proposed process flow diagram and a ground drawing. Mr Dunn couched his fax in general terms, but the figure for dry weather flow ("the DWF") was clearly identified. At the time, a scheme that included an SBR was only one of a number of possible options that were being considered by Taylor Woodrow/Biwater for the Alternative Scheme. In fact, at this stage, it was the Taylor Woodrow/Biwater intention to put forward an Alternative Scheme that did not include an SBR.
172. On 6^th June 1997, Mr Davis produced an outline quotation for an 8 tank SBR at a cost of $4.5m. (C7/83). In preparing the quotation, Mr Davis had taken the figure for the DWF, but had then proceeded to describe it as the "average flow". Apparently it is and was then standard practice in most parts of the USA to use the figure for the DWF as the average flow figure. This contrasts with what is regarded as good practice in the U.K., where average flow is taken to be 1.3 DWF (see paragraph 194 below).
173. Mr Dunn noticed that Mr Davis had used a BOD load of 12 tonnes per day for the purposes of his quotation, whereas the correct figure was 8t/day. Mr Dunn pointed this out to Mr Davis, who agreed to produce a revised quotation for a 6 tank SBR.
174. On 9^th June 1997, Mr Davis produced a revised quotation for a 6 tank SBR (C7/177). However, the overall physical volume of the plant remained the same, i.e. there were fewer tanks, but each tank was bigger. Again, the figure for the DWF was used to describe the average flow. However, Mr Dunn did not consider this to be significant because it was his understanding that SBRs were always designed to deal with the maximum flow, thus rendering average flow of academic significance only in his opinion.
175. Following a meeting and other discussions on 12^th June involving Yorkshire Water, Earth-Tech, Taylor Woodrow and Biwater, Mr Dunn wrote to Mr Patrick Lavelle ("Mr Lavelle") of Elga on 13^th June 1997 to the effect that Earth-Tech and Yorkshire Water were happy to go along with an SBR process for Knostrop, setting out fuller details of what was required and requesting an urgent meeting (C7/220). In this letter Mr Dunn clearly identified the DWFs and Peak flows for various different schemes, as well as stating that "…the plant must be able to maintain treatment of full flow and loads … with one unit out of service."
176. Mr Dunn's meeting with Mr Lavelle took place on about 16^th June. Mr Lavelle explained that he would have to send the relevant information to the USA for USF/JetTech to carry out the necessary design calculations. In due course, on 17^th June Mr Lavelle provided a proposal for the SBR (C7/251) in which the figure for DWF that had been provided by Mr Dunn was once again treated as the appropriate figure for average flow.
177. On 24^th June, Mr Dunn sent further information to Mr Lavelle (C7/358) for transmission to and consideration by USF/JetTech in the USA. Once more, Mr Dunn identified the DWF clearly and also pointed out that " …the plant must be able to maintain treatment to the consent level with one unit out of operation (Maintenance, Repair, Inspection etc.)". I accept that Mr Dunn added the words in parentheses because he believed that was what the Specification meant. He did not have in mind any particular period during which the plant would be required to operate with 1 unit out of operation, except that he believed that it would be a relatively short period (Transcript, Day 19/146 and 151-2).
178. On 25^th June 1997, Earth-Tech provided further information about design loads. In effect, the specified flows and loads were confirmed, apart from the ammonia load for the low-level system, which had been found to be 0.5 tonnes greater than that specified and, as such, required an appropriate revision (C7/399). As a result, Mr Dunn carried out a review of the designs and this resulted in the production of (inter alia) the Process Flow Diagram dated 16^th July 1997 (C8/133), which was included in the process submission that formed part of Taylor Woodrow's tender to Yorkshire Water.
179. As a result of various discussions that had taken place at Biwater during this period, it was decided that the Sidestream Plant proposed for the Alternative Scheme should be designed to treat a combination of appropriate proportions of the high-level and the low-level flows. It was not considered that the proposed combination of high and low-level flows would result in sewage that would present any particular difficulty in treatment and appropriate calculations were carried out to ascertain the relevant loads.
180. On 9^th July 1997, Mr Dunn faxed details of the various changes to Mr Lavelle, who acted (in effect) as a conduit between Biwater, Elga and the US designers of the SBR, USF/JetTech. Mr Dunn's fax gave brief details of both the proposed Compliant Scheme and the Alternative Scheme, together with the relevant flow and load figures for each. On this occasion, Mr Dunn's letter referred to average flows (not DWF) and included the words "NB Plant must be able to treat full flow and load with 1 unit out of operation".
181. On 22^nd July 1997, as a result of a specific request from Ms Henderson, Mr Dunn asked Mr Lavelle for further information about the USF/JetTech SBRs, together with "sequence profiles" for the operating cycles, based on figures for DWF, average flow (said to be 1.25DWF) and peak flow (C8/200). This request for information was duly sent by Mr Lavelle to Mr William Green ("Mr Green"), an applications engineer with USF/JetTech (C8/195).
182. On 25^th July 1997, Mr Green provided Mr Lavelle with the Cycle Time Distribution Charts ("CTCD"s). From these it is clear that USF/JetTech had continued to use the figure for DWF as the figure for average flow (C251- 253).
183. On 30^th July 1997, Biwater made a formal request that Elga tender for the supply of the SBR (C8/333).
184. USF/JetTech's/Elga's Design for the SBR. As I have already indicated, the SBR was designed by USF/JetTech in the USA, based on the information supplied by (in particular) Mr Dunn. However, the detailed design manual/spreadsheets were neither disclosed nor made available for use at the trial of this action, although Mr Norcross was able to give a certain amount of evidence as to how the matter progressed at USF/JetTech.
185. It appears that Mr Green was the person primarily responsible for the design of the SBR. During the summer of 1997, Mr Green asked Mr Norcross whether it was possible to omit USF/JetTech's proprietary IDSC (Influent Distribution and Sludge Collection) manifolds from the design of the Knostrop SBR in order to save costs. Mr Norcross told Mr Green that the IDSC manifold did not provide any particular advantage in an SBR system involving 6 or more tanks and that, provided that there were no fill/decants and that this could be accommodated by the software (which he believed was possible), the IDSC manifolds could be omitted from the design of the SBR (Transcript Day 23, 74-5).
186. It is apparent that, at some stage before any formal bid was submitted, the USF/JetTech designer (probably Mr Green) decided that IDSC manifolds were not required in the SBR design for Knostrop, because the submitted design made no provision for any such manifold. I am satisfied that this was a deliberate design decision, not an accidental omission.
187. The Tenders and the Formation of the Contracts. On 20^th August 1997 USF/JetTech sent its formal offer to Elga, together with a "Cost Breakdown between USF/JET TECH" (C10/103). On the following day (i.e. 21^st August), Elga made its formal offer to Biwater (C10/129) and, on 28^th August 1997, Taylor Woodrow submitted its tender to Yorkshire Water (C11/37).
188. Yorkshire Water's letter of intent was issued on 26^th November 1997, with an outstanding need to collate and agree certain terms and conditions in detail. The time taken to complete this process resulted in the main contract being signed in August 1998.
189. The main contract was a "performance" contract: see paragraph 16 of my judgment dated 2^nd May 2003, in which I considered a number of Preliminary Issues with regard to the terms of the main contract ("the Main Contract Preliminary Issues": see also paragraphs 30 to 77 above). It incorporated a Specification and Schedules that stipulated that the Plant should comply with certain stated performance requirements, to be verified by the taking of a series of four 28 day performance tests, to reflect the seasons, over a maximum period of 52 weeks as directed by the Project Manager. Accordingly, stated in broad terms, Taylor Woodrow was obliged to design, build and commission a process system that achieved the stated performance requirements in order to comply with its contractual obligations under the Main Contract.
190. As already indicated, the main contract also provided (inter alia):

(i) that the SBR should comprise a 6 basin SBR plant capable of treating 49% of the flow and load from the low-level ASP and 33% of the high-level primary tank effluent flow and load (see Clause 0.2.6.3 of the Specification: paragraph 26 above);

(ii) that the SBR should be capable of treating a maximum flow of 136 thousand cubic metres per day (tcmd) (see Clause 0.2.10 of the Specification: paragraph 26 above); and

(iii) that the Plant should be capable of treating the design flow and load to the specified effluent standard with one tank out of service (see Clause 0.2.6.3 of Section 0 Particular Specification: paragraph 26 above). So far as concerns the issue between the parties as to the proper construction of this particular provision in the Specification, see paragraphs 92 to 107 above.

191. As I have already stated (see paragraphs 7 and 8 above), the subcontract between Taylor Woodrow and Biwater was also the subject of an earlier preliminary issue ("the Sub-contract Preliminary Issue"), in which I decided that the subcontract was concluded by conduct, the relevant conduct having occurred some time after 7^th October 1999 but before February 2001.
192. The subcontract between Biwater and Elga was concluded on 31^st March 1998. As between Elga and USF/JetTech, agreement was reached in early March 1998 as to the appropriate split of margin and, on 27^th March 1997, Elga sent a letter of intent to USF/JetTech. Beyond that, it is not clear to what extent any formal agreement was actually concluded between Elga and USF/JetTech.
193. Development of the SBR Design. The detailed design of the SBR was developed during late 1997 and 1998. Fortunately, it is not necessary to deal with every aspect of that process. However, evidence in some detail was given in relation to the following two particular design matters: (i) the mass balance calculations and (ii) the method/equipment for introducing the influent wastewater into the SBR.
194. (i) Mass balance calculations. It appears that these were first produced by Biwater in late 1997 (C17/263). It was expressly noted that the "average" flowrate had been assumed to be the same as the DWF (which was specified as 65tcmd). The calculations were duly submitted to Earth-Tech in early 1998 and considered by Ms Henderson, who was not at all satisfied with them. One of the reasons for her concern was that, in the United Kingdom, it is regarded as good practice to design to an average flow that is taken to be 1.3 DWF (c.f. the position in the USA, see paragraph 172 above). Ms Henderson was therefore worried that the Biwater mass balance calculations, in which average flow had been equated with DWF, would result in an under-designed plant (i.e. as I understand it, a plant with insufficient physical capacity to carry out treatment to consent standards at all specified flows and loads).
195. On 19^th January 1998, Ms Henderson's concerns were duly passed on to Taylor Woodrow (C19/25), who agreed to carry out a recalculation. On 22^nd January 1998, Earth-Tech was provided with revised mass balance calculations, in which the average flow was equated with 1.3 DWF (C19/74), in accordance with UK best practice. As a result, when she saw the revised calculations, Ms Henderson said that she felt "comforted" (D1/73, paragraph 51).
196. The Design for introduction of the Influent. In the event, the SBR was actually designed so that the introduction of influent ("influent feed") into each tank was by means of a "dump fill" down a chute situated at one end of each tank/basin and without the use of any ID (Influent Distribution) manifold to distribute the influent evenly across the bottom of the tank.
197. Yorkshire Water attached considerable significance to the alleged shortcomings of this particular design feature. In broad terms, it was Yorkshire Water's case that the influent should have been introduced into each tank by means of an ID manifold (i.e. the USF/JetTech proprietary IDSC), so as to distribute the influent evenly across the bottom of each tank, where the activated sludge in the settled sludge blanket would be most concentrated, and with minimum disturbance of the sludge blanket, thus maximising the SBR's selector capacity: see, for example, the first report prepared by Dr Chambers, dated March 2003, at paragraph 6.3.7, which is in the following terms (E1/22):

"The selector capacity of the SBR is severely compromised by the absence of the IDSC manifold. The IDSC allows the influent wastewater to be admitted into the SBR in a uniform manner along the tank floor where the concentration of sludge will be highest. The mixing action is gentle and excessive turbulence is avoided. In my opinion, the IDSC is necessary to optimise whatever selector capacity exists during the SBR filling period. The actual wastewater inlet consists of a vertical chute in the corner of each tank. The wastewater is not brought into contact with the sludge in a uniform manner and the mixing is turbulent and localised. I cannot accept that the selector capacity of the SBR is optimised, or even predictable, as a result of this inlet arrangement."

198. Thus, it was said by Yorkshire Water that the designed omission of the ID manifolds meant that, for the reasons summarised in the previous paragraph, the SBR's selector capacity was significantly impaired, as a result of which the necessary predominance of flocculating microbes was not achieved and an excessive number of filamentous micro-organisms developed instead (in particular, microthrix parvicella), which in turn led to the poor sludge settleability that so adversely affected the performance of the SBR: see Dr Chambers' first report at paragraphs 7.82 and 7.84 (E1/34). It was also Yorkshire Water's case that the significant disturbance of the settled sludge blanket, caused by the turbulence initiated by the dump fill process, resulted in an inevitable carry-over of solids whenever a fill/decant occurred – a problem that was compounded by the proximity of the decanter arms to each influent chute.
199. However, according to Mr Norcross, USF/JetTech's main competitor had built "hundreds" of successful SBRs without using any USF/JetTech-type proprietary ID manifold and, moreover, many of USF/JetTech's own highly successful industrial SBRs did not incorporate any ID manifold. It was Mr Norcross' evidence that USF/JetTech's method of introducing the influent in that type of SBR had been to "chuck it over the side" with a "pipe" to the bottom (Transcript, Day 23/105). Furthermore, as I have already stated, Mr Norcross' design advice to Mr Green had been that an ID manifold was not required in an SBR system involving 6 or more tanks, provided that there were no fill/decants (see paragraph 185 above).
200. On 16^th April 1998, Biwater sent to Elga details of two proposals for introducing influent into the SBR tanks and sought advice as to whether these were "…acceptable for (Elga's) process requirements" (C24/157). Proposal A was for a chute (C24/159) and Proposal B was for a pipe (C24/161). In each proposal, the method of influent feed, whether by chute or by pipe, was located at a corner of the tank wall and was so designed that the whole of the incoming wastewater would be introduced at the bottom of the tank immediately below the point of entry (i.e. by means of a "dump fill" or "slug feed").
201. On 17^th April, Elga passed on Biwater's influent feed proposals to USF/JetTech for its advice and, on 21^st April, Mr Joe Harbison (who was either a Project Manager or Director of Engineering with USF/JetTech at the time) replied in the following terms:

"The influent arrangement transmitted in your April 17, 1998, facsimile is not acceptable. It is not a good idea to slug feed the influent into one corner of the basin. For best mixing, in the absence of a USF/JT influent distribution manifold, it will be necessary to bring the influent in close to the mixing pump suction bell."

202. However, on or about 1^st May 1998, a meeting was held between Biwater, Elga and Mr Norcross to consider (inter alia) the appropriate method of influent feed for the SBR (C26/1). At that meeting both Biwater's proposals for the influent feed were agreed to be acceptable. In the course of his evidence, Mr Norcross confirmed that he had approved the design that was eventually adopted (i.e. the chute design), despite the advice given by Mr Harbison in his 21^st April fax (Transcript, Day 23/109). Mr Norcross explained that the views expressed by Mr Harbison in that fax had subsequently been "overridden" because Mr Harbison had been "trying to be a process engineer", something which he was not (Transcript, Day 23/107).
203. Mr Holmes, Yorkshire Water's experienced and highly qualified Waste Water Technical Support Manager, has been employed by Yorkshire Water or its predecessors, Yorkshire Water Authority and Yorkshire River Authority, since 1973 and was appointed to his present position in 1995. As Technical Support Manager, Mr Holmes has participated in many projects and technical initiatives for Yorkshire Water. He has been involved with the works of improvement at Knostrop from the outset and he continues to be involved with the remedial works relating to it.
204. After the Main Contract had been awarded to Taylor Woodrow in November 1997, Mr Holmes attended a series of meetings with Taylor Woodrow/Biwater at which details of their design for (inter alia) the SBR were discussed. It was during this detailed design process that Mr Holmes first became aware that the SBR design omitted the USF/JetTech's proprietary IDSC manifold. It was and remains Mr Holmes' belief, based on USF/JetTech's technical literature (see paragraph 109 above), that the Omniflo proprietary IDSC was a key element of the USF/JetTech SBR design that ensured that the influent was properly introduced into the settled sludge blanket without excessive turbulence.
205. It was Mr Holmes' view that the proposed "dump fill" chute arrangement for the influent feed not only lacked appropriate provision for mixing the feed with the sludge, but was also likely to cause such disturbance of the sludge blanket that sludge solids would be decanted, particularly if a fill-decant occurred, with a consequent high risk of compliance failure (D1/87-88). Mr Holmes considered this to be all the more likely, having regard to the designed proximity of the decanters to the chute inlet.
206. Mr Holmes was well aware that Yorkshire Water was not in a position to insist upon the inclusion of an IDSC (see paragraph 114 above). However, Mr Holmes was so concerned about the omission of the IDSC that, in a meeting held in late January 1998, at which Yorkshire Water, Earth-Tech, Taylor Woodrow, Biwater and Elga were all represented, Mr Holmes raised the subject for discussion, only to be reminded that the Main Contract was a performance contract and that, since it was Taylor Woodrow/Biwater that were guaranteeing the process, it was for Taylor Woodrow/Biwater to decide how the guarantees were to be met.
207. Nevertheless, Mr Holmes asked for an assurance that fill/decants would never occur with this particular SBR whilst in normal operation. I am satisfied that Mr Lavelle then gave Mr Holmes the assurance that he sought and that Mr Holmes' handwritten note of the meeting (G5/1-2) accurately records the assurance given, as follows: "Fill-decant is a possibility – but only if flows are over & above design flow or if tanks are OOS". I accept that "OOS" stands for "out of service" and I also accept that the expression "tanks", as used in his note, was a reference by Mr Holmes to more than one tank (see Mr Holmes' evidence about this meeting and the contents of his note (D1/88, paragraph 32 and Transcript Day 3/106-108). Furthermore, Mr Holmes was also subsequently reassured when the Cycle Time Charts for the SBR were issued as part of the design information supplied to Yorkshire Water, because these appeared to show that no fill/decants would occur even when only 5 tanks/basins were in operation.
208. Similar concerns were expressed in a query that was raised by Earth-Tech in the formal approval sheets passing between Taylor Woodrow and Earth-Tech on 29^th May 1998, as follows: "(1) Is there any time that the basins would fill and decant at the same time, how many basins would need to be out of service before it occurred since the inlet is very close to the skimmer and would result in solids carry over." The query was passed on to Biwater who, in its response dated 17^th June 1998, gave a reply which, as it seems to me, was to the same general effect as the assurance that had been given to Mr Holmes by Mr Lavelle during January meeting, as follows:

"Fill-decant is not a normal operational circumstance. A combination of factors would have to exist for fill-decant to be undertaken i.e. basins out of service, with peak flow/load conditions existing for a sustained period. The decanter is several meters from the edge of the tank wall where the feed enters, and the feed enters at a low velocity several metres below the decanting arm. Fill-decant would be an exceptional circumstance and whilst short circuiting cannot wholly be ruled out it is expected that it would be minimal given the above."

209. Yorkshire Water's concern about the omission of the "Omniflo" SBR proprietary IDSC manifolds was perfectly understandable (although, in the event, misplaced), given the content and terms of the USF/JetTech technical literature on the subject. I am satisfied that Yorkshire Water's concern was made abundantly clear to Taylor Woodrow/Biwater/Elga and USF/JetTech during the design stage of the SBR. Because of its concerns, Yorkshire Water was given assurances to the effect that the only time there would be a risk of a fill/decant occurring would be if the SBR's operational circumstances were abnormal, e.g. a prolonged period of peak flow/load conditions, with two or more basins out of operation. In other words, Yorkshire Water was assured that there was no risk of any fill/decant occurring, as long as the SBR was being operated in normal circumstances, i.e. operation of the SBR up to peak flows/loads, including doing so with only 5 out of the 6 tanks actually in operation (i.e. with one tank out of operation).
210. As I have already indicated, the decision to omit any provision for an ID manifold in the design of the influent feed for the SBR was a deliberate design decision, made in the light of advice from USF/JetTech that none was required in a 6-basin SBR. However, as I have already indicated, this advice was expressed to be subject to the following two provisos: (i) that there were no fill/decant events and (ii) that this could be accommodated by the computerised control system of the SBR in question.
211. Process Commissioning and the Take Over Tests. The works of improvement started in or about December 1997. In April 1999, commissioning of the Plant commenced and, on 30^th September 1999, the various take-over tests that were required by the terms of the Main Contract (as to which, see below) were started.
212. Mr Robert Molloy ("Mr Molloy") is employed by Biwater as a Consultant Process Engineer. He has had over 26 years' experience in water and sewage design and commissioning. Although he had a certain amount to do with designing the works of improvement, Mr Molloy was primarily concerned on behalf of Biwater with the process commissioning stage (a stage which involves the three distinct phases of dry testing, wet testing and process start up: see below), during which he was given appropriate support and scientific advice whenever he required it by other specialists in Biwater, such as Dr Paul Zuber ("Dr Zuber": a principal process scientist and Biwater's Technical Services Manager) and Dr Spencer Greenhalgh ("Dr Greenhalgh": a consultant process engineer and Biwater's Engineering Director until January 1999). Mr Molloy's evidence was therefore mainly concerned with the period April to November 1999, during which the process commissioning took place and which culminated in the SBR's satisfactory completion of its take-over tests.
213. During the dry testing phase, all individual pieces of equipment are tested in the absence of any fluid to ensure that they are fully functional. This particular phase of the process commissioning at Knostrop had largely been completed before Mr Molloy became fully involved with the project.
214. The wet testing phase involves the introduction of fluid to the plant and is largely concerned with proving its hydraulic capacity and functionality. On successful completion of this phase, the plant is ready to receive the process fluids and chemicals safely and thus to proceed to process start up.
215. During the process start up phase, influent is introduced in a manner that enables the process for the plant to be started. For a sewage treatment plant which uses activated sludge, like the SBR at Knostrop, it is necessary to build up the biomass that will be used to treat the influent sewage over a period of time. The amount of the biomass is measured in mg/litre as "mixed liquor suspended solids" (commonly abbreviated to "MLSS": see also paragraph 134 above). It is important to note that the MLSS is one of the 2 parameters (the other being the BOD load) that are used to control the F/M ratio and the sludge age (the sludge age being inversely proportional to the F/M). Since it is not possible to control the BOD load in the influent to a plant, the only way of ensuring that a stable F/M ratio (or sludge age) is maintained is to control the in-basin MLSS. As Dr Chambers conceded, if a plant is operated to the designed F/M ratio, it should work (Transcript, Day 27/17).
216. At Knostrop, the build up of the biomass was started by the introduction of an appropriate amount of "seed" sludge, from which the required biomass was to be generated. Surplus activated sludge from the existing ASP at Knostrop was used to provide the "seed" sludge for the SBR. The process of "seeding" took about 8 days and was completed in mid-August 1999.
217. At the process start up of an activated sludge plant like the Knostrop SBR, a specific target value for the MLSS is set, which is derived from the process design and is a function of the nature of the specified sewage influent. The duration of the start up period is a direct function of the time required to build up the biomass to the prescribed level of MLSS and achieve full treatment. However, broadly stated, the overall stability of the plant is dependent on both achieving and maintaining the appropriate MLSS levels required for the effective treatment of the sewage in question. If the necessary MLSS levels are not achievable and/or cannot be maintained for whatever reason, the plant is regarded as unstable. Once the target MLSS has been achieved and the plant is stable and producing effluent within the specified consent standards, the process take-over tests can be commenced. The successful completion of these tests marks the end of the commissioning phase, at which point the plant is handed over to the client (D3/125, paragraph 18 - 20). In the present case, that took place on 26^th November 1999.
218. As Mr Molloy observed (D3/125, paragraph 21), the commissioning phase is the phase during which the components of the plant and the process are tested to locate and iron out problems. The failure of a plant to achieve the specified levels of effluent and/or to meet the relevant consent standards from time to time during the commissioning phase is an entirely normal part of making the plant operate properly. Mr Molloy's understanding was that the original target MLSS for the Knostrop SBR was 3,000 mg/l, based on the nature of the specified sewage influent (Transcript, Day 20/66). At one of the regular commissioning meetings held on 26^th August 1999, it was anticipated that the target MLSS would be achieved by day 38 (i.e. 38 days after the start of seeding).
219. In the event, it proved difficult to achieve 3000 mg/l MLSS within that anticipated time frame (Transcript, Day 20/67). The influent flow was significantly lower than expected and, in early October 1999, an influent sampler was installed that led Biwater to believe that the influent BOD load being received by the SBR was also significantly different from that specified (D3/131, paragraphs 51-52). In the year following take-over (i.e. November 1999 to December 2000) flows to the SBR ranged from 23.15 tcmd to 107.32 tcmd. The average flow for the same period was 59.99 tcmd, compared with the suggested 85 tcmd (see paragraph 162 above and also paragraph 87 of Yorkshire Water's Re-Re-Reamended Reply A1/tab 5/45-47). Biwater expressed their concern about these matters with both Yorkshire Water and Earth-Tech, as well as discussing the implications internally.
220. I accept that, as Mr Molloy observed (D3/131, paragraph 53), any attempt to build up the MLSS to an inappropriately high level, having regard to the flows and loads being experienced at the time, would have caused a high sludge age and an undesirable low food to biomass (F/M) ratio (see also paragraphs 133 and 142-143 above). This, in turn, could ordinarily be expected to result in poor sludge settleability (initially, mainly because of a higher than usual proportion of dead microbes in the biomass) and a consequential risk of the consent standard for suspended solids in the final effluent (in the present case, 50mg/l) being exceeded. In fact, the MLSS level that was being achieved at the time (approximately 2000 mg/l) was considered by Biwater to be more than adequate for the treatment of the influent at the flows and loads then being experienced (D3/131, paragraph 53).
221. Although Yorkshire Water was concerned about the decanting of solids (sometimes referred to as "solids carry over") during this period, its main concern was that an MLSS level of 2000 mg/l might not be sufficient to achieve full nitrification, particularly during the winter months (see paragraph 134 above). As Mr Molloy accepted, it is always necessary to achieve a satisfactory balance, so that the MLSS level is high enough to maintain nitrification during the winter, but not so high that it causes other problems such as an increased loss of solids in the decanted effluent.
222. Mr Molloy confirmed that a problem with regard to decanted solids was experienced from time to time during the process commissioning. Elga's proposed solution to the high solids content at the end of the decant phase was to reduce MLSS levels by increasing the amount of sludge being wasted, together with staggering cycle times. It is important to note that, at this stage, filamentous bacteria (the presence of which in excessive amounts would adversely affect sludge settlement: see paragraphs 138 to 140 above) were not regarded as a problem (D3/133).
223. It is common practice to commence the process of take-over testing at an early stage, even before it is clear that the plant in question will pass. As Mr Molloy explained (Transcript, Day 20/78-79), by adopting such an approach the take-over test requirements will be satisfied by the first 14 day period in which satisfactory performance is achieved (see clause 34 of the General Conditions and Schedule 6: paragraphs 24 and 25 above). This was the approach adopted at Knostrop. In the event, the satisfactory 14-day period commenced on 12^th November 1999 and the SBR passed the take-over tests 14 days later (i.e. 26^th November 1999: see paragraphs 227 and 228 below and the terms of the Take-Over certificate issued by Earth-Tech on 25^th February 2000, C1/2).
224. The daily flows and BOD loads of the influent treated by the SBR were measured and recorded during the period of the take-over testing and averaged about 51 tcmd and 4,300kg/d (i.e. 4.3 tonnes/d) respectively (D1A/21). Thus, during the period of the take-over testing, the average daily flow was below the 65 tcmd DWF suggested by Earth-Tech (see paragraph 182 above) and the average daily BOD load was below its specified average BOD of 6.89 tonnes/d (see Section 0.2.10 (iv) of the Specification, paragraph 26 above). As a result, the MLSS levels were also deliberately kept below the design figure of 3,000 mg/l, in order to maintain a satisfactory F/M ratio (i.e. to all intents, the same ratio as that produced by the equivalent design figures) (Transcript Day 20/80).
225. Earth-Tech's representative on site was Mr William Coulthard ("Mr Coulthard"). It was Mr Coulthard who formally notified Taylor Woodrow, by letter dated 30^th November 1999, that the SBR had satisfactorily completed both the Reliability Test and the Effluent Test (the former on 11^th November and the latter on 25^th November), although he also made it clear that the take-over certificate was to be heavily qualified by the exclusion of a number of items of plant/equipment and "crucial aspects" of plant operation: see C58/388-389 and D1/32).
226. In the same letter Mr Coulthard notified Taylor Woodrow that, for the purposes of satisfying the Take-Over Procedures as detailed in Schedule 6 of the main contract, the date of the Take-Over of "the Section of Plant – SBR" would be deemed to be 26^th November 1999. He also confirmed that a Take-Over Certificate would be issued in due course, in accordance with Clause 34 of the General Conditions of Contract, and that it would exclude certain specified items of plant. In due course, Earth-Tech did issue a "Take-Over Certificate for a section of Plant", dated 25^th February 2000, which was eventually sent to Taylor Woodrow in November 2000 (see paragraph 374 above).
227. For his part, Mr Holmes accepted that the SBR had passed the take-over tests. However, he was still concerned about the SBR's performance. He felt that it had struggled to pass the take-over tests and it was his view that it was not possible to conclude that the SBR was therefore capable of maintaining treatment to the required standard throughout the entire year and through all seasonal variations and different weather conditions (D1/92, paragraphs 48-49).
228. It was Mr Molloy's view (Transcript, Day 20/73-75) that passing the take-over tests had established that the plant was capable of meeting the requirements of the contract. He felt that, in some respects, a contractual take-over regime can be more demanding than performance tests because, for example, more samples are taken. Mr Molloy said that, because of this, he had never felt worried about the performance tests with regard to any plant that he had managed to get through its take-over tests (Transcript, Day 20/75).
229. In my view, the fact that the SBR passed its take-over tests is compelling evidence that it did not suffer from any fundamental or obvious defect or design deficiency and, to that extent, I accept that there is considerable force in what Mr Molloy had to say about their significance. However, in my view, the fact that the SBR did pass its take-over tests successfully does not mean, ipso facto, that the SBR was therefore capable of satisfying the more stringent performance tests. For that reason, the following more circumspect assessment of the significance of the take-over tests by Mr Holmes is equally valid:

" Therefore to suggest that, because the plant passed the effluent quality tests for a period of 14 days when operated in conditions that were unusual, it was capable of good performance in all other circumstances is not a sound conclusion. A much longer period of testing is required to be confident that a plant will perform, which is why four 28-day periods are specified by Yorkshire Water for performance testing of nitrifying plants." (D1/92, paragraph 49).

230. The Post Take-Over Period: December 1999 – January 2000. Once the SBR had passed its take-over tests, Yorkshire Water took possession of it in accordance with the terms of the main contract. For the first month of the SBR's operation (i.e. until Christmas 1999), Yorkshire Water's operatives were assisted by Mr Richard Gibson ("Mr Gibson"), Biwater's commissioning engineer. Mr Molloy also remained involved during the same period, by attending the site for short visits once a week.
231. Accordingly, as from 26^th November 1999 Yorkshire Water took possession of the SBR and became responsible for (inter alia) its operation and, in due course, the carrying out of the necessary performance tests: see Clause 34.7 of the General Conditions of Contract. However, it is Yorkshire Water's case that, at all material times, the Plant as designed, constructed and installed by Taylor Woodrow did not and could not comply with the Performance Requirements stated in the Schedules to the Special Conditions and/or the Specification. In the event, for the reasons given later in this judgment, I am satisfied that the SBR was capable of meeting the contractual performance requirements at all material times.
232. As already indicated, it is Yorkshire Water's case that, at all material times since commissioning, the SBR, as designed and installed by Taylor Woodrow/Biwater, failed to operate to the required capacity and/or to produce effluent of the required quality. In particular, it is said that:-

(i) the SBR was deficient in volumetric treatment capacity;

(ii) as designed, the SBR was incapable of treating the maximum design flow (FFT) of 136 tcmd to the required standards;

(iii) the SBR did not have sufficient capacity to treat the design loads and flows to the required standards; and

(iv) the SBR was incapable of operating to the required standards with one tank out of service.

At this stage it is worth noting that, at various points later in this judgment, I consider each of these various criticisms and give appropriate reasons for my conclusion that none of them is justified.

