Case No: 1159
IN THE HIGH COURT OF JUSTICE
QUEENS BENCH DIVISION
COMMERCIAL COURT
Royal Courts of Justice+
Date: 16th November 2001
B e f o r e :
THE HONOURABLE MR JUSTICE LANGLEY
- - - - - - - - - - - - - - - - - - - - -
| |
AMOCO (UK) EXPLORATION COMPANY |
Claimants |
| |
- and - |
|
| |
BRITISH AMERICAN OFFSHORE LIMITED |
Defendants |
- - - - - - - - - - - - - - - - - - - - -
- - - - - - - - - - - - - - - - - - - - -
Mr M. Barnes QC, Mr D. Unwin QC, Miss F. Parkin and Mr J. Howells (instructed by Messrs Herbert Smith) for the Claimants
Lord Goldsmith QC, Mr R. Siberry QC, Mr A. Griffiths, Mr C. Graham and Mr D. Toledano (instructed by Messrs Freshfields Bruckhaus Deringer) for the Defendants
- - - - - - - - - - - - - - - - - - - - -
JUDGMENT: APPROVED BY THE COURT FOR HANDING DOWN
The Honorable Mr Justice Langley
INDEX
Paragraphs
INTRODUCTION 1 -2
THE PARTIES AND PLAYERS 3 -9
THE RGV 10-23
THE CONTRACT AND ITS CONSTRUCTION 24-100
TERMINATION AT COMMON LAW 101-105
THE LEGISLATIVE FRAMEWORK 106-108
GENERAL CHRONOLOGY 109-204
1995 TO DEPARTURE FOR ROTTERDAM 109-143
DECEMBER 1998 TO TERMINATION
(A) AMOCO AND BAO 144-173
(B) AMOCO AND COVENTURERS 174-199
SUBSEQUENT EVENTS 200-204
THE EVIDENCE 205-211
CONTRACTUAL ISSUES 212-221
THE ROLE OF ADC 222
THE JACKING SYSTEM 223-369
Description 223-255
Amoco's Case 256-257
The Evidence/Chronology 258-304
Jacking History 305-309
Pinion Loads 310
Expert Opinions 311-368
Fracture Mechanics 312-321
Metallurgy 322-329
Magnetic Particle Inspection 330-333
Marine Operations 334-343
Overview 344-356
Operator Issues 357-368
Conclusions 369
BOPS/WELL CONTROL SYSTEM 370-491
Description 370-379
The Contract 380-387
Amoco's Case 388
The Evidence/Chronology 389-465
Expert Opinions 466-484
Operator Issues 485-487
Conclusions 488-491
PRS/TRAINING 492-509
Description 492
Amoco's Case 493-497
The Evidence 498-507
Expert Opinions 508
Conclusion 509
OVERALL CONCLUSIONS 510-514
REMEDIES 515-517
Mr Justice Langley :
INTRODUCTION
- By a contract dated September 26, 1997 the Defendant ("BAO") as "Contractor" agreed with the Claimant ("Amoco") to provide a Jack-up Drilling Unit known as the Rowan Gorilla V ("RGV") to "drill, test (as required), complete suspend, and/or abandon" one or more wells in the North Sea. The contract was to continue for at least one year. The operating hire rate was $175,000 per day. At the time the RGV was under construction in America. The contract contemplated that the RGV would arrive in Rotterdam between August 1 and October 15 1998. In the event it arrived in Rotterdam on December 15, 1998. BAO claims that the RGV came on hire under the contract from 18.30 on December 25. Amoco says it never came on hire and the condition of the RGV together with the delay was such that Amoco was entitled to terminate the contract. Amoco did purport to terminate the contract on January 19, 1999
- The central question in these proceedings is whether Amoco was entitled to terminate the contract either under its express provisions or at common law. Amoco claims a Declaration that its termination was valid. If it was, Amoco claims damages for wasted costs (the quantification of that claim being excluded from the present trial). Amoco makes no claim for damages for any breach of contract on any basis other than entitlement to terminate the contract. BAO counterclaims for Declarations that the termination was invalid and for a daily rate under the contract from December 25, 1998 to October 22, 1999 (the date when BAO accepted what it says was Amoco's repudiation of the contract) and for damages at the same rate thereafter until 25 December, 1999, or the expiry of the 1 year term. That claim is for a principal sum of approximately US $65m. In addition BAO claims the balance of a fee payable under the contract for mobilisation of the RGV from America to Rotterdam and reimbursement of costs incurred in mobilisation from Rotterdam to the first intended drilling location which in the event was to be the Arbroath field. The Arbroath platform is situated some 130 miles east of Aberdeen. The RGV cost some $225m to design and build
THE PARTIES AND PLAYERS
- At the date of the contract Amoco was a subsidiary of the American oil company, Amoco Corporation. In August 1998 a merger was announced between Amoco Corporation and BP plc. The merger was subject to regulatory approval but was completed on December 31, 1998 shortly before Amoco terminated the contract. Thereafter the merged group was known as BP Amoco.
- Amoco had various interests in the North Sea. It was the "operator" of the Arbroath field. Its co-venturers in the Arbroath field were Enterprise Oil plc and Amerada Hess Limited. "Enterprise" had the largest share in the field (41.03%), Amoco had 30.77% and "Amerada Hess" 28.2%. Attached as Appendix C is a list (in alphabetical order) of the individuals mainly involved in the events with an indication to show those who gave evidence. The principal players for Amoco who gave evidence were Mark Gillis, the Central North Sea Drilling Team Leader, who was responsible throughout for the Arbroath project, Morty Denholm, the Arbroath delivery manager following the merger, and John Mogford, then technology Vice- President of BP Amoco, who was responsible for safety matters and at least in part for the decision to terminate the contract. Where it is material to do so, those who were originally employed by Amoco or BP are referred to as "Amoco heritage" or "BP heritage" as the case may be. Mr Gillis was Amoco heritage. Mr Denholm and Mr Mogford were BP heritage
- The Arbroath field had been producing for some years from the existing platform. The platform had no drilling facilities. From about 1996 consideration was given to an "infill development" of the field designed to exploit the reserves better by drilling further wells from the existing platform by the use of a drilling rig.
- In September 1997 the RGV was under construction by a company called LeTourneau Inc which was building it for Rowan Companies Inc. ("Rowan"). Rowan is and was a well-known drilling company and the parent company of both LeTourneau and BAO. The RGV was to be owned by Rowan and operated by BAO. It was largely built at LeTourneau's yard in Vicksburg, Mississippi. LeTourneau's Head Office is in Longview, Texas. LeTourneau has designed and built some 160 jack-up rigs about 40% of the jack-up rigs in operation world-wide
- Daniel McNease is and was executive vice-president of Rowan and president of its drilling subsidiaries. Robert Rimlinger was Rowan's project manager for the construction of "super Gorilla" class rigs such as the RGV (which was the first of its class). Michael Marcom is and was the managing director of BAO. Mr (Bryan) Quinn is and was the RGV Rig Manager
- Much of the equipment on the rig to which it will be necessary to refer was supplied by other companies
i) Cameron (an operating division of Cooper Cameron Corporation) provided most of the well control or blowout prevention (BOP) equipment and related systems;
ii) Varco Industries Inc provided the Pipe Racking System (PRS);
iii) Pacific Steel Casting Company (PSC) manufactured and supplied the pinions for the jacking system.,
iv) National Oilwell supplied the power swivel for which the controls were provided by Louisiana Electric Rig Service (LERS); Ross Hill (the project engineering division of National Oilwell) provided power and control systems; Rigserv provided the driller's chair. The significance of the equipment supplied by these companies however was removed as a result of concessions made by Amoco on Day 34 of the trial.
- The certifying authorities for the rig were the American Bureau of Shipping (ABS) and Det Norske Veritas (DNV). Amoco engaged independent specialists in connection with the certification and testing of the rig called Aberdeen Drilling Consultants (ADC)
THE RGV
- It is not necessary to give more than a brief description at this stage of the RGV and those parts of it which are the subject of particular dispute. No less than 17 expert witnesses have given evidence on various aspects of the RGV and the issues to which that gives rise will be addressed later to the extent I consider it necessary to do so. Attached to this judgment as Appendix D is a list of the experts who gave evidence and the areas of expertise which they addressed
- The complaints Amoco pursue about the state of the RGV fall into three broad categories: the jacking system; the well control (or BOP) system; and the drill floor equipment now limited to the Pipe Racking System (PRS). A fourth complaint that the crew was not fully trained is also now limited to the PRS and can be considered in that context. Some complaints are said to have had serious safety implications, others to have affected efficiency; some to have been readily remediable, others not. To quote Amoco's opening submissions "in addition, BAO's persistent refusal to recognise that anything was wrong is of importance to the repudiatory nature of the breaches"
- The scale of the RGV and the proposed drilling programme in the Arbroath field is impressive. The wells were to be drilled to a length (they were to be substantially horizontal) of 10,000 to 20,000 ft. (2 to 4 miles); the drill string was to be made up of 30 ft. sections of pipe so requiring up to 600 lengths of pipe to be screwed together. Drilling requires machinery to rotate the drill, "mud" to be continuously passed down the drill string and out of the drill bit to return via the annulus (the space between the drill pipe and the well bore or casing) with the cuttings to be cleaned in the shakers and re-cycled. The drill string has to be withdrawn ("tripped") every few days to renew the drill bit and case the hole. The casing provides a secure passage for recovered formation fluids. It consists of sections of steel pipe which are cemented in place. Each section of casing, as the well gets deeper or longer, is of smaller diameter than the one preceding it, so that the ultimate effect is similar to a telescope
- The RGV itself dwarfed the Arbroath platform and hence the drilling equipment on it could be cantilevered out over the platform. The RGV was supported on three triangular legs which could be jacked down to the seabed and then as jacking continued the hull could be lifted up out of and to the chosen height above the sea. The RGV’s elevated weight was designed to be approximately 42,725,000 lbs and it was designed to operate in a water depth of 408 feet
The Jacking System
- The technology is relatively simple. The scale is huge. The 3 triangular legs are jacked up and down using a rack and pinion mechanism. The racks are fixed to each of the 3 corners of each of the 3 legs and corresponding ¼ pitch pinions rotate within 9 elevating units fixed to the deck, one at each of the leg chords on the 3 legs. Individual pinions and their associated motors are identified by letter (for the rack) and number. Each elevating unit has 8 gear units arranged in 4 opposing pairs. During normal operations the jacking system was capable of raising or lowering the hull at a rate of 1.5 feet per minute. The design ratings are stated in Appendix 1 to the Contract, paragraph 1.4.11.10. (Appendix A, at page 36)
Well control equipment
- The purpose of the well control equipment is to prevent blow-outs and to control the pressures that build up in the wells. In simplistic terms the first and main control is provided by the mud itself: so long as its pressure is greater than the pressure of the formation fluids the fluids cannot enter the wellbore. Drilling mud can be weighted according to the pressure anticipated
- There is a particular risk when drilling the early stages into the seabed that the formation will be such that pockets of low-pressure gas may be penetrated. The diverter, installed immediately under the drill rig floor, is used for well control at those stages (in the North Sea commonly until drilling reaches a depth of 1500-1800 feet). The diverter is designed to divert the resultant flow (or "kick") through an overboard line and away from the rig. The reason is that closure of the well by use of a BOP could result in pressures which would fracture the weak rock formations in shallow wells and risk the stability of the legs of the rig on the seabed. The diverter is installed in a housing with a path for the drilling mud to travel via the flowline to the shakers as it returns to the surface. This path can be closed and the flow diverted overboard. The diverter system is an assembly of valves and locks which can be moved out of the way when it is not needed. The valves are opened and closed hydraulically by the BOP control system. The RGV diverter was designed for a normal operating pressure of 750 psi with a maximum of 2200 psi. That is the pressure at which the system could maintain a seal to ensure no leakage on to the rig and diversion to safety overboard
- The Blowout Preventer (BOP) Stack is installed under the diverter once the early drilling stages have been completed. It is the means of well control in high pressure sections above the pressure that the diverter is rated to withstand. The Stack includes large valves which may be closed to contain wellbore pressure and fluid should formation fluids enter the wellbore. The valve (or valves) are closed, the well shut in, and the fluids can then be pumped out of the well under planned and controlled conditions. The BOP control system (sometimes referred to as the Accumulator or Koomey Unit) is operated by hydraulic pressure. The major item of the hydraulic control equipment is the Accumulator Unit in which hydraulic fluid is held under pressure ready for activation. The RGV had two types of blow out preventer. An Annular BOP placed under the diverter and four Ram BOPs placed under the Annular BOP, together making up the "BOP Stack". The BOPs seal the annulus or the entire wellbore if there is no pipe in the well. The Annular BOP (sometimes referred to as the Hydril) is rated to control pressures to 10,000 psi. It has a large doughnut shaped rubber packing element which is hydraulically pushed towards the centre of the wellbore and will close and seal on whatever size of pipe is in use. The Ram BOPs were rated to control pressures to 15,000 psi. Each has two steel blocks (rams) which include large rubber seals. The blocks are pushed into the wellbore by opposing hydraulic cylinders until they contact each other and form a seal. The four Ram BOPs consisted of two prepared with blocks to seal round the size of drill pipe in use; one which could seal round a range of pipe sizes, and one (the blind shear ram preventer) which can cut through any pipe in use and form a seal or seal an open hole. Choke and Kill valves are fitted to ports on the side of the Ram BOPs. They are opened to circulate mud out of (choke) or into (kill) the wellbore when a BOP is closed
The drill floor equipment
- The drill floor equipment is situated on that part of the RGV which can be cantilevered out over a platform and consists of a derrick 170 ft high, a Rotating System, a Pipe Handling System, a Hoisting System and a Mud Circulating System. There is a driller's cabin from which the various functions can be controlled from the driller's and assistant driller's chairs
- The Rotating System rotates the drill string and bit. It consists of a Power Swivel (or Top Drive) and the Rotary Table. The Power Swivel also supports the weight of the drill string for all normal hoisting operations and is a link in the mud system as it delivers mud into the drill string via the wash-pipe assembly. It is essentially a large motor attached to the drill string which can be raised or lowered in the derrick by the drawworks. It was powered by two large DC electric motors. The Power Swivel can also be used to make or break pipe connections
- The Pipe Handling System consists of an Iron Roughneck and a Pipe Racking System. Drill pipe is stored vertically in stands of 90 ft. lengths formed of 3 x 30ft. sections of pipe held in "fingerboards" set back from the drilling area. The pipe racking system (PRS) is used to carry pipe to or from the stands from or to the well centreline to add or remove stands of drill pipe from the drill string. The iron roughneck is a mechanical substitute for the human "roughnecks" who torqued the pipe joints using tongs. The iron roughneck grips the drill string and a stand of pipe presented by the PRS and spins the pin end of the stand into the box end at the top of the drill string torquing the two together. The PRS on the RGV was semi-automatic and computer assisted controlled from the assistant driller's chair in the driller's cabin. It consisted of a vertical column traversing the rig floor from the setback stands to the well centre driven by upper and lower drivers. It had upper and lower arms, the upper arm had a hydraulic grip assembly to clamp, hoist and guide the stand of pipe, the lower arm had a pneumatic claw assembly which provided only stability and guidance at the lower end of the stand of pipe
- The Hoisting System consists of the drawworks, blocks and drilling line. The drawworks is essentially a very big winch or fixed vertical crane used mainly to get drill pipe in and out of the well hole as rapidly and safely as possible. It consists principally of a hoisting drum, brakes and clutches. Power is transmitted through the drum to remove the drill string which is spooled on or off the drum as required. When pipe is being pulled out of or lowered into the well hole the weight of the drill string is held in "elevators" a hinged collar which closes around the pipe and suspends but does not grip it at the pipe joints
- The Mud Circulatory System. Drilling muds are water or oil based slurries with controlled physical and chemical properties. The mud serves to balance the pressure of formation fluids, removes drilled cuttings from the well, cools the drill bit and lubricates the drillstring. For well control purposes, the mud system is also connected to the choke manifold and the BOP stack by the choke/kill lines
Power and control
- Power on the RGV was supplied by 5 generators plus an emergency generator. These provided 3-phase AC power distributed through a main switchboard and an emergency switchboard. Most of the equipment was powered by DC motors. The power to these was controlled by Silicon Controlled Rectifiers in turn controlled by a number of computerised controllers provided by Ross Hill. The relevant instructions came from the chairs and consoles in the driller's cabin
THE CONTRACT AND ITS CONSTRUCTION
- To facilitate the reading of this judgment I have attached as Appendix A the material clauses and appendices of the contract but it is necessary to set out here and comment on those which are material to the dispute
- The recitals and Clause 1 establish three matters of some significance. First, the contract was for the drilling of wells by BAO using the RGV, not for example for the use by Amoco of the RGV. BAO was to supply and pay the crew and do the drilling. Second, although it was known by both parties at the time that Amoco intended to use the RGV for the Arbroath Infill Project, the contract made no reference to Arbroath but only to "locations in waters of the United Kingdom Continental Shelf" which Amoco might designate. Third, Clause 1 defines the drilling of wells as "Well Operations"
- Clause 2 is important. Clauses 2.1 and 2.2 provided for the commencement date and duration of the contract and the same commencement date for charges for the hire of the RGV
- Clause 2.1 provided
"Contractor shall mobilize the Drilling Unit from the United States Gulf of Mexico to Rotterdam, Netherlands with arrival in Rotterdam anticipated to occur sometime between August 1, 1998 and October 15, 1998. The Contract shall commence twenty-four (24) hours after delivery of the Drilling Unit at Rotterdam or when the Drilling Unit commences tow to Amoco's first drilling location, whichever occurs first. This Contract shall continue in effect for a term of one (1) year or until completion of well operations on any well in progress subject always to the provisions of Clause 28. Amoco shall have the option to extend the term of this Contract for a further period of one year ...."
- Thus
i) The clause provides only for an "anticipated" window for "arrival" of the RGV in Rotterdam. There was no "drop dead" date and no express provision for either mobilisation of the rig from America or completion of the rig or delay in delivery at all. The use of the word "anticipated" is I think inconsistent with a time commitment. It is not suggested that at the time of the contract arrival in the time window stated was not honestly and reasonably "anticipated" by BAO. If Amoco wanted and BAO had been prepared to accept a time commitment it would and should have been spelt out. Amoco must (or at least should) have known that "waiting on weather time" following delivery at Rotterdam would be for its account. But it is no surprise that with a new build rig under construction (by LeTourneau not BAO) and the need to transport it across the Atlantic no time commitment was agreed. That is not to say that time for mobilisation or delivery of the RGV might not have been made of the essence in given circumstances nor that delay or further delay might not be material to other breaches of contract if such occurred. In the event, Amoco does not contend that either the time of delivery or the time of mobilisation from America ever became of the essence of the Contract. Although the option to extend for a further year was dependent on further agreement, the one year term itself began on "delivery" and was automatically extended until completion of well operations on any well in progress on its expiry "subject always to the provisions of clause 28" which is the "Termination" provision. The extension is also consistent with the indefinite nature of the time provisions in the contract and reflects the commercial realities of drilling operations. Amoco had the protection that whenever delivery in fact occurred it could complete work in progress when a year expired. Neither start nor finish of the term of the contract was fixed. The rate was fixed. That was in my judgment the bargain, albeit one which Amoco and the co-venturers came to regret.
ii) Mobilisation "to Rotterdam" was expressed in obligatory terms ("shall"). See also paragraphs 53 to 58 of this judgment.
iii) The commencement of the contract was to be the earlier of 24 hours after "delivery" of the RGV "at Rotterdam" or upon the commencement of tow to Amoco's first drilling location. There is a dispute as to whether "delivery at Rotterdam" is to be construed to mean "arrival at or in Rotterdam" (Amoco's primary case and occurring on 15 December 1998) or to mean when BAO made the rig available to Amoco at Rotterdam for the purpose of performing the contractual services (BAO's primary case and which BAO contends occurred on 24 December 1998) or when the rig was fully in accordance with the contract (Amoco's secondary case and which it contends never occurred) or in any event when the RGV commenced tow to Arbroath, which it did at 23.00 on January 9, 1999 (BAO's secondary case). Amoco also submits that even if BAO's primary case on construction be right the rig was in fact made available to Amoco on 15 December in any event.
iv) The clause uses both the expressions "arrival in" and "delivery at" Rotterdam which at least suggests they are used as different concepts. If the contract was intended to commence 24 hours after "arrival" of the RGV in Rotterdam it could readily have said so. The word "delivery" can and, I think, more usually would, involve a tender or handing over of the object to be delivered.
v) The rationale for commencement of the contract to follow only 24 hours after "delivery" rather than at once upon delivery is not obvious. The suggestion of counsel was that it was to allow a short period after "arrival" or "delivery" to elapse for the benefit of Amoco to gear up its operations and before the contract period and so liability for charges (see clause 2.2) began. If that were so (despite Amoco's submissions) it would make equal sense whichever construction was preferred. In contrast no elapse of 24 hours applies to commencement of tow. The suggestion does not to my mind greatly assist the construction issues: both "arrival" and "delivery" would be expected to be well flagged in advance whatever the right construction. But I do think it unlikely that the parties would have agreed either that charges or that the term of the contract were to commence simply on arrival as distinct from delivery of the RGV in the sense on which BAO relies (with or without the 24 hours) at least without any clearly expressed (and I see no basis to imply) contractual commitment that the RGV should be ready for operations on "arrival".
vi) The alternative time of commencement (commencement of tow) is said to be of some significance. The contract very arguably requires and certainly contemplates that the RGV would go to Rotterdam from America and leave from Rotterdam on tow for its first drilling assignment. The only doubt arises from the fact that the alternative commencement time can be read as contemplating that tow to the first drilling location could precede delivery at Rotterdam. If, however, the RGV went to Rotterdam and if "delivery" was co-terminus with "arrival at" an alternative of commencing tow to the first drilling location could never in practice arise save in the 24 hour window following arrival. That would be to give no meaning to the whole clause save that the contract should commence 24 hours after arrival or "delivery" of the RGV in Rotterdam unless it began tow to the first drilling location before the elapse of that 24 hours. Whilst that is a possible meaning of the words and one for which Amoco contends, if it had been the intention it could have been more simply expressed. On the other hand it is also difficult to conceive that commencement of tow from Rotterdam to the first drilling site could practically occur without "delivery" or could have been intended as a commencement time even if a tender of "delivery" plus 24 hours had not already occurred. Neither approach is to my mind wholly satisfying.