233. Mr William Parker ("Mr Parker") is a graduate environmental scientist with over 11 years' relevant experience in the UK water industry. From June 1999 until September 2002, Mr Parker was Yorkshire Water's Waste Water Operations Manager for the Leeds area. In that capacity, Mr Parker was responsible for (inter alia) the sewage treatment works at Knostrop. It was Mr Parker's overall responsibility to ensure that the works as a whole conformed to the consent standards set by the Environment Agency. After Take-Over, the person with the immediate responsibility for the operation of the SBR on behalf of Yorkshire Water was Ms Amanda Schofield ("Ms Schofield"), an honours graduate in Applied Biology.
234. Ms Schofield first became involved with the operation of the SBR on 8^th November 1999, when she started to "shadow" Mr Gibson during the period of take-over testing. Part of Ms Schofield's limited training for the SBR consisted of Mr Gibson's explanation of the various operations that he carried out on the SBR. These covered matters such as maintenance of the gravity belt thickeners (part of the sludge wastage system), increasing the speed of the thickener feed pumps and the cleaning and calibration of the Dissolved Oxygen probes (D1/195, paragraph 7).
235. Mr Gibson also showed Ms Schofield the way in which the various "set points" on the MMI (Man Machine Interface) could be manually altered/adjusted. The MMI is part of the computerised system for controlling the automatic operation of the SBR. As I understand it, in simple terms the "set points" are the means by which the SBR's computerised control system is provided with information and/or criteria by which operational variables (e.g. top water levels, bottom water levels, waste sludge volumes, waste delay times, anoxic multipliers, aeration times etc, etc) are set for each treatment cycle.
236. It is obvious that the MMI set points are an extremely important element in the successful operation of an SBR with a computerised control system, like the one at Knostrop. For example, the MLSS levels in such an SBR (an important factor in its overall treatment capacity) are determined by (inter alia) the amount of sludge that is wasted during each treatment cycle and that particular process is itself automatically controlled by the set point for the volume of sludge that is to be wasted in each cycle.
237. However, despite the obvious importance of the set points, Ms Schofield was given little or no explanation by Mr Gibson as to how to decide whether any (and, if so, what) change or changes to the set points were required for the proper operation of the SBR (D1/195). It may be that Mr Gibson had no prior experience of commissioning an SBR. Whether or not that was the case, I accept that he was himself heavily dependent on advice and guidance on matters such as this from, amongst others, Elga's Dr Cannon.
238. Ms Schofield herself had very little training in the operation of an SBR prior to her involvement with the one at Knostrop (apparently she had had only one half-day session with USF/JetTech : see D1/195, paragraph 9). Ms Schofield did gain some experience whilst shadowing Mr Gibson and a certain amount of training was provided "on the job", including the provision of a "training pack". However, important matters such as sampling and analysis of the plant and appropriate responses to changes in the influent flow and load were not covered in any detail (D1/196, paragraph 10).
239. Almost as soon as the SBR was taken over by Yorkshire Water (i.e. from early to mid-December 1999), there were a number of problems with its performance. These included fill/decant events, episodes of solids carry over, ammonia failure and a steady apparent loss of mixed liquors from basin 1 of the SBR.
240. Unfortunately, there is a significant lack of data relating to the operation of the SBR during the period December 1999 to January 2000. The weekly reports prepared by Ms Schofield for Mr Parker for the period 29^th November 1999 to 31 January 2000 reveal that there were many occasions when flows and loads were not measured, samples were not taken and/or microscopic analysis was not carried out and so forth.
241. Although Ms Schofield did give various plausible explanations for the manifest failure to monitor the SBR's performance adequately during this period, I agree with Mr Streatfeild-James's comment that the reasons that the failure occurred are less important than the fact that it happened (Biwater's closing submissions, paragraph 76(3)). Systematic and routine monitoring of the SBR is obviously an important aspect of its proper operation and performance.
242. Thus the routine monitoring recommended by USF/JetTech in respect of the SBR was (inter alia), to the following effect (C58/143-144):

"Operation and Monitoring of the Knostrop SBR

Overall Strategy

Initially, the plant should be operated so as to maintain a sludge age that is only slightly longer than is required for nitrification. This can be gauged by monitoring the effluent ammonia concentration. …

Control of Sludge Age

The effluent ammonia concentration should be monitored every two – three days. …

If the effluent ammonia concentration is on average less than 0.5 mg/l, the sludge age is longer that necessary, and problems with sludge settleability may result. Conversely, if the effluent ammonia concentration is rising steadily, the sludge age is too short.

The sludge age can be varied by altering the MLSS. Once the decision to alter the sludge age has been taken, the MLSS concentration should be increased or decreased as appropriate by 300 mg/l. The change should not be made in one step but should be phased in gradually at a rate of not more than 100 mg/l/day. …

Monitoring of Dissolved Oxygen Concentration

The "historical trend" facility on the SCADA should be reviewed every two weeks …

The accuracy of the DO probes should be checked each week against a portable instrument. If significant differences are found, the probes should be recalibrated.

Monitoring Biomass Condition

Particularly in the early months following Take Over, a microscopic examination of the mixed liquor should be carried out every two weeks. Similarly, the sludge SVI should be measured every two days. This can be accomplished in a graduated cylinder using the conventional method, although a better result can be obtained by measuring the sludge blanket … using a sludge judge or similar equipment. …

Routine Monitoring

The following monitoring activities are recommended as a minimum:

Every Two Days. Check SVI in a graduated cylinder or preferably by measuring the sludge blanket level via a sludge judge. Check MLSS. Test effluent ammonia concentration.

Weekly, Check calibration of DO probes against a portable probe and calibrate/clean as necessary.

Every Two Weeks Microscopic examination of sludge to check for excessive filamentous population and healthy mix of protozoa. Check the DO set point is achieved on most cycles. If not contact USF for advice on changing set points."

243. There is some confusion as to the precise identity of the document quoted in the previous paragraph. However, it does appear in Yorkshire Water's version of the SBR's Operation and Maintenance (O&M) manual (J1/205) and a copy of it was provided to Ms Schofield as part of her "training pack" (see paragraph 239 above and Transcript, Day 5/22).
244. The apparent loss of mixed liquors from basin 1 of the SBR (see paragraph 240 above) started in early December 1999 and continued until early January 2000, when basin 1 was taken off line. There is a dispute as to what caused this apparent loss of MLSS. It was Yorkshire Water's case and Ms Schofield's evidence that it was the result of an inter-basin transfer of mixed liquors, i.e. that MLSS from basin 1 had escaped into basin 2. It was suggested that this had resulted from the installation of a faulty gate-valve in the cross-connecting pipe between the two basins (the valve shutter had an inappropriate bevel on one side, i.e. the basin 2 side).
245. However, I am not persuaded that such was the case. Even if the valve shutter design was unsatisfactory and the shutter did flex very slightly as a result (which is a possibility), I am satisfied that any resulting leakage would have been far too small to account for the apparent loss of MLSS in basin 1. In any event, as Mr Streatfeild-James observed, the actual MLSS figures for basins 1 and 2 do not correlate in such a way as to support the suggested inter-basin transfer and appear to have varied a great deal from day to day (Biwater's closing submissions, paragraph 76(2)).
246. Furthermore, there was confusion as to the sludge wastage regime that was being operated at the time, because there was some misunderstanding about the appropriate sludge wastage set point. Again, the reasons for that confusion are not important, so much as the fact that the confusion existed. In my view, other than to say that this phenomenon (either then or later – see below) did not result from the unsatisfactory design of the gate-valve, it is not now possible to identify any particular reason for the apparent loss of MLSS from basin 1 during this period or when the inter-basin transfer of MLSS was suspected later in 2000 (see paragraphs 352 and 382 below).
247. According to Mr Molloy (Transcript Day 20/90-91 and D3A page 137, paragraphs 91-95), the incidents of fill-decant and solids carry over during this particular period had mainly been caused by the unsatisfactory way in which Yorkshire Water's operatives had been operating the SBR. He referred, in particular, to an incident that occurred on 2^nd December 1999, in which the decant valve in tank 2 had failed and had been left unattended, with the result that the tank had been out of operation from midnight until after 7am, when the problem was addressed and dealt with by Mr Gibson. However, whilst I readily accept that the inexperience and lack of know-how of the Yorkshire Water operatives were a factor in the operational problems experienced with the SBR poor performance during this period, in my view they provide only part of the reasons for the SBR's operational shortcomings during this period (as to which, see paragraphs 257-258 below).
248. As a result of these various problems, a meeting was convened on 12^th January 2000 to consider the SBR's performance since take-over. Various representatives from Yorkshire Water, Biwater/Elga and Earth-Tech attended the meeting, including Mr Holmes, Ms Schofield, Mr Coulthard, Dr Hemphrey, Dr Cannon, and Mr Molloy. On behalf of Yorkshire Water, Mr Holmes made it clear that he was mainly concerned about the incidence of fill-decants, the fact that the decanters appeared to be clipping the sludge blanket at the end of the decant cycle (possibly as the result of a suction effect caused by the flow through the decanter orifices) and the need for longer term proposals for improving sludge settleability (Transcript Day 3/26-27).
249. During the course of the 12^th January meeting, Mr Molloy endeavoured to obtain more detailed information about the frequency, timing and location of the main operational problems that had occurred since take-over. I accept his evidence that it soon became apparent to him that, despite the terms of the recommended USF/JetTech monitoring routine (see paragraph 243 above), very few effluent samples had been taken, recorded and analysed by Yorkshire Water's staff during the period in question, thus making it difficult to operate the plant satisfactorily and to identify the cause or causes of the various operational problems that had arisen (D3A/139 paragraph 100).
250. I also accept Mr Molloy's evidence that, in the circumstances, it was difficult to formulate a remedial action plan at that stage that went further than making the various adjustments that were recommended by Elga to reduce the possibility of fill-decants and taking steps to obtain certain further specified information: see Earth-Tech's letter of 13^th January 2000 (C60/149), which recorded the "salient points" agreed at the 12^th January meeting, as follows:

"1.0 SSVI calculations to be corrected/correlated to 2500 mg/l values.

2.0 (Elga) to supply 1No. Sludge Judge (a device for measuring the depth/level of the sludge blanket) in order that the characteristics/levels of the sludge blanket can be determined.

3.0 Velocities through the decant nozzles to be calculated at top & bottom decant water levels.

4.0 (Elga) to confirm if the decant valve closing time is included in the maximum decant time of 45 minutes …

5.0 DO setting to be reduced from 5 to 4 mg/l.

6.0 Settlement Preparation Time to be reduced from 15 minutes to 5 mins.

7.0 Sludge age to be reduced.

8.0 Settle Safety Time to be increased from 60 to 75 mins after fill/decant software modifications have been completed. …"

251. There is no doubt that, by the time of the 12^th January meeting, Yorkshire Water was extremely concerned about the various operational problems that it had experienced since its take-over of the SBR in the previous November and had made those concerns very clear at the meeting: see Mr Molloy's internal memorandum dated 13^th January 2000 (C60/176-177), in which he summarised the position as follows:

"Just a quick note to advise you of the latest situation as I see it.

1) The client (YW) believes that there is a fundamental problem with the SBR decant. When the mlss values approach 2000 mg/l there is significant solids loss at the end of a decant. He believes that if the load was increased, and hence the mlss, to the design value of 3000 mg/l the problem would be even worse.

2) The client does not readily accept that the low loads and high sludge ages are contributing to the problem.

3) The client does not believe that we have set the plant up correctly for him to continue to run it.

4) The client believes that the plant is unfit for purpose because of the fill/decant facility.

5) The client now believes that the removal of the fill decant is not a VO and the consequential (small) increase in load to the high level filters to be the best of a bad job and is now suggesting an inlet diffuser may be an alternative solution.

6) The client would like to carry out a performance test at full flow, full load, and low temp with only 5 cells operating. He is not happy that we tell him this cannot be done due to the time required to build up the load and stabilise the plant.

7) The client and Earth Tech are making noises about extending the performance test period into next year since they have not taken over all the equipment yet.

8) The latest results handed over look very bad but there is insufficient information as to when and how samples were taken.

9) There is no doubt they are struggling with the plant evidenced by the fact that No. 1 cell has less than 1000 mg/l solids in it."

252. On 31^st January 2000, two further "Plant Performance" meetings were convened to consider how matters were progressing. These two meetings had the advantage of more detailed information being available about the SBR's performance in operation. Furthermore, the more significant representation of the Elga/Jet Tech personnel at these meetings (see e.g. the list of attendees at C62/46) meant that the various discussions that took place had the important benefit of their specialised knowledge and experience (e.g. that of Mr Norcross). Thus, although it was acknowledged that the sludge blanket was being "clipped" by the decanters (thus causing excessive levels or "carry-over" of solids in the effluent), it was Jet Tech's advice that Yorkshire Water's concern about decanter nozzle velocities and the possible suction effect on the sludge blanket was misplaced.
253. At the same time as seeking to allay Yorkshire Water's concerns about the decanter nozzle velocities, Jet Tech also pointed out that the reported SVI levels were too high, thus indicating poor settlement of the sludge blanket. In order to achieve an early (though not necessarily long term or permanent) solution to this particular problem, Elga advised that the period for sludge settlement should be increased and the MLSS levels reduced to 1600 mg/l by increasing the sludge wastage rates (C62/43-45). In addition, Mr Norcross took samples of the MLSS back with him to the USA for microbiological analysis.
254. So far as concerns Yorkshire Water's operational difficulties, it was agreed that Mr Gibson (who was working for Elga by then) should resume his on site assistance and further training for the Yorkshire Water operatives was also offered. As Mr Streatfeild-James observed (Closing Submissions, paragraph 84), the position that had been reached by the early part of February 2000 was that, notwithstanding Yorkshire Water's continued serious misgivings about the design, performance and perceived shortcomings of the SBR, as expressed by Mr Holmes in particular (Transcript, Day 3/35-39 and C62/43-45), Biwater/Elga/Jet Tech were confident that the plant could be stabilised relatively quickly and steps were being taken to analyse the available information, to provide operational support and to resolve anomalies/deficiencies in the operating manuals.
255. The question of the Performance Tests was also considered at the 31^st January meetings. In early January, Yorkshire Water had indicated an intention to carry out the first of the Performance Tests, using 5 tanks, but had been told that more time was required to build up the load and to stabilise the plant (C60/177), which would require a sludge age of about 25 days (Transcript, Day 20/112). In addition, there were other reasons that meant that the proposed Performance Test (which Yorkshire Water had hoped to commence on 7^th February) would not be possible, i.e. the need to deal with the qualifications to the take-over certificate, the continuing problems with mechanical reliability and the need to install the fill-decant inhibit software (see paragraph 259 below). All these matters were discussed during the 31^st January meetings, with the result that the proposed Performance Test was postponed.
256. Summary of the SBR's Performance: from Take-over to January 2000. There is no doubt that the SBR's performance deteriorated relatively rapidly after Yorkshire Water became responsible for the operation of the SBR, following its take-over at the end of November 1999. There were a number of reasons why this occurred and, in my view, these are fairly and accurately summarised in paragraph 339 of Biwater's closing submissions, as follows:

"(1) Fill decants were occurring. When this happened there was inevitably solids carry over. This was one of Yorkshire Water's most serious concerns by 12 January (Transcript, Day 3/26-27). Those who witnessed them at the time noticed a correlation between fill decants and solids carry-over in the effluent at the end of a decant (C59/60) and a "severe and uncontrolled disturbance of settling sludge": Schofield paragraph 80 (D1/220). A significant number of the solids carry over events can thus be related to the occurrence of fill decants.

(2) Mechanical failures were common and the response to them was not always adequate, or prompt. Mechanical defects meant that tanks were taken off-line for short periods and key pieces of equipment such as decant valves remained out of action for longer than strictly necessary: see Transcript, Day 20/90. Other defects such as air locks in the recirculation pumps were also witnessed to be causing solids loss (C62/1). It is obvious that mechanical failures such as these would have hindered the performance of the SBR: indeed, Mr Holmes referred to their effect as being quite "brutal" (Transcript, Day 3/82).

(3) Basin 1 was experiencing problems (with regard to MLSS levels) for reasons which were unclear…

(4) There was inexperience in operation. Although Mr Gibson had remained on site to assist the Yorkshire Water operators, he was only in an advisory role, and it is clear that there was a lack of understanding about how the SBR operated. One particularly clear example of this was the wastage regime in basin 1 …

(5) Part of this early teething problem was that there was not yet an adequate experience of what information was required to operate the SBR … This was undoubtedly not helped by the problems with the O&M manuals.

(6) MLSS levels were allowed to rise beyond those maintained at take-over. At the meeting on 12 January 2000 Yorkshire Water w(as) advised by Mr Molloy that reducing them would assist in controlling any in-basin settleability problems. See C60/127; C60/149.

(7) As became clear from Elga/Jet Tech's involvement, there were adjustments required to the set points."

257. I agree with Mr Streatfeild-James submission that these various matters provide ample explanation for the problems that were experienced during this immediately post take-over period (Biwater's closing submissions, paragraph 340). As Mr Holmes put it in the course of his evidence, "there were a lot of interactions at the time" (Transcript, Day 3/31). In my view, the available evidence also suggests that microthrix parvicella had not yet become a major factor in any sludge settlement problem at this stage, although there were already some indication that microthrix parvicella might well cause or contribute significantly to such a problem in the future. The most compelling evidence with regard to the presence and effect of microthrix parvicella in the Knostrop SBR at this early stage is that contained in a report prepared in February 2000 by Dr David Jenkins of Berkeley University ("the Jenkins' report"), which indicated that there was sufficient microthrix parvicella present in two out of five tanks at the end of January 2000 to cause "moderate settling problems": see paragraph 263 below.
258. February to April 2000. By about the end of January 2000, Mr Holmes had become very concerned about the performance of the SBR at Knostrop (Transcript, Day 3/39). As it happens, Yorkshire Water was also experiencing problems with another Jet Tech SBR at Adwick. The Adwick SBR had been installed and commissioned before the SBR at Knostrop. It was recognised that there was a serious problem with filamentous bacteria (including microthrix parvicella) at Adwick and that this was causing bulking sludge (i.e. poor sludge settlement), with resulting poor effluent quality.
259. However, at this stage, it was not suggested by any of the parties concerned that Knostrop might also be afflicted with a microthrix parvicella problem. The contemporaneous microbiological reports prepared by Ms Schofield and a late-1999 monitoring exercise carried out by Dr Cannon had not revealed the presence of microthrix parvicella at Knostrop. So it was that the continuing problem of solids carry-over events at Knostrop was still considered to be mainly linked to fill-decant events and to the decanting process itself, albeit in the context of a generally expressed concern about the need to improve sludge settleability at both plants (C62/106).
260. During this period of the SBR's operation, "fill-decant inhibit" software was installed and commissioned as part of the computerised control system in order to deal with the problem of fill-decants. The result was that, whenever and for as long as a fill-decant was threatened, the incoming influent was diverted from the SBR to the high-level filters. As a result of this particular measure being taken, fill-decants ceased to have the solids' carry over effect that they had previously had, but at the cost of diverting the influent from the SBR and imposing an additional albeit insignificant burden (as it happens: see paragraph 381 below) on the high-level filters for the duration of each such diversion. However, I accept Mr Streatfeild-James's submission that the fill-decant/fill-decant inhibit issue is something of a red herring so far as concerns the central issue in these proceedings (i.e. whether Taylor Woodrow/the SBR were able to meet the Schedule 7 performance requirements). Fill-decant events are a feature of the flexible software that controlled the operation of the SBR. It was assumed that fill-decants would be very rare and Yorkshire Water was given assurances to that effect (see paragraphs 207-209 above). In the event, there was a problem with fill-decants that was resolved by modifying the software so as to provide for fill-decant inhibits. It is clear from Dr Cannon's evidence that the volume of flow actually diverted between September 2000 and November 2001 as the result of fill-decant inhibits was only 1% of the total flow treated over the same period (see Transcript, Day 22/63-64, 74 and exhibit "Elga 1"). I accept Dr Cannon's evidence that additional changes to the software could have significantly reduced the frequency of fill-decant inhibits even further (Transcript, Day 22/64). In the event, it was common ground that no harm was actually caused to the high-level filters or the humus tanks as the result of SBR influent being diverted as a result of an fill-decant inhibit: see, for example, Dr Chambers' evidence at Transcript, Day 28/48. It is clear that fill-decant inhibits were relatively infrequent, usually of very short duration and triggered by occasional sudden, significant changes in flow. I accept the evidence of Dr Cannon to the effect that fill-decant inhibits had an insignificant effect on the treatment capacity of the SBR. There is therefore no evidence that provides any real support for the suggestion that fill-decant inhibits would have prevented the SBR from receiving and treating influent up to the maximum of 136 tcmd, whether it was being operated on 5 tanks or 6 tanks. Consequently, I am satisfied that the ability of the SBR to meet the Schedule 7 performance requirements was not diminished in any material respect by the modification to the control system to provide for fill-decant inhibits.
261. On 4^th February 2000, Yorkshire Water held an internal meeting to consider the operational problems that had arisen with the Jet Tech SBRs at both Adwick and Knostrop. Part of the discussions at that meeting concerned the various possible methods of improving sludge settleability. The suggestions for Knostrop included the retrofitting of an IDSC and the modification of the influent chute design (C62/107). Despite Mr Streatfeild-James' submissions to the contrary (Closing Submissions, paragraph 90), I see no reason to doubt Mr Holmes' evidence that, by the time of that meeting, it was already the Yorkshire Water/Earth-Tech view that the omission of the IDSC and the influent chute design had resulted in the Knostrop SBR having inadequate selector capability: see Transcript Day 3/47-49, where Mr Holmes put the matter thus:

"A. …we at Yorkshire Water and Earth-Tech felt the IDSC should be included in the Knostrop design in the first place and there is a lot of correspondence to that effect …

Q. Was your thinking … in relation to the need for an IDSC because of the possibilities of fill/decants or was it something different?

A. Fill/decants and sludge selection.

Q. At this stage you thought IDSCs might affect sludge selection?

A. Yes. …

Q. … At this stage, can you remember what your view was in February 2000? … in relation to … (an) IDSC to improve sludge settlement?

A. My view was the chute design that had been produced by Biwater, is basically just a bulk fill … If one then considers what it should be doing is providing intimate mixing between the influent and the sludge, my view is that that design does nothing of the sort and therefore cannot be described in any form as a selector. …

A. … the conditions you need (are) a high … floc loading, which is … an instantaneous F/M ration, but you also need intimate mixing so that the microbiology is in contact with the sewage. In this case there was no intimate mixing because of the chute design."

262. However, although none of the concerned parties in the United Kingdom yet suspected the presence of microthrix parvicella at Knostrop, it is reasonably clear that by February 2000 it had started to become a factor in causing a certain amount of poor sludge settleability in the Knostrop SBR, although results of the Jenkins' report suggest it had not yet become a significant factor and, in the event, it does not appear to have definitely become such until the winter of 2001/2002 (see paragraph 428 below). The Jenkins' report was concerned with the samples that had been taken back to the USA by Mr Norcross at the end of January (see paragraph 254 above). Mr Norcross then sent the samples to Dr Jenkins for microbiological analysis. In due course, Dr Jenkins had produced a report in the form of a letter dated 5^th February 2000 (C62/159-162) in which he identified microthrix parvicella as the main filamentous micro-organism present in the samples from the Knostrop SBR and went on to observe that there was enough microthrix parvicella in two out of the five tanks to cause a "moderate settling problem" (C62/161), as follows:

"In all samples Microthrix parvicella is present as the major filamentous organism. With the exception of SBR Tank 3 the Microthrix parvicella is present both in the flocs (most of it) and free-floating in the bulk liquid. This filamentous organism can cause foaming and its presence in free-floating form may indicate foam trapping in the SBRs. There is enough Microthrix parvicella present in the SBR Tanks 5 and 6 to cause a moderate settling problem. The amount of Microthrix parvicella is less in Tanks 2 and 4 and least in Tank 3. Microthrix parvicella causes both settling and foaming problems but because of its coiled growth form that keeps it in and around the flocs, it takes a lot of it to severely impact settling rates. Any foam in the SBRs is undoubtedly caused by the Microthrix parvicella. The other filamentous organisms are all low F/M (high sludge age) types and they are not present in levels that could cause any impairment of settling.

So, like may people who employ low F/M systems that include anoxic/anaerobic zones/periods, you have "public enemy No 1" Microthrix parvicella as the filamentous organism that is causing any settling and foaming problems that may exist."

263. It would seem that the results of Dr Jenkins' microbiological analysis ("the Jenkins' report") were not seen by Mr Norcross until shortly before 7^th March 2000, when he sent an email to Elga's contract manager, Mr John Jennings ("Mr Jennings"), stating that the Jenkins' report had revealed the presence of microthrix parvicella at Knostrop and that it was likely that this was to blame for the settling problems, as follows (C64/153):

"The samples collected during my last visit (Jan. 31) were sent to Dr. D. Jenkins at Berkeley for analysis. All samples showed varying quantities of Microthrix Parvicella. This is a common filament in nutrient removal systems, though we do not often see it as problematic in our SBRs in the States. It is causing us problems at Adwick and probably at Knostrop. At some point in time we hope to understand why that is. But first, we want to establish good settling, and good consistent effluent, which requires reducing the M. Parvicella proliferation. However, Dr. Jenkins concluded that the concentration, or abundance of M. Parvicella in those samples was not enough to cause severe settling problems. I think we have gotten to that point now, and it is likely that this organism is to blame. This is an assumption that we need to confirm, so we need another set of samples analyzed (sic) and the results compared to the in-basin settling curves Amanda (Schofield) provides. For now I will assume that M. parvicella is the culprit. It is critical that we be sure of this."

264. At this stage, the information provided by the Jenkins' report remained internal to Jet Tech/Elga. However, as a result of the views expressed by Mr Norcross, Dr Cannon made arrangements for the analysis of further samples to be carried out by Hyder Consulting Limited, trading as Hyder Laboratories & Sciences (" Hyder"). Hyder's first report ("the first Hyder report") is dated 17^th March 2000 (C65/23-31). It confirmed that moderate to large numbers of filamentous bacteria had been found in the Knostrop samples. However, it did not identify the specific filamentous types involved and went on to state that the response to dosing with chlorine had been poor (C65/24, 26, 28 and 30). A copy of that report was duly sent to Biwater on 20^th March (C65/22).
265. In a subsequent report dated 29^th March 2000 (F9/14: "the second Hyder report"), Hyder stated that the filamentous bacteria that had been found were nostocoida limincola (a filamentous species which, unlike microthrix parvicella, can be controlled to some extent by anoxic metabolic selection and which does not normally give rise to settlement problems in any event: Transcript, Day 31/8). Dr Cannon was concerned about the presence of significant numbers of filamentous bacteria, as reported in the first Hyder report. However, he thought it unlikely that the filaments in question were nostocoida limincola, as stated in the second Hyder report. He therefore decided that a definitive identification was required (Transcript, Day 22/10). This involved taking further samples on 27^th April 2000, which were then sent to Mr Eikelboom for analysis.
266. In his resulting report dated 4^th May 2000 (C68/375), Mr Eikelboom identified a moderate sludge quality due to a large filamentous population, the predominant species being microthrix parvicella. As a result, Dr Cannon concluded that Hyder's earlier identification of nostocoida limincola had been incorrect and that, on balance, the filaments present were microthrix parvicella. When giving his evidence, Mr Eikelboom also expressed the view that it was likely that Hyder had mistakenly identified microthrix parvicella as nostocoida limincola in its second report (Transcript, Day 31/9). It appears that Mr Eikelboom's initial report relating to the Knostrop samples ("the initial Eikelboom report") was sent to Biwater's project manager, Mr David Rome ("Mr Rome"), about a week later (C68/359).
267. I am therefore satisfied that, from May 2000, Dr Cannon and Mr Rome were aware that significant numbers of microthrix parvicella were present in the SBR at Knostrop. However, I accept the evidence of Mr Holmes that, despite the findings in the Jenkins' report, the first and second Hyder reports and the initial Eikelboom report, the possible significance of the presence of significant numbers of filamentous bacteria (in particular, microthrix parvicella) was not something that was actually considered by any of the concerned parties at the time, i.e. during the first half of 2000 (Transcript, Day 4/130).
268. As it happens, despite the findings of the Jenkins' report and the views that he had expressed in his email to Mr Jennings dated 7^th March 2000 (see paragraph 262 above), the general lack of awareness of the possible significance of the presence of microthrix parvicella in the SBR at Knostrop was further compounded by Mr Norcross, who expressed the opinion at a process meeting held on 15^th March 2000 that he did not believe that there was a filamentous bulking problem at Knostrop, although he did go on to recommend that the MLSS should be monitored once a week (C1/84). It appears that Mr Norcross based this opinion on the results of the earlier monitoring carried out by Yorkshire Water that appeared to indicate that there was no filamentous problem (Transcript, Day 22/8).
269. So it was that, during the first half of 2000, the parties persisted with the strategy, that had been proposed by Elga/Jet Tech at the 31^st January meetings, of seeking to improve effluent quality by reducing MLSS levels and thus getting control of solids' carry-over events (the optimum target figure for the reduced MLSS levels was 1600 mg/l by this time). It was appreciated by all concerned that this strategy was maintained "at the detriment of the ammonia levels" (C1/114), i.e. that the strategy put at risk the SBR's ability to maintain nitrification (see also paragraph 134 above).
270. In the event, the SBR lost nitrification completely from 25^th January to 17^th May 2000, i.e. the SBR was consistently unable to nitrify the ammonia content of the influent to consent levels during that period: see the daily spot sample results for ammonia, as recorded in exhibit ALS1 to Ms Schofield's 3^rd witness statement dated 27^th May 2003. However, as indicated above, the strategic decision had been taken that it was important to sort out the continuing problem of solids carry-over events, before seeking to restore nitrification. In the circumstances, as Mr Holmes readily accepted in the course of his evidence (Transcript, Day 3/74), it would not be fair to attach too much significance to the loss of nitrification during this period, because the control strategy that had been adopted consciously accepted that a loss of nitrification might occur as the result of the reduced MLSS levels.
271. Achieving the desired reduced MLSS level of 1600 mg/l proved to be very difficult. Yorkshire Water and Earth-Tech formed the view that this was due to a deficiency in the sludge-wastage system (i.e. a lack of capacity to waste enough sludge). However, this proved not to be the case. In February 2000, Mr Steve Kirkwood ("Mr Kirkwood": one of Earth-Tech's process engineers and an SBR specialist) carried out an investigation into the SBR's sludge production system. This revealed (inter alia) that the daily amount of sludge being produced was about 7 tonnes per day, whereas the estimated sludge yield, as calculated by reference to the original Yorkshire Water data, was about 2.74 tonnes per day (C63/100).
272. Although it appears that there was some discussion about the possible reasons for this phenomenon (e.g. high solids content of influent and/or the concentration of the sludge being wasted), no conclusion was actually reached as to why the sludge production figures were so high: see, for example, Mr Holmes' evidence about this aspect of the matter (Transcript, Day 3/63-65). However, the fact that the sludge handling system appeared to be wasting more sludge volumetrically than it had been designed to waste was the main reason that Biwater rejected the suggestion that it lacked capacity at the monthly progress meeting (H3/248 and see also the evidence of Mr Molloy at Transcript, Day 21/13). In my view, Biwater was right to do so.
273. As already indicated, on 25^th February 2000, Earth-Tech issued the Take-Over Certificate, with certain exclusions and a list of the incomplete minor items, and sent it to Yorkshire Water, who signed it on 3^rd March 2000.
274. On 14^th March 2000, Earth-Tech wrote to Taylor Woodrow, giving notice of a defect ("the 14^th March Defect Notice") pursuant to Clause 36.1 of the General Conditions of Contract, as follows (C64/339):

"Section of Plant -SBR -Process Defect

In accordance with Clause 36.1 of the General Conditions of Contract, I hereby give notice of a Defect in the Section of Plant-SBR with respect to the Plant not achieving the final effluent quality as required under the Contract.

With reference to Clause 36.5 of the General Conditions of Contract, I would advise that any rectification works to the Plant which are deemed necessary in order for the Plant to achieve a sustainable final effluent quality as required under the Contract, should be undertaken as a matter of urgency. I therefore request details of your proposals for completing such works/modifications to the Plant for discussion at the proposed process workshop dated 22.03.2000."

275. On 17^th March 2000 Taylor Woodrow wrote and challenged the validity of Earth-Tech's 14^th March Defect Notice (C1/38-39), as follows:

"In order that your notice given pursuant to clause 35.1 (sic) complies with the requirement of sub-paragraph (b), it needs to provide particulars of the defect. We do not believe that the general statement, with respect to the plant not achieving the final effluent quality as required under the contract, is sufficient to identify the defect.

As you are aware, the performance of the plant did deteriorate after Take Over. There is insufficient factual evidence to explain the deterioration at this time, but investigations are continuing. …

In accordance with clause 36.5 we confirm that we will make good any defect to which this clause applies, within a reasonable time, but we must first identify and particularise the defect. …

As we advised in our letter to Mr R Stringer dated 15 March 2000, Taylor Woodrow, Biwater and USF/JetTech are committed to providing assistance to the Client, in order that the current problems can be identified and resolved as quickly as possible. …"

276. For its part, by March 2000 Yorkshire Water had come to the firmly held view that there were deficiencies in the SBR's sludge handling system, problems with fill-decants and the SBR's fill arrangement and a hydraulic defect in the decanters that created a suction effect on the sludge blanket. It would appear that there was at least a measure of support from Earth-Tech in respect of all three matters. Furthermore, as I have already indicated in paragraph 262 above, Yorkshire Water and Earth-Tech had also formed the view by then that the omission of the IDSC and the influent chute design had resulted in the Knostrop SBR having an inadequate selector capability. On the other hand, the Biwater team did not consider that there was any design problem with the SBR and that it was simply a question of getting the SBR balanced and operating it correctly (C64/341). I am satisfied that, broadly speaking, events proved Biwater to be correct (see below passim).
277. As Mr Streatfeild-James observed (Biwater Closing Submissions, paragraph 98), such was the background to Earth-Tech's 14^th March Defect Notice. In effect, as I have already stated, the 14^th March Defect Notice specified a single defect, namely the failure of the SBR to achieve the requisite final effluent quality (see paragraph 79 above). In turn, that related to the proposal to hold a workshop on 22 March 2000 at which all these matters could be further discussed and addressed (C64/341 and Transcript, Day 3/71).
278. The 22^nd March 2000 Workshop ("the March Workshop"). One point that emerged in the period immediately before the March Workshop was the suggestion that the SBR was undersized because the design figure for 100% flow (i.e. average flow) had been based on the figure for DWF (incorrectly so, by reference to UK best practice which treats average flow as 1.3 DWF: see paragraphs 171 and 193-194 above). This suggested reason for the possible lack of capacity was apparently first raised by Yorkshire Water at a site meeting held on 15^th March 2000 (C1/115) and appears to have resulted from a consideration of the design details by Earth-Tech's project engineer, Mr Graham Copnall ("Mr Copnall"). It seems that Mr Copnall had come to the conclusion that the physical capacity of the SBR might be 30% too small because average flow had been equated with DWF instead of being calculated as 1.3 DWF: see his exchange of emails with Ms Henderson on 27^th March (C92/7-8).
279. Given that the SBR had been designed with sufficient capacity, in purely physical terms, to deal with flows up to the specified maximum flow of 136 tcmd (see, for example, the evidence of Mr Norcross at D4/63, paragraph 48 and paragraph 6.31 of Dr Horan's first report), there appears to be nothing in the point apparently being made by Mr Copnall, so far as concerns the purely physical capacity of the SBR and that seems to have been the view taken by Ms Henderson at the time (C92/8). However, it is important to bear in mind that it does not follow from this that the SBR had sufficient capacity for effective treatment at all flows and loads. In the event, Mr Copnall also passed on the information to Dr Hemphrey, with the suggestion that, if he was right, the SBR would be 30% undersized and that changes to the settlement times would have little effect (C92/9).
280. The March Workshop (referred to as a "Value Analysis Performance Evaluation Workshop": R1/9-20) was a facilitated meeting, conducted with the assistance of a facilitator. Its objectives were to review plant performance to date, brainstorm the potential reasons for the performance shortfall and to develop a strategy for addressing short and long-term compliance (R1/12). Although a total of ten issues/problems were originally identified and listed, four of them were not dealt with in the event. The reason for this was that, as Mr Holmes explained (Transcript, Day 3/82), the workshop was concerned with design issues and the omitted issues related solely to the separate matter of mechanical equipment malfunctions.
281. The workshop consisted of a series of presentations (R1/14), starting with an Elga/Jet Tech presentation on the plant design and continuing through a discussion of all potential problems to a list of investigative tasks and the "champion" who was to be responsible for each (R1/17-19). It is worth noting that there is no reference to microbiology and/or microthrix parvicella in any of the listed actions relating to sludge settleability, although there are references to an alternative selector strategy (R1/18/SS14) and to consideration of an IDSC and a modified inlet design (R1/17-18/SS4 and SS15). It was also agreed that tank 1 should be brought back into operation (R1/19/G1), having been off line since early January (see paragraph 245 above). This was something that Biwater had been seeking for some time as a necessary measure, as Biwater saw it, in achieving the overall stabilisation of the SBR.
282. The result of the March Workshop was that it effectively set the agenda for the next few weeks, with a series of "Champions Meetings" and the completion of the various actions that had been identified at the workshop. In relation to the DWF issue (see paragraph 279 above), Dr Cannon had been asked to discuss with Jet Tech whether the plant design would have been different had the correct figure for average flow been used (i.e. 1.3 x DWF, not 1 x DWF) and to clarify the cycle times. In due course the information relating to cycle times was provided (C66/179) and the particular action was recorded as having been completed (C66/333).
283. That there might be a problem with the presence of filamentous micro-organisms in the SBR at Knostrop appears to have been raised for the first time at a site meeting held on 28^th March 2000, at which various process issues were discussed. At the meeting Elga tabled the first Hyder report of 17^th March and it was noted that:

"there now appeared to be problems of an abundance of filaments in the mixed liquors particularly in Tank No 05 …It was acknowledged that the actions taken to date may have been responsible for the development of the filaments, i.e. sacrificing anoxic fill time for aerated fill …"(C66/31-32).

284. At this stage, the species of filamentous bacteria involved was not known and was currently being investigated by Dr Cannon, the presence of microthrix parvicella not actually being established until early May (see paragraphs 266-268 above). I accept that this ongoing investigation is probably the reason why little more was said or done about the presence of filamentous microbes in the meantime.
285. On 3^rd April 2000, Earth-Tech wrote to Taylor Woodrow (C66/26-27) stating that its letter of 14^th March did fulfil the requirements of Clause 36.1(b) of the Contract. For the purposes of clarification, it was made clear that the defect in question was the deterioration in the performance of the Plant. Earth-Tech then continued as follows:

"It is also considered from the information available in the O&M Manuals that the poor performance of the Plant with 5 No tanks in service could be attributable to designing the Plant on the basis of DWF and not average flow (average flow = 1.3 DWF) and I still await confirmation from yourselves on this matter.

Further more, I …would confirm pursuant to Clause 36.1(c) that the Plant is at your disposal for (making good the defect)…

Finally, the Plant has been extensively modified and adjusted by yourselves since Take-over and continues to be modified/adjusted and mechanical failures rectified resulting in delays to the commencement of the Performance Tests. I would therefore request confirmation in due course of completion of your modifications and adjustments to the Plant in order that the Performance Tests can be progressed."

286. On 5^th April 2000, Biwater's Operations Director, Mr Roger Ingham ("Mr Ingham"), wrote an irritated letter to Earth-Tech's Managing Director, Mr Bill Clarke ("Mr Clarke") in response to Earth-Tech's 14^th March Defect Notice, in which he accused Mr Coulthard of trying to score contractual points and continued as follows (C66/83-84):

"This is in addition to his attempts to carry out a "contractual" performance test, which is one of the principal reasons why the process is in the mess that it is.

I explained at some length, to Roy Stringer, at the recent meeting with Malcolm Thomas, that it was pointless tying to run the works on five lanes (sic) to replicate the performance test requirements on (sic) the contract when the average BOD load to the works would have to be increased by some 50% to satisfy the requirements of the contract. The decision to run the works at full flow on five lanes, when Yorkshire Water were losing control of the process in January 2000, purely to try and satisfy a contractual requirement for a process test, must rank as one of the worst decisions ever made. …

Yorkshire Water are extremely keen that all parties work together to reach a solution to the problems at Knostrop. The only party who does not appear to be on board is Earth-Tech as they are still trying to score these contractual points. If we want to be contractual we could walk off this job until you provide us with the correct load at the contract flow rate and have built the correct bio-mass …"

287. Mr Clarke asked Mr William Jones ("Mr Jones"), who was then Earth-Tech's Client Services Manager and later became Project Manager for the Knostrop project, to respond to Mr Ingham's 5^th April letter. In order to do so, Mr Jones discussed the matter in some detail with both Mr Coulthard and Earth-Tech's Project Manager, Mr Roy Stringer ("Mr Stringer"), and then replied to Mr Ingham by letter dated 7^th April 2000 (C66/172-173 and Transcript, Day 8/139). I am therefore satisfied that Mr Jones' letter sets out accurately the views then current in Earth-Tech.
288. As Mr Jones explained it, his understanding was that the SBR was operating in aerated fill mode, which affected selector capacity (Transcript, Day 8/139). This is the reason why his letter refers to actions DF1 and DF2 from the March workshop, i.e. the actions that were concerned with the relevance of the use of DWF in the cycle time charts and the need to provide the cycle time charts. In essence, what was being suggested was that, because of the lack of information from Elga/Jet Tech on these points, Earth-Tech suspected that the SBR was being operated in a manner that reduced its selector capacity.
289. On 7^th April 2000, following discussions with Yorkshire Water (i.e. probably between Dr Hemphrey and Mr Holmes: Transcript, Day 3/91) Earth-Tech wrote and informed Taylor Woodrow that it had reviewed the SBR plant design and had concluded that the Plant was defective for a number of specified reasons ("the 7^th April Defect Notice": C66/175), as follows:

"1. Given that aerated fill is not recommended then the plant does not have enough capacity for treatment of design loads and flows.

2. The plant does not appear to have sufficient selector capability.

3. The plant should be operable without recourse to fill decant.

4. Nitrification is not being achieved to the standard required."

Earth-Tech then requested Taylor Woodrow's proposals for remedying the defects by 17^th April 2000.

290. In the course of his evidence (Transcript, Day 13/152), Mr Stringer confirmed that the review referred to in the previous paragraph had been carried out by the process experts at Earth-Tech (i.e. by Dr Hemphrey). As it happens, Dr Hemphrey did not give evidence and no documentation relating to any such review was ever produced or disclosed in these proceedings. However, it is clear that by the date of the 7^th April Defect Notice, the real issue had become settleability. In effect, it is that issue that is raised by items 1 and 2 of the notice. The problem of fill-decant events (item 3) had been dealt with by installation of the fill-decant inhibit software (see paragraph 258 above) and the loss of nitrification (item 4) had been the result of the strategic decision to reduce MLSS levels in order to deal with the problem of solids carry over events (see paragraphs 270-272 above).
291. So far as concerns the problem of settleability, the attention of Yorkshire Water and Earth-Tech was focussed on the use of DWF as average flow and the fact that this perceived design error might affect the SBR's selector capacity, because of the need to use more of the filling stage for the aeration stage (see paragraph 147 above), i.e. aerated fill would form a much greater proportion of the overall filling stage.
292. However, on 12^th April 2000, Taylor Woodrow wrote to Earth-Tech, refuting the various alleged defects, as follows (inter alia) (C1/43-45) :

"We were surprised and disappointed to receive this notice since … it was our understanding that we would put any contractual issues to one side while we worked together to achieve Yorkshire Water's prime objective of ensuring that the effluent is in consent by the end of June…

To answer the four points you raise, we make the following comments:-

a. We do not understand how you have concluded that an aerated fill is not recommended or that this in turn means that the plant has insufficient capacity. Information provided to you during the tender period shows clearly how the operating cycles of the proprietary Jet Tech system change at different flow rates. It is clearly evident that as the flow increases the anoxic fill time decreases and the plant moves into "aerated fill mode". This is all normal and an integral part of the Jet Tech system. …

2 … As noted above, the operating cycles of the SBR vary considerably depending on the flow rate and at the higher rates … anoxic time is traded for aerated time. However, the Jet Tech system operates on the principle that the biomass population is selected not by the very specific conditions at high flow but by the overall average conditions. There is therefore sufficient selector capability.

However, and probably of much more importance, the selector capability is being seriously compromised by the loading conditions. Selectors … all work on the principle of providing very high loadings to the system biomass. At Knostrop the loadings currently being provided and designed for are only about half that anticipated and designed for, is having the effect of seriously reducing the selector capability.

3. We can find no reference in the contract specification to the plant not being allowed to operate in a fill decant mode. Fill decant is a standard feature of the specified Jet Tech design. …

4. The plant did nitrify when it was started up and produced extremely high quality, well nitrified effluents during the take-over tests.

The loss of nitrification may be due to inhibition but is much more likely to be caused by poor operation and control of the process.

Therefore, we do not agree that the plant is defective. As we have stated above, we are working with you to try and identify why the plant is not running as expected or as it did during take-over."

293. Summary of the SBR's Performance: February to April 2000. Although some improvement was achieved in reducing incidents of solids carry-over, this was at the expense of nitrification. However, this improvement was relatively short lived. By the end of March 2000, incidents of high TSS concentrations in the effluent were increasing and, for the first time, a problem with "an abundance of filaments in the mixed liquors" was identified (C1/121-122). In my view, the overall position during this period is accurately and fairly summarised in paragraph 350 of Biwater's closing submissions, as follows:

"…microthrix may well have been affecting settleability during March and April. However, there were other factors involved as well: MLSS levels were running at between 2000 – 2300 mg/l; there were high levels of mechanical breakdowns; the SBR was still operating on 5 tanks; and further control set point changes were made (C1/121) all of which might well have contributed to settleability problems."

294. April to July 2000. As I have already indicated (see paragraphs 117 and 154 above), the new consent standards were due to come into force on 1^st July 2000. Consequently, Yorkshire Water was anxious to get the plant stabilised as soon as possible. At a Yorkshire Water/Earth-Tech/Taylor Woodrow/Biwater Performance Review Meeting on 20^th April, it was decided that contractual issues should be put on hold, that the management control of the project should be reviewed and refocused and that the Champions Group from the March workshop should be wound up (C67/96-97).
295. So it was that, on 8^th May 2000, following a further internal discussion with Earth-Tech, Yorkshire Water wrote to Taylor Woodrow in the following conciliatory and constructive terms (C68/178-179):

"Following the Performance Review Meeting on 20 April 2000, I have met with the Project Manager, Roy Stringer, in order to understand more fully the basis of his decision to issue a Defects Notice. It was clear from that discussion that the Project Manager has significant concerns regarding the design of the SBR Plant to the point that he felt it necessary to raise those concerns within the context of a Defects Notice.

Despite the potential contractual implications of this situation, I confirm that it is Yorkshire Water's view that the best way forward is to work together to achieve satisfactory and stable performance of the SBR Plant in the shortest possible time.

…

However, Yorkshire Water's key objective at this time remains to stabilise the SBR Plant ahead of the implementation of the new consent for Knostrop STW. I therefore confirm that Yorkshire Water does not intend to pursue any further action resulting from the Defects Notice during this period. It will then be possible to assess more clearly the performance of the SBR Plant, after the process has been stabilised.

I trust that you will continue to work with us to achieve a successful outcome."

296. Accordingly, in May 2000, an urgent Recovery Plan ("the 2000 Recovery Plan" - as to which, see below) was put in hand. Biwater resumed operational support of the SBR in order to recover control of the Plant, restore nitrification and improve the settleability of the sludge blanket. Effectively, Biwater's Dr Zuber took the lead in addressing the various performance issues. Broadly speaking, although there continued to be a number of problems with the operation of the SBR, nitrification was regained by the end of May and the SBR produced good quality effluent during the summer and autumn of 2000.
297. As will be seen, this improved performance was achieved when the flows to the SBR were capped and, for that reason, the success appeared to be somewhat limited in nature. It was Yorkshire Water's case that the reason for the low flows (and thus the low BOD loads) to the SBR was that flows to the SBR had been capped. However, I am not persuaded that the caps had any real impact on the flows to the SBR, nor do I accept that this was the reason for the low BOD loads. I accept Dr Horan's evidence that the fact that the flows to the SBR were capped had no real practical effect on the flows or the BOD load actually received by the SBR during the relevant period. He suggested that the flows to the SBR, and thus the BOD load, would have been much the same if there had been no cap on the flows. In my view, he was correct in that assertion. Graph 6 appended to the end of Dr Horan's first report (E3) compares the SBR inlet flow, when capped, with the flow that would have been available if there had been no cap. As it seems to me, Graph 6 clearly shows that the imposition of the flows cap had only a marginal, if any, effect on the available flows to the SBR during 2000. As Dr Horan said, in the course of his cross-examination:

"The flow cap has made no significant difference to the BOD load, Mr Slater. You have mentioned it a number of times that the reason the load is low is because of the cap. I would like to point out that the flow cap has made no significant difference to the BOD load received by the SBR." (Transcript, Day 30/62)

298. In my view, there is no doubt that the BOD load received by the SBR in the year following take-over (i.e. 2000) was consistently low when compared with that specified. The specified BOD load to the SBR was 6.89 tonnes/day (see paragraphs 26 and 224 above). This compares with the following actual average BOD loads received in the year following take-over: (i) 4.35 tonnes/day for the period end-October 1999 to 11^th May 2000, (ii) 3.7 tonnes/day between 12^th May and 13^th September 2000 and (iii) 2.115 tonnes/day between 14^th September and 13^th December 2000: see Table 4 to the second Experts' Joint Statement ("EJS": E5/49 and see also Graph 5 appended at the end of Dr Horan's first report E3).
299. As I have already indicated (see paragraph 298 above), I accept Dr Horan's evidence that the flow caps were not responsible for these low BOD loads. As he suggested, the real reason was that the influent was much weaker than that specified. I am fortified in that view by the agreed information in table 4 of the second EJS (E5/49). This shows a steady decline in the BOD influent concentrations, during the course of the year, from 75mg/l to 31mg/l. I accept Mr Streatfeild-James's submission that, if the reductions in the BOD load been the result of the imposition of flow caps, one would have expected the influent BOD concentrations to have remained static over time (load being the product of flow multiplied by concentration). Instead, as table 4 shows, the BOD influent concentrations showed a steady and significant decline as the year progressed.
300. As the winter 2000/2001 approached, the performance of the SBR deteriorated once more. Although the SBR continued to nitrify satisfactorily from June 2000 onwards, with very few failures (four in all) between then and February 2002, there were many suspended solids failures after November 2000, particularly during the winters of 2000/2001 and 2001/2002: see Graphs 1 and 2 annexed to Dr Horan's first report (E3). I will deal with the reasons for that deterioration at a later stage of this judgment (see paragraph 400 below onwards). Before doing so, it is necessary to consider the events of the summer and autumn of 2000 in greater detail.
301. The Elga/Jet Tech Design Review. Starting in April/May 2000 and continuing until July 2000, Elga and Jet Tech carried out their own internal review of the SBR design, mainly with a view to considering the Yorkshire Water/Earth-Tech main design criticisms to the effect that the plant was undersized and that IDSCs should have been provided. It is clear that, from the outset, the review was to be strictly limited to Elga/Jet Tech and that there was no immediate intention of passing on whatever views or conclusions were reached to either Biwater or Yorkshire Water: see the email dated 13^th April 200 from Mr Derek Workman ("Mr Workman"), the Head of Projects for a group of companies of which Elga was one (C66/327), who was responsible for supervising and coordinating the project managers and conducting regular project reviews in relation to a number of projects including that at Knostrop.
302. The Elga/Jet Tech design review consisted of an exchange of views between those in the United Kingdom and those in the United States and a review (C69/209-210) by a related but separate company called OTV Birwelco Ltd ("OTVB"), culminating in series of meetings in mid-July 2000 (C73/21-23, C73/89 and C73/117-118).
303. The approach adopted by the UK representatives, e.g. Dr Cannon, was to put forward and, to some extent, support the criticisms made by Yorkshire Water and Earth-Tech. The United States' representatives then commented on these various matters, agreeing to some but not to others. Unfortunately the OTVB review was carried out using inappropriate software and modelling and, in consequence, was not particularly helpful and thus largely ignored.
304. After about two months, Elga/Jet Tech held three meetings over a period of two days in mid-July 2000. The conclusion reached at this stage was (inter alia) that there was a design problem with the SBR because of the omission of ID manifolds (i.e. the IDSCs), which were considered to be necessary and required to be installed (C73/89). However, it is important to note that the absence of IDs was not perceived to give rise to a selector problem. Rather it was seen to be the reason for the problems that arose as the result of the occurrence of fill-decant events (Transcript, Day 22/28, evidence of Dr Cannon and Transcript, Day 24/138-140, evidence of Mr Norcross). In any event, Elga/Jet Tech decided to keep their conclusions to themselves for the time being.
305. It is worth noting that, despite the fact that the initial Eikelboom report had been prepared and considered some two months before (see paragraph 265 above), there was still no suggestion in the course of the Elga/Jet Tech design review that the Knostrop SBR suffered from any microthrix parvicella problem. However, as Mr Streatfeild-James observed (Biwater's closing submissions, paragraph 142), a possible explanation for this is that by the date the Elga/Jet Tech review was concluded (i.e. July 2000), the SBR appeared to be working well and was producing good quality effluent. This relatively successful period of operation during the summer of 2000 was achieved as a result of the 2000 Recovery Plan to the facts of which I now turn.
306. The 2000 Recovery Plan. The new management strategy took effect from the beginning of May 2000, with Dr Zuber effectively taking the lead in addressing the various performance issues. The minutes of the first Process Meeting of the new project management team took place on 2^nd May 2000 and recorded the following overview of the prevailing conditions (C68/31):

"Ammonia removal performance has not improved as quickly as expected. Results of ammonia removal to date are to be expressed in terms of kg of NH4/kg MLSS per day and related to average aeration times, to give a better indication of performance in terms of nitrification.

Problems with thickeners have reduced sludge wastage capacity and mixed liquors are higher than needed, resulting in decant problems.

Plant breakdowns and the resulting reduction to 3 tank running have made things considerably worse."

307. Yorkshire Water made it clear to Dr Zuber that the main priority was to recover nitrification in time for the implementation of the new consent standards in July (D3A/326). In order to achieve this, Dr Zuber had various discussions with Dr Cannon and he considered all the available operational data. He was surprised at the number of mechanical problems that still remained. As Mr Holmes agreed in the course of his evidence, the various mechanical problems with the SBR had persisted throughout and had been adversely affecting the treatment process to a significant degree (Transcript, Day 3/118-119).
308. Dr Zuber gave careful thought to formulating an appropriate strategy to stabilise the SBR and to restore nitrification. He discussed his various proposals with Yorkshire Water at the early process meetings and, on 18^th May 2000, sent a memorandum to Mr Alan Gibson, Taylor Woodrow's Contracts Manager, in which he summarised the strategy upon which he had decided (C69/58-59).
309. The two principal features of the operational strategy that Dr Zuber proposed to adopt and which he appears to have implemented from early May 2000 onwards consisted of (i) increasing aeration, by making various adjustments to the computerised control settings, and (ii) in the short term, replacing the earlier strategy of reducing the MLSS with one of increasing the sludge age, so that the nitrifying bacteria would have sufficient time to establish themselves (D3A/326 and see also paragraph 137 above). In addition, Dr Zuber took appropriate steps to ensure that a Biwater commissioning engineer was based on site full time in order to deal with the mechanical problems (D3A/326-327).
310. Dr Zuber's strategy was one of the matters discussed at a process meeting that took place on 19^th May 2000. The strategy was further developed by deciding that a full hydraulic load should not be reintroduced as soon as nitrification had been restored (although, in view of the flows actually available, this precaution was essentially an academic one: see paragraph 298 above), that the problem of the discharge of solids should then be the first matter to be considered and that, in any event, the hydraulic load should only be increased gradually after 30^th June 2000 (C69/61-63).
311. Following the meeting of 19^th May and in the light of the discussions that had taken place between the various concerned parties, Taylor Woodrow then issued a written "Knostrop Compliance Action Plan", dated 25^th May 2000 ("Recovery Plan 1"), which incorporated Dr Zuber's strategy in full. So far as material, Recovery Plan 1 provided as follows:

"1. Process Recovery Strategy

1. Due to the inability to control nitrogen removal by any other means than by biological nitrification, the priority is to re-establish nitrification. This is being achieved at the reduced flows by:

(a) Reducing flows to 5/6 of design flows.

(b) Increased aeration period by extending the maximum aeration time to 220-240 minutes (minimum 200 mins).

i. Increasing the aeration set point to 5.5mg/l. this is to be reviewed with the potential of temporarily removing the aeration set-point control.

ii. Monitoring of MLSS and sludge age. The initial approach being to not increase sludge wastage as a means of enhancing sludge age.

2. The next stage will be to stabilise and improve sludge settlement within the tanks in order to ensure solids discharge consent. The initial approach, having re-established nitrification, will be to reduce the MLSS to enhance the F:M ratio. This will be controlled and will ensure that the sludge age remains at approximately the design 28 days. The work to further define the Knostrop settlement characteristics is on-going and includes:

(a) Investigation of floc formation. This will include microbial identification, and factors potentially influencing good floc formation and maintenance.

(b) Aeration. This will include considering the aeration patterns and what is influencing the aeration demand.

(c) Effect of anoxic/aerobic fill on plant operation/settlement. The anoxic period is currently reduced to a minimal level in order to provide maximum aeration for nitrification re-establishment. With the return of nitrification the anoxic period may require extending in order to minimise the potential effect of rising sludge.

(d) Evaluation of the settle-prep period on the settleability of the sludge.

(e) Consideration why periodically the settlement deteriorates in individual tanks.

…

3. When the solids carry-over problem has been resolved and after the 30^th June 2000, the current reduced flows will be slowly increased until the full hydraulic level has been introduced. This will enable nitrification to be enhanced for full nitrogen load.

4. With the process stable at design flow, consideration will be given to increasing, gradually, the biological loading on the plant which, we consider, will enhance the plant performance. …

2. Plant Reliability

We have a supervisor and fitter based on site to address plant issues as they arise …

3. Contingency Plan

There are two contingency actions being considered for use if the effluent standards are not achieved.

1. Chemical Dosing

It has been identified from use on other plants and laboratory testing that dosing with Talc will be effective. Further tests are being progressed to assess any benefits of using ferric or alum in conjunction with Talc. Yorkshire Water are to advise the E(nvironment) A(gency) of the proposal to get their comments. A decision on if and what to order will need to be made by 2 June.

2. Divert effluent through existing works.

This was done during commissioning and initial assessments indicate that, if the flow through the SBR is limited, it may be a workable option. …"

312. In essence, as can be seen from the terms of Recovery Plan 1, Dr Zuber's strategy sensibly involved taking one step at a time. In my view, the approach adopted in the strategy was both logical and consistent in the prevailing circumstances. However, so far as concerns the contingency plans, I am satisfied that these were (understandably) only ever considered by Yorkshire Water to be short term or emergency measures, intended to protect Knostrop from compliance failure in the event of any continuing or recurring problems with the SBR. It will be necessary to return briefly to the question of chemical dosing at a later stage, because it was Elga's case that there is "no concrete and steel solution to the Microthrix problem" at Knostrop and that appropriate chemical dosing was and is the only practical and sensible solution to the sludge settlement problem that continued to affect the SBR, particularly during the winter 2001/2002 (Elga's closing submissions, paragraph 126).
313. In the event, the SBR's performance soon showed a marked improvement once Dr Zuber's strategy was implemented. By 18^th May 2000, nitrification was restored (D1A/35 onwards). By 4^th June 2000, solids compliance was being achieved on a regular and consistent basis (D1A/37 onwards).
314. It was common ground that, by early June 2000, the SBR was producing excellent quality effluent, as Mr Holmes agreed very frankly and without hesitation in the course of his evidence (Transcript, Day 3/165). The apparent success was such that, towards the end of May 2000, Yorkshire Water's current project manager, Mr Philip Sawyer ("Mr Sawyer") expressed "guarded optimism that (the Plant) will achieve stability and compliance by the end June 2000 at the reduced max flows" (C69/302-303).
315. However, Mr Sawyer's optimism was not shared by Mrs Sue Williamson ("Mrs Williamson"), who had taken over responsibility for Knostrop when she was appointed as Yorkshire Water's West Area Manager in March 2000. In an email to Mr Sawyer dated 24^th May 2000, Mrs Williamson made her position (and that of Yorkshire Water) very clear in the following forceful and uncompromising terms:

"I am afraid I don't share your optimism that the plant will comply and treat the design flows and loads. Chemical dosing should only be seen as a short term expediency it cannot be accepted as a permanent solution due to the high OPEX costs.

The end of June is the end date for the plant to be in a stable operating mode and achieving consent. As the consent applies from the 1^st July and this is a biological process which should really be given four sludge ages to stabilise my deadline is mid June and I am taking a risk there because two weeks is not sufficient time to stabilise an ASP.

I don't want anyone to be under any illusions – I will be pulling the plug on any further changes to this plant from mid June. As far as I am concerned the contractor will have failed to meet his obligations.

Knostrop will not be allowed to fail consent. We will then be in the situation of raising a new need statement to make any further changes to Knostrop works to enable it to meet its consent – this could be additional final tanks to allow the filters to take more hydraulic load. The contractual issues will have to be resolved outside this."

316. Once nitrification had been restored and the SBR was producing effluent that complied with the required consent standards, the next stage in the recovery strategy was to try and identify what was or had been causing the various operational problems with the plant.
317. On 4^th May 2000, the Biwater team (Dr Zuber, Mr Molloy, Mr Ingham and Dr Greenhalgh) met representatives of Elga/Jet Tech (Mr Workman, Mr Norcross, Mr Paul Christiansen and Mr Raza Shamskhorzan – the latter three being process engineers/scientists/experts from USF/JetTech) for the specific purpose of discussing the various alleged design deficiencies that Earth-Tech had put forward the month before (see paragraph 288 above) (D3A/329-330). The agenda for the meeting included matters such as plant sizing, aeration, sludge settleability, the control system, the recovery strategy, the further information/analyses that were required, staff training requirements and the responses to the defect notices (C68/67).
318. The general thrust of the advice given to Biwater by Mr Norcross and the USF/JetTech team at the 4^th May meeting can be summarised as follows.

(i) The use of DWF as the average flow in the design calculations (as opposed to an average flow calculated as 1.3 DWF) would have made no difference to the physical size of the SBR basins: their volumetric capacity would have been the same, although the treatment cycle times might have been altered.

(ii) It was not possible to make any firm recommendations about the control parameters for operating the SBR, but that Mr Christiansen would visit Knostrop to consider the situation.

(iii) The nature of the influent that was being treated by the SBR was significantly different from that which could have been anticipated from the information made available to the tenderers (for the basic significance of this point, see paragraphs 127 and 159 above). In particular, when the SBR's apparent lack of aeration capacity was being discussed (see paragraphs 323-324 below), Mr Norcross claimed that the ratio of COD:BOD had changed significantly, that there was a much higher proportion of readily biodegradable COD present in the sewage than could have been expected from the data provided and that this was the explanation for the apparent lack of aeration. This phenomenon of COD that did not form any part of the BOD5 (i.e. the BOD as measured by the usual 5 day test – "5-day BOD": see paragraph 125-126 above) but was nevertheless readily biodegradable, became known at the trial as "the BOD disguised as COD" (i.e. COD that appeared to be as readily biodegradable as BOD, but which (contrary to normal expectation) had not been measured as and included in the figure for the 5-day BOD). In an internal memorandum dated 11^th May 2000 (C68/343-348), Mr Norcross summarised his views on this aspect of the matter, as follows:

"More important, the COD/BOD ratio was 3.4 compared to the 2.0 on which our design was based. Effluent COD values indicate 30 to 40mg/l as soluble COD; this means the "extra" COD is readily biodegradable. Therefore, the SBR is removing much more organic load (70% more) than was expected. And in most cycles, each tank still ends aeration because the D(issolved) O(xygen) setpoint is reached. And this is occurring in spite of dramatically reduced aeration times due to the poor settling rates; and in spite of low airflow rate …"

319. Following the meeting of 4^th May, Elga/Jet Tech provided further information from which Dr Zuber was able to carry out his own review of the design. As a result, he satisfied himself that the SBR did have a sufficient volumetric capacity and that it had a nitrification safety factor of 250% on 6 tanks. However, if only 5 tanks were operational, Dr Zuber's calculations suggested that, although the SBR would nitrify and produce effluent to the required standards for a short period of time, the SBR's ability to nitrify could not be sustained indefinitely at full flows and loads and that there "might well be a sudden drop in performance of the Plant after 2-3 weeks of continuous 5 tank operation" (D3A/331, paragraph 5.22). As will be seen, the results of Dr Zuber's calculations were consistent with the opinions expressed later by Dr Horan on this aspect of the matter (see paragraph 527 below).
320. Dr Greenhalgh had also carried out similar checks to those of Dr Zuber and, as a result, he was satisfied that the Elga figures for oxygen demand looked reasonable (D3A/82). Thus, by late May/early June 2000, having regard to what it had been told by Elga/Jet Tech and the results of the calculations that had been carried out by its own experts, it seemed to Biwater that the various operational problems that had been experienced with the SBR had not been due to any deficiencies in its design. In my opinion, that view was not unreasonable in all the circumstances and events proved it to be right.
321. The "BOD Disguised as COD" and the COD:BOD Ratio. The second stage of Recovery Plan 1 had envisaged defining the settlement characteristics of the sludge by considering (amongst other things) aeration, aeration patterns and what was influencing aeration demand (C69/307). As part of that process, during May 2000, Biwater identified that the SBR was not producing a residual Dissolved Oxygen (DO) at the end of most react stages (see paragraphs 147 to 148 above), thus indicating that all the dissolved oxygen had been used up during the react stage. This, in turn, was indicative of a lack of aeration in the react phase of the overall treatment cycle.
322. That there appeared to be anomalous DO results and that this called for some satisfactory explanation had been appreciated at the 4^th May meeting (see paragraph 319(iii) above). The subject was also mentioned in the overview of the current position at Knostrop, as recorded in the minutes of the process meeting held on 26^th May 2000 (C69/368), which was in the following terms:

"All tanks now appear to be nitrifying, which is extremely good news.

It is now very important that we get the solids down quickly. Commencing on Tuesday 30^th May, therefore, sludge wasting will be increased, with the objective of getting MLSS down to approx. 1800-2000 within 2 weeks. In order to achieve this, both belt thickeners will be run together and we can also waste to the existing filters if necessary, to reduce the overall period of wasting solids.

Overall, the results are fairly encouraging and all the signs are good, apart from some confusing DO figures, which are being investigated, but at the moment defy any logical explanation" (emphasis supplied).

323. At some stage in early May 2000, Dr Zuber asked Dr Greenhalgh to investigate the apparent problem of aeration deficiency in the SBR, whilst making an observation to following effect: "Something strange is happening here. The system is running out of oxygen and yet apparently is very underloaded. Can you try and find out what is happening (D3A/81 and Transcript, Day 16/142)."
324. On 25^th May (i.e. the day before the process meeting of 26^th May: see paragraph 323 above), Dr Greenhalgh gave instructions for tests to be carried out to determine the oxygen uptake and dissolved oxygen levels in SBR tanks 3 and 6. The tests involved removing samples from the tanks and measuring the rate of oxygen uptake and monitoring the dissolved oxygen level during a period of one day (D3A/81-82). Following the tests and on the same day (i.e. 25^th May), Dr Greenhalgh produced some initial handwritten notes in which he recorded the results of the tests. Unfortunately, only one page of those notes (relating to tank 6) has survived.
325. However, on 29^th May 2000 and following the process meeting on 26^th May, Dr Greenhalgh produced another set of handwritten calculations (C69/328/329) in which he again recorded the results of the 25^th May tests. Based on the results of the tests, Dr Greenhalgh's calculations suggested that something was present in the SBR influent that had not been measured as part of the normal 5-day BOD but which was, nevertheless, exerting a 25% to 30% additional oxygen demand and apparently indicating inadequate aeration (Transcript, Day 16/137). A meeting was therefore arranged for 23^rd June, so that the possible aeration inadequacy of the SBR could be specifically discussed by all concerned parties (i.e. Yorkshire Water, Earth-Tech, Taylor Woodrow and Biwater).
326. Prior to the 23^rd June meeting, Biwater also held an internal meeting to discuss the SBR on 7^th June 2000 (C70/184). Dr Greenhalgh explained that it was necessary to check the sludge guarantee figures and relate them to the COD, because sludge production was higher than would have been expected. As I have already indicated, the apparently high sludge yield had been a matter raised and discussed previously, but without reaching any conclusions as to the reason for it (see paragraphs 272-273 above).
327. At the meeting on 23^rd June (C71/119-121), it was common ground that the figures for DO appeared to be anomalous because, with the low BOD loads that were being experienced at the time, there should have been no oxygen deficiency, although it was Mr Holmes' view that it was later established that the low DO had been caused by a short term shortening of the react stage (Transcript Day 3/138). However, it was common ground that the low DO profiles (and the associated "BOD disguised as COD" phenomenon) were a temporary feature that did not persist after early June 2000 (Transcript, Day 16/148). It was also common ground that the amounts of sludge that were being produced were very high and this particular aspect of the matter did persist. The issue was as to why this was the case.
328. Dr Greenhalgh had prepared some further handwritten calculations relating to oxygen demand that he distributed to everybody at the meeting (D3A/84). A typed version of those notes appears in the court bundle at C72/71. Dr Greenhalgh's calculations were based on two separate equations that will give identical results if the COD:BOD ratio is normal or, in the case of a high COD:BOD ratio, where the COD that is inflating the normal ratio is not degradable and passes out with the treated effluent (Transcript, Day 16/138).
329. In the course of his evidence, Dr Greenhalgh gave a detailed and careful explanation of his calculations and underlying reasoning which is, in my view, fairly and accurately summarised in paragraph 156 of Biwater's Closing Submissions, as follows:

"(1) The Boon and Chambers equation uses BOD to calculate oxygen demand.