- In my judgment, really because of the choice of language, Amoco's primary case that "delivery" occurred simply on arrival of the RGV in Rotterdam fails. Amoco's secondary case that delivery occurs only when the RGV is fully in accordance with the contract also fails. It is not what the clause says. It is also an improbable criterion because of uncertainty and the likely existence of some defects (latent, serious or otherwise) or at least disputes about them
- I think BAO's primary case on the construction is correct. "Delivery" is a recognisable concept generally and in the context of shipping in particular: see The Madeleine [1967] 2 Lloyds Rep 224 at 238. What is or may be defective may yet be and often is nonetheless delivered. The defects may or may not of course give rise to a right in the other party to claim damages or reject the goods. Nor do I think the wording (paragraphs 57 and 58) or the facts relating to payment of the second half of the mobilisation fee (paragraph 146) affect this conclusion. Other considerations may apply to such a payment or the invoice may have been premature. Nor is it improbable, to my mind, that it would be agreed that delivery would be a matter for BAO. It was plainly in BAO's commercial interests to deliver the rig as soon as possible so that the meter for charges started to run
- Although the issues addressed in the previous sub-paragraphs remain of importance (in particular to the operation of Clause 28) it is also important to note what is not (or at least no longer) in issue. Amoco relies on delay in getting the RGV to Rotterdam not as a basis for termination of the contract but as a factor which made what it alleges was the lack of readiness and poor state of the RGV thereafter the more serious because it meant further prolonged or indefinite delays. Amoco does not submit that it would have been entitled to terminate the contract if the RGV was fit and ready for drilling operations on January 19, 1999. Indeed it is accepted by Amoco that it affirmed the contract as regards mobilisation of the RGV from America to Rotterdam which in fact took place only in November 1998. Nor do these issues affect the amount of BAO's counterclaim. Whether delivery occurred on 15 or 24 December or 9 January, the RGV remained without work for at least one year after each date. Moreover whilst Amoco does not even accept that the contract commenced under Clause 2.1 on 9 January it does describe the question as "a red herring" in its reply submissions because "the effective choice is between delivery on 15 or 24 December with commencement occurring 24 hours thereafter". In my judgment whatever the construction of the clause and whether or not delivery had occurred prior to the commencement of tow on 9 January in the circumstances of this case the tow also involved the commencement of the contract as it must have encompassed "delivery" in any acceptable sense of the word
- Clause 2.2 provided for the commencement and termination dates for charges. The former reads
"The commencement dates for charges shall be the point in time following the expiration of twenty-four (24) hours after the delivery of the Drilling Unit at Rotterdam or when the Drilling unit commences tow to Amoco's first drilling location, whichever occurs first."
- Thus charges commence at the same time as the contract commences under Clause 2.1. It follows, in my judgment, that they commenced 24 hours after BAO made the rig available to Amoco at Rotterdam for the purpose of performing the contractual services or at the latest on 9 January
- Clause 3.1 is entitled "The Drilling Unit". It contains the express obligations concerning the fitness and performance of the RGV
- Clause 3.1 relates to "Equipment" :
"The Drilling Unit shall be fully equipped as set out in Appendix 1 and shall be adequate to conduct the Well Operations at the locations specified by Amoco. The Drilling Unit and all other equipment, materials and supplies hereinafter specified as being supplied by the Contractor shall be certified and maintained in first class working condition, shall be suitable for their intended purpose, shall conform to all relevant laws and specifications and, together with the personnel, shall be furnished and maintained by Contractor at its sole cost ...."
- Appendix 1 contained a description of the RGV, its ratings, design criteria (see Clause 3.2) and short particulars of the equipment on it as appears from the extract in Appendix A to this judgment. It is not alleged by Amoco that any of the specified equipment was not supplied nor is any complaint now made about its certification. It is alleged that some of it, in particular the jacking system and well control system, was not maintained in first class condition or suitable for its intended purpose
- Clause 3.2 relates to "Performance"
"Contractor represents that the Drilling Unit shall be capable of moving from location to location, and that subject to seabed conditions, the Drilling Unit shall be capable of performing satisfactorily in United Kingdom Continental Shelf waters .... The design criteria for the Drilling Unit are defined in Appendix 1. The ratings for locations shall be established by Contractor upon Amoco's selection of location ...."
- There is an issue on the pleadings as to when these representations and the obligations in Clause 3.1 were required to be fulfilled. The competitors are: on mobilisation from America, on "delivery" or on commencement (Amoco) or when the relevant piece of equipment was required for well operations alternatively on commencement (BAO). The resolution of this issue is not to my mind as important as it might seem. Amoco makes no claim for damages for breach of either obligation even if it was broken at the earliest of the alternative dates. The crucial issue is not whether there was a breach or breaches but if there was whether it or they were such as to entitle Amoco to terminate the agreement. Whilst that issue no doubt encompasses matters of time, such as likely delays in rectification, it provides a different focus. That said, the emphasis in Clause 3.1 is on the "adequacy" of the RGV "to conduct Well Operations at the locations specified by Amoco" (in the event Arbroath) and in Clause 3.2 on "performing satisfactorily" in the North Sea but again with some specific reference to the chosen location. That, in my judgment, fixes the criterion by which fulfilment of the clauses is to be judged, but they are effective from commencement of the contract. I think this reflects the basic commercial objective of the agreement. Amoco was entitled at commencement to a rig which would be suitable for moving to and drilling at Arbroath when called upon to do so and which was equipped as agreed. That is what it agreed to pay for. On no basis, in my judgment, could either obligation be broken before the contract commenced under Clause 2.1. If it had been intended that the RGV should comply with the provisions at the time of mobilisation from the United States Clause 2.1 would not have been in the form it is. Nor is there any reason why Amoco should have been concerned whether the rig was completed during the tow the duration of which was itself uncertain
- Clause 3.3 provided for "Certification"
"Contractor warrants that, prior to commencement of operations under this Contract the Drilling Unit, having been examined by inspectors of the relevant Certifying Authority, conforms in all details of structure, accommodation, machinery, equipment and in every other manner with all current laws or statutory instruments setting out regulations to be observed in the conduct of offshore operations ...."
- At various stages in the proceedings both parties have raised issues in respect of certification. BAO has relied on the certificates which were issued by ABS and DNV to suggest that the RGV was fit for purpose. Whilst Amoco no longer relies on the alleged invalidity of the certificates it submits that they were unreliable because they were issued without knowledge of the full facts and that in any event as expressions of opinion of their authors they are immaterial and inadmissible
- Clause 3.3 itself is notable in providing for fulfilment of the certification requirements "prior to commencement of operations" and in referring to "offshore operations" and not "arrival" or "delivery" or commencement of the contract. I think, despite not using the defined expression "well operations", the words are to be construed to mean, in effect, drilling or at least preparation for drilling at the chosen location. That accords with the construction I prefer of Clauses 3.1 and 3.2. Moreover if certification and regulatory conformance is required only "prior to commencement of operations" it would be expected that the same criterion would apply to the obligations under Clauses 3.1 and 3.2
- Clause 5 provided for the "compensation" to be paid to Rowan. The provisions are important not only because of the counterclaim and because Amoco contends that even if its termination be held to have been repudiatory only "the breakdown rate" would have been recoverable, but also because they provide pointers to the basic commercial obligations undertaken and the degree of importance the parties attached to them
"5.1 Amoco shall pay Contractor for work performed, services rendered and material, equipment, supplies and personnel furnished by Contractor, a sum computed at the rates specified in this Clause. The period of time for which each rate shall be applicable shall be computed from and to the nearest half hour. The rates ... are based on Contractor's operations being conducted on a seven day week and twenty-four hour work day. The rates ... shall be fixed and firm for the duration of the Contract and are subject to the provisions of Appendix 10."
- Appendix 10 provided for rate adjustments in the event of the RGV not performing to the full extent of some of its efficiency-enhancing features or indeed performing better than expected. Appendix 10 is of some importance as an indication of how the parties themselves "priced" the various features and will be considered later
"5.2 Operating Rate. The operating rate as specified in Appendix 5 shall apply per twenty four hour day and pro rata in respect of any part of a day, payable each and every day except as otherwise provided herein."
- The operating rate in Appendix 5, subject to any adjustment in accordance with Appendix 10, was $175,000 per day
"5.3.1 Standby Rate. The standby rate as specified in Appendix 5 shall apply per twenty four hour day and pro rata ... and shall be payable except as otherwise provided herein :
a) during any period of time that Contractor is unable to conduct operations hereunder due to adverse sea or weather conditions, an act or omission of Amoco ... or failure of Amoco to issue instructions.
b) during any period when the Drilling Unit's primary equipment such as hoisting rotating or pumping equipment is not in use ....
c) ....
5.3.2 Reduced Standby Rate. In the event that operations hereunder are continuously suspended for a duration of seven (7) days or more due to adverse weather conditions, a Reduced Standby Rate shall apply as specified in Appendix 5"
- It is the Standby Rate pursuant to Clause 5.3.1.a) (and/or mobilisation rate: 5.7) which BAO claims in its counterclaim. In Appendix 5 both the Standby and Reduced Standby Rates were $169,750 per day, that is 97% of the Operating Rate
"5.4 Breakdown Rate. The breakdown rate as specified in Appendix 5 shall apply per twenty four hour day and shall be payable pro rata after the first twenty four (24) cumulative hours per calendar month that operations hereunder are suspended due to mechanical breakdown of Contractor's equipment or due to any failure by Contractor to furnish any item which Contractor is obliged to furnish hereunder and shall continue for a maximum period of twenty four (24) hours after which no remuneration shall be paid until the resumption of operations. The time required for conducting routine maintenance such as , but not limited to, ... testing and surface maintenance of blow-out prevention system (hereinafter 'BOP') well control equipment ... and routine top-drive maintenance shall be excluded from periods of suspension under this Clause. In the event that BOP and well control test equipment fail to test, repair time shall fall within the breakdown rate and commence when the first attempt to test has failed and shall continue until an acceptable test has been completed."
- The breakdown rate in Appendix 5 was the same as the Standby Rate, namely $169,750. Thus the effect of this provision was that in each month this rate would become payable after a cumulative 24 hours in which "operations hereunder" were "suspended due to mechanical breakdown of Contractor's equipment" and would then be payable for a maximum of a further 24 hours after which no remuneration would be paid to BAO. In effect the first 24 hours of "breakdown" in each month would not affect the payments to BAO which would remain at the full operating rate, the next 24 hours would be payable at the breakdown rate and after 48 hours no payments would be due. That was the agreed protection to Amoco in the event of "breakdown", and (see Clause 6.1.2) unsatisfactory or materially reduced performance. The provision also has to be read with Clause 28.1.b) which provides for a right to terminate if a breakdown of equipment results in BAO being unable to perform its obligations for 30 days
- There is a dispute about the meaning of the clause. Amoco contends that it applies for so long as there were defects in the well control equipment on the RGV even if (as was the case) the rig never arrived or operated at the Arbroath platform. BAO contends that the clause only applies when drilling operations are or are sought to be undertaken and have to be suspended because of mechanical breakdown, so that in the event the clause never became operative. In my judgment, BAO's submission is plainly right. Operations can only be "suspended" if they have begun or at least would otherwise have done so. The clause is addressing breakdown in use or, in the case of the BOP, a failure to test before drilling such testing being a regular requirement of all drilling operations (paragraph 64). Because it is part of Amoco's case that the BOP and well control equipment did "fail to test" at one point (24 December) at Rotterdam and no acceptable test was completed at least before 3 February 1999 (paragraph 460), Amoco also submits that whether or not it was entitled to terminate the contract the last sentence has a separate life and meaning of its own which results in its application to that situation, so producing "a nil rate" for at least some part of the one year period of BAO's claim. I reject that. The Clause has to be read as a whole. It exempts from "suspension of operations" testing and maintenance time but not if the test fails. That makes perfect sense. Operations (if about to start or started) would have to be suspended if the BOP failed to test. Amoco's construction makes no commercial sense. It would produce a nil rate because of a test failure which had no impact on any relevant operations and which would endure despite any practical need for urgency to put it right
"5.5. Force Majeure Rate. The force majeure rate ... shall be payable pro rata after the first three days of any period in which operations hereunder are suspended due to a force majeure cause as set forth in Clause 27 and shall continue until resumption of operations hereunder or until this Contract is terminated in accordance with the provisions of Clause 28."
- The force majeure rate was $166,250 per day. Clause 27 is set out in Appendix A
"5.7. Movements of the Drilling Rig. During mobilization from Rotterdam ... the standby rate ... shall apply. Amoco shall provide, at its expense, towing and anchor handling vessels, crews ...."
- This sub-clause therefore envisages that the RGV would be mobilised to and from Rotterdam and in accordance with Clause 2.2 that charges will be payable at least on and from mobilisation from Rotterdam. Amoco's obligation to provide tow boats could, at Amoco's option, be passed to BAO on payment to BAO of cost plus a 3% handling charge under Clause 7.2. Amoco exercised the option and BAO claims payment accordingly
"5.7.1 Mobilization. Amoco shall pay Contractor the lump sum mobilization fee specified in Appendix 5 for the mobilization from the Gulf of Mexico to Rotterdam. Mobilization from Rotterdam to Amoco's first drilling location shall begin when the drilling unit commences tow as specified in Clause 2.2 and ending when the Drilling Unit is safely jacked-up at the Well location and Contractor is ready to commence drilling operations with all necessary spud equipment onboard the Drilling Unit ...."
- This Clause therefore also envisaged mobilisation to and from Rotterdam.
- Appendix 5 provided that Amoco should pay BAO a "lump sum" of $3.5m for mobilising the Drilling Unit from the Gulf of Mexico to Rotterdam as follows
"the lump sum is earned immediately upon departure of the Drilling Unit from the US Gulf of Mexico and ... 50% of the sum shall be due and payable immediately upon departure. The balance shall become due upon delivery of the Drilling Unit to Rotterdam."
- Thus the lump sum was "earned" as soon as the RGV left the Gulf of Mexico but was payable as to 50% on departure with the balance due upon "delivery ... to" Rotterdam. There is a dispute as to whether this is referring to arrival at Rotterdam or delivery in one or other of the senses mentioned in referring to the words "delivery at" in Clause 2.1 (paragraph 28iii) above. As stated in paragraph 30 I do not think there is any need to resolve this dispute
- Clause 6 provided for the materials, supplies, equipment, services and personnel to be supplied by BAO
"6.1.1. Contractor shall furnish all items designated as "Furnished by Contractor" in Appendix 4.
6.1.2. Contractor's equipment shall be maintained in sound efficient operating condition at all times. Should the performance of Contractor's drilling equipment become unsatisfactory or the general performance of work hereunder be materially reduced because of Contractor furnished drilling equipment, Amoco shall give Contractor written notice specifying the cause of its dissatisfaction and Contractor shall have a reasonable opportunity to correct the specified deficiency. If Contractor fails to correct or to commence to correct such deficiency within such reasonable time Amoco shall have the right to direct Contractor to cease operations until such deficiency has been corrected and in that event Clause 5.4 shall apply as if such cessation of operations had been a suspension of one of the kinds referred to in Clause 5.4"
- Thus, as appears in a number of Clauses, a notification procedure for "poor performance" was expressly provided for with its own sanctions; in this case, where there is unsatisfactory performance of equipment or material reduction in general performance. On a failure by BAO to correct the deficiency in a reasonable time, the sanction is the operation of the breakdown rate provisions, in effect leading to no payments for the RGV if the "breakdown" persists for 48 hours (see paragraph 49)
- Section I of Appendix 4 under the title "Drilling and Associated Operations Services Performed Under the Contract", sets out by which party and at whose cost the various specified services were to be provided. Mobilisation to Amoco well locations (meaning mobilisation from Rotterdam as distinct from mobilisation to Rotterdam from the Gulf of Mexico) and Demobilisation "operations" are referred to and I think they, together with "Drilling Operations" or "Well Operations" may fairly be characterised as the "operations" which (subject to the context) the Contract addresses when it uses such language
- Clause 6.1.5 relates to the Well Control Equipment.
"6.1.5. Contractor shall demonstrate to Amoco that a sound preventative maintenance programme has been and will be carried out on the BOP and that all well control equipment provided by Contractor hereunder shall meet the requirements specified in Appendix 6.
Contractor shall ensure that all BOP elements ... are new or like new at the commencement of drilling operations hereunder .... at Amoco's discretion a third party inspection may be undertaken of the BOP system with tolerance and wear duly noted and any repairs required will be made prior to spud of the well, said third party inspection shall be for the account of Amoco. Repairs or replacement of parts to bring the BOP to the manufacturer's and/or certifying authority's specification as identified during such inspection will be for Contractor's account."
- Appendix 6 set out detailed "Acceptance Standards" for the Well Control Equipment. The introduction to the Appendix stated
"The well control acceptance standards that follow will be used to determine if the Drilling Unit is acceptable to Amoco for the UK Continental Shelf drilling and completion programme. The check list that follows indicates additional checks that will be made by Amoco well control auditors. In this instance the audit may be completed by an independent consulting company specialising in the audit and acceptance of all types of well control equipment, including the close out of all corrective action requests raised from the audit .... The results of the audit may be presented to Contractor and all agreed corrective action requests closed out in an agreed time frame prior to Amoco acceptance of the Drilling Unit."
- These provisions point up the importance of the BOP or well control equipment. The reason is safety. A blow-out on a rig or platform is potentially an enormous danger to all personnel present. The horror of the Piper Alpha disaster is well known. Hence Amoco had the right to "audit" the equipment. Hence, also, they had the express right to inspect it, a right Amoco did not exercise as such. There is a dispute whether ADC were acting for Amoco as an independent consulting company under Appendix 6. There were no comparable acceptance tests or standards in the contract for any other equipment on the RGV. Nor were there any provisions in the contract requiring BAO to commission the RGV in any particular way nor entitling Amoco to participate or be involved in commissioning. So far as material, the substance of the tests set out in Appendix 6 is addressed later. Clause 11.5.2 also provided that BAO should test the blowout prevention equipment "as often as and in the manner instructed by Amoco's representative"
- The same timing questions arise in relation to Clause 6.1.5 as arise in relation to Clause 3. In my judgment, the answer to the questions is also the same. The Clause refers to the well control equipment "provided" by BAO. As it seems to me the equipment is not "provided" until the contract commences by delivery of the RGV. The Clause also refers to BAO ensuring that all BOP elements are new or like new "at the commencement of drilling operations". That means what it says. Not mobilisation, not commencement of the contract, but commencement of drilling. "Prior to spud of the well" has the same connotation (see paragraph 70). It also accords with all the evidence that the well control equipment must be and is tested prior to drilling, and at no more than fortnightly intervals thereafter
"6.2.1. Personnel. Contractor shall furnish properly trained, qualified and competent personnel in the numbers and classifications set forth in Appendix 3 ...."
- Appendix 3 gives a detailed list of the personnel required, their job descriptions and qualifications. It is not alleged that BAO was in breach of this Clause, save as regards the lack of training on the PRS
- Clause 8.2 provided for Amoco to have
"the right at any time to inspect and reject for valid cause any items furnished by Contractor and Contractor shall replace, at no additional cost to Amoco, such items so rejected with items free of defects or if Amoco agrees, repair such items."
- Amoco did not at any time seek to exercise this right
- Clause 10 provided for Amoco to provide BAO with a well-drilling programme prior to spudding of the well. "Spudding" is the commencement of drilling
- Clause 11 provided for various performance criteria
- Clause 11.2 provided that
"The work contemplated herein shall be performed in a workmanlike manner and in accordance with good oilfield practice and shall be subject to the general rights of inspection and supervision herein provided to Amoco to secure the satisfactory completion thereof."
- Clause 11.7 required BAO and its staff to comply with the provisions of "the safety case" for the RGV and "the safety management system" established by BAO (see paragraphs 106 to 107). Amoco had the right (11.7.5) "at its complete discretion" to monitor and audit compliance with these provisions
- Clause 15.1 obliged Amoco to provide BAO with "sufficient rights of ingress" to the location where the well was to be drilled "for the performance by Contractor of all work contemplated by this Contract". BAO alleges that the termination of the contract was a breach by Amoco of this clause
- Clause 23 is one of the Clauses (the other is Clause 28) on which Amoco relies in support of its contention that it was entitled to terminate the contract pursuant to its express provisions. The title to Clause 23 is "Unsatisfactory Performance"
- Although the sub-Clauses in Clause 23 are all prefaced "23.1" there is no 23.2. Indeed it is obvious from the wording that what was agreed was the result of deletions to or changes in a draft or form carried out rather inelegantly
- Clause 23.1.1. reads
"Without prejudice to the provisions of Clause 28, if Contractor has, in the opinion of Amoco, failed to conduct its operations hereunder, including any part of its operations, in a diligent skilful and workmanlike manner for reasons within the control of Contractor, Amoco may give Contractor written notice in which Amoco shall specify in detail the cause(s) of its dissatisfaction. Should Contractor fail or refuse to remedy the matter(s) complained of or, having so commenced, fail to proceed diligently to remedy to Amoco's satisfaction the matter(s) complained of within a period of seven (7) days after the written notice is served by it, Amoco shall have the right, without prejudice to any other rights specified hereunder, at its option either (my underlining) to terminate this Contract as provided in Clause 28.1(d)."
- There is no "or" to the "either" which is another illustration of the rough nature of the drafting. It is not suggested (rightly, despite the language) that the Clause provides that a failure to remedy in 7 days is itself a breach of the clause. The 7 days qualifies the start of the activity not its result. The provision, which must be read with Clause 28.1.d), therefore requires that
i) In the opinion of Amoco BAO fails to conduct its operations or any part of them in a diligent skilful and workmanlike manner for reasons within its control;
ii) a notice which specifies "in detail" the causes of Amoco's dissatisfaction; and
iii) Refusal by BAO to remedy those matters or failure by BAO to proceed to that end within 7 days or, having commenced to remedy them, failure to proceed diligently to do so "to Amoco's satisfaction".
- The only matter on which Amoco relies under these provisions is the well control equipment and particularly on the alleged contamination in it. That is because it is the only remaining matter of complaint for which it is said notice was given by a letter dated 8 January 1999, the only notice relied upon by Amoco under Clause 23.1.1
- There are disputes as to whether notice was ever properly given by Amoco under this Clause and whether the Clause stands alone or depends on fulfilment of Clause 28. The final words in my judgment provide an answer to the last dispute. Clause 28.1.d) must be satisfied also: Clause 23.1.1 simply triggers its application. That construction also accords with the cross-reference to Clause 23.1.1. in Clause 28.1.d) itself and with the reference to Clause 28 but not Clause 23 in Clause 2.1. The practical effect of the two provisions is in any event I think the same: see paragraph 84. The "operations" which BAO is to "conduct" under the contract are, I think, the Well Operations referred to in the Recital and Clause 1 and the Mobilisation and Demobilisation operations referred to in Appendix 4 (that is from Rotterdam to Arbroath and thereafter). The requirement for the notice "to specify in detail" the matters complained of and the severity of the sanction (termination) are I think intended to ensure that BAO is left in no doubt as to what it is which it is required to rectify
- Clauses 23.1.2 and 1.3 provide for Amoco to take over responsibility for the RGV and to continue operations in the event of insolvency of BAO and for certain payments to be made in that eventuality
- Clause 27 provides for force majeure and by 27.3 that should any circumstance of force majeure continue for more than 30 days then, unless Amoco chose to extend its well drilling programme for a period equal to the period of the delay, Amoco might "terminate this contract as provided in Clause 28"
- Clause 28 is entitled "Termination"
"28.1 Termination by Amoco
Amoco may at its option terminate this Contract forthwith, with no liabilities accruing except Amoco's responsibility for payments to Contractor as set out in this contract up to the date of termination if:
a) a force majeure condition prevails for a period of thirty (30) consecutive days or more;
b) a breakdown of Contractor's equipment results in contractor being unable to perform its obligations hereunder for a period of thirty (30) consecutive days or more;
c) as provided under Clause 22, Contractor becomes insolvent;
d) Contractor, having been advised in writing by Amoco of unsatisfactory performance under the provisions of Clause 23.1.1 or that Contractor is in default of any of the terms and conditions of this contract, fails or refuses to take action to rectify the fault or fails to commence to rectify the fault within seven (7) days;
e) the Drilling Unit develops a major fault which has the effect of suspending operations hereunder for more than thirty (30) consecutive days;
f) ...."
- The two provisions of Clause 28.1 on which Amoco relies are sub-paragraphs b) and d). The latter has to be read with Clause 23.1.1. So read, whilst the language of the two sub-clauses is different, it is rightly accepted that the effect is the same. BAO is in breach if it fails or refuses to take action to rectify the notified default or fails to commence to do so within 7 days. It does not apply when BAO has commenced to take and is taking action to rectify the default. The one difference in the language of the two clauses which it is suggested gives rise to a potential conflict is the reference in Clause 23.1.1 but not Clause 28.1.d) to a failure to proceed "diligently" to remedy the faults "to Amoco's satisfaction". Amoco submits that if BAO commences to remedy the fault but does not proceed diligently to do so to its satisfaction then it can terminate the contract. BAO submits that Clause 28.1.d) is "the governing clause" and it precludes such a route to termination. In my judgment Amoco is right and BAO wrong in these submissions. The two Clauses are to be and can be read together and so read the criteria provided for by Clause 23.1.1 can readily be read into Clause 28.1.d) which otherwise would leave such matters to implication. That said, "Amoco's satisfaction" does not give rise to a subjective test any more than proceeding "diligently" does. At the least if it were the case that Amoco required some particular or specific action to be taken in order to be satisfied of a remedy to a specific fault in my judgment the rationale of Clause 23.1.1. is that it should be spelt out so that BAO could be in no doubt of what was required against the risk of the contract being terminated
- Amoco does allege (by a re-amendment of the claim served on November 22, 2000) that BAO refused to strip down or inspect the BOP system within the meaning of these Clauses
- Sub-clauses a) and e) of Clause 28.1 deal with events which in effect cause a 30-day cessation of operations. Sub-clause b) requires "a breakdown" of BAO's equipment which results in BAO "being unable to perform its obligations" under the contract for 30 days. Thus, in commercial terms, the contract contained an express regime for the consequences of "breakdowns"; a nil breakdown rate after 48 hours, termination after 30 days
- Clause 28.1.b) gives rise to a number of issues. Again Amoco only relies on the well control equipment under this Clause. That is presumably because no other extant matter of complaint could even arguably be said to fall within its terms. The Clause requires (i) a breakdown of BAO's equipment, (ii) a breakdown which results in BAO being unable to perform its obligations under the contract and (iii) the inability to perform those obligations to endure for at least 30 consecutive days. Each of these three requirements is very much in issue. The first raises questions such as whether equipment which has never been installed or put to the intended purpose can suffer "a breakdown", and whether the fact that equipment needs to be tested or operates below specification can be said to amount to "a breakdown". The second raises the same question but in a wider sense. BAO is I think plainly right in the submission that the breakdown must be such as in effect to prevent BAO performing its obligations viewed as a whole and so substantially to prevent mobilisation or drilling or at least to prevent mobilisation or drilling safely or with reasonable efficiency. It would not suffice that the breakdown merely involved a breach of an obligation under the contract; it would have to have the consequence of BAO being "unable to perform its obligations under the contract" which, I think, is directed to substantial non-performance of the mobilisation or drilling obligations in the contract as a whole by BAO. There is also a timing question. The present tense is used which at least suggests that the obligations addressed are those presently in view at the time the question is to be asked
- Whilst I do think (contrary to BAO's submissions) that the Clause can encompass breakdown in a number of items of equipment considered together I do not think there can be a "breakdown" of equipment which "results" in BAO "being unable to perform its obligations" in the sense indicated unless and until the item or items of equipment alleged to have that consequence is or are sought to be used or called upon in or for the performance of those obligations. As will be seen that was in fact never the case with the well control equipment. Such a construction does not mean, as Amoco submitted, that if the equipment was unserviceable from the outset and remained so for 30 days Amoco would be without a remedy. It does mean that Clause 28.1.b) looks to the consequence of the "breakdown" and it is only if that consequence results in, say, an inability to drill for 30 days either when the equipment is first sought to be used or on any subsequent use that the Clause bites. There is nothing uncommercial about that. A breakdown without consequence in terms of the performance of the substantial obligations under the contract would not normally be expected to give rise to the ultimate sanction of termination. Nor does it necessarily mean that Amoco is obliged to take the rig on location and wait for 30 days before the Clause bites, albeit I do think the Clause looks to a failure in use or attempted use. If the condition of the rig were such that following delivery at Rotterdam it could not properly be mobilised to Arbroath and/or commence drilling there for 30 days after that was properly required of it I think the Clause could bite. The question ultimately is simply one of applying the chosen words to the facts as found. But the point remains that the proper functioning of the well control equipment was required for the performance of BAO's drilling obligations and it is for Amoco to establish that defects in the well control equipment resulted in BAO being unable to perform those obligations for 30 days or more
- The sub-clause can have no operation unless the breakdown lasts for 30 consecutive days. 30 days from when? As will be seen this point is also of real significance because when Amoco purported to terminate the contract on 19 January whilst over 30 days had elapsed from the arrival of the rig in Rotterdam (which is the primary period on which Amoco seeks to rely), 30 days had not elapsed from the date on which BAO wrote to make the RGV available to Amoco and thus from the date when BAO submits (and I have decided as a matter of construction) the contract commenced (see paragraphs 28(iii) to 30). Hence the importance of Amoco's submissions that BAO did in fact make the RGV available and so deliver it to Amoco by a fax sent on December 15, 1998 (read with Amoco's fax sent on December 8 and another fax sent by BAO on December 14) and that the 30 day period could start even before the contract itself commenced under Clause 2.1.
- The latter submission raises a further question of construction. The sub-clause requires the breakdown to result in BAO being unable to perform "its obligations hereunder". The only express "obligations" which can arise before commencement of the contract under Clause 2 relate to "mobilisation" to Rotterdam. That obligation is itself specifically addressed in the first sentence of Clause 2.1. No case is made by Amoco that BAO failed to perform it whether through breakdown or otherwise. The concept of terminating a contract (which is the subject of Clause 28) when the contract has not commenced is both legally and linguistically improbable. I also think the 30 day period is plainly intended to be a period within the one year contract term. It follows that I do not think Clause 28.1.b) was available to Amoco as a basis for termination of the contract unless Amoco can establish a case on the facts that "delivery" in the sense to which I have referred took place no later than 20 December 1998 (30 days before 19 January)
- Clause 28.2 provides for termination by BAO:
"This Contract may not be terminated by CONTRACTOR except for non-payment by AMOCO ...."
- The non-payment has to persist for 30 days after notice. Amoco contrast this provision with the provisions for termination by Amoco which are not expressed in such exclusive language. So, it is submitted, the contractual regime is not exhaustive, and Amoco retains the right at common law to terminate the contract for what may loosely be termed a fundamental or repudiatory breach of contract. BAO does not contend that the common law right is wholly excluded but that it has to be read subject to the contract, including its express termination provisions, setting the parameters for what could amount at common law to a repudiatory breach of it and to a submission that if the matters of complaint do fall within the ambit of Clause 28 then Amoco is indeed limited to establishing strict compliance with its terms: see paragraph 104
- Clause 30 gave Amoco rights to assign the contract. Clause 31 included English law and exclusive jurisdiction clauses (31.1); a non-waiver clause (31.4); a clause providing that in the event of conflict or inconsistency between clauses of the contract and the Appendices, the former should prevail (31.6) and an "entire contract" clause (31.7)
- Appendix 7A provided for the observance of stated minimum safety, health and training requirements
- Appendix 10 provided for "Performance Criteria Rate Adjustments". The detailed provisions are of some significance in consideration of the complaints about the PRS and training as they place them in the context of the "price" the parties appear to have put on operation of the equipment. The Appendix records that the RGV has features which were critical to "achieving maximum operational efficiency". In general terms Clause 1 of the Appendix sets out a number of these features and provides that if they do not operate or operate properly "a variance" will be recorded. Clause 2 provides for the parties to agree whether an event constitutes a variance and its duration and if they do so agree that the Operating Rate shall be reduced for the duration of the variance over 1 hour by $15,000 for all the features set out save one, namely the pipe handling system, where the reduction is $5000. As Mr Gillis and Ms Turton acknowledged, the practical effect of these provisions was that if the RGV did not perform to its high specification, the rig rate would be reduced to the market rate for a standard rig at the time the contract was negotiated
- Clause 4 of the Appendix provided that a variance reduction should only be made against the operating rate an
"regardless of cause, the time of any variance agreed to pursuant to the provisions of this Appendix 10 shall not be considered as a mechanical breakdown or failure on the part of the contractor as set forth in Clause 5.4.1 of the contract."
- Clause 31.7 provided that the contract could only be varied by a specific form of written agreement. There were two such agreements. The first (Appendix A pages 23 to 24) signed by BAO in March 23, 1998 and by Amoco on April 21, 1998 provided principally for two cantilever extension beams to be provided so that "cantilever skid out" could extend to 85 feet. That was to enable the RGV to drill all the expected wells from a cantilever extended out over the Arbroath platform and to do so without having to move position.
- The second variation (Appendix A pages 25 to 26) provided for the replacement of the Derrick shakers referred to in Appendix 1 to the Contract with Thule shakers to be supplied by and at the expense of Amoco
- Both variations provided that "except as specifically amended herein, all terms and conditions of the contract remain unaltered and in full force and effect"
- I refer to this provision because BAO alleges in the pleadings that the two variations caused delays in completion of the RGV for which Amoco was responsible. That is disputed by Amoco both in fact and because this provision, it is said, means that the "anticipated" arrival date in Rotterdam in Clause 2.1 remained in full force. Mr Marcom, however, said that as the contract contained no delivery date it did not occur to him to think of changing it. In the event this is an issue I do not feel it necessary to address further
TERMINATION AT COMMON LAW
- Although detailed submissions have been made on this issue I think, so that the consideration of the technical issues and evidence can be seen in context, I only need to set out briefly what Amoco would need to establish in order to succeed in its case that it was entitled at common law to terminate the contract
- In opening Amoco's case, Mr Barnes accepted that the language used by Diplock LJ in Hong Kong Fir Shipping Co v Kawasaki Kisen Kaisha [1962] 2 QB 26 at pages 72 to 73 was an appropriate description of the question which the court had to ask itself, namely, adjusting the words to the circumstances of this case, whether, at the date Amoco purported to rescind the contract, any delay or default due to BAO which had already occurred and any further delay which was likely to occur in rectifying that default, and the conduct of BAO in relation to those defaults, were "when taken together such as to deprive" Amoco "of substantially the whole benefit which it was the intention of the parties Amoco should obtain" from the contract
- As I have already indicated (paragraph 92), BAO did not in the event (and rightly in my judgment) submit that the right to terminate at common law was wholly excluded by reason of the contract containing its own termination provisions. BAO (and Amoco) did of course submit that those and other provisions of the contract were both definitive of the "benefit" which it was intended that Amoco should obtain from the contract and indicative of the extent of the gravity of any default which would be necessary to justify termination at common law. As it was put in Amoco's Closing Submissions "the contractual provisions give a pointer to what the parties regarded as the commercial realities". The benefit to Amoco was the provision for a period of one year commencing on the commencement date of a drilling unit equipped and manned so as to be capable of drilling in the Arbroath field to seek to achieve further production from the field
- The Contract itself contained its own scheme for compensating Amoco for reduced efficiency or performance by reduction in the operating rate for the period of the reduction in efficiency or performance (Appendix 10) and by a breakdown rate reduced to a nil rate after 48 hours. It also contained in Clause 28.1 its own provisions for termination which were effective after a breakdown of sufficient gravity to cause BAO to be unable to perform its obligations under the contract lasting 30 days or a major fault causing a suspension of operations for more than 30 days. Those provisions themselves must in my judgment form part of an appreciation of the benefit the parties were intended to derive from the contract. Thus circumstances otherwise within the scope of the termination provisions but falling short of the precise terms would in my judgment not give rise to the right to terminate at common law for the very reason that the parties agreed when and how such circumstances should have that consequence: see Lockland Builders Ltd v Rickwood (1995) 46 Con LR 92 (CA). The provisions are ones for the benefit of both parties not just for the benefit of Amoco involving as they do time and notice constraints. For present purposes loosely expressed what I think that comes to is that to justify termination at common law something "worse" or not addressed by those provisions would be required. No doubt it was in recognition of that requirement that Amoco's case sought to emphasise the safety considerations involved in the jacking system and well control equipment and the alleged refusal or failure of BAO to explain matters or rectify them. As Mr Barnes put it in Opening, Amoco's case was that on 19 January 1999 it was not safe to take the RGV out to the Arbroath platform and BAO was refusing to make it safe to do so. In Amoco's written closing reply submissions the "core reasons why Amoco terminated the contract" are said to have been
"The reoccurrence of the well control and jacking system failures - despite BAO's assurances - meant that there were serious issues to be addressed before the rig could be used. Both operator experts agreed that Amoco had legitimate concerns. But BAO maintains, even now, that it had met its obligations and that there was no more to be done."
- This formulation illustrates another area of difference in the parties' submissions. BAO submits that repudiation by "renunciation" of a contract is different in principle from what may be termed repudiation by non-performance. Amoco submits that the two at least overlap. The question is discussed in Chitty on Contracts Vol. 1, para 25-017 and following. I do not think it necessary to analyse this further in the circumstances of this case. An express refusal to perform a contract or part of it may be repudiatory; likewise in my judgment a refusal to remedy or acknowledge a breach of contract may (as Amoco submits) be repudiatory. But in either case to be so it must amount to conduct which objectively is characterised as evincing an intention no longer to be bound by the contract such as to deprive the other party of substantially the whole benefit which it was to obtain from the contract. In that context, I think, for example, questions may arise as to the other party's failure to invoke contractual rights such as a right to require work to be done and as to whether a refusal to acknowledge or remedy a breach is a reasonable assertion of rights under the contract rather than a repudiation of it. Certainly it cannot be right to seek to characterise a statement of belief that equipment is fit for its purpose as a refusal to perform obligations unless perhaps it can be shown to have been dishonest or unreasonably held: see Woodar v Wimpey [1980] 1 WLR 277. Ultimately it is a question of degree requiring consideration of all the circumstances. That is, I think, the effect of Federal Commerce & Navigation Co Ltd v Molena Alpha Inc [1979] AC 757 where the "individual demerits" of the owners conduct were such that it was held to go "to the root" of the particular contract and particularly so because it went to an immediate commercial need of the charterers for the bills of lading which the owners were refusing to provide
THE LEGISLATIVE FRAMEWORK
- The vital importance of safety in the North Sea and its precedence over operating and commercial considerations is of course not in issue. The governing law is the Health and Safety at Work Act 1974. The specific regulations which govern the safety of and responsibility for offshore working in UK waters are the Offshore Installations (Safety Case) Regulations 1992/2885 to which I shall refer as “the SCRs”. The basic approach is to make the owner/operator of an installation responsible for identifying risks ("major accident hazards") and determining how to minimise them to “as low as reasonably practicable” (“ALARP”). The Health and Safety Executive (HSE) is given authority to approve Safety Cases and Safety Management Systems and to audit for compliance with them. As Mr Poss put it
“In short, the regulatory authorities do not tell offshore operators what or how to do something, but they require the operators to tell them what they plan to do, assure them that the plans are safe, and then expect the operators and relevant personnel to live by them.”
- An approved Safety Case is required for both fixed and mobile installations. There is also a requirement where (as would have been the case had the RGV gone to work at the Arbroath platform) two installations with different owners and Safety Cases are working together to prepare and have approved a Combined Operations Safety Case (COSC). The COSC and any Safety Case must identify the major hazards, quantify the risks and seek to reduce them to ALARP. Safety-critical elements ("SCEs") have to be identified and written procedures provided to ensure they are and remain in good condition. There is, as one would expect, a continuing duty to keep safety matters under review. There is no dispute that both the jacking system and the well control system were SCE's
- There is no need to set out any further detail about the legislative framework. Amoco does not allege any breach of any legislation or of Appendix 7A to the contract. Nor was it put to any witness who gave evidence on behalf of BAO/Rowan that the company or the witness was or would have been in breach of any legislation or the Appendix. Nor is there a single document which refers to such a probability. The highest Mr Barnes sought to put it was that had the RGV gone to work at Arbroath in the state in which Amoco say it was at 19 January 1999 then the safety risks would not have been reduced to ALARP. It is also worth stating that these obligations apply to both BAO and Amoco and the SCRs impose a duty to co-operate
GENERAL CHRONOLOGY
1995 TO DEPARTURE FOR ROTTERDAM
- Rowan contracted with LeTourneau for the construction of the RGV on 24 October 1995. Delivery at that stage was to be on or before 30 June 1998. In about May 1996 Rowan made a presentation of the RGV to various companies, including Amoco. The RGV was promoted as "the most technologically advanced mobile platform unit constructed to date". In July 1997, following an approach from Amoco, Rowan submitted a proposal for a one year programme starting in the second/third quarter 1998. There was some dispute as to whether the day rates quoted and in the event agreed were "high" but in the event there was not much in issue. The rate was "high" in the sense that it was some degree in excess of market rates (themselves high at the time) but it could be justified by the supposed advanced capabilities and so efficiency of the RGV. The extent of the efficiency was reflected in Appendix 10, the adjustments to the rate agreed in the event of "variances" affecting the performance of the RGV (paragraph 95)
- At the time the Contract with Amoco was signed on 26 September 1997, Rowan had informed Amoco that delivery was currently expected to be in September 1998 with the possibility of slippage to mid-October. The contract provided for the "anticipated" window of delivery from August 1 to 15 October: Clause 2.1
- Amoco's estimate in 1997 of the time required to drill the Arbroath Infill Project was about 9 to 10 months. There was therefore a commitment to BAO for an "overhang" period for such part of the one year term of the Contract as might not be required for Arbroath. Moreover that overhang commitment was a commitment for Amoco alone and not an obligation of or to be shared with the co-venturers. The documents also show that it was Amoco's intention (communicated to Rowan) that the RGV should be used first to drill what is called a "shakedown well" at a location other than Arbroath. As its name implies, a shakedown well is an uncomplicated well which can be drilled so as to establish the working of a new rig and the abilities of the crew to use it to optimum effect. Mr Gillis said and the documents show that at this time Amoco envisaged using the RGV to drill such a well on the Appleton field which was then believed to be an exciting prospect. Mr Craig said, and the documents show, that Enterprise made the drilling first of a shakedown well elsewhere a condition of agreeing to the Arbroath Infill project. In September 1997 Amoco was looking to the Arbroath project using the RGV "commencing in November 1998". In January 1998 it was recognised that the project might not start until 1999.
- The rig was too large to pass under the bridges on the Mississippi between LeTourneau's yard at Vicksburg and the sea. The intention was that it would be part built at the yard, "walked" to the river for launch and then sailed downriver to Sabine Pass where the work on it would be completed, including most of the work on the legs and the installation of the drill floor equipment.