(2) There is an alternative equation, the Ekama and Marais equation which uses COD to calculate the oxygen demand.

(3) Dr Greenhalgh used these 2 equations to produce calculations of oxygen demand. The oxygen demand based on the COD method of calculation (Ekama and Marais) was about 30% greater than on the BOD (Boon and Chambers) methodology, because so little of the COD was going out in the effluent.

(4) Normally the 2 methodologies would be expected to give the same result.

(5) It followed that there was an oxygen demand being exerted by something which was not being picked up by the standard 5 day BOD test."

330. At the same time, there was the phenomenon of the high sludge yields. This established fact, taken together with the fact that an unexpectedly high COD was apparently being treated and removed from the influent, led Biwater to suggest at the time (not unreasonably, in my view) that there was a large proportion of readily biodegradable COD in the influent that was not picked up by the 5-day BOD test (i.e. that there was a higher than measurable load in the influent to the SBR) and that it was being degraded, thus exerting an oxygen demand and creating or contributing to excessive sludge. In effect, this was the same point as that made by Mr Norcross during the 4^th May meeting (see paragraph 319(iii) above) and, in my view, is probably the correct explanation for this temporary phenomenon. Moreover (and more importantly), it was also Biwater's view that, rather than an anticipated COD:BOD ratio in line with the normal 2:1, the actual COD:BOD ratio of the influent was nearer to 4:1 (C71/120): see also paragraph 0.2.2(b) of the Specification, paragraph 26 (v) above.
331. On 21^st June 2000, Dr Zuber summarised Biwater's overall opinion on the matter in a succinct paper entitled "Knostrop STW – COD Loading" (C71/73). In that paper, he expressed the essential nature of Biwater's views in the following terms:

"The Knostrop plant was designed based on a BOD load … of 8,560 kg/d, and an assumption made of a COD:BOD ratio of 2:1. The oxygen requirement was determined using the BOD values and the following assumptions: 1.2 kg O2 required/kg BOD removed and 4.6 kg O2 required/kg nitrogen oxidised …

Analysis of the Knostrop feed and treated effluent has revealed a significant difference to these assumptions, which will have a dramatic effect on oxygen demand and ultimately sludge production (and potentially sludge settleability). …

Collectively these results demonstrate that 1) the COD:BOD ratio is greater than 4:1 and 2) that the COD is relatively degradable. As such the COD is exerting an oxygen demand and producing a sludge yield to a greater extent than is assumed in the traditional Boon and Chambers method …

This value, contrary to the low BOD loading evaluation, suggests the plant is at or close to the design loading in terms of total degradable organic material. If the BOD load were ever realised while maintaining the COD:BOD ratio there would be insufficient oxygen to sustain nitrification. At this stage it is unclear as to why the COD ratio is relatively high compared to the BOD load … and yet the COD is degradable.

… As previously commented it is believed that this organic load expressed as high CODs, low BODs is having a selective effect on sludge characteristics which may reduce overall sludge settleability."

332. The COD:BOD ratio was not expressly specified as an influent parameter in the contract. However, it was calculable from the data provided in the Specification ("the specified COD:BOD ratio") and, as Mr Holmes and Ms Henderson readily accepted, was a ratio that is normally used in the wastewater industry for assessing the nature and treatability of the sewage in question (see paragraph 127 above). During the summer of 2000, Yorkshire Water and Earth-Tech carried out their own internal investigation into what the specified COD:BOD ratio was. Mr Copnall carried out some calculations, using the GPSX report that suggested a ratio of 3.1:1 (C71/262). At a meeting held on 4^th July 2000, it was said that the ratio derived from the contractual documentation was 2.85:1, whereas the actual ratio appears to have been calculated to be 3.5:1 (G5/123).
333. However, at trial there was a dispute as to what the specified COD:BOD ratio was. It was Biwater's position that the data given in the Specification showed a COD:BOD ratio of 2.74:1 for the SBR influent. It is worth noting that Mr Arthur Boon arrived at the same ratio in his report in September 2000 (C77/149: as to the Boon report, see paragraph 338 below). For its part, Yorkshire Water maintained that the ratio of 2.74:1 related to settled sewage only and that it was not the appropriate ratio for the influent to the SBR. According to Yorkshire Water, the COD:BOD ratio of the SBR influent was nearer to 5:1, once proper account had been taken of the contribution of the ASP effluent (49% of which was to form part of the SBR influent: see paragraph 0.2.6.3 of the Specification, paragraph 26 above).
334. In my view, the correct figure for the specified COD:BOD ratio was that put forward by Biwater (i.e. 2.74:1). I accept the evidence of Dr Zuber and Dr Greenhalgh (Transcript, Day 14/55-61, Day 16/77-80 and 139-141) that the ratio of 5:1 asserted by Yorkshire Water (as calculated mainly by Dr Chambers and by Taylor Woodrow's expert, Dr David Johnstone, "Dr Johnstone") resulted from a process of calculation that was fundamentally flawed. Dr Zuber and Dr Greenhalgh both accepted that it was important to take account of the loads in the ASP effluent, when calculating the appropriate COD:BOD ratio of the SBR influent, but both rejected (correctly, in my view) the suggestion that the effect of doing so would be to raise its COD:BOD ratio to 5:1. As both Dr Zuber and Dr Greenhalgh pointed out, the Yorkshire Water calculation involved the mathematically impermissible exercise of averaging the various ratios, the fallacious nature of which was graphically illustrated by Dr Greenhalgh in the course of his evidence (Transcript, Day 16/140).
335. There is no doubt that, as claimed by Biwater, the actual COD:BOD ratio of the SBR influent was not 2.74:1. The influent data for the SBR for the period 2000 to 2003, as collected and summarised in exhibit TW5, gives a mean average COD:BOD ratio of 4.3:1 and a maximum value of 17.3:1. In the course of cross-examination, Dr Chambers accepted (albeit with some reluctance) that there was a difference (meaning a significant difference) between a COD:BOD ratio of 2.74:1 and one of 4.3:1. As it seems to me, it is clear that the COD:BOD ratio of the SBR influent was significantly and unexpectedly different to that which could have been ascertained from the influent data in the contract (in particular, from the tender information and the Specification: i.e. it was a much higher COD:BOD ratio). It follows that the influent to the SBR was much more difficult to treat than could reasonably have been expected from the specified data.
336. Due to the non-availability of Dr Hemphrey, one of Earth-Tech's other process engineers, Mr Peter Davies ("Mr Davies"), became involved in the internal investigation during this period. Following the meeting of 4^th July, Mr Davies produced a list of actions that included consideration of the contractual position and identifying the need to check the further calculations that Dr Greenhalgh was to provide following the 23^rd June meeting (C92/17). To a limited extent, Mr Davies also carried out his own review of the SBR design as a result of which he stated that if he had designed the plant, the volume would have been similar to that produced by Elga/Jet Tech, but that he differed as to oxygen demand (C92/20). On 10^th July 2000, Mr Davies wrote a memorandum, probably at Dr Hemphrey's request, in which he broadly agreed with Biwater's calculations on oxygenation, but disagreeing with the amount for a denitrification credit. So far as concerns the COD:BOD issue, Mr Davies made the following observation (inter alia):

"It is well documented that, even for a pure domestic waste, an incoming COD:BOD ratio of 2:1 will increase as treatment progresses and that the corresponding ratio after, say, activated sludge will be in the region of 8:1. I would not therefore be surprised that a flow blend of 48% ASP effluent and 52% settled influent should have a ratio close to 4:1. …"

337. The Total Suspended Solids ("TSS") loads. Another matter for serious concern with regard to the actual influent received by the SBR, compared with what could have been expected from the data provided by Yorkshire Water at the tender stage, relates to the TSS loads in the influent. There was no express specification of the influent TSS concentrations in the main contract. This is not unusual. I accept Dr Horan's evidence that a TSS concentration is only specified for sewage plant designers where a Purchaser wishes to draw the designer's attention to the fact that the TSS concentration is outside the range normally to be expected in influent sewage (see Dr Horan's 2^nd supplementary report at paragraph 4.2: E3A/tab2/15). In other words, since there was no expressly specified TSS concentration in the main contract, the designers of the SBR were entitled to expect that the TSS loads in the SBR influent would be within normal limits, that is to say within the normally expected range of plus or minus 25% of the of the specified influent BOD concentrations (paragraph 4.2: E3A/tab2/15).
338. Mr Eikelboom's evidence was to a like effect. On the basis of the specified BOD concentrations (see paragraph 26 above), Mr Eikelboom would have expected the SBR influent to have a TSS concentration of about 75mg/l to 100mg/l (see paragraph 54 of Mr Eikelboom's first report: E4/17), which translates into TSS loads of between 6.375 tonnes/day and 8.5 tonnes/day at average flows. Given that the experts were agreed that the BOD concentration of the SBR influent during the relevant period was 81mg/l, Mr Eikelboom's figures fall within the band identified by Dr Horan. I am thus satisfied that the weight of expert evidence supports Dr Horan's views (see also Biwater's closing submissions, paragraphs 287-288). Accordingly, in the present case, the normal expectation with regard to the TSS loads in the SBR influent would have been plus or minus 25% of the specified BOD loads.
339. However, as is apparent from Graph 5 appended to Dr Horan's first report (E3), the TSS in the influent to the SBR did not bear the normally expected relationship to the influent BOD loads. Moreover, I agree with Mr Streatfeild-James' submission (Biwater's closing submissions, paragraph 289) that an analysis of Graph 5 demonstrates the following: (i) from February 2000 onwards, influent TSS levels were highly erratic, (ii) between May and August 2000, the SBR was regularly receiving TSS loads in excess of 10 tonnes per day, with loads in excess of 20 tonnes per day being recorded in July and August 2000, (iii) TSS loads of between 10 tonnes and 20 tonnes per day were still occurring in September and October 2000, despite the BOD load having fallen to about 2 tonnes per day, and (iv) the high influent TSS loads were not seasonal, but occurred randomly throughout the year.
340. It was Yorkshire Water's case that the cause of the high TSS levels lay at Taylor Woodrow/Biwater's own door, having probably originated in the recycle stream from the SBR sludge treatment system: see paragraph 3.2.5 of Dr Chambers' first supplemental report (E1A/tab 1/5). In paragraph 3.1.6 of his third supplemental report, Dr Chambers explained the basis for that conclusion in the following terms (E1C/tab 3/3):

"I do not consider that the composition of either the H(igh)L(evel) settled sewage or the ASP effluent can account for the high TSS concentrations and my opinion on their origin is based on elimination of these two sources, which leaves only the sludge treatment recycle stream."

341. All three experts (i.e. Dr Chambers, Dr Horan and Mr Eikelboom) accepted that there were only three possible routes of entry for the high TSS loads received by the SBR, namely (i) through the high level works, (ii) from the low level ASP effluent and (iii) from the internal sludge return liquors. However all the experts were agreed that the possibility that the high TSS loads originated with the ASP effluent could be discounted altogether. Unlike Dr Chambers, Dr Horan and Mr Eikelboom also discounted the internal sludge return liquors. They both considered it likely that the high TSS loads were being added to the high level stream either above or below the primary tanks from other parts of the Knostrop works, before becoming part of the SBR influent (Transcript, Day 30/28 and Transcript, Day 31/61).
342. I am satisfied that Dr Horan and Mr Eikelboom are correct in their view as to the source of the high TSS loads. Despite Dr Chambers' insistence to the contrary, it is clear from the mass balance calculations set out in pages 4 to 6 of exhibit TW6 that it is simply impossible that solids' loads of the magnitude recorded in the SBR influent could have originated from the sludge return liquors. I also accept the force of the various other points made in paragraph 599(2) of Biwater's closing submissions. In any event, as it seems to me, the proposition that the sludge return liquors were not the source of the high TSS loads is self-evidently true. As Dr Horan said in evidence (Transcript, Day 30/27): "there is not 10 tonnes of solids leaving the G(ravity) B(elt) T(hickeners) even on the belt, so we know they are not from the GBTs". Mr Eikelboom put it even more succinctly when he said (Transcript, Day 31/60): "Your total mass is simply too much."
343. In my view, the evidence clearly shows that the high TSS loads did not originate from the internal return sludge liquors, as suggested by Dr Chambers. By process of elimination, the source must have been somewhere in the high level stream, although it is not possible to say precisely where. However, the fact that the precise source of the high TSS loads cannot be identified does not matter. For the purposes of this case, it is sufficient to find, as I do, that the unexpectedly high TSS loads were not caused by any aspect of the SBR's design or of its operation or system of control.
344. In my judgment, the scale of the SBR influent TSS loads during this period (i.e. 2000/2001) was such that Dr Horan was entirely justified in his characterisation of them being "absolutely unheard of" and (in the case of the daily loads of about 26 tonnes in July 2000) "an unprecedentedly large number" (Transcript, Day 30/25). I am therefore satisfied that the following point by Mr Streatfeild-James is well made (see Biwater's closing submissions, paragraph 291):

"The fact that TSS levels when looked at on their own are high is obviously important, but what is of equal importance in terms of plant performance is that the influent TSS:BOD ratio does not follow normally accepted levels: see Transcript, Day 30/24. During the periods of the highest spikes in TSS (June-August 200-), TSS:BOD ratios were around 4:1. This is way in excess of what designers would have expected from the influent information in the Specification."

345. The 10^th July 2000 Design Review Meeting. The various investigations and discussions all took place in advance of a high level design review meeting that took place on 10^th July 2000 and which brought matters to a head. At the meeting, Mr Holmes maintained on behalf of Yorkshire Water that the figures relating to the COD:BOD ratio were similar to those that had been provided at the tender stage whilst, on behalf of Biwater, Mr Ingham insisted that there was a significant difference. The parties therefore agreed to go to an independent party to obtain an opinion as to whether the influent quality then being received by the SBR was different from that which could have been expected at the tender stage. The result was the appointment of Mr Arthur Boon who eventually produced a report dated 25^th September 2000 ("the Boon report" C77/148). I will deal with the details of that report later in this judgment. However, before I do so, I turn to consider the performance of the SBR during the summer (i.e. from May to August 2000).
346. Summary of the SBR's Performance: May to August 2000. It is apparent that the parties maintained a deliberate and clear distinction between the various discussions and meetings that related to design/contractual issues and the process meetings at which the primary objective was to make the appropriate operational decisions that would ensure that the new consent standards would be met, i.e. at this stage, by implementation of Dr Zuber's strategy.
347. It is common ground that, as the SBR came back under control in May and early June 2000, it was producing good quality effluent (Transcript, Day 3/165). However, it is important to bear in mind that, notwithstanding its ability to produce good quality effluent during this period, the SBR had still not been stabilised (see paragraphs 350-352 below) The progress being made on effluent quality was reported at the various process and/or project review meetings and duly recorded in the minutes (see, for example, C1/171, C1/187, C1/192, C1/199 and C1/214). The marked improvement in the quality of the effluent during this period is clearly and dramatically shown in Graphs 1 to 4 annexed to the end of Dr Horan's first report (E3). Graph 1 shows that there were no consent failures for Ammonia from mid-June until towards the end of January 2001. Graph 2 shows that there were only four suspended solids failures between the beginning of June 2000 and mid-December 2000. Graphs 3 and 4 give a very similar picture with regard to the SBR's effluent COD and BOD respectively.
348. This general improvement in the performance of the SBR meant that consideration came to be given to increasing flows and moving towards performance tests. This question was raised by Earth-Tech on 12^th June (C70/270) and was discussed at a project review meeting on 29^th June (C1/187-188). However, there was still a significant dispute as to whether the plant had been correctly sized: i.e. the DWF question (see paragraphs 279, 280 and 319(i) above), the aeration issue was still continuing and the parties were still anxious to ensure that the plant was stabilised, before embarking upon any performance tests: see the following overview, as recorded in the minutes of the process meeting held the day before (28^th June):

"The criteria for the successful operation of the plant are being met at the present time. Ammonia results are all OK and sludge settleability is stable.

Solids levels in tanks 1, 4 & 5 have dropped to a level which is of some concern and sludge wasting has been adjusted to try and recover to the target MLSS levels. We need to monitor the MLSS levels for another week before deciding if the plant is stable.

Following a review of next weeks (sic) results, the plant will be considered to be stable if the MLSS levels have remained at 1800 +/- 200 without any of the effluent parameters showing variability. …"

349. The relatively low MLSS target figure of 1800 mg/l (with a range of error of +/ - 200), for establishing that the plant was stabilised, had eventually been adopted on Biwater's recommendation that it was appropriate to the low incoming BOD load (D3A/166). However, at the process meeting held on 17^th July 2000, it was recorded that the plant had still not been stabilised, as noted in the following overview (C1/169):

"In terms of treatment, the plant continues to perform reasonably well. Settlement is still fairly stable. We now have a fairly high sludge age. Ammonia is good, with sufficient capacity to deal with peaks in load. There is no evidence of solids being decanted.

Tanks 1,4& 5 now have MLSS around the 1800 mark, so surplussing from these will now recommence. The MLSS in tanks 2 & 3 are presently still too high, at approx. 2300, so sludge wasting will be increased to get the figure down to 1800. …

The Plant cannot yet be considered to be stable as the criteria for maintaining the MLSS at 1800 +/- 200, has not yet been met."

350. By the time of the process meeting held on 9^th August 2000, the SBR was continuing to perform well, although it was acknowledged that it had not been possible to maintain the MLSS at the target figure of 1800 +/- 200 (C1/214). There is limited information about the presence or effect of microthrix parvicella during this period. However, the SBR's greatly improved performance strongly suggests that the numbers of microthrix parvicella had diminished, possibly as a result of increased aeration and/or its seasonal growth pattern.
351. Throughout this period it was thought that one explanation for the difficulty in achieving and maintaining the target MLSS level might be some form of inter-basin transfer, possibly as a result of leakage due to faulty or defective valves in the connecting pipe work (see also paragraphs 245-247 above). However, at the SBR Review Meeting held on 17^th August 2000, it was acknowledged that the recorded data did not support the occurrence of any such inter-basin transfer (although there had been instances of a valve being wrongly opened or left open), but it was decided to investigate the possibility further (C1/229). At the same meeting it was agreed that it was not practical to carry out meaningful performance tests on the SBR in the prevailing circumstances, although Yorkshire Water indicated an intention to undertake the four 28 day tests as soon as circumstances permitted (C1/230).
352. The High-level and Low-level Filters: Summer 2000. By May 2000, from what she had been told and based on the various problems that had been reported to her by various members of the Yorkshire Water/Earth-Tech teams, Mrs Williamson had developed serious doubts as to whether the SBR was capable of treating the design flows and loads. She had also come to the conclusion that there was something fundamentally wrong with the SBR (D1/272 and Transcript, Day 13/18). Although Mrs Williamson was aware that Biwater disputed that this was the case, she had been advised that Earth-Tech shared Yorkshire Water's view that the SBR was fundamentally defective and she firmly believed this to be the case. It is no exaggeration to describe this as a critical watershed in the history of the original SBR. Inevitably, this firm belief on Mrs Williamson's part had a significant effect on her approach to matters and decisions relating to the SBR and to how matters progressed from the summer of 2000 onwards, despite the marked improvement in the SBR's performance, which strongly suggested that there was actually nothing fundamentally wrong with it.
353. On 7^th July 2000, Yorkshire Water found that its portion of a sample of final effluent from the Knostrop sewage treatment works (i.e. the effluent from the entire works, not specifically from the SBR effluent), which had been taken by the Environment Agency for testing, did not meet the ammonia compliance standards. Not surprisingly, this was of considerable concern to Yorkshire Water and Mrs Williamson made this clear at an SBR meeting held on 11^th July 2000 (R1/45). I have no doubt that this consent failure confirmed Mrs Williamson's worst fears and her belief that the SBR was fundamentally defective, although the failure could not be directly attributed to the SBR.
354. So it was that Mrs Williamson made it clear that no significant changes were to be made to the operation of the SBR during July and that any future significant changes to the operation of the plant would have to be agreed and signed off by her, with any such change being supported by information on the performance of the whole of the works, the potential impact of the change on the whole works, a risk assessment of the possible impact of any such change and the planned mitigation measures. These precautions were deliberately extended to the whole works, because it was Mrs Williamson's expressed belief that the SBR's poor performance had adversely affected the high and low-level filters (C73/34). At the same meeting, Taylor Woodrow's Environmental Client Services Director, Mr David Lax ("Mr Lax"), outlined the action plan for improving SBR performance in the following terms:

"1. Investigate the cause of the cross connection, which is allowing flow to pass from one basin to another. The initial view is that this is a software problem that should be resolved by the end of July. Once this is resolved it should stabilise the MLSS.

2. Taylor Woodrow claim that the feed quality has changed since the contract was let. This is to be resolved by appointing an expert to establish if the feed has changed beyond the limits of the contract. If a suitable expert can be agreed the report could be produced by 24^th August.

3. Taylor Woodrow claim that the current screw pump operation is confusing the software because there are wide fluctuations in the flow when the assist pump cuts in and out. A meeting is planned for next week to develop a way forward. This will be agreed by the end of July.

4. Taylor Woodrow are investigating what additional aeration capacity could be provided and what modifications this would require. The investigation will be completed and a report produced by the end of July."

355. In my view, it simply cannot be said or assumed that this particular failure in effluent quality for the whole of the works was directly due to any failure on the part of the SBR to produce effluent to the required standards. As I have already pointed out (see paragraphs 314-315 above), the SBR was producing excellent quality effluent by 7^th July. However, as Mr Holmes explained (Transcript, Day 3/163), there had been a deterioration in the performance of the high and low-level filters during the summer of 2000, especially on the low-level side, and it was important to discover why this was so. This resulted in the setting up of a group within Yorkshire Water in July 2000, at Mr Holmes' suggestion, to investigate the performance of the filters (C73/113).
356. On 28^th July 2000, Yorkshire Water held its first "Knostrop Filter" meeting at which, amongst the various matters discussed, a number of factors affecting the performance of the filters were identified and rated as to their significance (C73/241-242), including (i) the fact that there were reduced flows to the ASP because it had been out of service (high), (ii) the fact that the transfer pumping station had been taken out of service in February 2000, reducing flows to the low-level filters (high), (iii) reduced flows to SBR (medium/low), (iv) SBR performance (medium/low) and (v) industrial effluents (medium).
357. Items (i) and (ii), which were both characterised as being of "high" significance, were entirely the responsibility of Yorkshire Water and nothing to do with the SBR. Although items (iii) and (iv) did relate to the SBR, they were both characterised as being of medium to low significance and in the context of the SBR's current effluent being expressly acknowledged to be of excellent quality (C73/241). In my view, the likelihood is that the problems with the high-level and low-level filters during this period were not attributable to the SBR.
358. Item (v) related, in particular, to the discharge of industry-related pollution by a company called "Global". At the time, the Global pollution was considered to have had a very detrimental effect on the Knostrop plant's nitrification capability. It is clear that this factor was considered to be sufficiently significant for Mrs Williamson to decide that it was "absolutely essential" that the information should not be made known to Taylor Woodrow (and Biwater), in case it was used as "an excuse for the SBR not working" (C92/50) in the contractual arguments that were currently taking place with Taylor Woodrow. In my view, this particular episode of industrial pollution during the summer of 2000 and Yorkshire Water's reaction to it, does show that, from time to time, treatment problems at Knostrop were and were perceived to be due (in part, at least) to the toxic and/or chemical effect of incidents of industrial pollution: see, for example, the overnight loss of nitrification in tank 4 of the SBR, observed by Dr Zuber on 19^th June 2000 (D3A/333), which was probably directly due to the Global incident. As it seems to me, the problem of intermittent industrial pollution is a factor that must be taken into account when assessing the overall performance of the SBR during the period 1999 to 2001.
359. August to November 2000. By mid-August 2000, according to Mr Parker, the action taken in relation to the filters had resulted in the recovery of the high-level filters, but it was thought that a further 3 to 4 weeks was required in the case of the low-level filters (C92/58). In the event, the low-level filters recovered somewhat more quickly than that (Transcript, Day 13/32).
360. As Mrs Williamson readily acknowledged during the course of her evidence (Transcript Day 13/35), there was no prospect of the SBR performance tests being carried out during the summer of 2000 until such time as the high and low-level filters were back in full operation and the SBR had been stabilised (C75/138), notwithstanding the greatly improved performance of the SBR during the same period. However, as Mrs Williamson made clear (Transcript, Day 13/34-35), by the end of August 2000, Yorkshire Water was already giving active consideration to terminating its contract with Taylor Woodrow, despite the greatly improved performance of the SBR. Mrs Williamson described how, in the event, the decision to terminate the main contract was taken by Yorkshire Water's Board Capital Investment Committee ("the BCIC"), as follows:

"A. My understanding (is that) the decision was taken to end the relationship with Taylor Woodrow because the plant was not meeting the requirements of the contract and it was not felt that Taylor Woodrow/Biwater were going to make any improvement. It was nothing to do with the rest of the plant.

Q. And at the time it was not possible to do the performance test, was it, because the filters still were not in consent?

A. No, but the plant had been handed over in November so there had been plenty of opportunity previously, if the plant had been performing correctly, to do the performance tests. …

The plant had been handed over in November (1999). The original plan was to do performance tests. I think we were talking about doing them when I actually joined, took over my new role in March (2000), and the plant had never been stable and fit to carry out those performance tests, so that is almost 12 months. That is, we normally allow six months for a contract, a large contract like this, we would normally allow six months before a consent comes in to stabilise the plant. …

The decision was actually taken by BCIC. It was a recommendation from Ian Brown … that we would be ending the contract. …

It was discussed with me and it looked like a sensible way forward, but the final authorisation would be from BCIC. …

I do not know when it was put to BCIC, but it would be around (September 2000) and, yes, I did agree with the strategy."

361. The Decision to Terminate the Main Contract. In early July 2000, Mr Sawyer wrote to Mr Jones, seeking reconsideration of Earth-Tech's opinion that the SBR was defective (C72/103). On 31^st July 2000, Mr Stringer wrote to Yorkshire Water and confirmed that it remained Earth-Tech's view that the SBR was defective for the following reasons (C73/314-315): (i) a shortfall in both hydraulic and treatment capacity, (ii) the omission of IDSCs, causing poor sludge settleability, (iii) fill-decant events/the frequency of fill-decant inhibits and (iv) the fact that the plant could only be operated satisfactorily on 6 tanks at greatly reduced flows and loads. At the same time a letter was sent making it clear that the performance test period had not yet started (C74/174).
362. By late summer 2000, Yorkshire Water had formed the view that the Knostrop treatment works could be made to comply with the new consent standards with the implementation of the planned maintenance scheme, even if the SBR did not achieve its full contractual requirements (C92/60). At a meeting with Earth-Tech, held on 31^st August 2000, Yorkshire Water's Solution Deliveries Manager, Mr Ian Brown ("Mr Brown"), indicated that Yorkshire Water's preferred strategy for bringing the main contract to a conclusion was to "draw a line" under the main contract and accept that the SBR would not be able to perform as specified. This strategy was said to be preferred to going down the "strict contractual route i.e. defect liability" (C92/60).
363. Following the meeting on 31^st August, Mr Jones was sent an email that set out the strategy that Yorkshire Water wanted to pursue (C76/31A). It was suggested that Yorkshire Water might accept a reduced treatment capacity in return for a significant reduction in the contract sum, in order that Yorkshire Water could develop an approach to ensure that the Knostrop works as a whole complied with the consent standards. Mr Jones was asked to indicate how this might be achieved within the next two weeks.
364. On 14^th September 2000, Mr Jones produced his report (C77/222d-f). Mr Jones' strategy was to issue a determination notice under Clause 43 of the Main Contract in about the second week of November 2000 (in fact, no such notice was ever given). The reasons given for this approach were speed and the fact that Yorkshire Water could take control of the remedial requirements and the necessary works. Other possible routes were considered, but put to one side (Transcript, Day 8/153 onwards). Mr Jones' report was attached to a report to BCIC dated 26^th September 2000 (C77/222a-c) and was duly adopted by BCIC (C92/87). This resulted in Mr Jones producing a summary of the actions that had been put in hand to achieve Yorkshire Water's strategy (C78/21). The timetable was clearly intended to lead to a termination of the main contract in November 2000, in accordance with the adopted strategy.
365. Accordingly, I accept as correct Mr Streatfeild-James's following observation (Biwater's closing submissions, paragraph 188):

"Thus Yorkshire Water decided in August 2000 that its preferred strategy was to take control of the Knostrop plant as a whole, and to address compliance issues on that basis, by "drawing a line under the contract"; and during September it identified and adopted a contractual strategy for achieving this result."

366. The Boon Report. The Boon report was finally produced on 25^th September 2000 (C77/148-195), somewhat later than had been originally anticipated. In it Mr Boon identified the brief that he had been given, which included "2.1.0.3 – Assess if the present influent quality is different to that which could have been expected from the information made available at the tender stage". In his Executive Summary (C77/149), Mr Boon stated that the average COD:BOD ratio, based on the data obtained from the contract documentation, was 2.74 to 1, whereas the average COD:BOD ratio of the actual influent received by the SBR since October 1999 was 3.6 to 1.
367. In the body of his report, Mr Boon stated that the influent was within the specified range for the various parameters specified, but that there were "significant and unexpected" increases in the COD:BOD ratio (C77/176) and that the difference in the ratio indicated that the sewage was more difficult to treat than was initially expected (C77/177). Mr Boon also expressed the view that it was probable that the high-level feed contained a substance or substances that were non-biodegradable or that inhibited biochemical oxidation and that further tests should be undertaken (C77/179).
368. There were a number of meetings and discussions in Yorkshire Water/Earth-Tech arising out of the Boon report, including consideration being given to converting the SBR into a CASS plant. For the purposes of this judgment, it is not necessary to go into any particular detail of these discussions/meetings. Suffice it to say that Yorkshire Water and Earth-Tech both accepted that the Boon report had identified peculiarities with the sewage in question in terms of the COD:BOD ratio and that this required further investigation. It appears that Dr Hemphrey did carry out some investigations, with the assistance of information provided by Ms Schofield. To the extent that it is possible to ascertain the outcome of those investigations, it seems clear that it was established that there was an excessive sludge yield and that there were unresolved concerns over the treatability of the influent.
369. Recovery Plan 2 and Ms Henderson's Presentation. As part of Yorkshire Water's overall strategy towards termination of the main contract, Earth-Tech had undertaken to produce a report for Yorkshire Water on the SBR by early November 2000 (C92/87). This involved Earth-Tech considering the SBR and its performance and included the drawing together of all the relevant and available information. These investigations appear to have been carried out in October 2000 and led to a meeting on 25^th October at which Earth-Tech was asked to prepare a "scoping document" to address what had been identified as the "main shortfalls" in the SBR, namely (i) the high/low level split, (ii) the provision of SBR influent balancing, (iii) MLSS recirculation modifications, (iv) additional sludge treatment capacity and (v) SBR control modifications (C79/254). The scoping document was to become Recovery Plan 2 and was to be written under the auspices of Mr Copnall, but with individual sections being prepared and written by different specialists in Earth-Tech (C79/245). The purpose of Recovery Plan 2 was to establish the works that needed to be done to both stabilise the SBR and to maximise its treatment capacity at the existing 85% of the original maximum capacity (C9/287).
370. Whilst the scoping document was being put together, Mr Jones decided to ask Ms Henderson to return and carry out a review of the SBR as a whole, because it was his view, based on his previous experiences, that it was a good idea to have such a review undertaken by somebody who was reasonably independent, but who had the necessary skill and a basic knowledge of the project (Transcript, Day 8/165).
371. Ms Henderson had been heavily involved in other matters and had been working abroad a great deal. Because of these other commitments, she had only a limited recollection of being involved in Recovery Plan 2 (Transcript, Day 7/74-75). It was Ms Henderson's evidence that the main process input into Recovery Plan 2 had come from Dr Hemphrey, Ms Henderson's successor as Earth-Tech's lead process engineer for the Knostrop project (Transcript, Day 7/72: and see paragraph 156 above). However, it is clear that Ms Henderson did visit the Knostrop works, where she was shown around by Mr Coulthard, she had a discussion with Ms Schofield about the SBR, she obtained various documents from the files and she gave a presentation. Although Ms Henderson does not appear to have been extensively involved in its preparation, it is nevertheless clear that parts of her presentation were subsequently incorporated into Recovery Plan 2, although written by somebody else (either Mr Copnall or Dr Hemphrey).
372. In my view, Ms Henderson's understanding of the SBR was somewhat limited as, in effect, she herself acknowledged. She had a CASS background. She had never seen a Jet Tech SBR before (Transcript, Day 8/58) and she emphasised how complicated it was (D1/75). As Ms Henderson explained in the course of her evidence, she found it difficult to be sure what the control software was doing because it was constantly changing the treatment cycle times. In this respect, it is important to note that the Jet Tech SBR is markedly different from the CASS plant/system, which has fixed treatment cycles, not variable ones (Transcript, Day 8/57-58). During the first two weeks of November 2000, therefore, Ms Henderson tried to gather what information she could from the O&M manuals, she gleaned such information as she could from the way the SBR was operating during the visit that she made with Mr Coulthard, she discussed the operation of the SBR with Ms Schofield and she applied her general knowledge of SBRs to the information that she gathered in this manner. The result was the presentation, complete with visual displays, which she gave on 15^th November 2000 (C80/186-204: "Ms Henderson's presentation").
373. In her presentation, the conclusions expressed by Ms Henderson were: (i) that the SBR was operating using the wrong design flow and decant depth, (ii) that the SBR could only treat 87% of peak flow, (iii) that the control software was incorrectly calculating the anoxic fill times, (iv) that the settle time was too long and (v) since there was no idle time and no margin for error, the result was fill-decant events (C80/203). Ms Henderson's recommendations were: (i) that Jet Tech should check the design and confirm operating water levels, (ii) that Jet Tech should investigate the anomaly in the anoxic fill times and (iii) that the SBR should be set up and operated with design parameters and set points (C80/204).
374. In the event, as it seems to me and despite Mr Slater's submissions to the contrary, Ms Henderson's presentation did not withstand the critical analysis to which it was subjected by Mr Streatfeild-James' cross-examination (see Transcript, Day 8/60 onwards). The extent to which it was demonstrated that there were important errors in Ms Henderson's presentation can best be illustrated by reference to her critical (and erroneous) conclusion that the SBR had been designed using only 87% of design flows (C80/194). Ms Henderson had reached that conclusion by carrying out (inter alia) a number of calculations of decant volumes, entirely based on a CTDC (Cycle Time Distribution Chart) that she had obtained as part of her general inquiries. Unfortunately, the CTDC used by Ms Henderson (C80/192) was itself based on dimensions that were not the same as the dimensions of the SBR as built, so the decant volumes were completely different. I accept that, when the correct figures were used to repeat Ms Henderson's calculations, the SBR was shown to have adequate capacity. As Mr Streatfeild-James observed (Biwater's closing submissions, paragraph 208):

"The simple position was that Ms Henderson was using a chart which did not represent what had been built. The correct figures when inserted into Ms Henderson's calculations showed that in fact there was adequate capacity (Transcript, Day 8/76-77)."