- Both Addenda to the Contract were agreed whilst construction was proceeding at Vicksburg. Walking the hull to the river had been scheduled at the time of the Contract to take place in April 1998. In the event it took place between 18 and 30 June. It is a somewhat remarkable procedure involving a series of jacking operations in which the rig is levered forwards on a fulcrum of earth, placed at the aft end behind the aft legs, and takes "steps" of a few feet at a time as the bow leg is jacked up causing the stern of the rig to pivot on the earth and the aft legs to ground a few feet further forward each time
- In June 1998 a Hazop (Hazard and Operability) study was carried out for the mobilisation of the RGV to Arbroath. A Hazop is a well established procedure for hazard identification. It also identifies the safeguards and, if they are not considered to be adequate, a recommendation is made accordingly. The hazop worksheets dated 18 June covered the move of the RGV to its final position alongside the Arbroath platform. It recorded that the platform and pipelines would be shut down, depressurised and demanned, and the use of tugs and positioning equipment to control the RGV. For jacking up operations and pre-load tests at the platform the hazard of equipment failure caused by "failure of jacking motors" with the consequences of "delays in jacking operations" was shown as addressed by the safeguards of "redundancy of jacking systems, thrusters, tug boats and anchors.
- The Hazop was conducted by a high level team which included Mr Gillis, Mr Hosie, Mr Steve Johnston, Mr Mark Jones and Mr Price of Amoco and Mr McDonald of BAO/Rowan
- The tow downriver to Sabine Pass began on 21 July and the rig arrived there on 26 July
- Amoco was generally aware of the delays to the schedule. Mr Cowie and Mr Gillis knew in March the rig would not arrive in Rotterdam before the end of October. By early July Mr Hosie and Mr Gillis knew it would not arrive before the end of November. No complaint was made; nor surprise expressed. Amoco (in particular in the documents prepared for and at a meeting held on 28 May 1998) sought to persuade the co-venturers that partly because of the risk of weather delays in getting the rig to Arbroath, arising from later delivery of the rig in the winter months, it would be preferable to take the rig straight to Arbroath and not drill a shakedown well elsewhere first. The fact was that by this time Amoco had no well which could serve as a shakedown well. Appleton had not lived up to expectations. Part of the argument used to persuade a reluctant Enterprise to support this course (and to satisfy the Health and Safety Executive) was a proposal by Amoco that it would put resources into and be closely involved in seeking to ensure that the RGV was fully commissioned on arrival to try to avoid startup inefficiencies. The Contract provided (Appendix 6) for acceptance standards for well control equipment but gave Amoco no rights in respect of commissioning or testing otherwise. The proposal led to the presence on the rig of ADC and Amoco personnel at Sabine Pass in September 1998 and thereafter until the rig was mobilised to move to Rotterdam in mid-November. It also led to some friction with Rowan. Rowan felt confident in its own abilities to commission the RGV and satisfy ABS and DNV and had never used or needed a documented process such as ADC and Amoco sought to evolve. Nor had such a procedure been used for the RGIV for which Amoco was also the first customer. Mr Gray, who was in day to day charge of Rowan's involvement with the rig, was wholly convincing in saying that Rowan knew what the equipment should do and how it should do it and, when it was ready, how to prove (or not) as much. Moreover DNV and ABS were also content with such a procedure. It was no surprise after Mr Gray's evidence that criticisms of the commissioning process were not pursued with others
- Thus, when Mr Palmer was supplied by Amoco with a "Paper" entitled "Rowan Gorilla V Well Operations at Arbroath" (prepared by Amoco for the May meeting with co-venturers) he wrote to Mr Gillis on 9 June 1998 stating that Rowan was fully capable of delivering the RGV "fit for purpose" and did not require the intense oversight suggested in the Paper. He also said that until Rowan was satisfied that the RGV was ready to operate there would be no financial obligation on Amoco. Nonetheless ADC prepared a substantial document finalised in July 1998 called “Drilling Systems Acceptance Test Report” which became the basic working document used by ADC and Amoco for their assessment of the RGV.
- On 1 September the HSE approved the RGV Safety Case which had been submitted on 1 June. Also in September 1998 the first issue of the Arbroath and RGV "Combined Operations Safety Case" (COSC) was finalised. The HSE approved the COSC on 13 November. It addressed the hazards and risks arising from a mobile rig installation drilling new wells through a fixed installation and was prepared to comply with the SCRs (paragraphs 106-107). It is a formidable document. It spelt out that prior to arrival of the RGV in the 500 metre safety zone round Arbroath the platform and RGV Offshore Installation Managers (OIMs) had full responsibility for the platform and rig respectively but once within the 500 metre zone they were to hand over responsibility for the safety health and welfare of personnel and the security of the installation to the OIM for the Combined Installation ("the Complex OIM"), who was to be Mark Jones of Amoco (or his alternate, Tony Mahoney)
- On 17 September at Sabine Pass a tooth on a ¼ pitch pinion broke on the gear unit at B8 while the RGV was jacking. On September 25 two more ¼ pitch pinions, E3 and E6 were found to have broken teeth by personnel working on the rig. The broken tooth from E3 was found lying in the housing of the gear train. The broken tooth from E6 was never found. These pinion failures form a significant part of Amoco's case on the jacking system
- On 1 November 1998 a Hazop review meeting was conducted. Those present included those present at the June Hazop (paragraphs 114 to 115) save for Mr Gillis. Mr Quigley was also present for Amoco on this occasion. There were no revisions or material changes to the worksheets for moving the RGV to Arbroath or jacking it up there to drilling height. That is, as Mr Bartle agreed, evidence of the lack of concern at the time about the pinion failures discovered at Sabine Pass. Mr McDonald said, and I accept, that the pinion failures were not referred to at all at the meeting. Mr McDonald was Amoco's main point of contact with Rowan/BAO in Aberdeen and he was not even aware of the pinion failures until after 12 January 1999
- There is no dispute that the RGV was not "complete" when it left Sabine Pass and that it was then contemplated that work would continue on it not only during the tow to Rotterdam but at Rotterdam. ADC had taken to issuing "Advanced Recommendations" to Rowan about the rig at Sabine Pass. These were reviewed at the beginning of November and a list prepared by ADC of those still outstanding which were divided into Corrective Action Requests (CARs) to be completed on the tow and Recommendations. Mr Sleightholme (the ADC teamleader) wrote to Mr Quinn and Mr Snow (copied to Mr Cowie and Mr Hay) on 8 November attaching a copy of what he called "the commissioning report updated to the current status, with actions recorded as completed or outstanding". I shall refer to this as the ADC interim report. Mr Sleightholme requested return of the documents with Rowan's comments added by 11 November. The letter added that "at Rotterdam, the latest commissioning status will be reviewed and outstanding activities determined." The ADC interim report is a document of some importance. So far as material its contents are considered under the particular headings to which they relate
- Mr Sleightholme provided the list of CARs and Recommendations to Mr Quinn and Mr Snow on the next day (9 November) accompanied by a letter which described the (54) CARs as "actions which require corrective action by Rowan prior to acceptance of the rig by Amoco" and the Recommendations as "not mandatory but, if completed, will improve the safety and operability of the rig". Rowan (Mr Quinn) objected to this wording substantially on the ground that Amoco had no contractual right to make such a stipulation. Mr Cowie agreed and the letter was re-phrased so that the CARs were described as identifying "conditions which we consider not to comply with UK legislation, safe design or good operability. Amoco request that whenever possible the listed items are rectified prior to the rig going on contract". The CARs are also important and where material are considered in the context to which they refer. Rowan did respond to the ADC interim report with comments on 12 November and to the CARs on 13 November
- The rig left Sabine Pass for Freeport, Texas on 16 November. At Freeport it was loaded on the carrier "Transshelf" which the evidence establishes was the only carrier in the world at the time capable of transporting the RGV by dry tow to Rotterdam. It left for Rotterdam on 20 November
- There were significant events during this period which had an impact on the commercial appraisal of the Arbroath Infill Project. BAO submits the events in fact explain much of what happened including what it submits was the wholly unjustified termination of the contract. Amoco submits that the events are irrelevant because, as it contends, termination was justified by the delays in delivery of the rig, its condition and BAO's conduct in relation to them. That termination was commercially attractive to Amoco and the co-venturers is, of course, indeed irrelevant if Amoco was legally entitled to take that course. On the other hand in my judgment Amoco's motivation may, for example, throw revealing light on some of the evidence and submissions such as Amoco's alleged loss of confidence in and the intransigence of BAO which form an important part of Amoco's case on repudiatory breach
- At the time the contract was entered into Brent crude oil prices had been around $18/bbl. By the beginning of 1998 they were around $16/bbl. In October 1998 they were $12/bbl falling further to $11/bbl in January 1999. As early as mid-March 1998 Enterprise had calculated that the Arbroath Infill Project would be marginal at £10/bbl and negative at £8/bbl. The rate of exchange at the time was of the order of $1.6 to £1. Lower oil prices also meant lower market rates for rigs
- On 1 August 1998 BP and Amoco announced the proposed merger of the groups. Until 1 January 1999 there were competition restrictions on the exchange of information between the groups
- On 6 October 1998 Amoco (Robert Johnston) wrote to BAO (Mr Marcom) expressing great disappointment that the RGV would not be arriving in the UK within the "time window" identified in the contract and that it would likely be mid-December before it would be available for deployment to Arbroath. The thrust of the letter was that the delay meant the rig would have to be positioned in the worst of the winter weather whereas BAO's "predicted" delivery time "would in all likelihood have meant the rig being positioned in calm conditions". The letter sought proposals from Rowan to "eliminate the economic consequences" of the delay. BAO criticises this letter as disingenuous. Quite apart from the terms of Clause 2.1, granted that Amoco intended at the time of the contract to drill a shakedown well first, in my judgment the criticism has considerable force. In the event that a shakedown well had been drilled the RGV would always have been deployed to Arbroath in the winter. Indeed there are internal documents of Amoco which noted just that expectation. Moreover the evidence is that even had the RGV been delivered at Rotterdam by 15 October the general risks of bad weather delaying jacking up by the Arbroath platform would already have existed. Further Amoco had known for some time that delivery would not be within the window and yet had made no previous complaint about it. Insofar as Mr Gillis intended to suggest otherwise, I reject his evidence. The documents, and lack of them, speak clearly and I prefer the evidence of Rowan's witnesses. Mr McNease understandably said that the letter raised serious concerns in his mind about Amoco's attitude to the whole project and whether it might be planning to terminate the contract because of the falling oil price
- The co-venturers met on 3 November. The minutes of the meeting clearly show the concerns about the costs of the infill project and its scope. The co-venturers were told that a meeting was scheduled on 5 November between Amoco and Rowan to discuss a reduced day rate. They each refused to authorise the expenditure required to proceed with the project.
- Enterprise's attitude had been made quite clear in a letter from Mr Craig to Mr Fowler, the Managing Director of Amoco, dated 29 October
"Enterprise Oil has for some time been concerned about the technical and economic viability of the five well infill drilling programme and following studies has concluded that the Amoco proposed programme contains technical risk and is economically unattractive.
It seems to us that the late delivery of the rig may give Amoco the opportunity to terminate the rig contract. We should be grateful for Amoco's views on the point in advance of the meetings. If we conclude that termination is not a realistic option then it is Enterprise's view that in the meeting with Rowan owners Amoco seek to do two things:
1) Place a deadline on the delivery date of 31 December 1998, such that the contract is considered null and void if not met.
2) In consideration of the late delivery of the rig renegotiate the day rate to a significantly lower figure in line with current market rates."
- Amerada Hess (Mr Gugen, the Managing Director) wrote to Mr Fowler in similarly emphatic terms on 4 November
"The Arbroath Infill Drilling Campaign has become, in this continuing low oil price regime, an economically unattractive proposition. It is AHL's view that we should indefinitely postpone this project or enable a significant reduction in CAPEX that would make the project attractive and allow the campaign to proceed.
Our view is that the only viable way of making a significant CAPEX reduction is via a reduction in rig day rates.
In order to achieve a reduction in day rate we must position ourselves such that Rowan believe cancellation of the contract could follow, should they be unwilling to negotiate.
To this end one option would be to immediately telex Rowan and instruct them not to load the rig for transportation to Europe.
Acceptance of the rig should be predicated on negotiating a suitable day rate. In the current climate $175,000/day is unacceptably high. The risks and rewards associated with the Arbroath Infill Campaign cannot support rates such as this.
Rowan must understand that we cannot afford to proceed on the current basis and we must be prepared to take a very firm stance on this."
- Within Amoco, also on 4 November, Steve Johnston was telling his superiors (Mr Wiggs and Robert Johnston) that "in terms of what to hit them with, cancellation seems to be the only option. The bottom line is that the delays, weather downtime potential, etc have made the original deal unattractive. There is more to this than just a reduced day rate. They got the deal as being the only show in town and failed to deliver"
- The meeting on 5 November was attended by Robert Johnston, Mr Wiggs and Mr Gillis for Amoco and Mr McNease and Mr Marcom for Rowan/BAO. Amoco (Gillis) minutes record that Rowan said a contract was a contract and "the rig will do a great job". Amoco's stance was that "if we cannot get a rate reduction then termination is a very real possibility"
- Following the meeting, Rowan (Mr McNease) wrote asking Amoco to confirm in writing the concerns expressed at the meeting. By letter dated 10 November Amoco (Robert Johnston) did so summarising "the primary issues raised during the meeting". The letter was copied to Mr Aquilina of Amerada Hess and Denise Dass Hewitt of Enterprise. It repeated the late arrival/winter weather point and continued
"Since the rig did not arrive in the appropriate time frame, Amoco and its co-venturers now wish to indefinitely postpone the drilling programme, or re-negotiate the drilling contract prior to granting their approval to begin drilling. The economics of this drilling programme have been negatively impacted by low oil prices, and the late arrival of the rig has worsened that situation, to the point of making the drilling programme undesirable."
- The punch line was that Amoco wished "to open negotiations of the Gorilla V contract including negotiation of rate and term". In the meantime "all rights" were reserved "arising out of late delivery" of the rig
- BAO makes the point that although the RGV was about to be prepared for departure from Sabine Pass at this time and ADC's list of outstanding matters had been supplied to Rowan on 9 November, the letter says nothing at all about the state of the RGV itself. The complaint is delay and only delay. Indeed it is apparent from an internal Amoco note (Robert Johnston to Mr Urban) dated 11 November that it was then believed by Amoco that the RGV would be ready to start work by mid-December and, granted Rowan was willing to negotiate, the drilling programme would move forward. Amoco was also seeking outside legal advice on the contract at this time.
- Rowan (Mr McNease) replied on 12 November. The letter stated (correctly) that Amoco had known "for an extended period of time that the delivery of Gorilla V would occur during the Fourth Quarter of 1998", stated that the rig was expected to arrive at Rotterdam "fully operational and prepared to commence operations for Amoco", and that there was no basis to renegotiate the terms of contract and Rowan expected Amoco to honour its commitment
- On 13 November, and apparently without reference to this exchange of letters, Mr Gillis, in a letter signed on his behalf by Mr Hosie, wrote to Mr Marcom. The letter referred to "the excellent working relationship" at Sabine Pass and to uncompleted work on the rig much of which was detailed in the CARs. The letter continued by stating that "additionally" Amoco was "specifically concerned" about five items which needed immediate attention. Rowan responded to these items on 24 November. The five items were
i) The need for a cantilever load test. This item forms no part of the present disputes and Rowan's response was that no such test was required.
ii) A technical query about an auxiliary brake known as a Baylor Eddy Current Brake which also forms no part of the present disputes and was fully addressed in Rowan's response.
iii) The implementation of a Rig Floor Equipment Zone Management System (ZMS) which was not likely to be completed before the rig arrived in Rotterdam. It is agreed that the contract did not include a requirement for a ZMS. Amoco's pleaded case in the proceedings was that without one there was a greater need for training and concern for safety. A ZMS is an electronically controlled automatic system which prevents equipment collisions or equipment straying into areas in which people may be working. Rowan's response was to say that the crews would be trained to deal with the situation and that it was ultimately intended to install a ZMS. In the course of the trial, Amoco abandoned any case relating to the absence of a ZMS.
iv) Rig Electrical Supply. The concern was "power spikes". It is not a concern pursued in the proceedings.
v) Crew Training. The letter stresses the importance of training and states that it was to be but would not be achieved at Sabine Pass, and a new plan would therefore need to be formulated. Rowan's response was to say that arrangements had been made with equipment manufacturers to run courses on the drill floor equipment and that it would provide documentation to support the training provided to date. As stated, the only allegation in relation to training now pursued relates to training on the PRS.
- Thus, when the RGV left Sabine Pass on 16 November and left Freeport for Rotterdam on 20 November, such matters as had been specifically raised about its condition, with the exception of training on the PRS, have all been recognised by Amoco at least to be of insufficient significance to be pursued in these proceedings
- On 20 November 1998 ABS issued an Interim Class Certificate for the RGV certifying that after a survey a report had been submitted that the RGV was fit for the service intended. The certificate was valid until 20 January 1999 pending rectification prior to commencing operation of an attached list of recommendations. The only material outstanding item to establish classification to drill in UK waters was "function test Koomey unit has ability to close all rams and valves on BOP, verify alarms/trips & closure times." DNV issued a certificate of Interim Class on the same day. It, so far as material, also referred to the same outstanding matter as ABS
- Amoco (Robert Johnston) replied to Rowan's letter of 12 November on 18 November. The letter stated that the anticipated arrival date of not later than 15 October was a "fundamental obligation" which BAO had failed to meet which failure had substantially undermined the value of the agreement. The letter asserted that "had we known that the unit would not be ready for mobilisation until mid-November we would not have entered into the agreement on the same terms, if at all". This assertion was, as BAO submitted, and I agree, not consistent with the contract and belied by the documents and evidence. Again, and despite the 13 November letter and the RGV having left Sabine Pass, no mention was made of any defect in the RGV as a reason to justify termination or renegotiation
- Rowan's response on 23 November was to offer "an alternative pricing mechanism" for the RGV, namely a floating rate equal to 13,500 barrels per day of Brent crude at the daily posted price, subject also to the contract being extended to a fixed term of 3 years. 13,500 barrels per day at $12.96 per bbl would amount to the Contract rate. Amoco requested by letter of 9 December a meeting "at a senior level" to achieve a resolution
- As it was put on behalf of Amoco in the Opening submissions, it is "correct" that "nothing was made of the state of the rig" in the "negotiations" in late November and early December. In particular no reference at all was made in this correspondence to the pinion failures. Amoco's case was said to be that it was relying on Rowan to "get the rig right" as Rowan's letter of 12 November had said would be the case when the RGV arrived in Rotterdam
DECEMBER 1998 TO TERMINATION
(A) AMOCO AND BAO
- On 8 December Amoco (Mr Gillis) wrote to BAO (Mr Marcom)
"As you are aware, Amoco is finalising plans for Rotterdam. It would be helpful if you could advise when the rig will be off-loaded from the dry haul carrier and is available for boarding. Upon boarding our intention will be to ensure that Amoco's requirements, primarily as outlined in the Drilling Contract and the Rowan Gorilla V Drilling Systems Acceptance Manual have been met."
- The reference to the Manual was a reference to the ADC interim report (Paragraphs 118 and 122). In fact, as Mr McDonald described, for some weeks Amoco itself had been planning to carry out substantial work whilst the rig was in Rotterdam. Amoco employed a Dutch contractor to do the inter-face work at Rotterdam and arranged to load equipment and supplies there. There was no prospect of the RGV leaving for Arbroath at once on arrival in Rotterdam. Indeed Amoco did not in fact board the the rig until over 2 days after it arrived
- On 14 December BAO sent a fax to Amoco advising Amoco that the rig was expected to arrive the next day in Rotterdam and to be available for boarding at 20.00 adding that "every effort should be made to move the rig to Arbroath at first available weather window". The rig did arrive in Rotterdam at 18.00 on Tuesday 15 December. BAO sent a further fax to Amoco at 18.30 advising Amoco of this and that the rig had been offloaded from the Transshelf and was jacked up in the harbour. Also on 15 December Amoco paid the first half of the mobilisation fee under Appendix 5 of the Contract. On Friday 18 December Amoco and ADC personnel boarded the RGV, BAO invoiced the second half of the mobilisation fee "due upon delivery ...to Rotterdam" (paragraph 57) and advised Amoco that the RGV was expected to be "fully operational" by 12.00 on 20 December. This, as Mr Marcom accepted, was to prove to be optimistic. It is these documents and events on which Amoco relies for its submission that "delivery" of the RGV in fact took place on 16 December. The invoice was not paid. On 21 December Rowan handed over the CARs which had been completed during the voyage. ADC noted that "most have been closed out but one or two not acceptable". The ADC team were expressly instructed that their role on the rig was only to observe and to report to Amoco through Mr Sleightholme. Amoco's team had also been instructed to stay at arm's length. The reason was to avoid doing anything which could be used to suggest that the contract had commenced. Unsurprisingly the change of attitude was noticed and remarked upon by Rowan/BAO
- On 19 December Mr de Quelerij of DNV signed a DNV Survey Report which recorded that all the outstanding “Conditions of Class” had been tested and inspected and found to be satisfactory and so had been deleted. One of the outstanding conditions (RC022) was “Function Test Koomey Unit”.
- On 22 December Mr Van Rooten of ABS wrote to Mr Wesselingh (Rowan) confirming a telephone conversation they had held that morning
“that the undersigned has contacted the attending DNV surveyor and that in lieu of attendance of a surveyor in this instance we will accept a statement from Rowan confirming that the Koomey Unit is operable and able to operate the BOP when same is satisfactorily tested according to the applicable API standard.
Please forward your statement to the local DNV and ABS office”.
- Whilst this concession may not have been unrelated to the approach of Christmas and the departure of the certifiers from the rig, it also reflects the undoubted confidence in Rowan which ABS and DNV had.
- The circumstances in which BAO came to inform ABS and DNV that the condition was satisfied have been the subject of criticism by Amoco. Indeed it was suggested (albeit not pleaded) that BAO had deliberately deceived both ABS and DNV by asserting to them on 24 December that the accumulator unit had been tested to the API standard when they knew it had not. The details of the tests and extent of any failure to meet the standards are addressed later (see paragraph 397 and following) but there can be no doubt that in some, albeit limited, respects the stated API timings were not achieved by 24 December yet ABS and DNV were informed on 24 December that the control unit had satisfactorily tested and accepted as much
- On 24 December BAO (Mr McDonald) wrote to Amoco (Mr Gillis, copy to Mr Johnston) "for the sake of good order ... to confirm that all certification requirements for Gorilla V have been fully satisfied as of 18.30 hours December 24 1998 GMT, which shall be the commencement date for charges under the Drilling Contract". The letter added that the RGV was "standing by awaiting further instructions from Amoco". BAO's case is that this constituted delivery of the RGV under Clause 2.1 of the Contract and that (despite the wording) charges therefore commenced 24 hours later under Clause 2.2
- Amoco's (Mr Gillis) response on 26 December was
"Amoco is of the opinion that the rig is not 'fit for purpose', principally demonstrated by the almost complete lack of drill floor equipment functionality. Further, and as a result of the above, your personnel have not had a proper opportunity to operate the equipment and are clearly unable to demonstrate competence in the use of the said equipment. In addition our acceptance testing with regard to other items has not yet been completed.
The events of the past 36 hours unambiguously support our view and as a result we do not accept that the commencement date for charges has occurred. Furthermore Amoco expressly reserves all its rights and remedies under the drilling contract with [BAO]."
- The complaint by Amoco was therefore directed to the drill floor equipment. That is the feature which Amoco now accepts to be at most something of a makeweight. Mr Marcom discussed the situation with Mr McDonald and Mr McNease and Mr Gillis on Tuesday 29 December. The context was a 36 to 48 hour weather window to move from Rotterdam forecast for Thursday. Mr Marcom made a file note of his discussions. The note records that he agreed with Mr McNease to tell Mr Gillis that the rig was under contract now and “we would test whatever as long as they want while on payroll”. Mr Gillis’ position was that as far as Amoco was concerned they would not be on contract until everything was working. Mr Marcom and Mr McNease agreed they would try to satisfy Amoco on all the items Mr Gillis raised, “which hopefully won’t take but 24-36 hours”, and then have “a definitive answer as to weather (sic) they really mean to take the rig or not”. Mr (Rob) Johnston then called Mr Marcom and told him that Amoco regarded BAO as in breach of contract. Mr Marcom noted that he and Mr McNease agreed that the best course “is to continue to try to do everything possible to show them that the rig is ready and by Monday we will know for sure if they intend to screw us”. I unhesitatingly accept that this reflects Rowan/BAO's approach. I should add that Mr McDonald was impressive and patently truthful in describing his relationship with Amoco as a good and open one in which his aim was to please a good customer. I should also note, because there were a number of conflicts in their evidence, that where Mr McDonald and Mr Gillis and Mr Hosie were in disagreement as to the events and conversations in December and January, I found Mr McDonald's evidence to be reliable and truthful and very much to be preferred, consistent with and supported by the documents as it was. Mr Gillis, in particular, was I think in an invidious position. I accept that he wanted the project to proceed but as will be seen others did not and certainly not if Rowan would not agree to renegotiate the rig rate
- On 30 December Mr Marcom replied to Mr Gillis’ letter stating that the rig was ready to commence operations and recommending that Amoco use the first available weather window to commence the move to the Arbroath field. Mr Gillis wrote again in reply (4 January 1999) stating that "although progress has been made" commissioning and acceptance testing remained to be completed. The drill floor equipment and training in its use were again referred to. Reference was also made to "the well control equipment" not meeting the specifications in Appendix 6 of the contract and to BAO taking steps to rectify it. Mr Marcom said he regarded this letter as "posturing" because right or wrong none of the matters referred to affected moving the rig as they could be dealt with on location before drilling began or were being dealt with at Rotterdam. In their closing submissions Amoco describe Mr Marcom's 30 December letter as a "typical example of Rowan intransigence". In my judgment that description is wholly misplaced. Mr Marcom's reaction was understandable. On Thursday 31 December Steve Johnston (Amoco) had himself informed Robert Johnston, Mr Gillis and others at Amoco that "progress had been made at Rotterdam such that the rig may take the opportunity to sail from Rotterdam in the next weather window". That window was then tentatively said to be on Tuesday or Wednesday, 5/6 January. Steve Johnston's information itself came either from Mr Gillis or Mr Hosie as Mr Gillis agreed in evidence. That puts in context as regards the perceived state of the rig this correspondence and the suggestion of intransigence
- On Friday 1 January 1999 the BP/Amoco merger was finalised
- On Wednesday 6 January BAO (Mr McDonald) wrote confirming the forecast of a suitable weather window for mobilising the RGV to the Arbroath platform between 8 and 12 January adding that the rig was "fully crewed and ready to undertake Well Operations" and was standing by awaiting orders from Amoco. The co-venturers met in London on 7 January. Amoco (Mr Richard Owen pp Mr T. Smith) wrote disagreeing. This letter (dated 8 January) was plainly drafted by or with the assistance of legal advisers. It is important as it was the precursor to termination. It is relied upon by Amoco as the notice required by Clause 23.1.1. of the contract. The first complaint was delay in mobilisation, citing Clause 2.1. The second point was a reference to the contract requiring the RGV "to be fully equipped as set out in Appendix 1 and to be fit for purpose", and an assertion that neither was the case. The letter continued
"Clause 2.1 of the contract provides that the Contract shall commence 24 hours after delivery of the Drilling Unit at Rotterdam or when the Unit commences tow to the first drilling location if that occurs within the 24-hour period. The Contract therefore commenced on 16th December, 1998. However, as notified to you on a number of occasions, the Unit has not been fit for purpose or otherwise in accordance with the Contract at any time since then and is still in an unacceptable condition. Without setting out an exhaustive list of the defects and test failures which have taken place and which have been the subject to discussions between our respective teams in Rotterdam, deficiencies have been identified in the following major items of equipment. These defects remain to be satisfactorily resolved before the Unit will be fit for purpose and otherwise comply with the requirements of the Contract:
rig floor equipment including:
- top drive
- pipe racking system
- iron roughneck
- driller's console
well control equipment including:
- hydraulic control equipment
- choke manifold and associated piping
- BOP
For these reasons, no day rate is payable in respect of the Rowan Gorilla V for the period from 16th December, 1998. Moreover, no charges will be payable until the Drilling Unit is operational. Having regard to the history of delays in constructing and commissioning the Rowan Gorilla V to date, we have no confidence in it being fully operational in the near future. These delays have already resulted in an overall delay of at least 3 months. Under Clause 28.1(b), we are entitled to terminate the Contract in the event of equipment breakdown which results in you being unable to perform your obligations for a period of 30 days or more. In our view, there has been such equipment breakdown (for the reasons set out above) since the Contract commenced on 16th December. We therefore put you on notice that, if such equipment breakdown persists, and the rig remains unfit for purpose for a further 7 days, we will be entitled to terminate the Contract under Clause 28.1(b) (as well as Clause 23.1.1).
We understand that you are continuing to work on the rig with a view to overcoming the problems. However, all your previous estimates as to when this would be achieved have proved over-optimistic. In the meantime, if advantage is to be taken of the next available weather window, the rig will have to commence tow from Rotterdam within the next 24 hours or so. As the next weather window might not be for another 4 weeks or more, we will consent to your securing tow vessels and line (as agreed between us) and commencing the tow to Arbroath as you have requested in your fax of 5th January. However, such consent insofar as it is necessary (if at all) is given strictly without prejudice to our rights to claim damages and, if appropriate, to terminate the Contract if the rig is not operational by the end of Friday, 15th January, 1999."
- The letter was copied to Mr McNease and to Enterprise and Amerada Hess
- There are a number of significant points about this letter
i) It asserts that the contract commenced on 16 December, that is when the RGV arrived in Rotterdam. Until it was written, Amoco's often stated position had been that the contract had not commenced at all. Indeed all Amoco personnel had been expressly instructed to do nothing while the RGV was in Rotterdam which could be used by BAO to support a case that the rig was on contract, and those on the rig continued to believe that was Amoco's stance up to and including 19 January. The same applied to ADC personnel, exemplified by their "terms of reference" prepared by Mr Sleightholme and issued on 17 December (paragraph 146).
ii) It refers to the right to terminate under Clause 28.1.b) in the event of equipment breakdown for 30 days. 30 days from 16 December would expire on Friday 15 January. The right was expressed to be dependent on the commencement of the contract. Despite Amoco's present submissions that commencement of the contract is not a predicate for the operation of Clause 28.1.b) in my judgment (Paragraph 90) it is and the letter was right to assume as much.
iii) The "notice" that if the alleged equipment breakdown continued for a further 7 days after 8 January Amoco would be entitled to terminate the contract "under Clause 28.1.b) (as well as Clause 23.1.1)" would also expire on Friday 15 January.
iv) The consequence of the commencement of the contract would have been that charges were due from 17 December but the letter asserts that no day rate is payable because the RGV is not fit for its purpose and does not comply with the contract.
v) The defects are stated in very general terms. There is an issue as to whether the letter identified the defects in sufficient detail to constitute notice under Clause 23.1.1. On any basis the defects, so far as relevant to the matters still pursued by Amoco, were limited to the PRS and the well control equipment. The source may well have been the report of Mr O'Brien and Mr Slater (Paragraph 179). Safety is not mentioned.
vi) It acknowledges that BAO was continuing to work to overcome "the problems". BP Amoco was being told at the time that the rig would be ready by mid-January (Paragraphs 179-180). The predicate of the letter however appears to be that this was immaterial to the operation of Clauses 23.1.1 and 28.1(d) and that Amoco was entitled to terminate the contract if the problems were not solved within the final 7 days of the 30 day period. The claim was to a right to terminate after 7 days if the rig was not then fully operational. That is not what the Clauses provide: paragraphs 78 and 84.
vii) The tow to Arbroath was consented to "without prejudice". However on 7 January Mr Hosie on instructions from Mr Gillis had orally authorised BAO to select and fix towboats for the move. BAO did so and confirmed as much the next morning acknowledged by Amoco. Mr Hosie agreed that there were no qualifications or reservations expressed in these earlier exchanges. For Rowan/BAO the consent to tow was understandably a key factor (paragraph 153).
- The RGV was de-manned for the move and left Rotterdam for Arbroath overnight on 9/10 January. Early on 11 January it became clear that the weather was deteriorating and the rig was taken to a standby location at Halley North, about 60 miles from the Arbroath platform
- There is some conflict in the evidence as to whether or not, weather permitting, the intention had been for the rig to go alongside the platform and jack up there. Some of the witnesses suggested that at most it was intended to go to a stand-off location 750 metres from the platform. Others that it might have gone to the stand-off location 150 metres away where the tow would in any event have had to stop in order to prepare for the final approach to the platform. The operational significance of any location within 500 metres of the platform is that the Combined Operations Safety Case would then apply. The general significance is that the intention puts in context the complaints Amoco makes about the RGV. In any event, I am quite satisfied that the documents and the more reliable oral evidence overwhelmingly establish that, weather permitting, the intention of all those involved both for BAO and Amoco was indeed to go to the platform and jack up the RGV there. The 8 January letter itself (paragraph 156) refers to "commencing the tow to Arbroath". The destination notified in Amoco's Rig Move Notices and Daily Time Reports is "the Arbroath platform". That is what the tow boats were chartered for. The Arbroath platform logs say the same. E-mails sent on 9 January by Mr Hosie are to the same effect, so is the evidence of the towmaster, Captain Broertjes and the warranty surveyor, Captain Pique. The relevant personnel and equipment for the approach to the platform were all on board. Preparations were made for the work required to interface the rig and the platform and to shut down the platform for that to occur. Moreover Captain Mallett and Captain May agree that the intended destination when the RGV left Rotterdam on 9 January was the Arbroath platform
- At about 01.57 on 12 January when the RGV was jacking the legs down at the standby location at Halley North a tooth on the 1/4 pitch pinion F7 cracked and jacking stopped. Parts of two teeth on the pinion were cut out and jacking resumed after some 8 hours. This failure has loomed large in Amoco's case about the jacking system
- On 13 January Amoco and ADC personnel returned to the RGV jacked up at Halley North. On 14 January Amoco (Mr Gillis) wrote to BAO (Mr Marcom) about the pinion failure stating that BP Amoco "is very concerned over the re-occurrence of such failures". The letter continued
"Note that, despite repeated requests, we have not been provided with any explanation as to the likely causes of the failures whilst at Sabine Pass. The failure on January 12th underscores the criticality and importance of understanding the cause of such failures.
As you are no doubt aware, failure whilst jacking adjacent to the Arbroath platform could lead to a major incident. Accordingly, we must understand the root causes of the failures (and the engineering solutions) such that we can systematically evaluate the risks commensurate with jacking up adjacent to the Arbroath platform."
- There is a dispute as to whether "repeated requests" had been made for an explanation of the pinion failures in Sabine pass
- The subsequent correspondence concerned with the termination of the contract is attached to this judgement as Appendix B. For present purposes it, and the more material contemporaneous events, can be summarised as follows
- Amoco made further criticisms about the condition of the rig in a letter signed on behalf of Mr Denholm dated Friday 15 January. This letter was sent at 17.35 by fax that evening. It, like the 8 January letter, was copied to Enterprise and Amerada Hess. The letter stated that the rig was "some weeks away from being operational even if no new problems emerge which is itself unlikely in our view". It warned that Amoco would be considering with the co-venturers whether to terminate the contract in the next few days. Mr Marcom said, and I accept, that he only received the faxed letter on Monday 18 January as he was out of the office on 15 and 16 January
- In relation to the well control equipment the 15 January letter stated that two regulators on the accumulator unit had failed and needed to be replaced, the equipment remained to be satisfactorily tested and added "in particular, the problems with the BOP continue and we remain of the view that, as previously advised by Cameron, the whole system will need to be stripped down and inspected for damage caused by contaminated hydraulic fluid before it can be commissioned". The pipe racking system was said not to be "fully operational". Reference was also made to "a recurrence of the unexplained pinion failures at Sabine Pass during jacking operations in the North Sea on 12th January" and to "questions to be resolved arising from the pinion failure as to whether it would be safe to jack-up the rig alongside a platform at least before the cause of the pinion failure is known". Yet the letter asserted that there was an immediate right to terminate which was to be decided upon in the next few days regardless of any attempt to resolve the issues or questions to which it referred
- On 15 January, Mr Palmer, the Chairman of Rowan wrote to Mr Fuller, a Co-Chairman of BP Amoco a letter also copied to Sir Graham Hearne the Chairman of Enterprise and Mr Laidlaw the Chairman of Amerada Hess referring to “the enclosed letter from Amoco”. It is uncertain but the reference is likely to have been to the 8 January letter. Mr Palmer expressed his displeasure and provided his "personal assurance" that "notwithstanding a few minor glitches" the RGV was "fully capable of drilling operations". The letter referred to the need to "adjudicate" the dispute properly but if necessary to resolving the matter through litigation
- At a meeting held on Monday 18 January Amoco decided to terminate the contract. At that time BAO had not responded to the more recent complaints (or the 8 January letter). But Mr Marcom did so by two faxed letters timed at 19.41 (pinion failure) and 23.23 (other matters) on 18 January. The letter about pinions speaks for itself. The thrust was that the failures in Sabine Pass had been attributed to various causes as previously advised to representatives of Amoco, an explanation (binding of the brake plates) was offered for the recent F7 failure and a remedy to prevent re-occurrence. It was also stated that the pinions had been fully tested and certified. The letter has been the subject of criticism and will be considered further in the context of the allegations about the jacking system (paragraphs 366 to 367). It was based on conversations Mr Marcom had had that day with Mr Quinn and Mr Rimlinger
- The second letter addressed the alleged defects and asserted that, subject to the possibility that some shakedown or fine tuning might be required, the rig was "entirely outfitted, fit for its intended purpose, fully certified and prepared to perform Amoco's drilling operations when presented with the opportunity". The letter also referred to "Amoco's intent to create a breach of contract scenario where such breach had clearly not occurred" and the "economic reasons" driving Amoco's correspondence. The letter concluded by proposing a meeting to be attended by Mr McNease in the UK on 28 January. As regards the specific matters which remain in issue, the letter stated that
"1. The regulators on the BOP accumulator unit have been replaced and the unit is working. A function test of the unit has been witnessed by Amoco appointed personnel.
2. The BOP has been pressure tested and is fully certified.
....
4. The pipe racking system is fully operational.
...."
- It is Amoco's pleaded case that this letter constituted a refusal by BAO "to strip down or inspect the BOP system so as to address properly the defects and contamination in the BOP system" and that it therefore amounted to a refusal to rectify the defect within the meaning of Clauses 23.1.1 and 28.1.d) entitling Amoco to terminate the contract. This submission is addressed later in the context of the allegations about the well control system (paragraph 491)
- Amoco (Mr Garforth-Bles pp Mr Clutterbuck) sent the termination letter at 17.14 on Tuesday 19 January. The letter described a new problem that "the top drive cannot achieve either the speed or torque ... specified in Appendix 1 to the contract" which was "so serious" as alone to make the rig "unfit to drill". It was said that BAO's letters of 18 January did not affect Amoco's view that "in summary" BAO had "persistently failed" to meet its obligations and that the rig was "still not operational or even commissioned". The right to terminate "with immediate effect" was asserted under both Clause 28.1.b) (repeating that the contract had commenced on 16 December) and on the basis that Amoco was "entitled to accept the Contract as repudiated". No reference was made to Clause 23.1.1. Amoco's case about the top drive was abandoned in the course of the trial. Apart from a reference to not having enough information about the pinion failures to form a considered view as to "whether the rig could be safely jacked up next to the Arbroath platform" nothing was said about safety considerations as a reason or justification for termination nor to the effect that BAO had refused to rectify the BOP system or anything else
- By letter dated 20 January Amoco replied in detail to BAO's second 18 January letter. The letter stated that a separate reply would be sent to the first letter about the failed pinions (albeit in the event no such letter was ever sent) but repeated that BAO's letter "did not provide sufficient information to enable us to make an informed decision as to whether it would have been safe to jack up the rig next to the Arbroath platform". In contrast to the letter of 15 January (Paragraph 166) this letter acknowledged that the BOP accumulator unit was now working (save for some flowmeters) and stated: "the BOP remains to be stripped down at least for partial inspection in order to ascertain whether any damage has been caused by fluid contamination. A more complete inspection may be necessary dependent upon initial results". The letter as a whole is targeted at Clause 28 not repudiatory breach and not serious safety concerns. It does not even suggest that the well control system is still contaminated. Nor, again, is it consistent with a belief that BAO had refused to rectify the BOP system or anything else
- Amoco had issued the Writ in these proceedings on the previous day. It was in fact issued shortly before the termination letter was sent. The major relief sought was a Declaration that BAO had repudiated the contract and that Amoco was entitled to accept the repudiation and to terminate the contract under Clause 28. Also, but later on 19 January, following receipt of the termination letter, Rowan and BAO issued proceedings in Texas against Amoco and a number of related companies claiming damages for breach of contract and tortious interference with contractual and business relations. There followed various and protracted jurisdiction battles. It is only relevant to this judgment to record that, in accordance with Clause 31.1 of the contract, all contractual issues are to be resolved in these proceedings but certain other claims involving other parties are to be tried in the Texan courts. The consequence is that in the Texan proceedings a number of depositions have been given by persons who also gave evidence in these proceedings and the contents of some of those depositions were therefore deployed during the course of the evidence
(B) AMOCO AND CO-VENTURERS
- BAO submits that it is important to see the decision by Amoco to terminate the contract in the context of Amoco's and the co-venturers documentary records leading to the decision. I agree. I shall therefore summarise here what I think to be the substance of the matters contained in that evidence and the oral evidence relating to them
- On 1 December the co-venturers held an "Opcom" meeting. Enterprise sought "ideally" a reduced rate for the RGV and only a two-well (not five) Arbroath programme; Amerada Hess wanted to call off the whole project. It was recognised that "delay" was not a justifiable basis for terminating the contract if only because any right to do so had been waived. The question was raised whether acceptance of the rig could be refused because it was not "fit for purpose". Mr Gillis (who was present) recorded in his diary that there was a "need to (on the basis of the contract) find things that are not fit for purpose". He also noted that the financial consequences of the delay in delivery were, in my word (with which he agreed), "peanuts"
- Enterprise (Mr Reilly) wrote to Amoco (Mr Garforth-Bles) on 4 December. The letter was critical of Amoco for not instructing Mr Gillis to ensure there was no relaxation in enforcement of the contract. It continued
" We also consider that now the Rowan Gorilla V is on the high seas Amoco should change tack from co-operation to looking for any failure on BAO's part and exploiting that mercilessly.
We propose that you sit down with Mark Gillis and explain in detail the implications of making reasonable allowance for BAO failures so that he fully understands the role he must play both in protecting the position and in finding the failure that gives the lever required.
Tim Smith agreed to put two of your best contracts people on the case to assist in finding and exploiting any BAO failure ..."