375. The final version of Recovery Plan 2 (R1/72-92) contained the recommendations that Earth-Tech considered necessary to achieve satisfactory treatment of a flow up to 115 tcmd, but did not cover what would be necessary to achieve the contractual maximum flow of 136 tcmd (R1/75). Inevitably, to the extent that Recovery Plan 2 was based on and/or incorporated the calculations and conclusions of Ms Henderson's presentation, its calculations and/or conclusions were, by a parity of reasoning, erroneous. However, it was Mr Holmes' view that achieving stability at 115 tcmd had been a realistic objective at this stage in the project and that achieving the full 136 tcmd still remained an aspiration (Transcript, Day 4/57).
376. In summary, Earth-Tech's recommendations were as follows (R1/75):

"This report identifies there are inadequacies in the provision of decant volumes which together with the software shortfalls combine to prevent the plant from meeting the contractual requirements.

The report recommends:

The software should be either modified/corrected or extensively replaced at the same time as increasing the set points to increase the decant volume

The movement of the liquors between pairs of tanks managed by replacement of valves and modifications to standby pump software

Modify the flow split to spread the hydraulic load to the humus tanks

Modify the sludge removal facilities to allow increased flexibility."

377. In the meantime, on 2^nd November 2000, Taylor Woodrow received the Take-Over Certificate, it having been duly signed by Earth-Tech and Yorkshire Water on 25^th February and 3^rd March 2000 respectively (C1/22). It is to be noted that, although the Take-Over Certificate had been signed in the early part of the year, Earth-Tech did not actually get round to sending it to Taylor Woodrow until 25^th October 2000 (C1/21).
378. Having regard to the process of investigation and recommendation that had taken place during the autumn of 2000 I am satisfied that, at this stage: (i) Earth-Tech had formed the view that a flow of 115 tcmd (i.e. about 85% of the specified maximum flow) could be achieved by carrying out the relatively limited remedial works identified in Recovery Plan 2 and priced at just under £150,000 (R1/80), (ii) it was Earth-Tech's view that the design deficiency identified by Ms Henderson (which was erroneous in my opinion: see paragraph 371 above) meant that treatment of more than about 85% of the contractual maximum flow could not be achieved and (iii) Earth-Tech recommended that there should be further investigation of the need for an IDSC/selector, the decant arrangement and the need for additional humus tanks/tertiary treatment (R1/82).
379. August to November 2000. The strategy adopted by Yorkshire Water involved trying to stabilise the SBR before the planned termination of the main contract in mid November 2000 (see paragraphs 352-353 above). However, during the period September to November 2000, as is clear from the minutes of the various process meetings, the SBR continued to perform well and produce good quality effluent.
380. As Mr Holmes explained, although settleability of the sludge blanket started to deteriorate somewhat with the approach of the winter, the main concerns discussed and recorded at the process meetings during this period were how to stabilise the MLSS (in particular, by solving the apparent problem of inter-basin transfer of MLSS) and how to increase flows and loads (Transcript, Day 4/59). The problem with fill-decants had been resolved by the fill-decant inhibit and the monitoring carried out by Yorkshire Water revealed that there was no obvious effect when, as a result of a fill-decant inhibit event, influent to the SBR was diverted to the high-level filters. Fill-decants/fill-decant inhibits were therefore no longer considered to be one of the main issues by this stage. Attention was largely focused on achieving stability of the SBR.
381. Dr Zuber's investigations into MLSS levels had shown that the suspected inter-basin transfer of MLSS was not occurring and, as a result, attention had turned to the sludge wastage regime that was being operated by Yorkshire Water. In a letter sent to Taylor Woodrow, dated 12^th September 2000, Mr Molloy gave expression to Biwater's belief that the problem with MLSS levels was due to poor solids' control, which was itself the result of Yorkshire Water's actions or failings, as follows (C76/201-203):

"We note also that the sludge system is capable of wasting more solids than dictated by the design. The limiting factor in the control of the MLSS is not one of sizing of the sludge system but one of correct usage.

Biwater has provided assistance to Yorkshire Water for many months, yet it would appear that despite the fact that the plant is being operated in a manner contrary to our advice, there is still the wish to blame Biwater for poor solids control.

The evidence presented from Yorkshire Water's own data demonstrates that the plant is not being operated in accordance with the advice given, and that Yorkshire Water are not interpreting the data which they are gathering in order to arrive at correct operating decisions."

382. The result was that Biwater was asked to state specifically what set points should be used for sludge wasting (D3A/187 and C76/206). The idea was that Biwater should return to site and demonstrate that the SBR could be stabilised. This new regime started on 11^th September 2000, and the procedure adopted was that the Yorkshire Water operatives would carry out the MLSS analyses in the mornings and that these, together with the previous day's sludge wastage results were sent to Mr Molloy. He would then decide upon the appropriate sludge wastage set points and send details of these back to site, so that they could be set by Mr Coulthard for operation on the following day. After some initial problems, this system appears to have worked well. It took about an hour every day to carry out (Transcript, Day 6/157) and its success is compelling evidence that the SBR could be operated satisfactorily if its flexible control system was managed correctly.
383. By the date of the process meeting held on 29^th September 2000, the SBR appeared to be performing extremely well in terms of MLSS as well as effluent quality, although a slight deterioration in settleability was noted. However, no changes to the flows and loads were proposed in the short term since the SBR was still not considered to be stable by Yorkshire Water/Earth-Tech, because they considered that various important issues still remained unresolved (C1/239): i.e. (i) the continued incidence of fill-decant inhibits, (ii) mechanical plant reliability, (iii) software performance, (iv) the recent deterioration in settleability, (v) the possible flow transfer between basins and (vi) the operation of the control penstock.
384. However, Dr Zuber gave evidence that he felt frustrated, because he considered that the SBR was both stable and working extremely well by this time (D3A/346). In the circumstances then prevailing and having regard to the success of the strategy that he had promoted and pursued, his sense of frustration was very understandable. Although I am not persuaded that Yorkshire Water had deliberately embarked on a conscious policy of holding back from carrying out the performance tests because it had recently decided to terminate the main contract in November (which appears to be the thrust of paragraph 221 of Biwater's closing submissions), the decision to terminate the main contract undoubtedly coloured Yorkshire Water's overall approach to the operation of the SBR. I have no doubt that Yorkshire Water's various concerns were genuinely held (although, as it seems to me, misplaced) and that it was these concerns that were the main reason for its decision not to increase flows and loads and proceed to performance testing at this stage. It is important to remember that the flow of sewage cannot be "switched off" in an emergency (unlike other public utilities and services). Thus, the daily flow of sewage into the Knostrop works would be bound to continue, whatever treatment problems/difficulties were being experienced at the time. Accordingly, the consequences of a major compliance failure by the Knostrop works were far too serious for any risk of such an occurrence to be taken. There were a number of reasons that appeared to justify Yorkshire Water's decision not to increase the flows and loads to the SBR and proceed to performance testing at the time. In particular, the improvement in maintaining appropriate MLSS levels was fairly recent (following a rather lengthy period of difficulty) and there were some slight signs of deterioration in settleability with winter approaching (Transcript, Day 4/59). However, for the reasons given later in this judgment, I am satisfied that the SBR was perfectly capable of meeting the contractual performance requirements if operated properly. By the autumn of 2000, Taylor Woodrow/Biwater were quite prepared for the performance tests to be carried out on the SBR and, in my judgment, there is every reason to believe that it would have passed those tests satisfactorily had they actually been carried out. In the event, it was Yorkshire Water (the party who had the contractual obligation to carry out the performance tests: see Clause 35 of the General Conditions) that decided to hold back from doing so.
385. I am also satisfied that the efforts being made by all the parties concerned to optimise the SBR's performance with a view to increasing the flows and loads (assuming that was possible) were genuine. To that end, an SBR Optimisation Group ("the optimisation group") consisting of a working party, comprising representatives from Yorkshire Water, Earth-Tech, Taylor Woodrow and Biwater, was duly formed. I am not persuaded that the formation and work of the optimisation group was a result of Yorkshire Water's decision to determine the main contract in November 2000 (Biwater's closing submissions, paragraph 222), although the group's existence and the future prospects for the project as a whole must be seen against the background of that decision. All in all, I have no doubt that Yorkshire Water was not prepared to take any sort of discernable risk of any failure of the consent standards caused by the SBR and, inevitably, its perception of such a risk was heavily influenced by its firmly held belief (shared by Earth-Tech) that the SBR was fundamentally defective and the fact that (essentially for that reason) it had been decided to determine the contract in the near future.
386. As Mr Holmes explained, the work of the optimisation group consisted of brainstorming sessions in which ideas with regard to performance across the whole of the Knostrop works were considered (Transcript, Day 4/63). The first such meeting was held on 30^th October 2000 and there were further meetings in November. Thus, at the monthly Progress Meeting held on 16^th November 2000, it was agreed that Taylor Woodrow/Biwater were due to submit their report for the optimisation of the SBR by 17^th November 2000 and that they were to provide details of the operating regime required to stabilise the plant and thereafter for inclusion in the Operation and Maintenance Manuals. These meetings culminated in the production of a consolidated report dated 17^th November 2000 (R1/110-114).
387. The optimisation group's consolidated report summarised what had been done and decided at the various meetings, identified the options that had been considered and came to the conclusion that the plant should be optimised at current flows (about 110 tcmd maximum), before increasing flows to design levels (R1/114). In order to achieve this it was recommended that: (i) the flow averaging period (Watchdog) should be left at its current setting, (ii) the TWL (top water level) should be increased by 200mm., (iii) if the adjustment to the TWL did not work, then the settle safety time should be adjusted to give increased buffering capacity for flow changes and (iv) if that also did not work, that the SBR feed pumps should be slowed down (R1/114). As it seems to me, the optimisation group's work and conclusions make plain a general acceptance by all concerned that, despite some apparently unusual features of the influent, the SBR system had sufficient flexibility to enable it to treat it satisfactorily at full contractual flows, after appropriate adjustment to the SBR's various operational parameters.
388. Moving towards Termination. As Mr Streatfeild-James observed (Biwater's closing submissions, paragraph 224), there is no doubt that Biwater was becoming increasingly irritated by the apparent lack of progress towards performance testing during September/October 2000. From Biwater's perspective, the following appeared to be the case: (i) it had assisted on site to recover performance, (ii) on its (perfectly reasonable) interpretation of the Boon report, its suggestion that there had been a significant change in the nature of the influent appeared to be correct, (iii) it had taken an active role in managing the SBR since 11^th September 2000, in the light of what it considered to be Yorkshire Water's various operational failings and (iv) now considered itself to be faced with what appeared to be an increasing intransigence about increasing the flows. In addition, there was the issue of the outstanding sums to which Biwater considered itself to be entitled.
389. From Mr Ingham's point of view, he had been told in September 2000 that it was accepted that the SBR was producing good quality effluent, that sludge settleability was acceptable and that the process was performing well (D3A/109). He was therefore understandably surprised to receive a copy of Earth-Tech's letter to Taylor Woodrow, dated 2^nd October 2000, stating that it was considered inappropriate to certify any further payments (C78/14). Taylor Woodrow's reply (C78/62), dated 6^th October, set out the Taylor Woodrow/Biwater position with regard to the perceived problems with the SBR and Earth-Tech's suggestion of non-payment and made clear that issues over extensions of time and the final account were outstanding, as follows:

"The perceived problems with the plant have been shown to be as a result of having to deal with an effluent (sic) that is significantly different from that which was expected from the tender information. The independent report issued by Mr A Boon confirms this and that there has been a dramatic change to the influent in April of this year, probably as a result of industrial discharge. …

In reality plant analysis has proved that the situation is worse than postulated, in that the COD is highly biodegradable with tests results showing an average residual COD of 40mg/l. This will create an oxygen demand in excess of the design at higher loading rates and the plant will use more power. There will also be an increased sludge production. The sludge system has already demonstrated that it is possible to achieve a greater throughput than the design.

The plant passed its take over test and was handed over and accepted by Yorkshire Water in a satisfactory operating condition, producing good quality effluent. We have demonstrated in the past month that, even with the different, more onerous, influent, the plant can be set up and managed to continue to produce good effluent and indeed, in achieving a throughput of sludge greater than the design.

In our opinion, backed up by the report, the contract guarantees relate to a wholly different influent to that which the plant is now receiving and therefore they are not applicable to the current situation.

In addition, discussions have been ongoing since July 1999 with yourselves, regarding an extension of time and associated costs. Agreement had been reached in principle in March 2000, with only one item, with a value of £24k, not being finalised. However, we have still not received any acknowledgement of this position or any interim payment, despite repeated requests. Furthermore, as we have recorded in meetings and correspondence, we consider that a significant proportion of the works and attendance that have been carried out since takeover, and are still ongoing, are additional work for which we expect payment.

We would therefore request that acknowledgement is made in principle of our entitlement to payment of the original contract sum. Payment is overdue for variations, additional attendance since take-over and the costs, as previously agreed, for the extension of time of 18 weeks. Unless significant progress is made in the near future we will have no alternative but to seek reimbursement through the Contract under Clauses 39.6 and 46.

Notwithstanding the above we confirm our commitment to work with Yorkshire Water in the immediate future and trust that we will make some progress in resolving our differences at the meeting arranged for …20 October…"

390. However, although Taylor Woodrow still felt able to give an assurance of its continued commitment to the project, as indicated in the final paragraph of its letter of 6^th October (which probably should have been dated 16^th October), Biwater's level of concern was such that it gave Taylor Woodrow notice of its intention to suspend work by letter dated 17^th October 2000 (C78/305). In the event, in the light of the apparent lack of progress in dealing with the extension of time and final account issues (as to which, see below), Taylor Woodrow wrote to Earth-Tech on 9^th November 2000, giving notice itself under Clause 39.6 of suspension of work in the event of non-payment (C80/98). Taylor Woodrow concluded its letter as follows:

"We regret taking such action but consider that there has been considerable efforts by all parties to resolve the difficulties at Knostrop and that withholding payment is unreasonable under the circumstances outlined in our previous letter (i.e. the letter of 6^th October: C78/62) …"

391. In the meantime, a meeting had taken place on 20^th October 2000, attended by both Mr Ingham and Mr Jones, at which the outstanding extension of time and final account issues were discussed (C79/89). Mr Jones, who claimed that his main purpose in attending the meeting was to familiarise himself further with the project, confirmed that Taylor Woodrow/Biwater were very keen that Earth-Tech should address these issues and that he gave a target date for doing so of the week commencing 6^th November 2000 (Transcript, Day 8/173).
392. Although the extension of time and final account issues were discussed internally by Earth-Tech in the period that followed, mainly by Mr Jones, Mr Coulthard and Mr Stringer, so far as concerns Taylor Woodrow/Biwater the matter was not dealt with by the promised target date of 6^th November. However, I am satisfied that Earth-Tech's views as to the merits of the Taylor Woodrow/Biwater claims were as set out in draft letters dated 15^th November 2000, prepared by Mr Coulthard, following an internal Earth-Tech meeting held on the same date (C80/165A).
393. I am also satisfied that the intention was that these letters were to be sent to Taylor Woodrow, once they had been signed by Mr Jones (Transcript, Day 6/122-123), and certainly so far as Mr Coulthard was concerned, it was intended that the letters should have contractual effect. As he put it, he would not have wasted his time drafting them had it not been the intention that they should be issued. In the draft letter dealing with the claim for an extension of time, a total of 17.5 weeks' extension of time was awarded, of which 16.5 weeks of the extension of time was said to be granted with costs, and 26^th November 1999 was stated to be the revised take-over date for the SBR (C80/212-214). After setting out the various additional sums that Earth-Tech proposed to award to Taylor Woodrow, the draft letter dealing with the final account put forward a provisional Final Account Valuation of £11,682,021.21 (C80/214-215).
394. Having regard to Mr Coulthard's evidence, the terms of his emails that accompanied the drafts when he forwarded them to Mr Jones for approval and signature on 15^th and 16^th November (C92/121 and 123) and the form and content of the letters themselves, I should make it clear that I do not accept Mr Jones' assertion that these drafts were prepared merely to crystallise the thoughts of Mr Coulthard and Mr Stringer on paper (Transcript, Day 8, 127-128). In fact, Mr Jones produced his own drafts of the extension of time and final account letters on the 23^rd November and 8^th December 2000 respectively (C92/125 and 127). Neither letter differed from the drafts prepared by Mr Coulthard in any significant respect, except that the provisional Final Account valuation had been reduced to £11,642,334.65, after adjustment of the power costs (Transcript, Day 9/17). Furthermore, on his final account letter, it can be seen that Mr Jones made a handwritten note dated 8^th December to the effect that the letter had not been sent at Yorkshire Water's request (Mr Sawyer), because there was a "contractual issue" (C92/127).
395. On 12^th December 2000, Mr Jones sent an email to Mr Sawyer in which he acknowledged that the final account letter was still "on hold" at Mr Sawyer's request, but expressing the view, in emphatic terms, that it should be issued as soon as possible (C82/46). Furthermore, on 11^th January 2001, Mr Jones wrote to Mr Sawyer effectively expressing his concern that the extension of time/final account issues had not been dealt with (C83/216). Attached to Mr Jones' letter was his brief written account and opinion with regard to the extension of time/final account issues in which he confirmed that, in Earth-Tech's opinion, an extension of 17.5 weeks (10.25 of which were stated to be at no cost) and a provisional Final Account Valuation of £11,642,334.65 were appropriate. It is worth noting at this stage that, as it seems to me, Earth-Tech's conclusions with regard to the extension of time/final account issues provide compelling evidence of Taylor Woodrow's entitlement to the 17.8 weeks' extension of time and further payment that it has counterclaimed in these proceedings. For the brief reasons that I give later, when dealing with that counterclaim (see below), I am satisfied that Taylor Woodrow is entitled to the extension of time claimed, which effectively disposes of Yorkshire Water's claim for liquidated damages.
396. Having regard to all these matters, I accept Mr Streatfeild-James's submission that, from October 2000 onwards, Mr Jones acted in accordance with the instructions of Yorkshire Water in not issuing the documents that he would otherwise have issued in relation to both extensions of time and the final account. The result was that no further payment was made by Yorkshire Water and, on 23^rd November 2000, Biwater gave notice to Taylor Woodrow suspending the execution of the works (C81/71). This brought to an end Biwater's active involvement in the operation of the SBR.
397. Summary of the SBR's Performance: August to November 2000. I accept Mr Streatfeild-James' submission that the general picture with regard to the SBR during August to November 2000 was one of improvement and good performance. However, as Mr Streatfeild-James went on to observe (see paragraph 356 of Biwater's closing submissions), it is important to note the following additional features of this period:

(i) By mid-August 2000, the average BOD load had increased from 3.622 tonnes/day in July (D1A/42) to 4.958 tonnes/day (D1A/45);

(ii) During the same period, the influent contained excessive TSS loads (often in excess of 10 tonnes/day, with some values over 20 tonnes/day (see E3 Graph 5);

(iii) The isolated suspended solids and BOD failures at the end of August (see D1A/45, D1A/91 and E3 Graphs 2 and 4) mirror the periods in which there were significant peaks, well in excess of 20 tonnes/day, in the TSS loads (as shown by a comparison of Graph 5 with Graphs 2 and 4 of E3);

(iv) By mid-August, MLSS levels were moving upwards to unacceptably high levels, with basins 2 and 3 exceeding 3000 mg/l by 27^th/28^th August, a trend that was remedied by following Mr Molloy's instructions with regard to the appropriate set points (see paragraph 379 above and see also D1A/162);

(v) This period was also one which saw a marked decline in the BOD load being received by the SBR; from an average of about 5 tonnes/day in mid-August, the BOD load dropped to about 3 tonnes/day in mid-September and to less than 2 tonnes/day in October, before rising slightly to around 2 – 4 tonnes/day in mid-November; during the same period the TSS loads fell from the mid-August peak to about 4 – 6 tonnes/day during mid-September to mid-October, before moving upwards again in November and

(vi) The absence of any direct evidence as to the presence of microthrix parvicella or its effect on the SBR's performance during this period.

398. Once more I find myself in agreement with Mr Streatfeild-James's analysis of the position during August to November 2000 (see paragraphs 357 to 358 of Biwater's closing submissions). As he pointed out, not only was the SBR's performance good throughout the period but, once Mr Molloy was in control of the sludge wastage regime, the SBR's apparent lack of stability in August was rapidly remedied and stability achieved. As a result, the SBR was now operating well, despite being affected by the dramatic variations in the TSS loads, the low BOD loads and the high COD:BOD ratio, all of which had the effect of making the SBR difficult to operate. However, as Mr Streatfeild-James observed, "these various difficulties could be – and were – being overcome" (paragraph 358 of Biwater's closing observations). Furthermore (and importantly), although it is clear that Taylor Woodrow/Biwater were perfectly prepared for performance testing to be carried out on the SBR from about late September 2000 onwards, it was Yorkshire Water that decided not to do so.
399. The SBR's Performance during the Winter 2000/2001. In the meantime, Earth-Tech gave Taylor Woodrow further particulars of "the alleged defect" that had been specified in the original 14^th March Defect Notice by its letter dated 22^nd November 2000 (C1/49-50). The main deficiencies notified by these written particulars were as follows: (i) that the Decant Volume was 0.2 metres less than the design requirements); (ii) that, as set up, the SBR could only deal with 87% of the maximum design flow of 136 tcmd and (iii) that each treatment cycle had a fixed period of 40 minutes for the anoxic filling stage, instead of the design concept of a progressive reduction in that time as the flow increased. Each of these points was itself the perpetuation of an error that originated in Ms Henderson's presentation: see paragraphs 373-375 above and paragraphs 207 to 211 of Biwater's closing submissions, which I accept.
400. On 24^th November 2000, Yorkshire Water wrote to Taylor Woodrow in the following terms (C1/52):

"We are in receipt of a copy of your letter dated 20 November 2000 to Earthtech.

We and Earthtech had been promised proposals for the remedying of the defect in the SBR Plant at Knostrop. We have given you ample opportunity to carry out the work to make good the defect notified to you pursuant to clause 36 of the Contract more than six months ago. As you are well aware, this work is of an absolutely essential nature. It has always been urgent and becomes more so as time passes. We and Earthtech are of the view that you have procrastinated to date. We have come to doubt the sincerity of your stated wish to resolve the problems at Knostrop.

We had hoped that the promised proposals would have removed such doubt. Instead, your report refers to the recommendations of the working group without any attempt to translate them into proposals for making good the defect. This is entirely unsatisfactory given that there is an urgent need to resolve this defect, you have known of it for months and yet you appear to have made little progress towards doing so. Furthermore, we are of the view that the defect is attributable to your breach of Contract. Accordingly, we write to you in the terms set out below.

We hereby give you notice, pursuant to clause 36.5 of the Contract, that we require the defect notified to you under sub-clause 36.1 by Earthtech's letter dated 14 March 2000 to be made good urgently, failing which we shall proceed to do the work in a manner approved by Earthtech. We therefore require, by 8 December 2000, confirmation that you intend to correct the defect urgently together with your detailed proposals with timescales and full method statements. If you fail to provide such confirmation, we will take this as confirmation of your inability and/or unwillingness to comply with our requirement within a reasonable time. The necessary defect remedial works are of an urgent nature. We cannot delay any further carrying out these essential works.

Without prejudice to such notice, in view of your continued neglect and refusal to make good as soon as reasonably practicable the above mentioned defect, for which you are responsible under clause 36.2 of the Contract, we give you notice, pursuant to clause 36.6 of the General Conditions of the Contract, of our intention, after 14 days from the date of this letter, to proceed to carry out the necessary defect remedial works unless we receive from you within that time your detailed proposals as set out above.

With regard to both notices, we will look to you for reimbursement of the costs of making good the defect. Furthermore, you are in breach of the Contract. We will proceed as indicated above without prejudice to our other rights under and arising out of the Contract.

You are well aware of the circumstances which force us to write in these terms. Nonetheless, we summarise them below.

On 14 March 2000 the Project Manager wrote to you giving notice, under clause 36.1 of the General Conditions of the Contract, of a defect in the SBR section of Plant with respect to the Plant not achieving the final effluent quality as required under the Contract. You were also informed that defect remedial works were required as a matter of urgency.

Clause 36.1 of the Conditions of Contract requires you to make good defects notified to you as soon as reasonably practicable. You admitted in your letter dated 17 March 2000 that the performance of the Plant had deteriorated after take-over. You queried whether proper particulars of the defect had been given and sought to characterise the instruction to remedy the defect as a variation. This issue is only relevant to the question of who pays for the remedial works and did not and does not affect your obligation to carry them out. However, for the avoidance of doubt, we confirm that we are of the view that the defect is attributable to your breach of the Contract.

Further details relating to the defect have been provided to you. For example, in Earthtech's letters of 3 April, 7 April and 2 May 2000. Also, the problems with the Plant have been witnessed on site and discussed in numerous meetings since March 2000. You have, at various points since March 2000, alleged that we and Earthtech agreed to put any contractual issues to one side. We dispute that any such agreement was reached. Also, Earthtech' s letters since March 2000 have made it perfectly clear that they have continued to require that the clause 36 notice be complied with.

Since the notice, we have adopted an extremely conciliatory approach towards you. We agreed to co-operate with and assist you to comply with your contractual obligations with regard to the defect. Despite considerable indulgence shown towards you, your attitude has throughout been one of procrastination. For example, in your letter of 3 July 2000 you stated that it had been agreed with us and Earthtech that we would "work together to solve the operational problems on the plant. We are continuing to work with you and Yorkshire Water on this basis". Despite such assurances this has not happened and no meaningful proposals have been received relating to the defect remedial work required. The Plant continues to fail to perform in accordance with the Contract despite the fact that so much time has elapsed both since take-over and notification of the defect.

Your letter dated 20 November 2000 provides us with no comfort that the problems will be resolved. Indeed, recently you have conveyed exactly the opposite message. Your letter dated 9 November 2000 threatens, illegitimately, suspension of performance of the Contract. On 23 November 2000, we were informed that Biwater had suspended its works.

The position is now both serious and urgent. We are very concerned that our compliance with UWWTD and NEP discharge consents is at risk because of Environment Agency effluent sample test failures, of which the Plant has suffered two already. You are well aware of the very serious consequences for Yorkshire Water should a compliance failure occur at such a prominent sewage works. "

401. On 26^th November 2000, the Defects Liability Period under the Main Contract expired. On 4^th December 2000, Taylor Woodrow responded to Yorkshire Water's letter of 24^th November, disputing Yorkshire Water's account of the facts and concluding as follows (inter alia):-

"The "Further Detai1s" provided in your letters 3 and 7 April and 2 May 2000 all refer to design or control issues which have been answered in detail in our letters of 12 April, 11 May and 3 July. We do not consider any of these grounds for a Defect Notice and as we have stated above the plant is performing very well and has done for six months. We wou1d accept that we have suffered from a lack of plant reliability but these issues have been dealt with as they arose and it has been acknowledged at the regular Process Meetings that reliability is now much improved.

…

We resent your allegations that we have procrastinated and consider it to be untrue and a gross distortion of the facts. Both ourselves and Biwater have tried to co- operate fully with you and Earth Tech to meet your requirements. The fact that we have maintained our involvement all year, in good faith on your assurances that the resulting cost wou1d be agreed fairly once the plant was stable, cannot be called procrastination. Indeed it is only the latest failure of Earth Tech to honour agreements and undertakings made to issue extensions of time and make appropriate payments that has initiated this current dispute situation.

In conclusion, we repeat that there is no defect in the terms of the notice of 14 March since the plant has been producing effluent of excellent quality for six months. Similarly, we are not in Breach of the Contract. On the contrary; we consider that Yorkshire Water are in breach for not providing the specified influent and for withholding of payment unjustly.

With the receipt of your letter we now consider that we are in dispute and we are preparing our claim to recover all the costs owing which will be in the order of £1.5m. Since matters have not been resolved with yourselves or Earth Tech we request that in accordance with Cl 46.2 of the contract, the issues are discussed with us at PCMG level."