- BP had identified by early December that the combined group appeared to have "a significant oversupply of drilling units based on anticipated capex reductions". By 9 December BP was seeking "urgently" to "address the RGV problem" and addressing the "options", noting that there was no internal requirement for the RGV and to sublet the rig would likely require a subsidy of about $100,000 a day. By 29 December Mr Denholm had concluded that BP would "likely" cancel the Arbroath programme altogether given an option to cancel the RGV contract
- When BP heritage people were in a position (on and after 1 January 1999) to act directly in relation to the Arbroath Infill Project and the RGV the matter was plainly given considerable priority. The BP people (in particular Mr Denholm) were provided with a detailed insight into Amoco's "current strategy for negotiation with Rowan" about the RGV. The "problems" with co-venturers were also tabled. Enterprise and Amerada Hess had only authorised expenditure up to spud of the first well. The current project economics were such that in the year 1999 a decision not to proceed with the infill project would achieve a capital expenditure saving of $25m, and a positive cash flow effect of $18m against a loss of net income of $2.7m. Enterprise and Amerada Hess had suggested a shortened infill programme of 3 not 5 wells. As it was, a 3 months period of unused rig time was likely and a shortened programme would lengthen that period. It was also stated that Enterprise' original agreement to a 5 well programme had been conditional on "the shakedown well not being drilled at Arbroath". Other options for use of the RGV were believed to be limited (none were ever suggested)
- Mr Denholm arranged for two very experienced BP rig auditors to visit the RGV in Rotterdam, Mr O’Brien and Mr Slater. Mr Denholm said he wanted an independent (from Amoco) second opinion on the rig. Mr O'Brien and Mr Slater reported back to Mr Denholm first by a visit note dated 4 January. The note remarks on the “awkward” position created by Amoco’s need to avoid acting as if the rig were “on contract” (“a position we also had to take”) and criticises Rowan's attitude (not acting "in a team manner") but also gives an overall view of the condition of the rig
“From our visit, discussions with individuals and direct request of views with ADC, who have the major equipment interface, if nothing serious goes wrong during further testing or if the additional work suggested later in this note does not cause a problem, our opinion is that the rig should be ready to leave Rotterdam by the middle of January”.
- The note also contained a reference to additional problems being experienced with the Koomey (BOP) control unit and a list of 13 items brought to BP Amoco’s attention only one of which now features in the trial (the alleged lack of a separate power supply to the BOP). It was also Mr Cowie’s view when he left the rig on 5 January that it would take another 2 weeks for everything to be complete including crew training and system acceptance testing
- By 7 January it is recorded that the co-venturers had agreed to what became the 8 January letter (Paragraph 156). Mr Denholm was dealing with the matter but his immediate superior (Mr Schrader) and Mr Mogford were also in the picture at this time. Mr Mogford was described in evidence as "Mr Safety" and his involvement attributed to that role. Yet at this time and for some days thereafter there is no suggestion in the documents that safety considerations played any part in BP Amoco's considerations
- On 8 January, Dr Feil produced a BP Amoco briefing note at Mr Denholm's request. The note flagged the business risks of the project, that Enterprise had suggested that some of the wells looked uneconomic and that "current Amoco estimates" indicated that even the 5 well programme might potentially be completed in 214 days "resulting in BP Amoco 100% exposure for 151 days". The estimate when the RGV was contracted was some 270 days, leaving a 100% exposure of 95 days. The new estimate was plainly seen within BP Amoco as increasing the risk to BP Amoco, even though Mr Routh of Amerada Hess gave evidence of his belief that the co-venturers were committed to a 9 month rig cost not just the cost of a 5 well programme. The reduced time estimate also does not lie well with a case that the RGV was not thought to be able to do the job or that a winter start was of any consequence at all. Indeed on 24 December Mr Gillis had circulated Amoco’s programme for the project which stated that drilling was expected to commence in December and last approximately 7½ months, and the reduced time estimate was, as Ms Turton (Amoco) said, primarily due to the capabilities of the RGV.
- By the evening of 12 January Mr Steve Johnston was supplying Mr Denholm with a draft of a document to contain a recommendation to a then unspecified decision maker about the Arbroath Infill Project and the RGV. The document referred to the pinion failure (which had only taken place that day) describing it as creating a "potentially hazardous situation", and (the version for BP Amoco internal use but not the version for dissemination to co-venturers) set out the net (i.e. after credit for the co-venturers' shares) financial exposure for the cost of the RGV for BP Amoco according to whether the 5 well programme was based on a 270 day or 217 day period. The exposure for 270 days was $31.2m and for 217 days $37.6m, on the basis that 100% of the cost was for the account of BP Amoco once the programme was completed and until the end of the one year term of the BAO contract
- Mr Denholm acknowledged in cross-examination that he had never informed the co-venturers of the estimated reduced amount of time required for the infill programme. As operators of the field BP Amoco were contractually obliged to give the co-venturers that information and whilst on the evidence they were supplied with documents which at least contained it, Mr Denholm nonetheless acknowledged that in failing to give this information to the co-venturers he had deceived them
- The draft recommendation document was the subject of further work and amendment with the target of finalising it by "close of play" on Thursday 14 January. In the morning of 13 January Mr Denholm informed Mr Mogford, Mr Schrader and Mr Horton that the RGV was at the standby location because of deteriorating weather conditions and that the " earliest next weather window to commence movement to Arbroath is Monday/Tuesday next week, but may well be later than this". He added
"Forward Plan. We need to get ourselves (and Partners) in a position by Friday close of play where we are all able to advise Rowan of termination of contract if appropriate. If poor weather conditions persist we may be able to delay this decision for a few days but worst case is we need to know what our plan is by midnight Friday."
- The 7 day period in the 8 January letter and the 30 day period under Clause 28.1.b) (assuming "delivery" of the RGV on 16 December) both expired on Friday 15 January (paragraph 158). The relevance of the weather conditions was of course that in practical terms this end date could be extended until the conditions permitted the final approach of the RGV to the Arbroath platform when a decision would have to be taken whether or not to permit the RGV to go to the platform. The relevance of 15 January in terms of the abilities of the RGV let alone safety was nil
- In the course of the further work on the document the "options available" were set out, analysed for risk and costed. It is of some significance, granted the importance ascribed to the pinion failure in Amoco's case, that these documents describe the potential for the termination option as "low" whereas options to buy out the contract or to stack the rig were described as "good". There were no options available for internal utilisation or sub-let
- By the afternoon of 15 January Mr Steve Johnston was informing Mr Denholm, with detailed figures to support it, that "in terms of 1999 numbers" the worst option for BP Amoco was to go ahead with the programme, worse indeed than terminating the contract and failing in the claims by Rowan to which it was anticipated that would inevitably give rise. Termination and failure in such a claim was "the worst" case only "full cycle". Of course the question is not just one of numbers. It also involves other levels of return available from other projects which may compete for capital expenditure but there is no doubt that BP Amoco and the co-venturers were concerned to reduce capital expenditure in 1999
- Mr O’Brien and Mr Slater returned to the RGV on 13 January. They reported back to Mr Denholm (copied to Mr Gillis, Mr Hosie and Mr Price) on Saturday 16 January. The note was critical of the “functionality” of the PRS (a complaint not now pursued as such), referred to the BOP control system still not being fully operational, and to a desire to see the results of “due process” on the pinion failure (conjecture heard was “insufficient clearances”) and concluded
“Without the results of a suitable integrated test process the rest of our present findings would, under BP’s processes, fall under categories that we would plan to risk manage at start up, these are being incorporated into our standard ‘Omissions Profile’ document.
The BP Amoco concerns about safety that you and I discussed were addressed at a meeting … held yesterday … Any safety issues raised have been left to Rowan to manage and decide time frames”
- The “Omissions Profile” referred to contained a number of items relating to the BOP all of which were shown either as completed on 17 January or to be completed “Prior to Spud” (including the supposed need for a separate power supply to the BOP). The pinion failures were referred to under Ref. No. 2.3 as follows
“2.3 There have been three failures of the jacking system to date – two being of a similar cause. [LeTourneau] are still to investigate and report, findings urgently required.
Completion by: Prior to going over platform.”
- As Mr O’Brien made clear “prior to going over platform” meant prior to cantilevering out the drill floor of the RGV over the platform not prior to jacking up the RGV adjacent to the platform. That puts the expression of concern in context. It was not seen by Mr O'Brien as a reason not to go to the platform and jack up there. The contrast with the termination letter and 20 January letter (paragraphs 171 and 172) is apparent. The safety concerns referred to in the 16 January report are of no relevance to the issues
- In the event the planned move of the RGV to Arbroath was delayed by weather until Tuesday 19 January and the meeting to consider termination was set for Monday 18 January. To that extent the urgency to reach a conclusion on 15 January was alleviated and the final documentation continued to be refined over the weekend and on the Monday morning. The 15 January letter (paragraphs 165 and 166) was sent in that context. The decision to terminate the contract was, as Amoco accepts, taken at a meeting on Monday 18 January, albeit the termination letter was only sent the next day. The meeting was attended by Mr Mogford, Mr Denholm, Mr Schrader, Mr Horton and Mr Wood. No one of Amoco heritage was present
- On 19 and 20 January, Mr O’Brien reported further to Mr Denholm (also copied to Mr Gillis). The report is, I think, significant. It gives a BP Amoco critical view of the state of the RGV at the time of termination which was prepared without an eye on legal or contractual niceties. As a rig auditor Mr O'Brien also said that he was expected to make comments about anything that related to safety. The decision to terminate the contract, unknown to Mr O'Brien, had been taken on 18 January. Omitting passages of no or no continuing relevance, the reports stated
19th January:
Today I expect to leave on the helicopter being brought in to de-man the rig. De-manning being prior to the coming weather window, that looks as if it will allow the rig to be towed to the platform. Allan (Slater) left yesterday and should debrief you today before returning home. I will pass the completion of any further notes to Mark Jones, the OIM onboard for the integrated operations over Arbroath.
The situation as of this morning at 1000 hrs, following a walk around is:
The rig has not yet been offered to the onboard BP Amoco drilling personnel for commencement of the proposed process of Drilling Systems Operational Acceptance Tests. A draft document re the tests has been created and offered for internal discussion at 1030 hrs.
….
The Top Drive is not fully functional – this is seen as a Ross Hill problem and is presently being addressed ….
The Varco PRS 3 pipe racking system’s parts for completion arrived, but came without a critical manifold, this is now being chased down as to whether it is on it’s way or was inadvertently not ordered. Training on this equipment was ongoing but has stopped, discussions indicated that some individuals are learning but are not yet operationally competent, others are struggling and causing comments of concern from the Varco specialists.
The BOP control system is hydraulically operational, problems now exist with electronic flow measurement equipment and a suitable Cameron electrical service engineer has been requested. I do not know if this problem is a knock on affect of the problems caused by the non-cleaning/pickling of the control hydraulics whilst the rig was in Sabine.
….
The Diverter system is presently not made up, nor in a state to flow through or test, we are hoping that this is possible today –several days later than originally planned. During the walk round there were control problems with the trip tank pump.
….
I still have no answer re the review on the jacking system and cannot categorically state that it has a clean bill of health.
Allan and I continued to update the “Omission Profile”, though it should not start in earnest until the integrated acceptance testing process commences.
There still have to be concerns re safety, training and competence, especially as rig floor training is going so slowly. Rowan have brought on board a consultant, Rab Morney to handle several issues, he complained to several of the BP Amoco people, including myself re lack of access to individuals, equipment problems (cranes especially) etc and was advised to deal with Brian Quinn the Rig Manager. Rab gave a talk at yesterday’s safety meeting re waste management, as he said he was using a 30 minute talk to replace what was a day’s course.
20th January:
At the time of leaving I had at last held my meeting with Rowan re the concentrated acceptance tests for the unit as an integrated drilling rig. Although the document shows around 72 hours, I believe that it could be done in a concentrated 48 hours if everything went well. In respect of when the rig would be truly fit to put forward for testing, my view based on if there was no tow, platform hook up etc. is that 5 to 7 days should see the rig at a point where successful continuous tests would be possible. This form of test is an unknown quantity to the Rowan guys so when asked on their view of times neither the Rowan manager, OIM or engineer replied conclusively.
Being on the rig itself, when there was so little urgency to complete, was for me very strange. .... I do not know if complacency because of weather window delays had crept in but the rig was not reacting like a rig that was several months behind it’s original start date.
Allan and I drew up a draft of what we term an “Omission Profile”, in this case based on what we were seeing around us, though normally this is drawn up during the integrated acceptance tests. By using the information we had to hand we were able (to) build a demonstration of the simple electronic tracking system and time based risk management decisions format used. The intention being that the guys on the rig who had not seen our process would be able to relate much easier. Ultimately this was not handed to Rowan as the one problem we brought up (bulk tank safety valves) caused them to suddenly deviate from tasks in hand and change direction to immediately address our concerns. We see this reaction of “taking the eye off the main ball of making the rig function”, often in our job. To us it was critical that Rowan concentrated on the job in hand so we withheld the “OP”, for BP Amoco drilling reps use after the tow. (The emphases are mine).
- I would comment
i) Plainly Mr O’Brien believed, as he said, that the RGV was about to go to the Arbroath platform because there was a weather window available. Mark Jones was on board for that reason. There is no hint of a suggestion that this was inappropriate let alone unsafe.
ii) The Top Drive is no longer any part of Amoco’s case.
iii) Amoco no longer relies on any problems with the PRS after 7 January (see Paragraph 496).
iv) The BOP control system is said to be “hydraulically operational”; the problem with the flow measurement equipment is agreed not to be a serious matter. There is no reference to any sort of strip down of even part of the system.
v) The Diverter would not be required until drilling was to commence when it would be tested.
vi) All Mr O’Brien noted about the jacking system was that he could not “categorically state that it has a clean bill of health”. Mr O’Brien is meticulous and is not, I think, a man given to categoric statements, and this must I think be read as a very limited expression of doubt especially in view of the "Omissions Profile" (Paragraphs 190 and 191)
vii) The concerns expressed about safety were directed at matters related to the drill floor.
viii) Mr O’Brien’s view at the time that 5 to 7 days should see the rig at a point where successful continuous tests would be possible was one of which Mr Denholm said he could not recall whether or not he was aware. Nor, it followed, could he recall giving the information to any of Mr Mogford, Mr Schrader or Mr Hayward. Mr Quinn and Mr Gray had told Mr O'Brien at their meeting that they were happy to do the tests and confident the rig would pass them.
- Mr Quance (who alternated 3 week shifts with Mr Price) arrived back on the RGV at 13.30 hours on 19 January believing that the rig was to be mobilised to the platform the next day but discovered it was actually being mobilised then and there. He said it came as “very much a surprise” to him when he was told the contract was to be terminated. Mr Price left Mr Quance some “handover notes”. The notes contain nothing of any relevance to the issues and are significant for that reason. As Mr Quance disarmingly put it no one told him it was unsafe for him to be on the RGV
- Against this background the final documentary record, dated 19 January, called an Authorisation to Negotiate (ATN) and headed "Privileged" was prepared for the signatures of Mr Mogford, Mr Schrader, and Mr Hayward. They were all BP Heritage. The signatures were obtained in the course of 19 January, Mr Mogford authorising Dr Feil to sign on his behalf. It
i) sought approval to serve notice to terminate the contract;
ii) referred to the delay in delivery and stated that at no point since the RGV arrived in Rotterdam had it been ready for full acceptance testing by Amoco; "equipment problems associated with the well control system, rig floor equipment and, latterly, the rig jacking system, have caused continuous cause for concern"; and
iii) addressed the options, noted that the DTI had informally indicated it would be comfortable with a decision not to drill the programme at the current time, stated that the view was that "reputation risks" could be managed, and recorded that the co-venturers had agreed to pursue termination and "to meet their share of all associated costs".
- Mr Denholm said the ATN was "an after-the-event document to be signed" and it was intended to establish proper authorisation for the decision to terminate not to set out the justification for termination. But the document was carefully drafted. It served little purpose if it was inaccurate or incomplete. It did address technical matters but did so emphasising delay and concerns regarding "fitness for purpose". Safety hardly gets a mention let alone in the strident terms which were advanced during the trial. The authors must have got their information from somewhere
- It is Amoco's submission that the pinion failure on 12 January and the alleged failures of the well control system after the flushing "transformed the situation into a crisis" and from then on "Amoco's main concern was safety and that was the basis on which the contract was terminated". The documents lend no support to that thesis. Indeed the opposite. Mr O'Brien's reports speak for themselves albeit he was on the rig at Mr Denholm's request to give his views on the state of the RGV. I regret to say that whilst both Mr Mogford and Mr Denholm maintained that the decision by Amoco to terminate was made on grounds of safety and loss of confidence in BAO neither was able to sustain that in evidence. Termination was expressed to be based on the rig not being operational and delay. Whilst Amoco submits that the evidence of the co-venturers "confirmed that safety was Amoco's driving concern in the decision to terminate" I also do not think that is so. The thrust of Mr Routh's (Amerada Hess) evidence in his witness statement was that what he was told was that the rig was "still not operable" and Amoco did not consider the rig to be "fit for purpose". Moreover Mr Routh was not responsible for the decision at Amerada Hess to support termination nor did he even discuss the matter with those who were who apparently made the decision on Friday 15 January on the basis of a brief which has properly not been disclosed. Mr Craig (Enterprise) reported to a management team meeting on 18 January the minutes of which state that termination was being considered "as the rig was late and suffering from a number of defects". Mr Craig's own information was derived from others in Enterprise as he had no direct contact with anyone at Amoco. The minutes are consistent with the terms of the ATN. Whilst both Mr Routh and Mr Craig made references to safety in their oral evidence I think the documents are more reliable. Amoco's letters dated 8, 15, 18 and 19 January were each copied to Enterprise and Amerada Hess and I have already commented on their terms: see paragraphs 156 to 158; 165 to 166; 171; and 172
- After 19 January ADC (Mr Sleightholme and Mel Thomson) prepared a final version of the ADC interim report (paragraph 122). They did so by going through the interim report, daily reports made thereafter, and the CARs. This final version was described as "Final Revision - showing status (of RGV) at North Sea Location -19th January 1999". The issue date is recorded as "February 1999" which, on the evidence, represents when it was subsequently approved by Mr Hay. It is an important and substantial document and will be referred to again in context as "ADC's final report". It was written in the knowledge that Amoco had terminated the contract. Mr Sleightholme described it as a good record of the state of the rig as seen through the acceptance testing as at 19 January. Mr (Mel) Thomson's evidence seeking to downplay its significance did him no credit. The "Key Points" were summarised. There were five of them. Only one is of any relevance. "Outstanding works" included the statement that "a major concern is the amount of acceptance testing work to be done, particularly in the area of BOP testing, the drill floor equipment ...." Reference was also made to the CARs under "Other Issues": "CARs cover some of the major concerns listed above. Good progress has been made in progressing CARs since arrival at Rotterdam and there are now 12 out of 58 still in the 'not done' category. Copies are attached ...." In fact only 1 of the 'not done' CARs relates to a matter complained of in these proceedings, namely No 56 described as "BOP Operating Chamber Contamination". No 56 is also and rightly described as "late CAR" albeit no exact date for its issue is given. Neither the Key Points nor the CARs make any reference to the Jacking System. No safety concerns were expressed about either the jacking or well control systems: see paragraphs 294 and 447 to 450
SUBSEQUENT EVENTS
- On January 26 (Appendix B) BAO wrote stating that it would treat the contract as continuing in full force and effect and that the RGV, for safety and operational reasons, would shortly be arriving at a standby location at Dundee. The RGV arrived alongside a wharf at Dundee harbour on 26 January and whilst it was there BAO continued to carry out various works upon it and to investigate the 12 January and earlier pinion failures. Amoco relies on the records of these matters to support its case on fitness for purpose and repudiation.
- A detailed response to Amoco's letters of 19 and 20 January was sent by BAO on 19 February. Amoco also places some considerable reliance on this letter as a further manifestation of what is said to be BAO's "intransigence" in the face of serious and justified complaints about the RGV and as evidencing a failure to recognise those complaints. That is addressed as regards the well control system in paragraph 491. The letter is also included in Appendix B. In summary
i) The purpose of the letter (which ran to 15 pages) as it stated and its terms exemplify, was to respond to the "contractual and technical points" raised in Amoco's letters;
ii) It included statements that:
"In the morning of 19 January representatives of Amoco attended on board the rig and requested an extensive test to demonstrate the operability of the rig. Although such a test was not required by the Agreement BAO agreed to perform such a test"; and that
"the alleged breakdowns are, on proper analysis, without substance. In any event it is absurd to suggest that BAO would not, and did not, seek to address each and every concern Amoco raised. To do otherwise would potentially jeopardise a contract under which BAO had spent over 2 years constructing a rig at a cost in excess of $200m, transporting it to the North Sea and arranging for it to operate for a period of at least 12 months. To assert further that BAO repudiated the Agreement in these circumstances is a nonsense."
iii) It addressed and refuted the allegations about the BOP including stating that: "The BOP and BOP accumulator system were thoroughly (and satisfactorily) flushed before your purported termination of the agreement. In those circumstances there is no reasonable justification for stripping down the BOP nor has the manufacturer's representative on board recommended this. This requirement is beyond any reasonable acceptance testing and is unnecessary as, indeed, was acknowledged by Amoco's ADC representative on board the rig".
iv) It stated that the access limitations of the PRS had no impact on the performance of the rig, and that BAO's personnel were properly trained, qualified and competent and had been and were in the process of developing increased proficiency in new equipment;
v) It asserted regarding the pinion failures that "our letter of 18 January 1999 clearly establishes the likely root cause of each incident and would allow a reasonable person to make an informed judgment";
vi) It concluded by stating that "as this letter demonstrates, none of the issues you have raised has any substance and could not properly form the basis for any notice of termination. Your actions in purporting to terminate the agreement are themselves a clear repudiation of the agreement. We shall however continue to treat the contract as subsisting for a further seven days .... in the event that Amoco wishes to reconsider the position we stand ready to resume the contract. In such event we would expect payment of all outstanding amounts due under the agreement ...."