402. On 23^rd January 2001, Taylor Woodrow referred the dispute to the Preferred Contractors' Management Group (the "PCMG") under Clause 46.2 of the General Conditions of Contract. Although a PCMG meeting was held on 22^nd February 2001, satisfactory resolution of the dispute was not achieved.
403. As I have already indicated (see paragraph 301 above), during the winter of 2000/2001, the performance of the SBR was poor. Nitrification deteriorated and there were many BOD, COD and solids failures against consent standards: see graphs 1 to 4 at the end of Dr Horan's first report (E3).
404. In my view, there is no doubt that a significant factor in the SBR's rapid deterioration in performance in November 2000 was precipitated by a serious error made with regard to the anoxic multiplier set point. Either Mr Coulthard or a representative of Elga changed the anoxic multiplier set point from 0.35 to 3.5, representing a tenfold increase. It is not clear who was actually responsible for making the alteration, but it was common ground that it was done and that it was not corrected until 22^nd November. Furthermore, on 15^th November 2000, Mr Coulthard increased the top water level (the "TWL") by 200mm (Transcript, Day 6/166), which had the effect of increasing the fill time, changing the volume of the influent admitted to each tank and increasing the decant depth. Unfortunately, it is clear that Mr Coulthard did not appreciate that these changes meant that appropriate adjustments to the cycle times would also be needed, a matter that required appropriate modifications to be made to the software of the computerised control system if the increase in the TWL was to be maintained (Transcript, Day 6/167-168).
405. From 14^th November 2000, the SBR began to experience excessive aerated fills (C81/202) and from 16^th November 2000, it began to discharge solids (C81/56). As it seems to me, Mr Coulthard's approach to remedying the situation, by making the various changes to the set points that he did, was largely one of trial and error (C80/A27, C80/A28 and Transcript, Day 6/170 onwards). It is apparent that he had only a partial and imperfect grasp of the complexities of the SBR control system, complexities that stemmed from the very flexibility of the Jet Tech SBR system.
406. The hit and miss nature of such an approach inevitably caused Mr Sawyer to express some concern as to whether there was a proper understanding of the changes that were being made to the set points: see the terms of his email to Mr Holmes, dated 29^th November 2000 (C81/207), in which he posed the question: "Are we managing the changes properly?" Mr Holmes' recollection was that it was the loss of control of the SBR as a result of the anoxic multiplier episode, coupled with the inevitable time that it took to recover and the fact that winter was starting, which caused the decline in performance of the SBR (Transcript, Day 4/69). Although this is not an unfair overall assessment of the situation, it does not take account of the important fact that there was a lack of a real and complete understanding of the SBR system (and, in particular, of the appropriate changes that needed to be made to the set points from time to time) on the part of the various concerned Yorkshire Water/Earth-Tech operatives at the time. It is important to note that there is no evidence (nor was there any suggestion at the time) that any of the operational problems during the winter of 2000/2001 were due to poor sludge settleability caused by microthrix parvicella.
407. Although Earth-Tech's investigations were still largely predicated on the basis, mainly derived from Ms Henderson's erroneous presentation, that the SBR had insufficient capacity to treat full flows on either 5 or 6 tanks (see, for example, Mr Copnall's calculations of plant capacity in C82/81A and C83/150), it also appears that, in December 2000, Dr Hemphrey had come to the conclusion that the SBR lacked adequate selector capacity (C83/150 and Transcript, Day 9/125). There appears to be no record of any contemporaneous investigation/analysis of the microbiology or of the SBR's selector capability and no way of knowing precisely how or why Dr Hemphrey came to express that conclusion at this particular stage.
408. However, as it seems to me, it is significant that the suggested lack of selector capacity was specifically linked to the absence of an IDSC and the design for introducing the influent into each SBR tank, i.e. by means of the inlet chute (C83/150). As I have already stated, from as early as February 2000 both Yorkshire Water and Earth-Tech had taken the view that the omission of the IDSC, coupled with the effect of introducing influent by means of an influent chute, would have a detrimental effect on the SBR's selector capacity (see paragraph 262 above). It therefore seems to me very likely that fresh interest was being shown in or greater importance attached to this long standing opinion, possibly because of the slight deterioration in settleability that had been observed from September 2000 onwards (see paragraph 381 above).
409. Mr Copnall set out the results of Earth-Tech's investigations in an email dated 18^th January 2001 (C92/132). In it, he stated that if a selector zone were provided, the SBR should have a capacity of 105 tcmd for 6 tanks (as opposed to a capacity of 92 tcmd on the basis of its then configuration) and that at that level there was (in Earth-Tech's view) a limitation in the capacity of the SBR by reason of the decant volume. Mr Copnall suggested that to correct this it would be necessary to increase the decant depth by 500mm. However, he went on to point out that, with a selector zone, it would be possible to increase the decant depth by 200mmm and this would take the SBR capacity to 117 tcmd. He also stated that there would then be a requirement for new humus tanks to accommodate the capacity shortfall between 117 tcmd and the contractual 136 tcmd. Mr Copnall put the estimated cost of the suggested "short-term" remedial works, i.e. excluding the new humus tanks, at £1.170 million (C92/132).
410. At the same time, Earth-Tech produced a draft report on the remedial works (C84/168). The summary of the short-term remedial works included the relatively small amount of work that had been identified earlier, together with the addition of a selector (C84/170). As Mr Copnall confirmed in evidence, Earth-Tech's approach was to increase the SBR to what was considered to be its maximum capacity and then to review the position further (Transcript, Day 9/160), with the carrying out of 4 performance tests (C84/177).
411. Earth-Tech's remedial works report was discussed at a workshop meeting on 19^th February 2001 (R2/108A). Mr Copnall gave the first presentation by way of background (Transcript, Day 9/164) and after discussion it was decided that the way forward was to adopt an external selector tank of about 900 m3 to overcome the poor sludge settleability (R2/108J and K). At the same time, it appears that the ambition of achieving 105 tcmd as a result of providing the selector was revised downwards to 100 tcmd (R2/108M).
412. These various activities then led to a Change Authorisation Report a few days later (R2/118). This report identified the estimated cost of the short term works as being £1.230 million, to include a new selector tank, with the aspiration of achieving 100 tcmd, and longer term works being identified to achieve the contractual maximum of 136 tcmd. In due course, the first stage became Phase 1 of the remedial works (see below). In March 2001, BCIC gave approval for carrying out the short-term remedial works to include the external selector. BCIC also approved the proposed procurement strategy, which was to be carried out under Yorkshire Water's Asset Management Plan 3, which covers the period 2000 to 2005 ("AMP 3").
413. Works under AMP 3 are carried out by one of the contractors on Yorkshire Water's pre-selected panel of contractors. The contractors on this panel are known as Capital Solutions Partners (CaSP). Works valued at less than £2 million are usually awarded to a CaSP without any tender process, although sometimes the work is put out to competitive tender. Works valued at more that £2 million are normally subject to a competitive tendering process, using three contractors from another Yorkshire Water pre-selected panel called the Framework Contractors List. The original selection of the contractors on these pre-selected panels was by a competitive tender process for the duration of AMP 3. The CaSPs are joint ventures between various contractors and consultants (see the evidence of Mr Kevin Smith: D1/229).
414. In the event, so far as concerns Phase 1 of the remedial works, the view taken by Yorkshire Water was that it was not feasible to go through a competitive tendering process, because of the shortness of time in which to design, procure and carry out the works before the onset of the winter 2001/2002 and the fact that the works were required as soon as possible (D1/229). So it was that, in due course, Yorkshire Water decided to appoint the west area Gleeson CaSP (led by M.J. Gleeson: "Gleeson") as the contractor for the Phase 1 remedial works, because Gleeson was already established on site at the Knostrop works, where it was carrying out investigation/feasibility works in relation to the Knostrop High Priority Maintenance Scheme. For the same reason, the consultant chosen to design the remedial works was TEAM, which is a joint venture between E.C. Harris, RKL-Arup and Montgomery Watson Hazra ("Montgomery Watson"). There was also the added convenience that Montgomery Watson happened to be part of the Gleeson CaSP (D1/229-230)
415. Phase 1 of the Remedial Works. Almost at once, it was decided that the size of the proposed external selector was insufficient (C86/84). It appears from Mr Holmes' evidence that, soon after the meeting on 19^th February, there was another meeting, attended by himself, Ms Henderson and Mr John Newman of TEAM ("Mr Newman"), at which it was decided that the volume discussed at the workshop meeting was too small (Transcript, Day 4/89). It seems likely that Mr Copnall's email dated 9^th March 2001 (C86/84) resulted from that meeting. It was also Mr Holmes' evidence (Transcript, Day 4/89-90) that he, Ms Henderson and Mr Newman of TEAM had joint responsibility for the design of the first remedial works. However that may be, it appears to have been Mr Holmes who produced the design for the Phase 1 remedial works, based in part on standard Yorkshire Water criteria (Transcript, Day 4/90).
416. On 3^rd April 2001, Yorkshire Water held a meeting with TEAM to discuss the design of the proposed remedial works. Yorkshire Water was represented by Mr Parker and Mr Kevin Smith ("Mr Smith"), the current project manager, who first became involved in the project in March 2001. The representatives of TEAM who were present at the meeting included Mr Christopher Maw ("Mr Maw") of Montgomery Watson, who was described as a key design engineer (R2/270B), and Mr Copnall of Earth-Tech.
417. At the meeting, Mr Maw stated that a specialist designer within Montgomery Watson would be required and that he required information about the influent in order to design the selector and to assess whether it would improve the process and by how much (R2/270C and Transcript, Day 12/65). In an email sent to Mr Copnall the following day (R2/271), Mr Maw repeated his request for further information. Mr Maw's reason for requesting the additional information was that he had expressed doubt as to whether the selector was the answer to the perceived problems with the SBR. The selector would not deal with the turbulence of the sludge blanket during fill-decant events. The only other purpose for which the external selector might be required was to improve settleability – but, as to that, there was no current microbiological information (Transcript, Day 12/67).
418. After the meeting on 3^rd April, Mr Copnall sent an email to Mr Jones dated 4^th April 2001 (C92/153), in which he summarised what had taken place, as follows:

"Kevin (Smith) explained that Earth-Tech were there in an advisory role and to assist in providing the necessary information.

I corrected Kevin that our role was far stronger in that as project managers we had to approve any works.

We then went through the works to be undertaken in the immediate and the short term; which Kevin advised was by November 2001

Immediate

i Instrumentation for MLSS monitoring and sludge blanket levels

ii Chlorine dosing to the SBR

iii Alum dosing to the humus tanks

short term

iv Selector provision

v change the software to fixed flow software

vi rectify the movement of liquors between the tanks/change the valves

vii increase the sludge handling capacity

Chris Maw agreed to go away and come back with proposals but from his "experience" he was not convinced that selector provision would affect (sic) any improvement."

419. In his reply to Mr Copnall (C92/154), Mr Jones expressed his displeasure at not having received any confirmation of the 3^rd April meeting (which is why he did not attend) and making clear that, so far as he was concerned, the remedial measures had already been identified and the decision taken that a selector was required. It was therefore Mr Jones' view that Mr Maw did not have to consider whether a selector was needed and, as a result, a far more limited set of data was sent to him (Transcript, Day 12/71-73 and R2/277).
420. Mr Smith's evidence was to the same effect: what Mr Maw wanted to do was to go through the design from the beginning, but he was told not to do so – the reason being that it would have taken far too long and, as a result, the remedial works would not be completed before the onset of winter, leaving Yorkshire Water (so it was believed) in a "precarious position" (Transcript, Day 13/79). In fact, as Mr Streatfeild-James pointed out (Biwater's closing submissions, paragraph 258): "no-one ever went through the exercise suggested by Mr Maw. … The result was unfortunate: as Mr Holmes accepted, the external selector was thus designed without Yorkshire Water knowing whether there were a lot of filaments in the SBR, and if there were, what species those filaments comprised (Transcript, Day 6/57)".
421. This obviously unsatisfactory state of affairs was highlighted by Mr Elliott in the course of his cross-examination of Mr Holmes (Transcript, Day 6/56-57), as follows:

"Q. In the winter of 2000 and 2001, there was a settlement problem with the tanks, and there was S(uspended) S(olids) in abundant quantities and BOD and COD going out in the effluent. Agreed?

a. Yes.

Q. And if necessary we can look at Miss Amanda Schofield's third statement where she sets it out in great detail. So it was known at that time that there was a settlement problem?

A. Yes.

Q. And yet no check-up was made as to the level of filaments in the plant and no check-up was made to discover that if there were filaments there, what particular filaments they were?

A. That is apparently so, yes.

Q. So is this right: Yorkshire Water decided to go ahead and to construct this external selector when they did not know (a) whether they had a lot of filaments in their SBRs and (b), if they did, what those filaments were? That is the position, is it not?

A. Yes."

422. On 23^rd April 2001, TEAM produced a Value Scoping Report (R2/214-244). This noted that no process investigations had been required as part of the study and that a prescribed solution had been provided by Earth-Tech to achieve 100 tcmd (R2/220), i.e. the result that had been discussed at the 19^th February 2001 workshop (see paragraph 400 above). The project need was expressed in the following terms (R2/220):

"The Project Need is to attain stability of performance of the SBR with a target of 100 tcmd hydraulic throughput (future target 137 tcmd for phase 3). The phase 1 and 2 works are required to be commissioned prior to the commencement of the Knostrop WwTW – HPM scheme which relies on acceptable SBR performance."

423. Since TEAM had been told not to look into the basis for the design, it was not in a position to endorse the Yorkshire Water scheme, as Mr Smith effectively accepted in the course of his evidence (Transcript, Day 13/80). On 26^th April, Mr Copnall emailed Mr Maw and expressed himself to be generally in agreement with TEAM's "overall concept and proposals" (R2/341). Mr Copnall then proceeded to make a number of comments on the report. Included in those comments was a requirement that a COD:BOD ratio of 3.38:1 be used (apparently instead of one of about 2.2:1), a ratio that Mr Copnall derived from the Boon report (Transcript, Day 12/77). Mr Copnall also included a note that the selector should be sized for 137 tcmd. As a result, on 11^th May 2001, TEAM produced a revised Value Scoping Report, which provided for an external selector with a volume of 2000m3 (R2/308), apparently as a result of further reconsideration and discussion by Yorkshire Water and Earth-Tech (Transcript, Day 13/82).
424. In the event, during the summer of 2001, in general terms the SBR once more performed satisfactorily. Graphs 1 to 4 appended to Dr Horan's first report (E3) show this very clearly. Apart from one spot sample failure in mid-January 2001 and a borderline spot sample in early October, there were no ammonia consent failures (Graph 1) during 2001. Graphs 2, 3 and 4 show that, by the end of March 2001, the SBR's performance was beginning to improve and that there were only very few isolated SS/BOD/COD failures during the summer of 2001.
425. However, despite the improved SBR performance and the absence of any prior process investigation of the type suggested by Mr Maw (see paragraphs 413 to 414 above), Gleeson commenced carrying out Phase 1 of the remedial works in the late summer of 2001, consisting mainly of the new external selector tank situated just upstream of the SBR, at a total cost of £2,707,243.
426. On 14^th August 2001, Yorkshire Water issued its Claim Form in this action, at which date (as already indicated above) none of the required performance tests had been carried out on the SBR. It is to be noted that, as originally formulated, Yorkshire Water's claim was simply for damages for breach of contract and/or negligence. However, under the terms of the Main Contract and for the reasons given earlier in this judgment (see, in particular, paragraphs 16 to 18 above), such a claim was not actually open to Yorkshire Water at that stage.
427. However, putting to one side the question of the correct basis for its claim, it was and is Yorkshire Water's case that, at all material times, the SBR, as designed and installed by Taylor Woodrow/Biwater, was incapable of passing the following performance tests: (i) TP1 – Effluent Quality; (ii) TP2 – Flow Capacity; (iii) TP9 – Reliability and (iv) TP11 – Sludge Production and Sludge Quality (see paragraph 81 of the Particulars of Claim).
428. It was and is also Yorkshire Water's case that, if the performance tests had been attempted at any material time, the SBR would undoubtedly have failed, with what might well have been catastrophic consequences. Mr Slater suggested that this very real danger had also been fully appreciated by both Taylor Woodrow and Biwater, both of whom realised that the SBR was incapable of satisfying the relevant Performance Tests, and that it was for that reason that they had not been prepared to agree to Yorkshire Water operating the SBR at full flows and loads throughout the entire relevant period. Mr Slater emphasised that, at all material times, Yorkshire Water had been anxious and willing to carry out the Performance Tests, but had been prevented from doing so because (i) it was common ground that the SBR was quite unable to meet the requirements of the relevant Schedule 7 Performance Tests and (ii) there was a significant risk of "catastrophic consequences" as the result of the inevitable failure to meet those requirements.
429. However, whilst I readily accept that Yorkshire Water were extremely and justifiably concerned to avoid any risk of a major consent failure, I am not persuaded that Taylor Woodrow/Biwater shared Yorkshire Water's misgivings throughout the entire period, as suggested by Mr Slater. Undoubtedly, Taylor Woodrow/Biwater accepted that taking performance tests was not appropriate whilst the SBR was operating at reduced flows and loads and/or whilst it was unstable (see my foregoing account of the facts generally). However, as I have already indicated (see paragraphs 385 and 399 above) there did come a stage, in the late summer of 2000, when Biwater was anxious to increase flows and loads and proceed to performance testing, in the light of the SBR's excellent performance at the time, and it was Yorkshire Water that decided not to do so. As I have also already made clear, by that stage Yorkshire Water had decided to terminate the main contract and was determined not to take any risk of failing the relevant consent standards.
430. By 10^th December 2001, Phase 1 of the remedial works was completed and the SBR recommissioned with its new external selector tank in operation. It is important to note that, as part of Phase 1 of the remedial works, the SBR was no longer required to treat any part of the effluent from the ASP. The influent to the SBR was and has thereafter remained confined solely to the originally specified proportion of the effluent from the high-level primary tanks (i.e. 33% of the high-level primary tank flow and load, up to a maximum of 76 tcmd). It is also important to bear in mind that, for these reasons (i.e. the addition of the external selector tank and the modified influent), the SBR was now markedly different to that originally designed and installed by Taylor Woodrow/Biwater.
431. However, notwithstanding the Phase 1 remedial works, the performance of the new SBR during the remaining period of the winter of 2001/2002 was poor. Nitrification was not satisfactory and there was a significant incidence of BOD, COD and solids consent failures: see graphs 1 to 4 attached to Dr Horan's first report. BOD loads remained low at under 2 tonnes/day in mid-January 2002, rising to around 2 – 4 tonnes/day thereafter (see Graph 5, E3). In my view, having regard to the extent and nature of the modifications to the original SBR (see paragraph 431 above), it cannot be assumed or inferred that the SBR's poor performance after recommissioning was attributable to any defect or design deficiency in the original SBR, nor does it demonstrate that the original SBR would have been unable to meet the Schedule 7 performance requirements.
432. Within a month of its recommissioning, it became apparent that (despite its external selector tank) the new SBR was suffering from serious problems, caused by the presence of large numbers of microthrix parvicella (Transcript, Day 6/58-61). The presence of significant numbers of microthrix parvicella was confirmed in microbiological reports, which had been commissioned by Yorkshire Water and started in November 2001 at about the time new SBR was recommissioned (F9/24).
433. At first it was thought that a diesel spillage might be responsible for the new SBR's problems. However, at an SBR review meeting held on 20^th March 2002, it was acknowledged that the microscopy analyses carried out at the time showed that the damage done by the spillage was not significant, although some adverse effect on flocculance and solids' levels was noted (C92/213A). At the same meeting, it was also suggested that the new selector tank might be responsible for the development of microthrix parvicella, because at the time the selector tank had been designed it had not been thought that filaments were a problem and that microthrix parvicella might have been there all the time (C92/213A-B).
434. Because of the concerns about how the new selector tank was performing, Mr Holmes went back to basics and did some research on selectors, something he had not done when he had been designing the external selector about a year earlier – rather he had relied on the fact that, at a meeting held in December 2000 to consider the microthrix parvicella problems at the Adwick and Swinton plants, the various experts had agreed that the appropriate solution was a selector (Transcript, Day 4/97-100).
435. The result was that the flows through the SBR were greatly reduced. By June 2002 it was said that the flow was limited to 60 tcmd (R5/109). By July 2002, it was being reported that the SBR could only take 50 tcmd (C91/164). Thus, far from enabling the flow to the SBR to be increased from 92 tcmd to 100 or 105 tcmd and possibly more (see paragraphs 405-407 above), the Phase 1 remedial works had apparently resulted in a significant reduction in the flow that could be taken.
436. The Phase 2 (Part 1) Remedial Works. Although it appears that Mrs Williamson may have been given the task (jointly with Mr Holmes) of investigating and producing a justification and scope for further remedial works in January 2002 (R5/27), she was unable to throw any light on the matter, other than to suggest that Mr Holmes would have "put together a paper. He would show me the paper, but I would not be involved in putting it together" (Transcript, Day 13/41). In the event, during the summer of 2002, further remedial works were carried out, i.e. the first part of Phase 2 of the remedial works, consisting mainly of the replacement of the decanter arms. At the same time, the SBR control software and the recirculation pumps were replaced and modifications carried out to the decant valves and actuators. These remedial works were designed and carried out by ETM, a joint venture between Earth-Tech and Morrison Construction Limited, another of Yorkshire Water's CaSPs, at a total cost of £2,076,720.
437. According to Mr Copnall, there had been investigation of the new SBR's perceived problems and appropriate workshops had been held in about April 2002 (Transcript, Day 12/90). However, no documentation relating to these investigations/workshops was forthcoming at the trial, although it was claimed that it was as a result of these that the further remedial works were identified. At all events, it appears that the decision to install new decanters had been made by 14^th June 2002 and Mr Copnall's recollection was that the decision to do so was made by the Yorkshire Water team, rather than by any particular individual (Transcript, Day 12/94).
438. At the time, the flow to the new SBR was apparently limited to 60 tcmd and the decant depth was 1.1m (R5/109). At this stage, it is not entirely clear what benefit Yorkshire Water/Earth-Tech thought would be achieved by changing the decanters. From Mr Smith's note of a meeting with ETM held on 21^st June 2002, it appears to have been generally accepted that the proposed new decanters would not result in any process improvement; rather it seems to have been thought that some increased capacity would be achieved by increasing the distance between the decanters and the top of the sludge blanket, thereby increasing the effective decant depth (R5/110F). It also appears to have been the intention to carry out the work to the decanters as rapidly as possible, so that it would be completed before the winter of 2002/2003.
439. The decision to go ahead with the Phase 2 (Part1) remedial works appears to have been made in principle at the meeting on 21^st June. Mr Smith submitted his Solution Authorisation Report at a BCIC meeting on 2^nd July 2002, seeking authorisation to go ahead with the Phase 2 (Part 1) remedial works (R5/142-145), and was told by BCIC to get on with the works as soon as possible and that formal written approval would follow later (D1/240-241). On the 2^nd July 2002, Mr Smith emailed ETM to confirm that BCIC had authorised the works and giving instructions for the work to proceed a soon as possible (R5/126). ETM then commenced the necessary design and procurement work (D1/241).
440. An accurate and helpful summary of what then took place is set out in paragraph 271 of Biwater's closing submissions, which I cannot improve and therefore gratefully adopt with some additions of my own, as follows:

"(1) On 9^th July 2002, Mr Aspinwall of ETM produced a document dealing with the design of the decanters (R5/155). Mr Copnall referred to it as a process statement (Transcript, Day 12/101). This addressed the additional depth which could be gained by reference to the CASS decanters which were proposed; it was said to be 100mm. (In addition, Mr Aspinwall suggested that, in the case of the CASS decanters, the area of flow nearest the sludge blanket was many times greater than for the Jet Tech decanters and that, therefore, the flow velocity nearest the sludge blanket would be proportionally less for the CASS system, further reducing the risk of solids carry-over).

(2) Mr Copnall commented on this on the same day (R5/152): "As written it reads as (if) there is no real advantage in replacing the decanters". He suggested some amendments (by a combination of deletions and additions), including the suggestion that the CASS decanter drew treated effluent "from just below the water level".

(3) The issue which was being wrestled with is clear from R5/170: the Jet Tech design (bottom of the page) drew water from 500mm below the surface, while the CASS design allowed water into a trough at surface level, but only after it had passed below a scum baffle which was below the water level (see also Transcript, Day 12/104 and ff). What was being discussed was the fact that the CASS system did not achieve a saving of 500mm (i.e. the difference between the surface water level and the point of draw of the Jet Tech decanter) because the scum baffle extended below the water surface.

(4) This is made clear by Mr Kirkwood's email of 11 July 2002 (R5/154) where he indicated that Mr Copnall's suggested wording was misleading: he wrote, "…In fact, the effective draw off for both types of decanter will be broadly similar in relation to top water level". (Mr Kirkwood went on to identify the main benefit of installing the CASS decanters as Mr Apinwall's additional point, namely "the draw-off velocities will be much less, greatly reducing the risk of entraining solids from the sludge blanket"). He suggested avoiding any references to specific depths, and leaving the process description vague.

(5) Surprisingly, and completely implausibly, Mr Copnall maintained that Mr Kirkwood was agreeing with him, as opposed to pointing out that his corrections were liable to mislead: Transcript, Day 12/108-110.

(6) Mr Copnall had a discussion with Yorkshire Water on 12 July 2002, but he could not say with whom or what he talked about: Transcript, Day 12/110-112. It is evidenced by his email to ETM the following Monday (15 July 2002; R5/160A), in which he recorded that the BWL (Bottom Water Level) should be set at 5.5m, and that Yorkshire Water had "taken into account other issues" in determining to use surface draw decanters. What those "other issues" were, Mr Copnall could no longer say (Transcript, Day 12/111).

(7) On 15 July 2002 Mr Smith contacted Mr Jones to check that a decision to replace the decanters was one which could be justified (R5/160A). this apparently caused the matter to be investigated (in any detail) for the first time: see C92/232 and C92/233.

(8) The result was a report which was produced by Mr Copnall on about 17 July 2002 (R5/169), justifying the (CASS) decanters on 2 grounds: additional decant depth of 500mm and reduced flow through the decanters."

441. The report, to which reference is made in paragraph 271(8) of Biwater's closing submissions (see above), appears to have been drafted by Mr Jones and produced by Mr Copnall (Transcript, Day 12/114). The intention was to provide the justification for the proposed installation of the new decanters (i.e. the CASS decanters). However, I am satisfied that the two reasons given as justifying the proposed replacement of the Jet Tech decanters are both unsound.
442. The suggestion that there would be a gain of 500mm in decant depth (the first reason) had already been shown to be erroneous by Mr Kirkwood (R5/154: see Biwater's closing submissions paragraph 271(4) quoted above). In reality, the draw-off point for both types of decanter was "broadly similar" (R5/154).
443. So far as concerns the flow velocity through the decanters (the second reason), the proposition that the Jet Tech decanters generated a draw-off velocity of 4m/second was basically incorrect. As Mr Copnall appears to have known at the time, Yorkshire Water's own investigations, carried out at some point during 2001, had indicated that the velocity of the flow through the Jet Tech decanters was about 1.35m/second at the decanter nozzles and that its effect rapidly diminished from a distance of about 0.1m from the decanter nozzles, until reaching a constant of somewhere just below 0.15m/second (I1/260: Transcript, Day 12/115-116).
444. On 3^rd July 2002, ETM and Yorkshire Water met to discuss the Phase 2 (Part 1) remedial works (R5/129-130). It was made clear that, since the design was Yorkshire Water's, ETM would not be required to guarantee any improvement in process performance (Transcript, Day 12/117-118), although ETM did agree to provide a statement concerning the likely effect on process, all of which was recorded in the minutes of the meeting, as follows (R5/130):

"1.3 Discussion on process improvements. It was noted that ETM would not be required to guarantee improvements in process performance, although YWS expected this to occur. It was agreed that improvements would be on a best endeavours basis. ETM to issue a statement regarding their view of likely effect on process performance following the installation …"

445. In the event, ETM provided the promised statement on 11^th September 2002 (R5/247-249). In my view, it is not a satisfactory or persuasive document. Little original or updated thought appears to have been given to its preparation. As Mr Streatfeild-James observed (Biwater's closing submissions, paragraph 275), it is a reworked version of the Aspinwall draft (see paragraph 441 above), together with a sketch showing the decanter arrangement dated 18^th July 2002.
446. At some stage during the period January to June/July 2002, Yorkshire Water decided that, after the Phase 2 (Part 1) remedial works were completed, the capacity of the recommissioned SBR was to be 76 tcmd (see, for example, R5/229) although, as Mr Copnall appeared to accept in the course of his evidence (Transcript, Day 12/123-124), the new decanters were sized to take flows of 136 tcmd. No explanation was forthcoming as to who made the decision as to the capacity of 76 tcmd for the modified SBR, or upon what basis.
447. However, it is to be noted that, as Mr Copnall confirmed (Transcript, Day 12/122), the procurement process had started in relation to the Phase 2 (Part 2) remedial works in July 2002. The main element of these further remedial works was a Biological Aerated Filter (a "BAF": see paragraph 133 above), sized at 60 tcmd. The BAF was intended and designed to deal with the ASP effluent that was to have been but no longer was being treated by the original SBR, following the completion of the Phase 1 remedial works (see paragraph 426 above). Of course, after allowing for the 60 tcmd to be treated by the proposed BAF, the figure of 76 tcmd represents the balance of 136 tcmd, i.e. the maximum flow for which the original SBR was intended to provide treatment.
448. The modified SBR process was re-commissioned during November/December 2002, using seed sludge containing a high proportion of nitrifiers to reduce the time required to reach acceptable levels of effluent quality. By mid-December 2002, the SBR was producing fully-nitrified effluent: see paragraph 6.1 of Dr Chamber's first supplemental report dated April 2003 (E1A/tab1/21).
449. On 18^th October 2002, the Claim Form was amended and reissued pursuant to the Order of His Honour Judge Humphrey LLoyd QC. By its amendment Yorkshire Water deleted the whole of the original claim for damages for breach of contract and/or negligence and substituted for it the following claim:

"Payment of all sums due under the terms of a written contract dated 19 August 1998 and made between the Claimant and the Defendant, namely the cost of certain remedial works pursuant to clauses 34.8 and/or 36.7 and/or 36.11 and/or 35.10 and/or Schedule 10 thereof; and/or A declaration that:

a. there was a Defect in the Plant provided by the Defendant (within the meaning of the aforesaid contract); and

b. the Claimant was entitled at 13 August 2001 and/or at some time thereafter (under the terms of the contract) to make good the Defect; and

c. in relation to each of those items which the Claimant proposes to carry out in order to make good the Defect, the Defendant is liable under the terms of the contract to pay the reasonable cost thereof once each such item has been completed; …"

450. Having regard to my conclusions with regard to a number of the contractual issues in this case (as to which, see above), it is important to note that, although an amendment takes effect from the date of the original document that it amends (see CPR 17.3.4), there is no absolute rule of law or practice which precludes an amendment so as to rely on a cause of action that has accrued only after the date of the original claim in circumstances where (but for the amendment) the claim could fail: see Maridive & Oil Services (SAE) v. CNA Insurance Co. (Europe) Ltd (2002) EWCA Civ 369, where the relevant principle is stated in the following succinct terms (per Chadwick LJ at paragraph 54):

"There is no absolute rule of law or practice which precludes an amendment to rely on a cause of action which has arisen after the commencement of the proceedings in circumstances where (but for the amendment) the claim would fail. The court has a discretion whether or not to allow the amendment in such a case; a discretion which is to be exercised as justice requires. In the present case I have no doubt that, had the claimants sought to amend their particulars of claim (so as to rely on the demand of 13 March 2000) within the period from 12 April to 30 August 2000, they should have been permitted to do so. There was no reason why they should have been required to commence new proceedings."

451. As already stated, it was not until February/March 2003 that Yorkshire Water carried out any form of "performance test" on the modified SBR. That test was carried out with 5 tanks in operation and showed the effluent quality to be within the standards required by the Main Contract: see paragraphs 6.2 to 6.4 of Dr Chambers' first supplemental report. However, in my opinion, that particular test cannot be said to be equivalent to a performance test carried out under the terms of the main contract. Not only was the modified SBR markedly different from the SBR as designed and installed by Taylor Woodrow/Biwater, but the maximum flow tested was 76 tcmd, as opposed to the maximum of 136 tcmd specified by the main contract.
452. The performance and capacity of the SBR as originally designed and installed. In this case, as I have already made clear, it is very important to bear in mind that this is not a claim by Yorkshire Water for damages for breach of contract. That approach was abandoned for good reason in October 2002 (see paragraphs 12, 16 to 20, 28, 87 and 450 above). For the reasons given earlier in this judgment and in my judgment in the Main Contract Preliminary Issues, this case is almost entirely concerned with Yorkshire Water's contractual entitlement to recover its relevant remedial costs, pursuant to the provisions of Schedule 10 of the Main Contract (see paragraph 86 above). However, even if (contrary to the way that I have construed the main contract) Yorkshire Water is entitled to rely upon any of the other contractual provisions identified in the Particulars of Claim, I am satisfied that the outcome of these proceedings remains unaffected for the reasons given in the latter part of this judgment.
453. Accordingly, the central issue is whether Yorkshire Water has proved that Taylor Woodrow was "unable to meet the requirements of Schedule 7", in respect of the relevant absolute guarantees (i.e. Guarantees TP1 (Effluent Quality), TP2 (Flow Capacity), TP9 (Reliability) and TP11(b) (Sludge Quality), so far as concerns the SBR, as originally designed and installed. I therefore accept Mr Streatfeild-James's submission (Biwater's closing submissions paragraph 280) that it is not enough to show that at certain periods the SBR performed poorly in terms of its effluent. The nature of the original SBR was such that any failure in terms of effluent could be due to any one or more of a wide variety of causes. The evidence of poor performance has to be considered and the reasons for its occurrence identified. Only then will it be possible to say whether Yorkshire Water has proved that, at all material times, Taylor Woodrow was unable to meet the relevant performance requirements of Schedule 7.
454. Accordingly, as it seems to me, Mr Slater's "Unanswerable Point" (see Yorkshire Water's closing submissions, paragraphs 461 to 474) is not as persuasive as at first it appeared to be. Although it can be argued that, in the light of the modified SBR's recent performance history (in particular, the results of the "performance test" carried out in February/March 2003), "full compliance" in terms of effluent quality can be and is now being achieved all the year round, this has only been shown to be the case in relation to the treatment of a differently constituted influent at reduced flows by a significantly modified SBR (in effect, a CASS with an external selector tank).
455. The matter only has to be stated thus for it to be apparent that it does not follow, from the fact that the modified SBR is performing satisfactorily with regard to a different influent at reduced flows, that the original SBR would have been unable to meet the relevant performance requirements, because it performed poorly in the winters of 2000 and 2001.
456. As it seems to me, in order to come to any reliable conclusion about the treatment capabilities of the original SBR, it is essential to identify the cause or causes of its poor performance during those two winter periods. With that in mind, I take the view that the following general conclusions, drawn from my detailed account of the facts as I find them to be (see above), represent the essential first stage in that process.

(1) Having passed its take-over tests successfully in November 1999, the SBR's poor performance between then and the early part of 2000 was due to the combined effect of a number of different factors, as summarised in paragraph 257 above. In my view, none of those factors can be said, ipso facto, to be suggestive of a significant and/or irremediable defect or fundamental design deficiency in the SBR.

(2) From February to April 2000, the main operational problems were due to an inability to reduce high MLSS levels and to the high sludge production that was being experienced, coupled with the consequences of the operational decision to pursue a deliberate strategy of concentrating on reducing MLSS levels at the expense of nitrification. However, there is also evidence that, by this stage, microthrix parvicella was present in sufficient quantities to cause some moderate settlement problem in two of the tanks.

(3) Microthrix parvicella was still present during the period April to July 2000, but the SBR's performance nevertheless improved steadily and nitrification was restored. However, there were continuing issues about maintaining target MLSS levels and achieving plant stability. As well as its continued very low BOD loads and high sludge production, further issues emerged with regard to the nature of the influent which showed that the influent was significantly different from that which had been specified, i.e. the anomalous DO values, the high COD:BOD ratio and the temporary phenomenon of BOD disguised as COD.

(4) From August to November 2000 the SBR performed well and produced excellent quality effluent. Although flow caps had been imposed, these had little practical effect. Dr Zuber's recovery strategy had clearly been successful. There is no evidence that microthrix parvicella was present in sufficient numbers to cause any settlement problems during this period. By the autumn of 2000, Biwater was anxious to proceed to performance testing, but Yorkshire Water/Earth-Tech were not yet persuaded that the plant was stable. However, if performance tests had been carried out, I am satisfied that there is every reason to believe that the SBR would have passed them satisfactorily (see below).

(5) During the winter 2000/2001, there is no evidence of significant numbers of microthrix parvicella being present. However, an operational error resulted in control of the SBR being lost in November 2000. Recovery of control was delayed by further operational errors and lack of operational experience and proper understanding on the part of Yorkshire Water/Earth-Tech operatives. By the end of March 2001, control had been recovered and the SBR's performance had started to improve.

(6) During the summer of 2001, with the flow caps still in place, the SBR performed well prior to the commencement of the Phase 1 remedial works.

(7) During the winter of 2001/2002 the Phase 1 remedial works were completed and the external selector commissioned. However, the recommissioned SBR performed badly from the outset and, for the first time, there is clear evidence of a significant infestation of microthrix parvicella.

457. As it seems to me, the foregoing conclusions clearly suggest that two of the main matters that need to be considered in relation to the operational problems with the SBR during the period in question are (i) the difficulty experienced in achieving and maintaining the design/target MLSS levels and (ii) the SBR's selector capability, together with the related question of whether microthrix parvicella was the cause of sludge settleability problems and, if so, to what extent. I therefore now turn to consider each of these matters in turn and, in particular, whether either was the result of defects or design deficiencies in the SBR.
458. Control of the MLSS. It was Biwater's case that it was essential to control the F/M ratio by matching the MLSS levels to the incoming BOD load (the appropriate F/M ratio being a vital aspect of kinetic selection, leading to good sludge settlement: see paragraph 143 above). It was Dr Zuber's evidence that there is a standard rule of thumb in conventional plant operation that the sludge settlement required for good performance will be normally be achieved by making appropriate adjustment to the MLSS levels in order to match the incoming BOD load and that this is done by monitoring and reacting as appropriate to the developing trends in the BOD load (Transcript, Day 14/52-53). Although this was challenged by Yorkshire Water, for the reasons given in paragraph 369 of Biwater's closing submissions, I am satisfied that Dr Zuber's evidence is correct and I accept it as such.
459. I also accept Mr Streatfeild-James' submission that MLSS control is closely related to sludge production. The amount of MLSS wasted is dependant on the amount of sludge produced. If plant stability is to be maintained, MLSS wastage will increase as sludge production increases (Biwater's closing submissions, paragraph 373). I am satisfied that, in the year following take-over, sludge production exceeded all expectations of those operating the SBR. Sludge yields were well in excess of normal values to be expected from an activated sludge plant: see the figures given in the final column of exhibit TW6, which I accept accurately sets out the facts relating to sludge production/yields during the relevant period.
460. For the reasons given in paragraphs 376 to 379 of Biwater's closing submissions, I do not accept Dr Chambers' rather non-specific criticisms of exhibit TW6 and the data it contains. Nor do I accept the suggestion that the large volumes of sludge resulted from the wastage of dilute sludge, caused by a process known as "rat-holing". As Mr Streatfeild-James observed, "rat-holing" was a very late suggestion and there is no evidence that such a phenomenon was actually observed at the time, nor was it addressed or treated as a credible explanation for the high sludge yields that were being experienced (Biwater's closing submissions, paragraph 380).
461. Accordingly, I accept Mr Streatfeild-James' submission (Biwater's closing submissions paragraphs 383-387) that the high sludge yields were obviously caused by the high and erratic TSS loads that were received by the SBR during 2000, which exceeded all expectation and for which Taylor Woodrow/Biwater were not responsible (see paragraphs 338-345 above).
462. I am therefore satisfied that it was the unexpectedly high TSS loads that caused the high sludge yields, which in turn made sludge wastage and the control of MLSS levels much more difficult, a difficulty that was compounded by the significant variability in the levels of the TSS loads and the failure of the Yorkshire Water operatives to conduct appropriate sludge wastage regimes (see the criticisms to the latter effect as set out in paragraphs 388 to 397 of Biwater's closing submissions, which I accept to be both accurate and fair).
463. As I have already indicated (see paragraphs 245-247, 352 and 382 above), the apparent phenomenon of inter-basin transfer of MLSS was not the result of any defect or design deficiency in the SBR (see also paragraph 401 of Biwater's closing submissions). Accordingly, the difficulty in controlling the MLSS levels in the SBR was not due to any defect or design deficiency in the SBR or its sludge handling system. It was the result of abnormally high and erratic TSS loads coupled with operational errors, inexperience and a lack of understanding of the Jet Tech SBR process. I am fortified in that conclusion by what happened when Mr Molloy took control of the decision-making in relation to the sludge wastage set points in September 2000 (see paragraphs 383-384 above). I am satisfied that, during the period that Mr Molloy controlled the appropriate wastage set points, the sludge handling system was found to have ample capacity to remove the volumes of sludge necessary to allow the MLSS in the SBR basins to be brought down to and maintained at the appropriate target levels, despite the additional challenge presented by the low BOD loads, the high COD:BOD ratio and the high and erratic TSS loads in the SBR influent.
464. In summary, therefore, on this aspect of the matter, I am satisfied as to the following.

(1) During 2000, the SBR's sludge yields were very high;

(2) The high sludge yields were caused by unexpectedly high and erratic levels of influent solids for which Taylor Woodrow/Biwater were not responsible;

(3) The high solids' levels also made it difficult to control the MLSS levels in the SBR basins;

(4) The difficulties in controlling the MLSS levels were compounded by Yorkshire Water's failure to waste in accordance with the regime set out in the O&M manuals;

(5) Despite the abnormalities in the influent, it was possible to control the SBR's MLSS levels and to achieve and maintain stability, as Mr Molloy demonstrated in September/October 2000;

(6) Although there were instances of inter-basin transfers of MLSS, they were infrequent and were the result of operational errors (such as failing to close an open valve). In any event, such inter-basin transfer of MLSS as did occur was not the cause of the difficulty experienced in controlling the SBR's MLSS levels and maintaining stability. As already indicated, this was entirely due to the high levels of TSS in the influent and to operational errors/inexperience.

465. Selector Capacity and Microthrix Parvicella. The first of the two main planks in Yorkshire Water's case that, at all material times, the SBR was incapable of meeting the relevant performance requirements, was its claim that that a root cause of the SBR's operational problems was the lack of adequate or sufficient selector capacity, as a result of which it suffered from poor sludge settleability: see paragraphs 78.3, 79.1 and 85.3 of Yorkshire Water's Particulars of Claim (A1/tab2/36-37 and 40), Response 90 of Yorkshire Water's Further Information (A1/tab3/35) and paragraph 101F(2) of Yorkshire Water's Reply (A1/tab5/66). In paragraph 2.2.1 of his first report (E1/3), Dr Chambers put the matter thus:

"In the year following take-over, it is my opinion that the SBR process failed to produce effluent of the required quality because of poor sludge settling characteristics and a lack of treatment capacity. These problems were associated with the absence of the IDSC, a lack of selector capacity, defective process control software, unreliable M&E equipment and inadequate sludge treatment capacity …"

466. Responses 89 and 90 of Yorkshire Water's Further Information (A1/tab 3/35) make it clear that two main reasons for the SBR's alleged lack of selector capacity are put forward, namely (i) the omission of the IDSC and (ii) as set up at an average flow (85 tcmd), the fill was not all anoxic, instead there were substantial periods of aerated fill. I will deal with each of these two matters before turning to consider the associated issue of whether and, if so, to what extent sludge settleability was adversely affected by microthrix parvicella.
467. As Mr Streatfeild-James observed, the starting point is one of definition. As I have already stated (see paragraph 139 above), a selector is concerned to provide competitive advantage for floc-forming microbes over filamentous species. There are two types of selection process: (i) kinetic selection (the feast/famine regime: see paragraphs 141 to 143 above) and (ii) metabolic selection (using nitrates to metabolise food in the absence of oxygen: see paragraph 144 above). Accordingly, I accept Mr Streatfeild-James' submission that selector capacity is not necessarily provided by the construction of a specifically identifiable region or feature (either within or outside the treatment plant itself) that can be termed a "Selector" (Biwater's closing submissions, paragraph 457). In the circumstances of this case, the significance of that submission cannot be overstated.
468. Certainly so far as concerns kinetic selection, the process of selection (which itself defines what is meant by the expression "a selector") is a regime that ensures the promotion of feast/famine conditions, or as more fully defined by Dr Johnstone in his first report (E2/24-25, paragraph 119):

"Some filamentous bacteria can be "selected" out of the biomass by application of a selector, which can either be an internal or an external selector. A "selector" is not necessarily a specific item of equipment, although it can be. A selector is a regime where the sludge is subjected to a rapid uptake of organic material from the sewage …usually followed by a relatively long period when the organic matter is oxidised to the extent that the sludge is "starving" and ready to accept another feed of organic matter. It is conveniently thought of as a "feast to famine" regime. The feast part can be designed to happen in a separate compartment but it does not have to be so, as long as the conditions allow a rapid uptake of "food". Selectors are aimed at providing conditions to "select out" filamentous bacteria. The expression is more qualitative than quantitative. There are no numerical values ascribed to selector capacity. …"

469. I therefore accept Mr Streatfeild-James' submission (Biwater's closing submissions paragraph 457) that it would be wrong to assert that a plant has insufficient selector capacity simply because an identifiable physical feature (such as an ID or an IDSC) is absent or, by a parity of reasoning, because a particular feature is present (such as the influent "chute") in the plant in question. I agree with Mr Streatfeild-James that what is required is an analysis of the presence or absence of conditions or regimes that promote either kinetic or metabolic selection in the plant in question or a combination of the two (Biwater's closing submissions, paragraph 458).
470. However, as Dr Chambers observed in his first report: "Selector design is not an exact science and the mechanisms involved in selection remain incompletely understood" (E1/21). Accordingly, because of the unpredictable nature of selectors, it may be difficult to identify the existence of an appropriate selector regime, particularly since the existence of one form of selection in a plant may make the presence of the other less important (see the evidence of Mr Eikelboom: Transcript, Day 31/46-48). I therefore agree with Mr Streatfeild-James' submission that where, as here, there is operational experience of a selector, the best evidence of its effectiveness is an analysis of whether the desired consequences of selection actually came about. In this context, it is important to stress the following two matters: (i) the fundamental purpose of selection is to select out the selectable filamentous bacteria (see paragraphs 139 and 463 above) and (ii) microthrix parvicella is a filamentous species that is not selectable (see paragraph 140 above).
471. As I have already indicated (see paragraph 141 above) feast/famine regimes are promoted by the creation of conditions in which an appropriate ratio of food (BOD) to micro-organisms (MLSS) is achieved. This is known as the "floc load" or "so/xo". The right conditions exist to create a feast/famine regime (i.e. a kinetic selector regime) when an appropriately high floc loading can be created. In this context, it is important to note that it was Mr Norcross' unchallenged evidence that a rule of thumb value for a beneficial floc loading is 75mg BOD per gramme of MLSS (D4/77 and Transcript, Day 23/99). As it happens, this was also the value used by Yorkshire Water in the design and construction of the external selector, a fact that strongly supports Mr Norcross' evidence on this point.
472. It is clear that the concept of floc loading was actually developed to assist in the design of defined selector regions, such as the external selector that forms the main part of the Phase 1 remedial works at Knostrop. Accordingly, as Mr Streatfeild-James observed (Biwater's closing submissions, paragraph 460), the absence of a defined selector region in the original SBR means that, although the theory of promoting a high floc loading still remains valid in principle for such a treatment plant, the calculation of an exact floc loading in respect of the SBR is not without its difficulties: see paragraph 59 of Mr Eikelboom's first report (D4/19). However, in the same paragraph of his first report, Mr Eikelboom expressed the view that: "Considering the loads mentioned in the contract and 10% mixing with the settled sludge, a floc load of about 75mg BOD/g MLSS can be calculated, a loading level that would have been high enough to create the selective pressure that favours floc forming bacteria." I accept that this particular calculation is sound in principle and that it indicates that the SBR did have an adequate selector capacity (i.e. that the SBR was theoretically capable of providing an adequate kinetic selector regime).
473. However, although it is possible to undertake a theoretical assessment of the as designed and as built selector capacity of a treatment plant, I accept Mr Streatfeild-James's submission that the question as to whether a particular plant has sufficient selector capacity will ultimately require empirical analysis (Biwater's closing submissions, paragraph 461). As Mr Streatfeild-James succinctly put it: "In short, did the SBR suffer from the presence of filamentous bacteria susceptible to a selector regime?" The real proof of the presence or absence of an adequate selector regime in the as built SBR is whether it prevented the proliferation of filamentous bacteria that would otherwise have been susceptible to selection at the time. Again, it is important to stress that microthrix parvicella is a species of filamentous bacteria that cannot be controlled by conventional selection.
474. It is also important to bear in mind the following points: (i) the mere presence of filamentous bacteria in sludge does not imply that a proper selector is missing; in fact, the presence of filaments in controlled numbers is a clear indication that there is a working selector regime in place (see Dr Cannon's evidence Transcript, Day 22/126 and the evidence of Mr Norcross Transcript, Day 24/96); (ii) the effect of a small population of filaments on sludge settleability is limited; they only become a problem when they reach a critical mass (Dr Cannon: Transcript, Day 22/141); and (iii) the presence of some filaments can be positively beneficial since they can strengthen the flocs which develop in the sludge (Dr Cannon: Transcript, Day 21/81).
475. I accept that, if a wastewater treatment plant does not work well (particularly if the apparent problem is one of poor sludge settleability), filamentous identification is almost indispensable for making a correct diagnosis (see paragraph 61 of Mr Eikelboom's first report: E4/20). Unfortunately, in the present case, the microbiological evidence during the operational life of the SBR before the remedial works is sparse and, in any event, the design and installation of the external selector (the Phase 1 remedial works) went ahead without any consideration of the microbiology, despite the advice from TEAM that a full process investigation should be carried out (see paragraphs 418-422 above).
476. Accordingly, I accept Mr Streatfeild-James' submission that Yorkshire Water's external selector scheme went ahead without any proper consideration as to whether it was required or what it was intended to achieve, much less whether it would achieve it: see Biwater's closing submissions, paragraph 464.
477. I also accept that, to the extent that there is any microbiological evidence, it strongly supports the suggestion (and the theoretical calculation) that the original SBR did not lack selector capacity. Stated shortly, the existing microbiological evidence shows the following: (i) there is no evidence of any significant quantities of selectable filamentous micro-organisms in the MLSS during 2000; (ii) the only filamentous micro-organism of which there is evidence is microthrix parvicella, which is not selectable; and (iii) there is evidence of floc-forming bacteria in the MLSS.
478. In addition, as Mr Streatfeild-James pointed out (Biwater's closing submissions paragraph 466), there were significant periods of good operation during 2000 (i.e. from June to November), during which the SSVI values were good and there was no concern about filamentous bacteria of any type. In my view, this could only have occurred because the SBR did have a selector capacity. Furthermore, I also accept that, given its non-selectable nature, the presence of microthrix parvicella during this period (to the extent that it did occur) is not evidence of a lack of selector capacity.
479. Accordingly, I am satisfied that there is no evidence of the presence of selectable filaments during this period and good reason to consider that they were not present. I therefore accept Mr Streatfeild-James' submission that the evidence from the operation of the SBR in 2000 does not support the proposition that it lacked selector capacity (Biwater's closing submissions, paragraph 468). In my view, the SBR's operational history during 2000 strongly suggests that the SBR did have adequate selector capacity and I reject Mr Slater's submissions to the contrary effect.
480. The IDSC. Dr Chambers was firmly of the view that the omission of the IDSC was a significant cause of what he believed to be the SBR's lack of selector capacity: see, for example, paragraphs 2.2.1 and 6.3.7 of his first report (E1/3 and E1/22: quoted respectively in paragraphs 466 and 197 above) and his evidence in cross-examination (Transcript, Day 27/32-34). However, in the light of my understanding of the relevant process principles relating to selection, it is not easy to understand the basis upon which Dr Chambers arrived at that conclusion. In the event, for the reasons that appear below, I have come to the conclusion that he is wrong and I reject his evidence on this aspect of the matter.
481. Although the IDSC is described as one of the "key elements" of the Omniflo system in the USF/JetTech technical literature (C11/9-14 and see also paragraph 109 above), its "key" role is not specifically linked to any selection process. Anoxic Fill is said to be via the IDSC, which ensures the even distribution of the influent (C11/12), whereas Aerated Fill and mixed Anoxic Fill are stated to be via the Vari Cant jet aerators (C11/12). So far as concerns the actual selector regimes that are identified in the same literature, these are clearly stated to be achieved by varying the "operating strategy" so as to achieve the appropriate selection conditions, as follows (C11/10):

"Varying the operating strategy enables aerobic, anaerobic or anoxic conditions to be achieved. Precise control of these conditions allows Organism Selection to take place – the proliferation of specific desirable micro-organisms is encouraged, while the growth of undesirable micro-organisms is inhibited."

482. As it seems to me, on a fair reading of its entire text, the USF/JetTech technical literature does not lend any real support for the suggestion that installation of the IDSC is essential to ensure that a USF/JetTech SBR has adequate selector capacity. Furthermore, as I have already indicated in paragraph 199 above, it was Mr Norcross' evidence (which I accept) that Jet Tech has built many successful industrial SBRs that did not incorporate the IDSC.
483. As Mr Streatfeild-James observed (Biwater's closing submissions, paragraph 474), Dr Chambers' overall opinion as to the design error constituted by the omission of the IDSC appears to be based on the following propositions (see also paragraphs 197-198 above): (i) the influent chute did not bring the "food" into contact with the sludge in a uniform manner; mixing is turbulent and localised, with the result that there was no predictable selector capacity (E1/22); (ii) there were no calculations or any other method of estimating what proportion of mixing was achieved as a result of filling an SBR basin via its influent chute (Transcript, Day 27/36); (iii) there was no or no sufficient contact between the feed and the MLSS (Transcript, Day 27/38); and a substantial degree of aeration was occasioned during periods of so-called anoxic fill as a result of air entrainment (Transcript, Day 27/45).
484. However, as part of his evidence, Mr Norcross produced calculations that clearly showed that the influent chute arrangements promoted the necessary high floc loading required for satisfactory kinetic selection (i.e. 78mg BOD/g MLSS at 1.3 DWF), particularly in comparison to that which would have been achieved by use of an ID (where the resulting floc loading for the design BOD would have been 16mg BOD/g MLSS and only 9mg/g for the actual influent as received: D4/77). Although Mr Norcross' calculations were not carried out contemporaneously, I am satisfied that they were based on reasonable estimates of the degree of contact between the feed and the MLSS and that they took into account all other relevant factors. In my view, the validity of Mr Norcross' calculations was not undermined or discredited in cross-examination. I accept that the calculations represent a fair theoretical assessment of the satisfactory selector effect of the dump fill via the influent chute, compared with the very unsatisfactory selector regime that would have resulted from the use of an IDSC in this particular SBR. Furthermore, Mr Norcross confirmed that the air entrainment effect was not observed at flows between 40 tcmd and 60 tcmd (Mr Norcross: Transcript, Day 23/111) and I accept his evidence to the effect that, to the extent that any air entrainment may have occurred, it was of no real significance to the SBR's selector capacity.
485. Having seen the chute feed in action and heard the evidence of others, Mr Norcross remained adamant that the design decision to use the chute feed was right. Mr Norcross is a very experienced process engineer, he was the original designer of the Omniflo SBR system and he had specifically approved both the decision to omit the IDSC (see paragraph 185 above) and the design of the influent chute itself: see Transcript, Day 23/109, where he said this:

"Q. As I understand it, you were happy with the feed arrangement that was in fact adopted?

A. Yes, once we had angled it so that again it had that influent distribution effective of jetting into the mixed liquor, the settled sludge.

Q. So you were content with the arrangement that was adopted?

A. Yes."

486. I agree with Mr Streatfeild-James that the views of the designer of the SBR system in question, particularly one of Mr Norcross' considerable experience, should not lightly be put to one side (Biwater's closing submissions, paragraph 478). Furthermore, Mr Norcross' opinion was supported by Dr Horan, who agreed that the influent chute design did promote a good selector effect in the SBR's basins, by providing a small region in which high concentrations of food were available to the MLSS. As Dr Horan pointed out, contact between the MLSS and the high concentration of food provided by the dump fill would be achieved as soon as the motive pumps were switched on once filling had finished, thus creating a good kinetic selection regime: see Dr Horan's evidence at Transcript, Day 30/123-124, where he said this:

"A. If that were the case (i.e. that the flow in the corner caused by the dump fill would push the sludge away from it), I think you have built there the best selector you could possibly build. Dr Chambers said what you require is a dump fill. What Dr Chambers is saying is that you are effectively dumping the fill in the corner and pushing the sludge from it, so you are getting your feed into the basin. No microbial contact with it, as Dr Johnstone said there. You are allowing the feed to accumulate in the corner. That inlet is sited next to the motive pumps. As soon as the motive pumps come, you are mixing that dump feed with the settled mixed liquor. If Dr Johnstone is right – I had not appreciated it – if he is right, that is the best selector you could have designed. What else is it but a zone in the tank where you have a high feed? You could not get a higher feed than putting all the sludge into the corner, all the feed into the corner like that. You turn on the motive pumps and you mix it with the mixed liquor. The perfect selector."

487. In my view, the omission of the IDSC did not prejudice the SBR's selector capacity. Rather, the evidence strongly suggests (correctly, as it seems to me) that, in the circumstances of this case and with the influent in question (both design and actual), the influent chute arrangements did ensure that the SBR had adequate selector capacity, whereas there would have been far less discernable selector capacity if the influent had been introduced via an IDSC. I accept the evidence of Mr Norcross and Dr Horan to that effect and am not persuaded by the evidence of Dr Chambers and Dr Johnstone to the contrary effect. In my judgment, the position is fairly and accurately summarised by Mr Streatfeild-James in paragraph 482 of Biwater's closing submissions, as follows:

"Putting it all together, Dr Chambers' point … comes down to the fact that the mixing regime is difficult to understand and is not uniform, and there are no calculations which allow it to be considered, together with the fact that there was a concern over poor settleability on site. However, when one stands back and takes into account:

a. That SBRs have been installed successfully by Jet Tech/Elga without ids; and

b. That there is just as respectable (an) argument that kinetic selection is being achieved as that it is not (because it is impossible to see how the mixing is taking place); and

c. That the SBR relied on metabolic as well as kinetic selection; and

d. That the SBR did operate for long periods with excellent quality effluent; and

e. That there is no evidence of any selectable filament being present

it is simply not sensible to say that there was no (or almost no) selector effect present. There clearly was as a matter of fact, and there is nothing in the theoretical objections to suggest that the evidence of (good) performance should not be taken at face value."

488. The "Average Flow" Issue. The second central criticism, made by Dr Chambers with regard to the SBR's selector capacity, concerns the use by Jet Tech/Elga of 1 x DWF as the average flow, instead of 1.3 x DWF (see paragraphs 182 and 194-195 above). This was the concern most often raised by Earth-Tech during the operational life of the original SBR (see, for example, paragraphs 279-280, 286, 289 and 349 above).
489. It is common ground that anoxic fill assists in controlling filaments through metabolic selection (see paragraph 144 above and Mr Eikelboom's first report at paragraph 35: E4/11). The essential thrust of Yorkshire Water's case on this aspect of the matter is that unless the SBR's fill was substantially anoxic at average flow (1.3 x DWF i.e. 85 tcmd) its selector capacity would be compromised. It is said that, because the average flow that was used for the design of the SBR was 1 x DWF (i.e. 65 tcmd), the result was that the period of aerated fill commenced much earlier than it should have done, the period of anoxic fill was reduced as a result and that this adversely affected the selector effect.
490. In paragraph 7.5.2 of his first report, Dr Chambers does refer to a reduction in selector capacity if there is inadequate anoxic fill (E1/31). However, although he made that point, he did so by simply noting it. Dr Chambers plainly regarded the omission of the IDSC as the main problem with the SBR's selector capacity (see paragraph 6.3.7 of his first report: E1/22 and paragraph 197 above). It is therefore not entirely clear how much importance he attached to the DWF issue in his overall criticism of the SBR's selector capacity. It is possible that Dr Chambers restricted himself to the general observation made in paragraph 7.5.2 of his first report, because he recognised that the point only arises at flows in excess of 65 tcmd and is expressed in terms of an average flow of 85 tcmd – whereas the actual average flows to the SBR during 2000 were about 60 tcmd (i.e. the actual average was below the DWF used by Jet Tech/Elga as the average flow for design purposes).
491. However that may be, I am satisfied that, as Mr Streatfeild-James pointed out (Biwater's closing submissions, paragraph 488), the argument falls away in the light of the flows actually experienced during 2000, because Yorkshire Water's case on this aspect of the matter turned entirely on the assertion that there was too much aerated fill at flows above 65 tcmd, as the result of the design error made in equating average flow with DWF, and the actual flows were consistently well below that figure.
492. In any event, for the reasons given in paragraphs 489-490 of Biwater's closing submissions, I am satisfied that the flexibility of the Jet Tech SBR control system was such that all the necessary adjustments could have been achieved by appropriate changes to the relevant set points. All in all, the suggestion that the use of 1 x DWF instead of 1.3 x DWF for average flow resulted in a loss of selector capacity, which in turn caused poor sludge settleability, is simply not sustainable. There is no evidence of the presence of selectable filaments, the average flows actually experienced were below the design average flow of DWF and, if the problem had arisen, it could have been addressed by appropriate adjustment of the set points.
493. The ASP Effluent. During the course of the trial, Yorkshire Water amended the Particulars of Claim to add the allegation that the Plant was "incapable of treating 60 tcmd from the Low Level ASP; i.e. 49% of the maximum flow" (see paragraph 76.2A of the Particulars of Claim: A1/tab 2/35). It was Dr Chambers' view that the inclusion of the ASP effluent in the SBR influent made selector design more difficult (because the ASP effluent diluted the sewage and thus reduced the BOD load of the SBR influent) and that no real consideration had been given to this aspect of the matter at the design stage: see paragraph 3.2.11 of his first supplemental report (E1A/tab 1/6).
494. I am also satisfied that there is nothing in this particular point. The fact is that the SBR did successfully treat influent that included the ASP effluent for significant periods of time from late 1999 until late 2001, as described above. Both Mr Norcross and Mr Eikelboom gave evidence that lower BOD concentrations than those experienced at Knostrop could be and are treated successfully (Transcript, Day 14/156, Day 15/25 and Day 24/35). More importantly, it was Dr Horan's evidence that the ASP effluent was treatable, did not compromise the SBR's selector capacity and did not explain the poor settleability experienced in the SBR: see paragraph 3.2 of his first supplemental report (E3A/19). I accept Dr Horan's evidence on this aspect of the matter in preference to that of Dr Chambers and Dr Johnstone because: (i) it is consistent with the operational history of the SBR, (ii) it is supported by the evidence of Mr Norcross and Mr Eikelboom and (iii) it is entirely supported by bench tests that Dr Horan carried out and which are fully described in Appendix 2 to his first supplemental report (E3A/50-56).
495. The latter point is important. Dr Horan was the only expert to have carried out such bench tests on the SBR influent. Details of the tests are set out in Appendix 2 of his first supplemental report and it is clear that they were carried out with great care. In my view, Dr Horan's evidence about the bench tests was not undermined or called into question in cross-examination. I am satisfied that Dr Horan's conclusion (which I accept as both reliable and fully supported by the results obtained in the tests) is of fundamental importance with regard to the selector capacity issue generally and the treatability of the ASP influent in particular. It is expressed as follows:

"4.1 The Knostrop influent is capable of good settlement and SSVI3.5 results of 50 ml/g have been achieved in the bench scale tests. This proves that the presence of ASP effluent in the Knostrop influent is not detrimental to settlement and that the Knostrop influent is capable of good settlement with proper operation."

496. Microthrix parvicella. I accept Mr Streatfeild-James' submission that there is some uncertainty about the extent to which microthrix parvicella was present in the SBR and, if present, whether it caused any problems with sludge settlement before the winter of 2001/2002 and the completion of the Phase 1 remedial works (Biwater's closing submissions, paragraph 403).
497. As I have already stated, it is clear that microthrix parvicella was present in sufficient numbers to cause some moderate settlement problems in the early part of 2000 (see paragraphs 263-269 above). However, the presence of microthrix parvicella in activated sludge plants does not mean, ipso facto, that there will be serious sludge settlement problems. In particular, as with other filamentous species, it is only when its presence increases beyond a certain critical mass that it does cause settleability problems. Although the presence of some microthrix parvicella may well be inevitable in nutrient removal plants, it need not always be fatal to their proper operation. The key is to operate the plant so that microthrix parvicella is controlled at manageable levels.
498. The performance of the SBR was very good during the summer and autumn of 2000, which clearly indicates that microthrix parvicella was not present in sufficient numbers to cause any settlement problems during that period. Furthermore, the winter of 2000/2001 and the greater part of 2001 passed without Yorkshire Water identifying any filamentous problem or seeing the need to carry out any microbiological investigation. As Mr Streatfeild-James observed (Biwater's closing submissions, paragraph 405), this strongly suggests that there was no apparent evidence of large numbers of microthrix parvicella or other filaments at Knostrop during that period, otherwise a proper investigation would have been carried out, particularly since Yorkshire Water had become aware of a significant microthrix parvicella problem at its Adwick SBR in the early part of 2000 (see paragraph 259 above).
499. Accordingly, there is no real evidence that microthrix parvicella was present in sufficient numbers to cause problems with sludge settlement until the SBR had been recommissioned with its new external selector in December 2001, shortly after which microthrix parvicella was found to be present in large numbers (see paragraph 433 above). In my judgment, this proliferation of microthrix parvicella was largely due to the installation of the new external selector: see the evidence of Mr Eikelboom at Transcript, Day 31/34-35, where he said this:

"Q. … what are the combinations of features in the design (of the SBR) that in your view will promote the existence of Microthrix?

A. First of all, the anoxic periods included in the design, because under anoxic conditions Microthrix parvicella only is able to take up the feed it absolutely needs for its growth. It has no competition from other bacteria. So long (as there are) anoxic or even … non-aerated periods (these) will always promote Microthrix. In the case of Knostrop, they made it far worse by the construction of the external selector upstream of the SBR. So these are the two … most important aspects. Then you have some other aspects of …(secondary)…weight. That is the type of feed … in the case of Knostrop, it is very likely that the influent is a very important aspect as well and then … you have aspects like maybe low oxygen concentrations during the start of the react period …"

500. However, although there is no evidence that the SBR, whilst operated as originally designed and built (i.e. from November 1999 to late summer 2001), suffered from sludge settlement problems as the result of the proliferation of microthrix parvicella, it is necessary to go on and consider whether that would have remained the case, i.e. whether the SBR was bound to suffer from serious sludge settlement problems due to microthrix parvicella (particularly during winter), thus calling into question its ability to meet the Schedule 7 performance requirements if a full programme of performance tests had been carried out.
501. It was Mr Eikelboom's view that there were two particular problems at Knostrop that had favoured the proliferation of microthrix parvicella: (i) the nature of the influent (high TSS and low BOD) and (ii) the anoxic fill. It was his opinion that the construction of the external selector had made matters much worse so far as concerns microthrix parvicella, because of the anoxic fill. He accepted that the position with regard to microthrix parvicella could be improved by substituting a wholly aerated fill for the anoxic fill, thereby eliminating the latter, but pointed out that this would mean the loss of denitrification (i.e. metabolic selection) and that the low BOD loads meant that kinetic selection would be insufficient on its own (Transcript, Day 31/37-49). Dr Horan's evidence was to a similar effect (Transcript, Day 30/29 and Transcript, Day 30/98-102), where he said this (inter alia):

"The sewage is so far removed from a normal sewage that we have such high TSS to BOD5 following primary settlement that that TSS is not removed and that is providing the food source for the Microthrix. And it is not an outlandish proposition, it is the proposition that Mr Eikelboom has also put forward" (Transcript, Day 30/98-99).

502. Mr Eikelboom also expressed the view that, in most configurations of the activated sludge process (including the SBR, as originally designed and built), chemical dosing would be required for a couple of weeks in the winter in order to control microthrix parvicella: see Transcript, Day 31/39-40, where he said this:

"Q. … if you have enough settlement time available, enough decant depth and you promoted and encouraged the good floc-forming bacteria, …do you say that you always end up having to dose?

A. For Microthrix you always end up with a requirement for dosing. … It is a temporary solution because Microthrix – if a plant is operated properly and if the influent quality is not too far from a common influent quality, then Microthrix is only causing serious problems during wintertime, and that means that dosing is required for a couple of weeks during winter. That is what I call temporary dosing.

Q. Do I understand, Mr Eikelboom, that you are saying that provided the plant is in all respects an activated sludge plant like an SBR, is in all respects designed and operated correctly for the volumes and loads that it is receiving, if the influent contains domestic sewage, it is inevitable that there will have to be chemical dosing for a couple of weeks during the winter to prevent proliferation of Microthrix and bulking problems that go with that?