- In March DNV issued a document entitled “Conditions (Recommendations) memo for Owners” which included as a condition to be dealt with by 20 May “Resolution of broken jacking pinions to be forwarded on completion of investigation”. That did not affect the validity of the RGV interim Class Certificate. On 30 July DNV faxed Mr Dowdy chasing up a number of “outstanding survey items”. One was this condition with its "Status” recorded as “LeTourneau to confirm to DNV Houston in writing that original theory re overloading was cause of broken pinions & that subsequent investigations with casting manufacturer revealed no manufacturing defect”. Mr Dowdy sent DNV Houston a copy of a letter he had already received from PSC dated 9 July (paragraphs 303 to 304) and wrote confirming that the failures were due to severe overload. DNV then agreed to delete the condition
- The RGV remained at Dundee until 22 October. On that day it left on the start of a journey to Canada and BAO's solicitors wrote stating that BAO now accepted what was said to be Amoco's repudiation of the contract. In the event, the RGV only started work offshore Nova Scotia Canada on 29 March 2000. The evidence is undisputed that the RGV has performed well and without significant problems in Canada. The Arbroath Infill Project had not been carried out at the time the trial began. A considerably revised programme was planned for this year (2001) but had not commenced by mid-February
- No chronology would be complete without recording that the trial began on Monday 22 January, 2001 and ended on 4 October after 85 days in court. The Closing submissions were provided in writing, over the course of the last 2 weeks of the Summer term. Amoco's ran to 361 pages and 199 pages of reply submissions. BAO's were presented in 3 substantial lever-arch volumes. That is not intended to read as a complaint but more to indicate the nature of the proceedings and to provide some excuse for the inordinate length of this judgment, and the frequent need to resort to cross-references. Further, some serious allegations have been made against individuals which, even if they do not affect the outcome, those concerned are entitled to have addressed. I have nonetheless at least sought to be selective and where I think appropriate brief. If ever there was a case in which I think it neither sensible nor practical to address all the submissions made let alone all the evidence adduced this is that case.
THE EVIDENCE
- The evidence both documentary and oral unsurprisingly does not fit neatly into compartments but I think it appropriate to express some general views upon it so that the specific issues to which I must now turn and my judgment upon them may be seen in that context
- First, the documentary evidence is almost without exception supportive of BAO's case on all the major issues. In many instances in which if Amoco's case was as it was advanced to be it would be natural to expect to find support for it in the documents there is no support to be found. Examples I have already touched upon are the pinion failures on the jacking system. Amoco's case is that it was unsafe to jack up the RGV alongside the Arbroath platform at least without further investigation. Yet the earlier failures in September were not the subject of any pressing enquiries; and the one which occurred on 12 January was not considered significant by ADC if the final report ADC made means what it says. Further the documents and the evidence of those on the RGV demonstrate that on 19 January the RGV was being prepared for its final approach to the Arbroath platform so far as both BAO and Amoco's rig personnel were concerned with apparent disregard for what is now put forward as a major safety issue. Similarly the complaints about contamination of the hydraulic system activating the BOP which are also said to involve potential major safety concerns do not find much real support in the documents at least after the system was flushed as it was before the contract was terminated
- Second, the evidence is really overwhelming that Amoco and the co-venturers were determined to come up with the most persuasive case they could to put BAO on the spot of renegotiating the rates under the contract or risk losing the expected first return from the massive investment by Rowan in the RGV. It was indeed a policy of looking for any failure on BAO's part and exploiting it mercilessly. But if you search for failures with such a pressing objective there is a risk you will find them where they do not in fact exist or exaggerate those that do exist to an extent which is unjustified. In my judgment the evidence demonstrates both those characteristics, as does the bare fact that the allegations themselves have undergone considerable change and in the case of the drill floor been substantially abandoned during the course of the trial
- Third, and with only limited exceptions, all the witnesses to the events who gave evidence on behalf of Amoco had a torrid time in cross-examination. That was a reflection of the difficulty they had in sustaining Amoco's case in the face of the undoubted facts and the documents. BAO also submitted that there were some surprising omissions in the choice and status of the witnesses to give evidence. There is some force in that criticism, albeit a large number of witnesses did give evidence, and covered all the salient issues. In particular, I regret to record that I found much of the evidence of Mr Gillis and Mr Hosie inconsistent and unsatisfactory as well as the evidence of Mr Fleming, Mr Thomson and Mr Sheen of ADC. The evidence of Mr Mogford and Mr Denholm was also unimpressive particularly in seeking to maintain a case which was frequently demonstrated to be in conflict with the documents. In contrast I should record that with only very limited exceptions, I found BAO's witnesses compelling, straightforward, and impressive. Indeed the quality, in particular of the Rowan, LeTourneau and BAO personnel, who gave evidence was remarkable. They had an obvious and attractive pride in what they did and a justified confidence in their ability to do it born of experience
- Mr Mogford and Mr Denholm gave evidence that the decision to terminate the contract was based only on safety factors and not economic factors. As stated (paragraph 198) I cannot accept that evidence. It is belied by the documents and the sequence of events which led to termination. Even the ATN (paragraphs 196 to 197) makes little mention of “safety” and both Mr Mogford and Mr Denholm were wholly unconvincing in their attempts to explain it away. The strategy was to rely on Clause 28 and delay not safety considerations. Insofar as the failed F7 pinion was a matter of concern the correspondence demonstrates that the termination strategy was well under way before the pinion failed on 12 January and remarkably those from Amoco , BP and ADC on board the RGV who believed up to and on 19 January that it was about to go to Arbroath were hardly asked for their views. Both Mr Mogford and Mr Denholm also sought to make a case that BAO’s “attitude” was one of “intransigence” in relation to the condition of the RGV. Again I reject that totally. The intransigence, if that is an appropriate word, was in relying on the contract and refusing to reduce the rate of hire when BAO saw no reason to do so. The evidence is all one way that BAO was making every effort to solve such problems as there were. Amoco and ADC were on board and well able to judge that for themselves. BAO was not so naïve as not to appreciate that Amoco was looking for grounds to escape from or renegotiate the contract. Indeed the correspondence acknowledges BAO’s efforts to get things right. The point made was more that the fact that BAO was moving heaven and earth to do so meant the problems were serious and would take time to resolve not that they were being ignored. Mr Quigley, who left the RGV for the last time on 9 January, readily accepted that the people on the rig were doing their very best diligently to get the rig into the best possible state. Mr Quance and many others who had direct experience of work on the rig had nothing but praise for the quality of Rowan personnel
- Fourth there was a very marked disparity between Amoco's case which was concentrated upon theoretical risks, complaints and supposed or possible concerns and the evidence of those who worked on and who would in fact have been operating the RGV. They were the last people to treat safety issues with anything other than total respect, their concern for the safety of those for whom they were answerable as well as themselves was utterly genuine and thorough, but seeing the complaints in the light of their realistic and on occasion disbelieving response was an eye opener to the real world of rigs and the qualities of those who work on them. Anything can go wrong in the sense of possibility and on rigs can have catastrophic consequences but there is still a need to keep one's feet at least near the deck. The fact that the dangers are great does not mean or require that the exercise of judgment about risk by intelligent, experienced and well-qualified people is precluded. In the accepted language and principles of the industry the requirement is to reduce risk "as low as reasonably practicable" (ALARP). That sets an appropriately stringent standard but also recognises that risk cannot be eliminated anymore than human error or mechanical failure
- Fifth, the expert evidence. Whilst I do not question the expertise and professionalism of the experts called on behalf of Amoco nor that in general each of them was properly seeking to assist the court in resolving the issues to which their expertise related, I do think much of the expert evidence exhibited the same contrast as much of the factual evidence between theory and reality. Further in many instances it became or in my judgment is apparent that the reports were built on mistaken factual premises
CONTRACTUAL ISSUES
- Although to a considerable extent my conclusions on questions of construction have already been expressed, I propose both to summarise them here and, so far as can be done, also to address certain ancillary or related matters of fact in the context of the documents and chronology to which I have also already referred. Not only are my conclusions important to the issues themselves but also to addressing the various allegations about the condition of the RGV which are considered in subsequent sections of this judgment
- Clause 2.1 did not impose any obligation on BAO as to the time for mobilisation of the RGV from the United States to Rotterdam
- The contract and charges "commenced" 24 hours after "delivery" of the RGV to Amoco at Rotterdam: Clauses 2.1 and 2.2. In my judgment, as a matter of construction, that occurred 24 hours after BAO made the RGV available to Amoco at Rotterdam for the purposes of performing the contractual services: Paragraphs 26 to 33
- As a matter of fact, I think delivery in that sense occurred at 18.30 hours on December 24, 1998: (paragraph 151). I reject Amoco's case that delivery in fact occurred prior to that date. The evidence is all one way that neither BAO nor Amoco considered that the contract had commenced until BAO wrote the letter dated December 24. Whilst Amoco has sought to distinguish commencement of the contract from an obligation to pay charges the distinction is not in my judgment sustainable when Clause 2 provides that both depend on "delivery" and whilst the motive may have been not to pay charges the tactic was plainly to do nothing which could enable BAO to say that the contract had commenced. Indeed until 24 December that was a shared understanding of Amoco and BAO. In Amoco's case express instructions were given that those who did board the rig at Rotterdam should not act in any way which could give rise to an argument that the rig was on contract. BAO's position (Mr Marcom) was that there was further work to be done to meet the requirements of ABS and DNV prior to drilling and that this work was to be completed before the rig was delivered to Amoco. Mr Marcom was not challenged on this evidence. Indeed there was no basis on which Amoco could challenge it. No more did any witness from Amoco even suggest that it had been understood by Amoco that the rig had been delivered before December 24
- Mr Barnes submitted that the parties intentions were irrelevant because "delivery" was a unilateral matter for BAO to be judged objectively and not from unexpressed intentions. Whether that submission be right or wrong the correspondence on which Amoco relies (paragraphs 144 to 146) cannot in my judgment be read as BAO making the RGV available to Amoco for the purpose of performing the contractual services, nor should it be read out of the factual context in which it was sent. Amoco's letter of December 8 was clear that the purpose of Amoco boarding the rig was not to take delivery of it but rather to ensure that it was, as Mr Cowie put it, 100 per cent ready before it went on contract. That was in effect no more than an extension of Amoco's activities at Sabine Pass where Amoco and ADC had also been on board the rig. Mr Marcom did say in the fax of 14 December that every effort should be made to move the rig to Arbroath at the first available weather opportunity but not only was that sent prior to arrival of the rig in Rotterdam but, before, on and after its arrival, and when Amoco boarded it, it was known that the rig would in fact remain in harbour for some time (paragraph 145). No reference was made to charges. I see no basis therefore in the correspondence not to give effect to what both parties considered and intended at the time, and how they conducted themselves, namely that the RGV had not been delivered in the relevant sense and the contract had not commenced
- Clause 28.1.b) can only apply in circumstances where at least 30 consecutive days have elapsed after the contract has commenced under Clause 2.1. At 19 January 1999 that condition had not been met and for that reason alone, in my judgment, Amoco cannot rely on Clause 28.1.b) as providing a contractual right to terminate the contract : see Paragraphs 89 and 90
- Amoco's case that it was entitled to terminate the contract pursuant to Clauses 23.1.1. and 28.1.d) depends (at least) on the establishment both of (i) a proper notice and (ii) a refusal or failure by BAO to take action to rectify the fault or a failure to commence to do so. The only "notice" relied upon is the 8 January letter (Paragraph 156). The only item in that letter now relied upon as notified by it is expressed to be "deficiencies" in "well control equipment including hydraulic control equipment ....BOP". This letter, far from asserting that BAO had failed to respond to date or at all, stated that BAO was "continuing to work on the rig with a view to overcoming the problems". That is the antithesis of a refusal to take action to rectify the fault required by both Clauses 23.1.1. and 28.1.d)
- The letter was wholly unspecific as to the "deficiencies". Nor did it refer to or state any requirements of Amoco as to how or what steps should be taken to remedy them. Nor did it state how if at all Amoco was dissatisfied with the steps BAO was taking to "overcome the problems". Indeed insofar as "problems" or "deficiencies" were referred to it was expressly said that BAO was working on them with a view to overcoming them. That was unqualified, save for an expression of doubt as to when not whether they might be overcome. No doubt conscious of that, Amoco has alleged that BAO was in fact aware of the deficiencies and that in its response on 18 January (Paragraph 169) demonstrated as much. But that (even if right) only serves to add emphasis to the statement in that letter that any deficiencies so far as known to BAO had been or were being addressed. As Amoco itself submitted "the whole point of the letter" was that "it should be acted upon" and I would add that such was also the point of the provision for notice itself
- Amoco's case is that BAO (i) refused to strip down or inspect the BOP system and (ii) did not proceed diligently to Amoco's satisfaction to remedy the well control system. As to the first (i) the allegation is, I think, hopeless as a basis for the operation of the clause. The 8 January "notice" does not begin to satisfy the requirement that it "specify in detail" the defect and it makes no reference at all to stripping down or inspecting the system. As will be seen the evidence is again sufficiently clear. BAO considered a strip down unnecessary. Amoco did not request one let alone require one prior to 19 January. The nearest Amoco came to it was in the letter of 15 January; but the contract was terminated within days of that and it cannot be relied upon as a notice under the Clause. Further BAO did continue to work on the well control system by flushing it and replacing the control fluid. Far from refusing to remedy the problem BAO was doing what it considered would cure it. If Amoco wanted more in my judgment the clause required it to say so. Personnel of the relevant contractors (Cameron) were on board the rig throughout. Mr Marcom said (without challenge) and I unhesitatingly accept that BAO did not refuse to do anything for Amoco. He also said and I accept that whatever BAO's views as to the need to do so, if Amoco had insisted on a strip down of the BOP BAO would have stripped it down. Mr Quigley and Mr Cowie readily and rightly acknowledged that "Rowan" was doing everything possible to get the rig fully operational. It is of course of some ironic significance in this context that because Amoco did not want to do anything to support BAO's contention that the rig was on contract Amoco, as Mr Gillis said, was deliberately avoiding giving instructions to BAO as to work to be done on the rig. That was so throughout the period up to termination on 19 January. To characterise, as Amoco's submissions sought to do, BAO's letter of 18 January and failure to agree to Amoco's 15 January letter (itself emasculated by the 20 January letter: paragraph 172) as a "refusal" by BAO to remedy the matter complained of is in my judgment close to absurd
- The reality was that Amoco had taken the commercial need to terminate the contract up to (and beyond) the wire which was represented by a suitable weather window to take the RGV alongside the Arbroath platform. The niceties of Clause 28 and in particular Clause 23.1.1. did not enter into the matter as a reading of the correspondence makes clear; and insofar as they did it could not be asserted that BAO was doing anything other than getting the rig ready to go to work. Amoco appears to have thought it was entitled to impose a seven day deadline by which the RGV must be operational, failing which it was entitled to terminate the contract. But that is not what either Clause 23.1.1. or Clause 28.1.b) provides
THE ROLE OF ADC
- I do not think it necessary or helpful to analyse the role of ADC in terms of agency or estoppel or Clause 6.1.5. There is no real issue that ADC was responsible for closing out the CARs and did so or should be taken as doing so with Amoco’s knowledge. Mr Cowie, who introduced ADC to Rowan in Sabine Pass in September 1998, in effect accepted that whilst ADC would sign off on the CARs Amoco would be aware ADC was doing so and would have sanctioned it. Mr Hay and Mr Sleightholme accepted that Rowan would have understood that if ADC witnessed tests and did not ask for them to be repeated then Amoco could be taken to have been satisfied. That was how Mr Perkins and Mr Snow saw ADC’s role also. The real importance of ADC's evidence and documents lies in the extent to which they support or otherwise the parties' factual and expert cases
THE JACKING SYSTEM
Description
- The system uses multiple drive trains housed in a structure attached to the hull of the rig. The output from each drive train is a 'climbing' ¼ pitch pinion which meshes with the leg rack, making a rack and pinion system for vertical movement. It is these ¼ pitch pinions and the four failures (paragraphs 120 and 161) which are at the heart of the issues concerning the jacking system. The RGV has a total of 72 climbing pinions with each pinion having a rated normal elevating capacity of 1000 Kips (equivalent to 1m pounds). Each pinion has 7 teeth. Each elevating unit comprises 8 drive trains or gear units arranged in 4 opposed pairs with one elevating unit on each of the 3 chords of each leg, making a total of 24 units on each of the 3 legs. The distance tip to tip of the teeth is 3 feet. Each tooth is about 10 inches thick at the root and 9 inches at the tip
- The Unit is a 3402.2 : 1 ratio speed reducing unit (i.e. it reduces the motor revolutions from very fast to very slow) consisting of an Electric Drive Motor, the motor pinion, the No 6 Primary gearbox and a Secondary gearbox. Holding power in the static condition is achieved by spring-set, electro-magnetically- released multi-faced disc brakes attached to the motors. The brakes are "on" so long as no power is applied to release them. The holding capacity of the brakes is at least some 3200 Kips and possibly more. The motor pinion is attached to each motor and protrudes into the No 6 gearbox which is also attached to the motor. It meshes with and drives the first reduction gear in the gearbox. The output pinion from the No 6 gearbox meshes with and drives "the red gear" which is a large diameter gear. The hub of the red gear is connected ("splined") to the shaft of a ½ pitch pinion which drives another large diameter gear, the bull gear. The bull gear is splined to the shaft of the larger ¼ pitch pinion, the teeth of which climb up the rack
- The normal elevating capacity rating of 1000 Kips was based on load application directly on the ¼ pitch climbing pinion. The system as a whole had six load ratings to provide design parameters for all operating conditions. The minimum design temperature was -20 C. The ratings were
i) normal lifting capacity 925 Kips per pinion;
ii) preload lifting capacity 1200 Kips per pinion;
iii) preload lowering capacity 1300 Kips per pinion;
iv) normal holding capacity 1300 Kips per pinion;
v) preload holding capacity 1800 Kips per pinion;
vi) severe storm holding capacity 1800 Kips per pinion
- "Preloading" is the addition of water into tanks in the hull of the rig in order to simulate the maximum loading of the leg footings (or spud cans) on the seabed. It is used in effect to establish a firm footing on the seabed to support the rig's elevated weight and resist severe storm reactions. The water ballast simulates the increased vertical load in such conditions. The North Sea has a hard sea bed and typically the spud cans will penetrate less than 5 metres into the seabed during the preload operation
- The RGV also has StormLOKs, a system of chocks and wedges which engage the leg rack when the rig is in an elevated position. StormLOKs take a proportion of storm induced loads to ensure that the pinions do not become overloaded in severe storm conditions
- The design parameters for the RGV were significantly in excess of the 50 year storm condition required for operations at the Arbroath field, meaning that even the worst possible storm predicted to occur at Arbroath once in a 50 year period could readily be withstood by the rig if the design was in fact met
- In December 1998 LeTourneau performed a site specific study of the Arbroath site for the RGV (see paragraph 37). The “critical check” was “leg chord usage factor for a maximum survival storm”. That is the estimated worst 50 year storm at Arbroath. The maximum pinion load with the StormLOKs in place (for a wave at a 90( heading to the leg) was calculated to be 1327 Kips. With the rig in what is called Operating Condition 1 it was 1842 Kips. Operating or Drilling Condition 1 is intended to represent the maximum condition in which the rig can operate without the StormLOKs engaged
- Of course if the rig was in Operating Condition 1 there could be expected to be ample warning before a “maximum” storm arrived to enable the StormLOKs to be engaged. They can be engaged at any time the rig is not jacking and take some 3 hours to install. Subsequent evaluation by LeTourneau in January 1999 concluded that the figure of 1327 Kips was a slight over-estimation of the load and it was reduced to 1289 Kips. Mr Mobbs said and I accept that both values were conservative: see also paragraph 310
- The principal reason for having multiple climbing pinions is redundancy. It is not uncommon to have one or more gear units out of service during normal operations. In the case of the RGV normal elevating capacity, assuming proper balance of the hull weight between the three legs, would still be obtained even with 8 of the 24 pinions on each leg out of service for repair or maintenance
- A consequence of the level of redundancy and the design of the No 6 motors is that the jacking system in fact had sufficient power to "jack through" jams in the gear trains, in effect to carry on jacking by forcing the jam to yield. On any view of the evidence that is a course of action which would not even be contemplated save in an extreme emergency because of the risk of damage to the rack, gears and motors. Whilst Amoco submits that BAO's witnesses "changed their tune" from their written to oral evidence on the practicality and ease of jacking through, and to a limited extent I think that is right, it is of little significance as they readily recognised the risks of doing so. The point made was that circumstances could not be envisaged where it would need to be considered but if they occurred it could probably be managed successfully (see paragraph 285)
- The earlier design for the jacking system was rated at 750 Kips. It was tested in 1991 but never put into manufacture. It was itself a development of a 375 Kip system used successfully on numerous LeTourneau rigs throughout the world. The results of the 1991 tests and other work led to the enhancement of the rating in 1993 to 1000 Kips. Both ABS and DNV approved the gear units. Testing was repeated on the 1000 Kips units which included duty cycle and ultimate loading tests as well as the normal holding tests required by ABS.
- The testing was carried out on a test stand utilising two opposed gear units. In November 1996 each of the units was tested to 1950 Kips for one complete revolution of a ¼ pitch-climbing pinion. Thus each tooth on both pinions was loaded. The test was witnessed and approved by both ABS and DNV. In fact during the test the units were subjected to a maximum load of 2105 Kips without failure and were then subjected to magnetic particle inspection (MPI) for cracks. No cracks were found. This was the test required by ABS for certification. The temperature at which the test was carried out was approximately 9 C. Dr Slater calculated that the equivalent at -6 C was 1800 Kips. Dr Manteghi's equivalent figure was 1717 Kips. In July 1997 ABS approved the system for capacities of 1800 Kips (static) at -20 C on the basis of this test (paragraph 225)
- In April/May 1997 a LeTourneau in-house 20 year duty cycle test was carried out on the same two test stand units. The test sought to subject the units to the amount of load per pinion to which it would typically be subjected during 20 years normal service but in a concentrated much shorter time period. 185 complete revolutions at mean load levels of 1300 Kips with peak values of 1388 Kips were observed. After the test MPI showed surface cracks on the compression side of the root of several teeth of both ½ pitch pinions. No other components were found to have cracks. The cracks to the teeth of the ½ pitch pinions were thought to be the result of a combination of the stresses from the November load test and the multiple stress cycles during the 20 year duty cycle test itself. The same pinions were subsequently subjected to the ultimate load test when they supported a load of 1.8 times the severe storm holding capacity
- The ultimate load test was a static test which applied a load of 3000 Kips to one tooth of each pinion. A load of 3218 Kips was in fact applied to each unit. Neither failed. Any further load risked failure of the test machinery itself
- Mr Dowdy was responsible at LeTourneau for the design development, shipyard liaison and regulatory approval of the Super Gorilla series of which the RGV was the first. He (like his “boss” Mr Bowes) was a most impressive witness, plainly well qualified by both education and experience. He drafted the specification for the cast steel components of the gear units, based on the original 750 Kips design parameters and test results enhanced to 1000 Kips. The specification was provided to PSC for review and approval and after agreement with PSC was approved by both ABS and DNV in early 1996. The specification in terms and as usual in the industry applied to test coupons (API 8C test blocks) of the steel to be cast, albeit it is agreed and was known that the small coupons would not be expected to have the same properties as the much larger components they represent. Mr Dowdy said the governing property for the design of the ¼ pitch pinion was strength. That was not really challenged and I accept it without hesitation
- The specification was originally dated 19 December 1995. This Specification, known as the External Source Product (ESP) Specification, included statements to the effect that
i) Mechanical testing would be conducted on “separately cast Equivalent Round API 8C test Blocks representative of the cast part”.
ii) The design atmospheric source temperature was -20(C (-4(F).
iii) Magnetic particle examination was to be carried out by the wet fluorescent method and “shall include 100% surface inspection of all toothed areas including ends of teeth.” Specific standards were referred to as well as the need to satisfy DNV and ABS.
iv) “Mechanical Properties: The following mechanical properties are required after final machining and heat treatment. Core properties of yield strength, tensile strength, Charpy energy, lateral expansion, elongation, and reduction of area requirements defined below are to be established by destructive testing of samples cut from API 8C Equivalent Round test blocks at ¼ below the surface…. All testing shall conform to ASTM A370. All values specified are minimum unless a range is provided.”