A. In most configurations of the activated sludge process that is true, yes."

503. However, it appears to have been common ground between Dr Chambers and Dr Horan that, in principle, it is possible to control microthrix parvicella (even during winter) by use of a suitably short sludge age. Both accepted that sludge age (and hence the F/M ratio) is a significant factor that mediates between microbes in the sludge. Based upon these principles, it was Dr Horan's evidence that any microthrix parvicella present in the SBR could be controlled at manageable levels (even during winter) by use of a suitable sludge age regime/operation at a suitable F/M ratio (Transcript, Day 29/21), i.e. a sludge age of about 12 days. As it seems to me, the fact that there is no evidence of any proliferation of microthrix parvicella during the winter of 2000/2001 (although there were other operational problems: see paragraphs 404-407 above), is evidence that Mr Eikelboom's view as to the inevitability of chemical dosing was too uncompromising and that Dr Horan's view (which I accept as correct) is to be preferred. As it seems to me, that particular finding and the paragraphs which follow effectively dispose of Elga's "chemical dosing" case (see paragraph 313 above).
504. The design F/M of the original SBR was 0.037, which is equivalent to a sludge age of about 13 days. Subject to the question of nitrification (as to which, see below), I am satisfied that, had it been possible to operate the SBR at that F/M ratio, any microthrix parvicella that was present in the SBR would have been controlled at manageable levels, even during winter. In fact, after take-over, because of the unexpected nature of the influent (i.e. in particular, its unexpectedly low BOD and high TSS, compared with that specified), the SBR was never operated at the design F/M: see generally Dr Horan's first report sections 7.33 to 7.116 (E3/53 onwards). The average operating F/M ratios never went above 0.028 and, by March 2001, they had sunk as low as 0.014. The steady decline in the operating F/M ratios coincided with the fall-off in the BOD load to the SBR. By October 2000, the operating F/M ratio (which had dropped to 0.02) equated to a sludge age of about 25 days. In 2001, it fell even further to levels that equated to an aerated sludge age of about 35 days.
505. I am satisfied that sludge ages of this length would have been very conducive to the development of microthrix parvicella. I accept Dr Horan's evidence that the low BOD loads in the influent received by the SBR (which caused the low F/M ratio/ long sludge ages) were a significant feature that would have caused or contributed to the development of microthrix parvicella in the SBR. However, the paucity of microbiological evidence is such that it is not possible to say precisely what the effect of the low BOD loads was in terms of microthrix parvicella over the entire operational life of the original SBR (i.e. from November 1999 until the Phase 1 remedial works).
506. It is clear that the BOD loads were very low going into the winter of 2000/2001 (see paragraph 502 above). If there were any microthrix parvicella problems during this period (and there is no microbiological evidence that such was the case), I am satisfied that the low BOD loads would have made a major contribution to them. In the following year, the BOD loads were even lower and, in my view, these were a significant cause of the microthrix parvicella problems that undoubtedly occurred in the winter of 2001/2002, after completion of the Phase 1 remedial works.
507. In the event, as it seems to me, the real issue between Dr Chambers and Dr Horan on this aspect of the matter (i.e. Dr Horan's suggested F/M control of microthrix parvicella) was whether the SBR could be operated at a high enough F/M (or low enough sludge age) sufficient to control microthrix parvicella that would, nevertheless, still also allow full nitrification (particularly during winter). Put another way, the issue was whether a balance could be achieved, in which sludge age was high enough for full nitrification but, at the same time, low enough to keep microthrix parvicella at minimum/manageable levels.
508. Dr Horan's evidence was that this balance could be achieved if the SBR was operated with a sludge age of about 12 days: see paragraphs 4.28 to 4.32 of his third supplemental report, where he said this (E3C/16-17):

"4.28 My views on the need to operate the SBR at the correct F/M ratio have been set out in paragraphs 4.5.1 and 8.14.9 and 8.16 of my first report. In summary, a Plant which operates with a low F/M ratio (or a high sludge age) is susceptible to the growth of Microthrix. The maintenance of an appropriate F/M is fundamental to the promotion of good floc biology. …

4.29. …

4.30. The operational issue at Knostrop as concerns sludge settleability is that when the aerated sludge age exceeds 10 days Microthrix is provided with a retention time which allows it to grow successfully. However, as a nitrifying treatment plant must provide at least 10 days sludge age in order for the nitrifying bacteria to grow at the minimum winter temperature, then conditions are also in place for Microthrix to start growing. The growth rate of the bacteria is fixed …"

4.31. The issue then is whether a sludge age (and therefore an F/M) is available at which nitrification is possible but the Microthrix population can be maintained at levels that do not significantly impact on sludge settleability. In my opinion, for most of the year it would be possible to operate the plant on a 10 day sludge age (or shorter) as in-basin temperatures would be high enough to maintain nitrification at this sludge age and this would be sufficient to prevent the growth of Microthrix. For minimum winter temperatures that might be expected at Knostrop, of 10 – 11C, a 12-day sludge age would guarantee nitrification with a large margin of safety for both 5 and 6 tank operation. This sludge age would also be low enough to ensure that Microthrix was not present in large numbers … These low numbers would not cause serious deterioration in settleability."

509. Dr Chambers disagreed that a 12-day sludge age was sufficient to ensure nitrification in winter. However, it is important to note that Dr Chambers did not take issue with Dr Horan's proposition that operation at a 12 day sludge age would control microthrix parvicella (see Dr Chambers' comments on paragraph 4.31 of Dr Horan's third supplemental report: E1C/23), nor was Dr Horan challenged on that point in cross-examination. Accordingly, I accept Mr Streatfeild-James's submission that the real issue was not whether a sludge age of 12 days would limit microthrix parvicella to acceptable levels, but whether a 12 day sludge age would be sufficient to produce a fully nitrified effluent at the plant (Biwater's closing submissions, paragraph 432).
510. It was Dr Chambers' opinion that a sludge age of 15 days was the reasonable minimum for ensuring that nitrification to the required standards would be achieved at all UK temperatures (including winter): see paragraph 6.2.8 of his first report (E1/19). The other experts did not consider that a period as long as 15 days was required at Knostrop. Both Dr Horan and Dr Johnstone considered that a sludge age of 12 days would be sufficient and Mr Eikelboom's view was that 10 days would suffice (a sludge age that would be in line with the Danish approach: see paragraph 2.26 of Dr Horan's first supplemental report, quoted in paragraph 511 below): see the 3^rd EJS, dated 9^th April 2003 (E5/tab 3/52). However, all the experts were agreed that: (i) sludge age is a useful concept for determination of nitrification capacity, (ii) a minimum sludge age is required in order to nitrify to the required standards and (iii) this minimum value depends on the temperature in the basin: see the 3^rd EJS (E5/tab 3/52).
511. It was Dr Horan's opinion that the provision in the Specification that the plant should be capable of treating the sewage over a temperature range of 7C to 17C (see clause 0.2.10 of the Specification: paragraph 26 above) referred to sewage temperatures and not to in-basin temperatures. It was his view that a sewage temperature of 7C would be reflected as in-basin temperatures of 10C to 11C and that sewage temperatures of 7C would be rare at Knostrop. In my view, his evidence on this aspect of the matter is correct and it is noteworthy that it was not the subject of challenge in cross-examination. In view of its importance, it is worth quoting the following passage from his first supplemental report (E2A/6-7):

"2.21 However, it is important to note that an in-basin temperature of 7C is produced by an influent sewage with a lower temperature. This is because the temperature of the sewage entering the treatment basins is raised as a result of:

i. The heat generated by the MLSS during the treatment process as part of the bacteria's normal metabolic processes, and

ii. The action of aeration of the sewage.

b. Thus, I would expect minimum in-basin temperatures of 10-11C to result from incoming sewage with a temperature of 7C.

c. In my opinion, for basins as large as those at Knostrop the minimum in-basin temperature would rarely drop below 10-11C even during winter, and, indeed for most of the time during winter would not drop below 12C.

d. I cannot accept that 7C should be regarded as the minimum winter in-basin temperature at Knostrop. As I said above, even if the incoming sewage could be expected to drop to that temperature, 10-11C would be the minimum resulting in-basin temperature at Knostrop. I note that the Contract Specification indicates that treatment should be expected to occur at a range of temperatures of 7C-17C. To my mind, this must relate to the sewage temperature not the in-basin winter temperature as this conforms with what would reasonably be expected at Knostrop.

e. I say this because in order for the temperature in the basin to fall as low as 7C, it would require a sustained period of flow derived from snow melt, and the last time there was sufficient snow in Leeds to produce such a low temperature was probably the winter of 1983.

f. Surprisingly in-basin temperatures are rarely measured in the UK, but figures for in-basin temperatures in Denmark (which has much colder winters than Leeds) are 6 to 8C. The Danish approach is to select a sludge age based on the sustained winter in-basin temperature of 10C and they recommend a value of 10 days for full nitrification. …"

512. I also accept Mr Streatfeild-James' submission that the evidence of actual operation supports Dr Horan's opinion and overall approach (Biwater's closing submissions, paragraphs 518 to 521). It will only be exceptionally during an event such as a snow- melt that the in-basin temperature at Knostrop will drop as low as 8C and that has happened on only one occasion to date (22^nd March 2001). On that occasion, although full nitrification was affected, the effluent spot samples did not breach the contractual standard of 8mg/l (see D1A/59).
513. Accordingly, I am satisfied that a 12 day sludge age would have been sufficient for the SBR to achieve full nitrification throughout the year and I accept Dr Horan's evidence to that effect.
514. I have therefore come to the following conclusions with regard to the issues relating to microthrix parvicella: (i) microthrix parvicella was present, but not dominant, for a period of time in 2000; (ii) microthrix parvicella did not become dominant until the end of 2001, after completion of the Phase 1 remedial works; (iii) the primary reason for its dominance in late 2001 was the installation of the external selector, which favoured the growth of microthrix parvicella; (iv) the evidence about the presence of microthrix parvicella during the operational life of the original SBR is unclear; (v) the unexpectedly high TSS loads and low BOD loads in the influent (and later the external selector) all contributed to the development of microthrix parvicella in the SBR; and (vi) problems caused by a proliferation of microthrix parvicella were not inevitable, because the presence of microthrix parvicella in the SBR could have been controlled by use of a suitable sludge age/F/M ratio that could have been achieved and maintained, particularly if the influent had been within the limits of the "likely variations in flows and loads" (see clause 0.2.6 of the Specification) which, for the reasons given above, I am satisfied it was not.
515. Treatment Capacity. As Mr Streatfeild-James observed (Biwater's closing submissions, paragraph 498), the second main theme of Yorkshire Water's case is that there was a deficiency in the SBR's treatment capacity. As it seems to me, Yorkshire Water's case on treatment capacity is inextricably interwoven with its case on the SBR's alleged lack of selector capacity. In particular, as already explained, it is a fundamental aspect of Yorkshire Water's case in these proceedings that the SBR lacked selector capacity, that this led to settlement problems as a consequence of which the SBR's treatment capacity was restricted. Accordingly, as a result of my consideration of the selector capacity issue, I have already dealt with various important aspects of Yorkshire Water's lack of treatment capacity case.
516. The following appear to have been the main features of Yorkshire Water's case on the SBR's alleged lack of treatment capacity: (i) the SBR was unable to treat design loads and flows to the required standard, because it was not designed for average flow (i.e. 1.3 DWF), (ii) the SBR was not able to nitrify at peak flows when ambient temperatures were low, (iii) the SBR was unable to treat any ASP effluent (and, as such, its capacity was limited accordingly), (iv) the SBR was unable to operate on 5 tanks indefinitely and (v) the SBR was unable to operate at full flows without resorting to fill-decant.
517. So far as concerns paragraph 516 (i) above (the "average flow" issue), I have already dealt with this in paragraphs 489 to 493 above and come to the conclusion that the SBR did not lack treatment capacity for that reason, as explained in those paragraphs. As for paragraph 516 (ii), I am satisfied that the SBR was able to nitrify satisfactorily with a sludge age of 12 days at low ambient temperatures for the reasons given in paragraphs 506 to 511 above. So far as concerns paragraph 516 (iii), I am satisfied that the ASP effluent was treatable by the SBR, for the reasons set out in paragraphs 494 to 496 above. As for paragraph 516 (v), for the reasons given in paragraph 261 above, I am satisfied that the fill-decant issue was resolved by the installation of fill-decant inhibit and that this had no significant effect on the treatment capacity of the SBR.
518. As it seems to me, an appropriate starting point for my consideration of the remaining elements of Yorkshire Water's case on the SBR's alleged lack of treatment capacity is the 3^rd EJS, in which the experts expressed the agreed view that, as set up (i.e. with a decant depth of 1.6m), the SBR was theoretically capable of producing effluent of the required quality at peak flows (136 tcmd) with an average BOD load of 6.89 tonnes/day, either as a 6-tank or as a 5-tank operation, provided that a certain level of settleability was achieved (as shown in tables 1 and 2: see E5/tab 3/52-53). Thus, for example, the agreed tables show that, in order to treat a flow of 136 tcmd with a BOD load of 6.89t/d: (i) the 6-tank SBR would require an MLSS of 1800mg/l and an SSVI of 110ml/g, if operated at an aerated sludge age of 14.1 days, (ii) the 5-tank SBR would require an MLSS of 2500mg/l and a SSVI of 90ml/g, if operated at an aerated sludge age of 13.2 days and (iii) the lower the aerated sludge age at which the SBR was operated, the less MLSS that would be required and the greater the SSVI that could be tolerated.
519. However, the experts then went on in the 3^rd EJS to make the following agreed statement (E5/tab 3/53-54):

"Table 1 and 2 have been prepared based on an assumed decant depth of 1.6m. We all agree that the larger the decant depth the greater the treatment capacity of the SBR. At the set-up depth of 1.6m there is controversy as to whether the treatment capacity is adequate. This disagreement centres around our differing interpretation and engineering judgement of appropriate values for aerated sludge age and sludge settleability to be used in design.

We do not understand why the decant depth of 1.6m was selected and why it was never increased."

520. Accordingly, I agree with Mr Streatfeild-James' submission (Biwater's closing submissions, paragraph 506) that the agreement reached by the experts, as recorded in the 3^rd EJS, indicates that the real issue on treatment capacity is whether the SBR could achieve an appropriate level of settleability, which in turn depends upon the aerated sludge age at which the SBR was operated. As it seems to me, given that I am satisfied that the SBR did have adequate selector capacity, the obvious outstanding question that this issue raises is whether the SBR was capable of nitrification throughout the year at full flows. It is to that issue that I therefore now turn.
521. As I have already stated, it was Dr Chambers' view that a sludge age of 15 days is required to ensure full nitrification (Transcript, Day 27/20). According to the jointly agreed tables, this would require the SBR to be operated with an SSVI of about 100ml/g. However, an SSVI of 100ml/g might well not be achievable during the winter months because, with a sludge age of 15 days, it would not be possible to control microthrix parvicella satisfactorily at manageable levels (see paragraphs 510 to 515 above), with resulting poor sludge settleability and a high SSVI. It was Dr Chambers' view that a sludge age of 12 days would be insufficient to achieve full nitrification during the winter period, although he did not dispute the proposition that microthrix parvicella would be successfully controlled by operating with a sludge age of that length (see paragraph 510 above). However, for the reasons already given (see paragraphs 510 to 515 above), I am satisfied that a sludge age of 12 days is sufficient for nitrification and the agreed tables show that, with a sludge age of 12 days, the appropriate operating SSVI (which will ensure that there is no solids' carry-over) would be increased to 120ml/g.
522. Between mid-May 2000 and the Phase 1 remedial works, the SBR produced fully nitrified effluent with only one or two minor exceptions. The problems with nitrification prior to mid-May 2000 were almost entirely due to the loss of control in late December 1999/early January 2000 and the deliberate strategy of bringing the solids' carry-over problem under control (see paragraphs 231 to 258 above). As I have already indicated, it was therefore common ground that it would be unfair to attach any significance to the nitrification problems experienced during this period (see paragraph 271 above). To the extent that there were a small number of failures thereafter, I am satisfied that the most likely explanation for those is that, from time to time, there were incidents of industrial pollution and/or some other form of toxicity in the SBR influent, for which Taylor Woodrow/Biwater were not responsible, that temporarily inhibited nitrification (see, for example, paragraph 359 above). Unfortunately, it is not possible to be more precise about this aspect of the matter, because the documentation relating to industrial pollution during the relevant period was only made available after the trial had started and a detailed investigation of the matters referred to in that documentation was not possible (see, in particular, the documents contained in C92).
523. Accordingly, the operational life of the SBR as originally designed and built does not provide any evidence that the SBR lacked capacity to achieve nitrification to the required standards. Yorkshire Water's case that it did not have such a capacity (particularly during winter) was almost entirely dependent on theory. As to that, for the reasons already explained (see paragraphs 510 to 515 and 520 above), I do not accept Dr Chambers' evidence that a minimum sludge age of 15 days would be required to achieve full nitrification throughout the year. As it seems to me, that leaves only the further point made by Dr Chambers on this aspect of the matter, namely that, when operating on 5 tanks at FFT, the available reaction time would only be about 10 minutes, which would be insufficient to achieve full nitrification: see paragraph 6.2.15 of his first report (E1/20).
524. In my view, it is important to emphasise once more that, in order to succeed in these proceedings, it is necessary for Yorkshire Water to prove that Taylor Woodrow was unable to meet the performance requirements of the contract. As Mr Elliott pointed out (see paragraph 15 above), the contract provides appropriate machinery for testing the SBR's ability to meet the performance requirements, i.e. the four 28-day seasonal performance tests. Although I have held that it is open to Yorkshire Water to prove its case, notwithstanding the fact that it did not subject the SBR to any of the contractual performance tests (see paragraph 38 above), this is obviously a difficult burden to discharge, particularly in a case where (as here) the Plant in question performed perfectly satisfactorily for significant periods of time (in particular, from June 2000 to November 2000).
525. It is clear from paragraph 6.2.15 of his first report (E1/20) that Dr Chambers' criticism of insufficient reaction time to achieve nitrification (i.e. because the react time would only be 10 minutes) was limited to those occasions when the SBR would be operating on 5 tanks and receiving 136 tcmd. Having regard to Dr Horan's unchallenged evidence (as to which, see paragraph 527 below), it is clear that this criticism by Dr Chambers is subject to a further limiting factor, namely that the risk of ammonia failure due to insufficient react time would only arise in winter, when operating at FFT and at or about peak BOD loads.
526. It was common ground that the peak flow of 136 tcmd (FFT) would be a rare event and short in duration (Dr Chambers suggested that it would be likely to persist for a couple of days at most and Dr Zuber described it as a transient condition: Transcript, Day 27/64 and Day 14/44). It was also accepted by Yorkshire Water that the peak BOD load of 7.58t/d would hardly ever coincide with FFT, because maximum flow normally results from heavy rainfall and this has the effect of diluting the sewage: see Mr Holmes' evidence to that effect: Transcript, Day 6/69-70).
527. I accept Dr Horan's unchallenged evidence that, provided the SBR could achieve the appropriate SSVIs, it would achieve nitrification to the required standards when operating on 5 tanks while receiving average BOD loads at FFT: see paragraphs 2.29 to 2.32 of his first supplementary report (E3A/8-9). I also accept his evidence that the SBR could operate successfully for about a week (although Dr Horan acknowledged that it "would struggle"), if it were to be subjected to full flows (136 tcmd) with the maximum BOD load at minimum temperatures (Transcript, Day 29/156-157). However, I am satisfied that the likelihood of such a conjunction of events would be so remote as to be, to all intents, non-existent. It is also worth noting that, as I have already indicated, Dr Zuber's contemporaneous calculations with regard to this aspect of the matter gave the slightly more optimistic answer of about two to three weeks (see paragraph 320 above).
528. In cross-examination, Dr Chambers accepted, in effect, that he had miscalculated the available react time for a cycle at FFT on 5 tanks (Transcript, Day 27/97). The correct figure is about 15 minutes. Furthermore, Dr Chambers had made no allowance for the fact that the fill period of 25 minutes at FFT would be entirely aerobic and that there would be some reaction taking place during that period. I therefore accept Mr Streatfeild-James's submission that both these factors meant that Dr Chambers took too gloomy a view of the difficulties of nitrification at FFT and full loads in winter (Biwater's closing submissions, paragraph 525 (5)). I therefore accept the evidence of Dr Horan on this topic.
529. In summary, therefore, I have come to the conclusion that the only circumstance in which, if operated correctly as originally designed and set up, the SBR would have "struggled" to nitrify would have been when being operated with 5 tanks and receiving peak BOD loads at FFT and at minimum ambient temperatures. As I have already indicated, the likelihood of such a conjunction of events would be so remote as to be virtually non-existent. However, even when operating on 5 tanks, the SBR would have been able to nitrify for a sufficient period of time to cover the duration of any such very unlikely event. Accordingly, I am satisfied that such a limitation in the SBR's treatment capacity (if limitation it be) does not provide any foundation for the suggestion that the SBR was unable to meet the performance requirements of Schedule 7, even if the performance testing had been entirely carried out on 5 tanks only.
530. In any event, the experts were agreed that the larger the decant depth, the greater the treatment capacity of the SBR would have been: see the passage in the 3^rd EJS quoted in paragraph 519 above. I accept the general nature of the improvement in treatment capacity that is summarised in paragraphs 540 to 541 of Biwater's closing submissions. I also accept Mr Streatfeild-James' submission that the decant depth could have been increased to 2.06m by altering the set points so as to lower the bottom water level. I accept Dr Horan's unchallenged evidence to that effect and his calculations demonstrating that there was no hydraulic restriction to prevent this from being done: see paragraph 2.35 and Appendix 1 of Dr Horan's first supplementary report (E3A/11 and tab1). As Mr Streatfeild-James observed, an increase in decant depth was perfectly possible and would have resulted in significant improvements in both selector and treatment capacity (Biwater's closing submissions, paragraph 543). In my view, this strongly supports my conclusion that Yorkshire Water has not proved that the SBR was unable to meet the Schedule 7 performance requirements. Furthermore, had that increase in decant depth been carried out, there is every reason to believe that the SBR would have been able to meet the Schedule 7 performance requirements without difficulty, even if operated throughout on 5 tanks. By a parity of reasoning, Yorkshire Water has failed to prove the contrary proposition.
531. Conclusions with regard the SBR's performance and capabilities. The operational difficulties experienced in 2000 were caused by a number of matters that had differing effects over time. The initial problems with set points, mechanical failures, issues over the operation of the SBR and so forth were eventually replaced by a period of conspicuously good operation and stability, where the only matter of concern was MLSS control. MLSS levels were difficult to control because of the unexpectedly high TSS loads, which persisted throughout 2000 and for which Taylor Woodrow/Biwater were not responsible. However, these control issues were perfectly capable of being addressed and were successfully addressed by Biwater in September and October 2000.
532. Microthrix parvicella may have been present throughout 2000. To the extent that it was present it was not there in any significant quantity, although its presence would have added to the difficulties being experienced. However, I am satisfied that microthrix parvicella could have been controlled at acceptable levels by use of a suitable sludge age/F/M ratio. Such a sludge age could have been maintained throughout the year without jeopardising nitrification. I am satisfied that there is nothing in the performance of the SBR as originally designed and built to support the suggestion that the SBR was fundamentally defective or lacked adequate selector capacity. I accept Mr Streatfeild-James' submission that Dr Zuber's assessment of the operational difficulties was entirely accurate and correct (D3A/341-343): i.e. the SBR was not defective, but was difficult to operate because of the varying and low BOD loads, the high and variable TSS, the high COD:BOD ratio and the high sludge yields (Biwater's closing submissions, paragraph 442). Unfortunately, as Mr Streatfeild-James went on to observe, Yorkshire Water had decided to determine the contract in the belief that there were fundamental defects with the SBR and that led to remedial works being undertaken that were not, as it seems to me, actually required to remedy any defect or deficiency in the design of the SBR.
533. I am satisfied that, whether by reference to its operational history or to appropriate theoretical calculation, the SBR as originally designed and built did not lack either adequate selector or treatment capacity. Had Yorkshire Water proceeded to performance testing in the autumn of 2000 (as Biwater was anxious to do) and had the SBR been operated properly, there is every reason to believe that it would have passed the performance tests in question, particularly if the influent had been within the limits of "the likely variations in flows and loads", which had not been the case for much of 2000. Furthermore, the prospect of a successful completion of the performance tests would have been even greater if the set points had been adjusted so as to increase the decant depth, a measure that could have been carried out without the need for any of the remedial works actually carried out. As it seems to me, it follows from these findings that Yorkshire Water has not proved that Taylor Woodrow was unable to meet the performance requirements of Schedule 7 of the main contract and its case, therefore, fails. By a parity of reasoning, I am satisfied that the SBR was not defective as alleged in either the 14^th March Defect Notice (see paragraph 275 above) or the 7^th April Defect Notice (see paragraph 290 above). It follows that, even if (contrary to my construction of the relevant terms of the main contract) Yorkshire Water is entitled to rely on any of the other contractual terms identified in the Particulars of Claim as well as the Schedule 10 absolute guarantees, the outcome of these proceedings remains the same.
534. Conclusion with regard to the Main Action. For all the foregoing reasons, I have come to the firm conclusion that Yorkshire Water's claims for financial relief, liquidated damages and declaratory relief must be and are hereby dismissed. It follows that the dependent Part 20 claims also fall to be dismissed.
535. The Counterclaim and other Claims. I now turn to consider the Counterclaim and other claims summarised in paragraph 14 above. These can be dealt with very briefly in the light of my conclusions in the main action, because there is a large measure of agreement as to figures and no question of any form of set-off by Yorkshire Water now arises.
536. Taylor Woodrow's counterclaim is in two parts, namely (i) claims made in respect of sums ultimately due to Taylor Woodrow and (ii) claims made by Taylor Woodrow that pass on claims made by Biwater. In effect, the Biwater counterclaims are made, defended and pleaded "back to back" through the relevant provisions of the sub-contract and the main contract.
537. In summary, Taylor Woodrow's claims are as follows:

i. payment of the Final Account including: (a) outstanding items included in the Main Contract Activity Schedule, (b) outstanding variations and (c) outstanding retention;

ii. payment of sums due in respect of post-takeover assistance provided by Taylor Woodrow to Yorkshire Water;

iii. Biwater's claims under the Taylor Woodrow/Biwater subcontract, adopted by Taylor Woodrow against Yorkshire Water for: (a) an extension of time of 17.8 weeks, (b) loss and expense incurred as a result of the delay of 17.8 weeks, (c) additional costs incurred as a result of changes to the software control package, (d) payment of sums due in respect of post-takeover assistance provided by Taylor Woodrow to Yorkshire Water and (e) sums due and owing under the Biwater final account; and

iv. a declaration of Taylor Woodrow's entitlement to a Final Certificate.

538. As Mr Streatfeild-James observed (Biwater's closing submissions, paragraph 760), whilst there are a significant number of discrete counterclaims, the substantive dispute (such as it was) and the majority of the evidence was confined to Biwater's claim for an extension of time. The other counterclaims are largely the subject of agreement.
539. (1) Taylor Woodrow's own Final Account Claims. Taylor Woodrow's Commercial Manager, Mr Heath James Bennett, gave evidence as to the outstanding Final Account sums. I turn to consider each in turn briefly.
540. (a) The Main Contract Activity Schedule. The outstanding sum of £44,226 is made up of 7 items, each recorded in the Activity Schedule. The figures have been agreed as figures and three of the items (totalling £8,000) are admitted as due, subject to set-off. The remaining four items are disputed. As to those items, I accept the submissions made in paragraphs 766 to 778 of Biwater's closing submissions and do not regard it as necessary to repeat them in this judgment. Accordingly, I am satisfied that Taylor Woodrow is entitled to the amount claimed in respect of the Main Contract Activity Schedule.
541. (b) Outstanding Variations. Taylor Woodrow claims £3,590.48 in respect of a number of variation orders (see Biwater's closing submissions, paragraph 779). These are admitted, subject to substantiation of the sums claimed. The quantum experts have considered the evidence substantiating the claim and have agreed the figure of £3,590, as set out in their 3^rd joint statement (E6/tab 3/15). Accordingly, I am satisfied that Taylor Woodrow is entitled to the sum claimed.
542. (c) Retention. Taylor Woodrow claims payment of a total of £324,440 retained in respect of the main contract works and variations. The figures that make up that total have been agreed by the quantum experts in their 3^rd joint statement. The only defence was set-off. Accordingly, I am satisfied that Taylor Woodrow are entitled to the total sum claimed in respect of retention.
543. (2) Taylor Woodrow's Post Take-Over Assistance. The amount claimed (£33,097) is agreed as a figure. I accept the submissions made in paragraph 784 of Biwater's closing submissions. I am therefore satisfied that Taylor Woodrow is entitled to the sum claimed.
544. (3) Biwater's claims, adopted and passed on by Taylor Woodrow. As I have already indicated (see paragraph 538 above), the main area of dispute at trial related to Biwater's claim for an extension of time. However, as I have made clear in paragraphs 392 to 396 above, the unambiguous view taken by Earth-Tech at the time was that Taylor Woodrow/Biwater were entitled to almost the whole of the extension of time claimed and, but for the intervention of Yorkshire Water, would have issued the extension of time letter that was already drafted. I therefore accept Mr Streatfeild-James's submission that the starting point for the consideration of the claim is that, but for the intervention of Yorkshire Water for its own commercial advantage, a full extension of time would have been granted.
545. (a) The extension of time of 17.8 weeks. Biwater's claim is made up of the following elements: (i) 4.4 weeks' delay to the provision of "power on" to blowers, (ii) 6.4 weeks' concurrent delay in the provision of software (i.e. an additional delay of 2 weeks), (iii) 3 weeks' delay in the cobalt chloride drain-down of the SBR and (iv) 8.4 weeks' delay to the process development, including the re-programming of reliability tests.
546. Paragraphs 792 to 854 of Biwater's closing submissions deal with the various elements in Biwater's claim for an extension of time. I am satisfied that the analysis of the evidence and of the factual circumstances that is to be found in those paragraphs is both accurate and reliable and I adopt them without repeating them. I accept as correct the submissions that are founded on that analysis. I am fortified in that judgment by the unambiguous view taken by Earth-Tech at the time that Taylor Woodrow/Biwater were entitled to an extension of 17.5 weeks (i.e. all but .3 of a week of the period actually claimed). Accordingly, having regard to the existing length of this judgment, I do not feel that it is either appropriate or necessary to give any further detailed reasons for my decision that Biwater and, by a parity of reasoning, Taylor Woodrow are entitled to the extension of time of 17.8 weeks claimed. However, before passing on, I would like to acknowledge the very considerable assistance that I gained from both the detail and clarity of this part of Biwater's closing submissions, which was such that I have been able to deal properly with the matter in the foregoing short form.
547. (b) Biwater's Loss and Expense claim arising from the 17.8 weeks' delay. The figures have largely been agreed as figures and there is no issue as to whether and if so what the appropriate contractual mechanism for this claim is. Accordingly, the two outstanding issues between the quantum experts relate to the following: (i) the inclusion of "electrical supervisors" Messrs Barrow and Heale and (ii) the relationship between staff, staffing levels and hours per week actually worked and their relationship to the periods of delay.
548. So far as concerns Messrs Barrow and Heale, the issue turns on whether they worked on site as electricians or whether, in view of their description as "supervisors", they should be regarded as forming part of the site overheads. In my view, the approach adopted by Biwater's quantum expert (Mr Wilde) is correct. Biwater's site diary for 12^th May 1999 (G2/tab 1/6) records that Mr Heale and Mr Barrow, together with a Mr McLeod, had been authorised to work on the electrical equipment/switching by Mr Richard Rennie of Yorkshire Water. I accept the submission that this strongly supports the claim that they were working on site as electricians at the relevant time and were not merely part of the site supervision staff.
549. As for the relationship between staff, staffing levels and hours worked and the periods of delay, I accept that, in his supplementary witness statement (D3A/371-373), Mr Keith Richards provides the necessary (and, in the event, unchallenged) evidence to substantiate the conclusions reached by Mr Wilde in paragraphs 9.4.7 and 9.4.8 of his expert's report. I am therefore satisfied that there is no substance in the reservations expressed with regard to this matter by Yorkshire Water' quantum expert, Mr Schmitt.
550. Accordingly, I am satisfied that Biwater is entitled to recover the agreed sum of £468,821.44 in respect of its costs associated with the delay of 17.8 weeks, as set out in Appendix 9 of the quantum experts' 3^rd joint statement (E6/tab 3).
551. (c) Additional costs resulting from changes to the software. The only issue on this aspect of Biwater's claims is that between the quantum experts with regard to the relationship between the staff, staffing levels and hours worked and the periods of delay. Again, I accept that Mr Richards' unchallenged evidence provides the necessary substantiation and that, accordingly, Biwater are entitled to recover the sum claimed.
552. (d) Post Take-Over assistance. The quantum experts have agreed the figures that make up this claim as figures. I accept that the unchallenged evidence of Mr Rome (D3A/297, paragraph 406) and Mr Ingham (D3A/105, paragraph 57) provides the necessary proof as to the hours booked and that the hours in question were attributable to Biwater's post take-over assistance. Having regard to my conclusions in the main action, I am satisfied that Biwater is therefore entitled to recover the sum of £121,895.27 claimed under this heading.
553. (e) Biwater's Final Account Claim. The quantum experts have agreed that Biwater is entitled, in any event, to a final account sum of £9,807 in respect of the outstanding balance for variations: see Appendix 10 to the experts' 3^rd joint statement (E6/tab3). As to the remainder of the balance, the experts have agreed a further sum of £50,874 as a figure. In the light of my conclusions in the main action and for the reasons set out in paragraph 879 of Biwater's closing submissions, I am satisfied that Biwater is entitled to this further part of the outstanding balance for variations.
554. The second aspect of Biwater's Final Account claim is its claim to recover electricity costs contra-charged by Yorkshire Water in respect of the period for which Biwater seeks an extension of time. Biwater has calculated the relevant amount to be the sum of £30,357. In their 3^rd joint statement the experts have agreed that the outstanding sum of £30,357 should be paid to Biwater proportionately to the period of extension of time decided by the court (E6/tab 3/19). Since I have decided that Biwater is entitled to the entire period claimed, it follows that it is also entitled to the whole of the sum of £30,357.
555. (4) The Final Certificate. Given that the works are complete and that the defects liability period expired on 26^th November 2000 (see sub-clause 36.1 of the General Conditions and paragraph 402 above), I am satisfied that Taylor Woodrow is entitled to be issued with a Final Certificate pursuant to sub-clause 38.1 and to a declaration to that effect.
556. Accordingly, I am satisfied that Taylor Woodrow is entitled to the relief counterclaimed in accordance with the conclusions that I expressed in the foregoing paragraphs, together with interest (where appropriate). I will hear further submissions from the parties as to the appropriate form and terms of the necessary Order.
557. In view of my conclusions in the main action, the only part of Elga's counterclaim against Biwater to which I need refer concerns the outstanding balance of the sub-contract price, less retention (the remainder of the counterclaim is a contingent one which does not now arise, as I understand it: see paragraph 289 to 296 of Elga's closing submissions). The outstanding balance of Elga's sub-contract price has been agreed at £136,301.25: see the quantum experts' 2^nd joint statement (E6/tab 2/22). After deducting retention, the amount claimed by Elga in this part of the counterclaim is £101,894.55 and Elga would therefore appear to be entitled to that amount. However, I will defer expressing any final conclusion on either aspect of Elga's counterclaim until I have heard such further submissions (if any) as Counsel may wish to make with regard to the effect of the settlement agreement between Taylor Woodrow and Biwater: see paragraph 12 above.