- For the ¼ pitch pinion the values of the core mechanical properties were
i) Core Strength (KSI) Yield 100/Tensile 115
ii) Min. Elongation 10%
iii) Area Reduction 35%
iv) Charpy Energy 15ft lbs @ -40(C
v) Lateral Expansion 0.0010 min.
- The ESP specification went through a number of revisions but there were no changes to these values. KSI is Kips per square inch. The test coupons are cast at the same time as the pinions. The core properties were to be established by destructive testing of the test coupons. The specification also required PSC to take surface hardness readings with a minimum required value
- The ESP Specification provided for Magnetic Particle Inspection (MPI) of the pinions after the normalizing and tempering process. Acceptance criteria were to be as indicated in API 8C 6.4.6.2. That provided “only those indications with major dimensions greater than 1/16 inch (1.58mm) and associated with surface rupture shall be considered relevant”. MPI is designed to detect surface breaking flaws and, as will be seen, it is surface flaws to which the thrust of Amoco's allegations is directed
- Yield strength and tensile strength are measurements of the ability of a pinion to withstand load. Yield strength is the force measured in KSI that will initiate "plastic" deformation of the material, meaning deformation which remains permanent upon removal of the force or load. The contrast is with "elastic" deformation which is deformation which disappears when the force or load is removed
- Tensile strength is the force needed to break the material. The maximum specified for the ¼ pitch pinion test coupon was 115 KSI
- Hardness. Hardness is a measure of the ability of a material to resist indentation and abrasion and so avoid wear. There is now no allegation that the pinions were defective because of poor surface hardness. Indeed the evidence is that the minimum value was achieved
- Minimum elongation and Reduction in Area are measures or indications of ductility.
- Ductility is the ability of a material to deform plastically before it fractures. A material that lacks in ductility will be more sensitive to the presence of surface imperfections where stress concentrations (higher loads) occur
- A material may elongate before fracture. A specified minimum elongation is the extension which is required before fracture occurs of a uniform section of a specimen expressed as a percentage of its original length. For the ¼ pitch pinion the minimum percentage specified for the test coupon was 10
- A material which is elongated will also become slightly thinner before it fractures. Area Reduction (or Reduction of Area), if specified, is the required minimum decrease before fracture occurs of the cross-sectional area of a uniform section of a specimen expressed as a percentage of the original cross-sectional area. The specified minimum for the ¼ pitch pinion test coupon was 35%
- Toughness is the ability of a material to absorb energy and deform without fracturing. It relates to both strength and ductility. It can be assessed by destructive fracture toughness testing or indicated by measuring impact strength. Toughness decreases as strength increases. It is also sensitive to temperature but decreases with it. It, too, is a measure of a material's ability to withstand brittle fracture
- "Charpy Energy" refers to an impact test in which a specimen of a uniform size with a V-notch in the middle, supported at both ends, is struck behind the notch by a striker mounted at the lower end of a bar that can swing as a pendulum. The lowest quantity of energy (measured by the Charpy impact machine) absorbed by the specimen before it fractures is the Charpy energy. Charpy testing is carried out at low temperatures (-40°C in the case of the ¼ pitch pinion) when the ductility and toughness of a material is low. The specified minimum level of Charpy energy for the ¼ pitch pinion test coupon was 15 ft lbs
- When a Charpy sample is struck it will expand laterally before it fractures. The minimum "lateral expansion" specified for the ¼ pitch pinion test coupon was 0.010"
- In general terms, the higher the measurements of minimum elongation, area reduction, and Charpy energy and lateral expansion of a sample the more "ductile" the sample. The aim of any casting is to achieve the right balance of strength, hardness and ductility for the particular usage to which it is to be subjected. The more strong and hard a material is the more brittle it is. The ¼ pitch pinion requires hardness to limit wear at the rack and pinion mating surfaces. Tensile strength and yield are important due to the applied loadings on the pinions, albeit they are normally loaded significantly below their rated capacity. The risk of impact loading is minimal and so is the risk of cycle fatigue because of the location of the pinions and their low rotational speed. The importance of ductility and toughness is a matter of dispute
- The casting of the pinions was in one of PSC’s most commonly used alloys. PSC has years of experience in producing large castings for the oil industry using the same alloy. Mr Neely said, and I accept, that PSC has never experienced a casting failure below design load for any such parts. The company is certified by both ABS and DNV. Mr Neely said it was a “fact of life” (albeit very hard to explain to customers) that final values will vary between the casting and its test coupon as well as other identically treated coupons or castings. That was the evidence of all the relevant witnesses (save Mr Loflin). I accept it. Mr Loflin was wrong. It is why dynamic testing of castings is the preferred method of and of more value in determining suitability. It is important to note that all the test coupons for the ¼ pitch pinions for the RGV in fact met the requirements of the specification. The significance of the coupons as explained by Mr Neely and Dr Slater is that if they correlate to the test stand coupons they establish a correlation to the test stand pinion which is known to have been proved by dynamic testing
- There is no dispute that PSC’s MPI procedures were of the best standards and those who carried them out very experienced and skilled. PSC MPI the pinions twice. Once prior to rough machining and a second time after the casting has been heat treated and prepared by shot blasting to produce a more clean and smooth surface. At each stage the casting is sprayed with a mix of fluorescent magnetic particle inspection powder and water whilst an electrical current is applied through the casting. An ultra violet light is shone on the part to which the current has just been applied. The operator then examines the surface by eye to check for any indications or flaws which show up under the light. The operator has a gauge with indications of various sizes which also react to the light and can be compared to any indications shown up on the casting. Mr Neely said that “with experience” an operator will know the approximate size of a defect one-sixteenth of an inch long (1.58mm) erring on the conservative side. Mr Neely described that as “the industry standard”. Any such defect is marked. If it obviously needs welding, it is welded. Otherwise, after the second MPI, the casting is machine-polished and then ground or welded as required. At each stage it is inspected by a qualified operator. If the casting is welded the whole MPI procedure is repeated. Thus every pinion was subjected to MPI over the entire surface twice and any that were welded three times
- Mr Costa was the MPI operator at PSC for the pinions. He gave evidence. His skills and pride in his job were apparent. He said he was trained to locate and assess indications of one-sixteenth of an inch or less but with his experience (he qualified for the relevant level in 1991) he said he could identify indications of a length of one-thirty-second of an inch (0.8mm)
Amoco's Case
- The allegations made in Paragraph 4C(9) of the Claim and the essence of Amoco's case on the jacking System are as follows
"(a) On 12th January 1999, there was a pinion failure on the starboard leg jacking system. There had been a history of earlier pinion failures, including three occasions which were known at the time to the Plaintiff, while the drilling unit was being fitted out in the USA. Furthermore, two further pinions (E7 and E8) had been changed out or modified in October 1998, at least five faulty brake heads had been found in October/ November 1991, eleven gearboxes in the jacking system (E1, E2, E4, E5, E7, D1, D2, D3, C7, A3 and H1) had given problems or had to be removed or changed in October/ November 1998, and a check of the jacking system on or about 6 January 1999 had revealed leaks on 14 seals and 5 bad bearings and/or thrust plates. This further failure on 12 January 1999 made it essential to investigate and identify the cause of failure before the Drilling Unit could safely be jacked up at its intended location adjacent to the Plaintiff's Arbroath platform. That had not been done satisfactorily by the Termination Date. Such an investigation would have included consideration of any earlier failures or malfunctions in any part of the jacking system (including but not limited to the components referred to above) so that the underlying causes of the pinion failures and the reliability of the entire jacking train could be established with confidence.
(b) Furthermore the Defendant (but not the Plaintiff ) was aware at the Termination Date that tests carried out after the earlier pinion failures had shown there was a discrepancy between the properties of the test coupons reported in the mill test certificates and the pinions themselves and that the pinions had properties as regards toughness and ductility which were inferior both to those stated in the mill test certificates and the LeTourneau specification. Those properties made the pinions more susceptible to brittle fracture. These findings related not only to the failed pinions but also to the comparison pinions and it is to be inferred that all of the pinions had similarly poor (though variable) properties. This reinforced the need for further investigation and risk assessment before the Drilling Unit could safely be jacked up adjacent to the platform. The Defendant failed to disclose these test results to the Plaintiff, but instead (in Mr Marcom's fax to Mr Gillis of 18 January 1999) attempted to reassure the Plaintiff that the pinion failures gave no cause for concern.
(c) Amoco will say that it is to be inferred from the pinion failures and from the state of the jacking system in the period before the Termination Date as set out in sub-paragraphs (a) and (b) above that there was an unacceptable risk of further failure.
(i) because at least two of the failed pinions (E3 and E6) had probably been subject to loads beyond those for which they had been designed, the source of which had not been reliably identified or (so far as identified) eliminated; and
(ii) because the low toughness and ductility meant that all the pinions were susceptible to brittle failure even at their design loads.
(d) Further in February 1999 LeTourneau recommended and later carried out the replacement of the damaged hubs to the 1/2 pitch pinions and recommended the opening for inspection of all primary gearboxes. LeTourneau also recommended on 9 February 1999 that there should only be "minimal jacking" until some of this work had been carried out. All 72 secondary gear trains were dismantled and the parts replaced in at least 17 gear trains. Additionally at least 4 further gearboxes were replaced. Amoco will rely upon these recommendations and works as further indications of the defective and/or unready or unfit state of the jacking system at the Termination Date."
- The Contract contained a short description of the Jacking System and its capacity. Amoco relies on the allegations made about it as part of its case on repudiatory breach. It forms no part of the case for termination under Clause 28. The contract provisions on which Amoco relies in support of its case that BAO was in breach as regards the jacking system are Clauses 3.1, 3.2 and 6.1.2. These may at least loosely be described, as Amoco described them, as the general obligations of "fitness for purpose" (paragraphs 35 to 37 and 59 to 60). But it is not (or no longer) alleged that the RGV was not validly certified nor that it did not conform to all relevant laws and specifications nor that there was a failure to comply with the safety case or any safety obligations, nor indeed that the system was not in fact capable of performing adequately at Arbroath. What is alleged (and was developed during the trial) is that there were sufficient doubts about the integrity of the system that the RGV "could not properly be used in the North Sea" until the doubts had been fully investigated and so that it was not fit for its purpose. It is also alleged that BAO repudiated the contract by not acknowledging the need for such an investigation and that the delay involved in carrying it out was such as would itself be repudiatory. There was some considerable debate as to whether Amoco's "investigation" case could properly be brought within these provisions if in fact the jacking system was fit for its purpose. Amoco submitted that if there were sufficient doubts about the system such that they required investigation before the rig could jack up at the Arbroath platform then it was, or at least could be, unfit for its purpose within the Clauses even if investigation would later have revealed it was in fact fit to do so. BAO submits that is not so. For my part I find this a somewhat arid debate and in any event one which is of no real significance on my findings of fact. I do not, however, think the issue is one which can or should be addressed as a matter of principle. To state the obvious, whether the equipment is fit for its purpose depends on an analysis of the purpose and matters of fact and degree. If it cannot be used for its purpose at the time it is required until further investigated I think it may not be fit. To that extent I agree with Amoco's submission. But it does not take anyone very far
The Evidence/Chronology
- The brakes were installed and set in Vicksburg by mechanics supervised by Mr Ford. They and the motors were wired up by electricians supervised by Mr Burris. The first pinion failure to be discovered was on B8. It was found on about 17 September. Mr Snow was jacking the rig as part of the leg up when there was a loud noise from the B chord, and the pinion was found to be broken. As the evidence in relation to the circumstances of the failure unfolded I think that both the probabilities of its cause and why they could not be stated with confidence became apparent. There is no dispute that B8 was first wired up immediately before the walk to the river at Vicksburg and that when jacking started the motor was seen by Mr Massey and others to smoke. The probability is, as Mr Quinn said and Mr Burris accepted, that it had been mis-wired. No other Rowan rig had its gear trains set in an opposing formation. The wiring was also “triple phase” but could in error be wired in single phase by connecting only two of the three wires to the same busbar. Although plainly it should not have happened it is not therefore difficult to suppose that the gear had been mis-wired so that it was seeking to turn against and not with the other gears or (if single phase) not turning or turning only slowly because it was not receiving the full three phases of power. In any event, following the smoking incident, the motor was unhooked. But at Sabine Pass something must have caused the failure, or at least the final and evident failure. Mr Snow said that both he and Mr Burris had thought at the time B8 failed that the brakes had been re-set in the erroneous belief that the new motor which had been fitted had been wired up again. If so, when jacking up began, no power had been applied. The brakes are 'on' when the power is cut. Power is required to release them. So the brakes had not released but acted to restrain the pinion as the other pinions turned freely. The lack of power would also account for the lack of any warning indication. Mr Burris' evidence was, however, that he had on later inspection found that the brakes were in fact released but wondered whether that had been done to cover up a mistake. Mr Ford was adamant the brakes could not have been set let alone released in such circumstances. Mr Snow's evidence is supported by the entry in his log and his evidence that he was told that it was only when the brake was released after the pinion broke that the torque unwound. The brake must therefore have been on when the rig was jacking. I should note that I did not think that Mr Ford's evidence was reliable in this regard. It was, I think, the product of a sense of loyalty to his staff and an obstinate refusal to contemplate that mistakes could occur under his command. What I do think is clear is that at an early stage it was concluded that it was not likely that it would be possible to achieve any greater degree of certainty as to the precise underlying cause of failure. On inspection it was found that there was no damage to the motor, brake or gearbox and they were not changed out. After the pinion was replaced B8 caused no problems
- The failures of the pinions on E3 and E6 give rise to much the same considerations. It is agreed that the pinions failed under reverse loading and severe over-loading in excess of 1800 Kips. Again there are a number of possibilities but all necessarily require a gear or brake or wiring fault. Although differing as to the precise operation in progress at the time, both Mr Quinn and Mr Snow said, and I accept, that they had heard a loud noise on the starboard side of the rig during the final stages of the walk at Vicksburg (Mr Quinn) or when the rig was being first jacked up in the river after the walk (Mr Snow). Both thought in retrospect that this could have been either or both of E3 and E6 breaking, albeit jacking was stopped at the time and an inspection had not revealed any breakage. There is, however, no evidence that any inspection was made from the leg ladders which would probably have been necessary to see any broken pinion tooth. Mr Snow said, and I accept, that when he later heard the noise on the B chord it reminded him of the noise he had heard at Vicksburg on the E chord.
- When the failures were found on 25 September (paragraph 120), Mr Massey's concern was more that they had not been found earlier than that they had occurred. They were not found as a result of any event that day but by workmen carrying out other operations on the leg. A full visual inspection of all the other pinions was then carried out and no other missing teeth or cracks were found. Mr Massey (another most impressive and experienced witness) readily accepted that no definite explanation could be offered. Mr Burris unsurprisingly favoured wrongly set brakes. Mr Ford again would not contemplate that he or the mechanics who worked for him could possibly have made a mistake. But brake binding can also be the result of mis-wiring. Whilst in the final analysis I do not think it affects the issues which arise, as a matter of probability I do think the burden of the evidence, in particular the evidence of Mr Quinn and Mr Snow and Mr Loflin, is that E3 and E6 in fact also suffered from mis-wiring and that the teeth failed or at least were fatally damaged at Vicksburg perhaps one during the walk and one during the jacking up in the river. The failures should have been discovered earlier and Mr Massey was right to be concerned that they were not. But he was also right, as was Mr Butler, that the problem as it then presented itself was not a serious one. Knowledge that the immediate cause was reverse overloading and subsequent proof of performance after repair by the successful jacking operations carried out in Sabine Pass after the pinions were replaced was enough. That would have revealed any further mis-wirings. The whole gear trains were inspected but again nothing else required to be changed out. It is true that it is not easy to explain how the failures were only discovered later, but the broken teeth could have fallen away leaving the pinion in effect to free wheel during subsequent jacking operations. One tooth was never recovered and one was found in the jacking case. Neither was obstructing the system. The only other theory that has credence is brake binding because of debris or rust or some intermittent fault in the brakes. Some witnesses favoured this theory. It cannot be excluded: see also paragraph 277
- Mr Cowie was at Sabine Pass and learnt of the pinion failures at the time. There is no reference to B8 (or the subsequent discovery of the failures of E3 and E6) in any of his reports (prepared twice weekly) nor in any of the records of site meetings. Mr Cowie did say that he had raised the matter orally with Mr (Bryan) Quinn and Mr Rimlinger about twice a week during October and November. He said he was told the pinions were overstressed during the walk. He also said Mr Quinn had referred to a metallurgical report which was "inconclusive". Mr Quinn denies that he referred to a report in those terms. There are also different recollections as to the frequency with which Mr Cowie raised the matter. Whilst these disputes appeared to be significant, in the result I do not think they were. It is not in dispute that Mr Cowie did refer to the failures nor that Mr Quinn agreed to hand over any report if and when he got one nor that he told Mr Cowie that the failures were caused by over-stressing during the walk to the river, were not a serious problem and that, as Mr Cowie had seen, the rig had been successfully jacked since the failures. Whilst I accept that Mr Cowie did raise the matter on a number of occasions, albeit not with the frequency Mr Cowie thought, and that someone told him of metallurgy tests, I also accept Mr Quinn's evidence that Mr Quinn did not refer to (because he was in fact unaware of) any metallurgical report. I think the totality of the evidence is clear and both Mr Cowie and Mr Quinn were honestly seeking to recall the events. The matter simply was not seen at this time as a serious problem or concern but rather one that had been resolved for practical purposes. If it had been otherwise I think it would have found its way into the documents much more and in much stronger terms than it did. Indeed Mr Cowie frankly said that in mid-November Amoco was satisfied with the explanations they had been given which seemed reasonable (see also paragraph 121). So did Mr Gillis (Paragraphs 273 and 275)
- Mr Dowdy of LeTourneau instructed LAW Engineering Services Inc to make and report on a metallurgical investigation of the B8 tooth failure. LAW were able to see the pinion and the failed tooth. LAW reported to Mr Dowdy by letter dated October 21, 1998. LAW had tested a sample of material from the failed tooth. The significant aspects of the report were that
i) "The majority of the fracture surface indicates a brittle fracture mode and there is no evidence which indicates that the failure resulted from a casting inclusion or defect although there are inclusions present on the fracture surface."
ii) "Inclusions are visible on the fracture surface and can be detected microscopically. Although similar inclusions are typical in castings and have been known to initiate cracks, there are no indications which conclude that the tooth fracture originated at such an inclusion."
iii) "The crack which is located on the main body of the pinion has the characteristics of a torsional crack which indicates that the pinion had a torsional load beyond the physical limits of the material. The initiation point and the propagation of the crack suggest that a force was applied on the left side of the subject pinion tooth ... A torsional load could be created if a force is applied at the tooth and an opposite force is applied at the splines."
iv) "The mechanical tests conducted conclude that the material near the tooth fracture surface possesses brittle properties. The elongation and reduction of area results of the tensile test illustrate the lack of ductility and toughness of the material in this area."
v) Mechanical testing showed that the yield strength of the sample was 139 KSI and the tensile strength 149 KSI (specification 100 and 115 respectively). Minimum elongation was 6% (10%), reduction in area 7.4% (35%) and Charpy V-notch 15-15-10 ft/lbs (average of 15). Essentially the material was stronger than the coupon specification but less ductile. Generally that would mean that it would break later but faster.
- The "Conclusion" of the Report was expressed in these terms
"The observations conducted and the data gathered conclude that the root cause of the failure was a mechanical overload of the pinion. Although fracture mode and crack propagation illustrate the brittle nature of the material, which may have contributed to this fracture, the pinion experienced loads which were beyond its physical ability to withstand."
- Calling to an extent on the evidence and explanations available to me from the trial what this comes to (in my terms) is that the pinion was reverse loaded, it was loaded beyond its physical strength, it therefore had to fail somewhere, it failed in a brittle mode (which it is agreed it would whatever its properties), the authors cannot put a figure on the load, but it was an "overload". In terms of the specification the material was stronger than the specification for the test coupon but less ductile. The report is not saying that "the brittle nature of the material" caused the failure in the sense that it failed below the specified strength of the material because it was brittle. Both Mr Bowes and Mr Dowdy said, and I accept, that they were satisfied that the strength of the sample tested was so high that the low ductility was not a factor in the failure
- All the experts are agreed that LAW had the best (if not the only) opportunity to examine and address the B8 failure. None of them really question their conclusions. The reference to inclusions on the fracture surface is not a reference to surface imperfections on the pinions themselves. Had surface imperfections played any part in the failure it is substantially accepted that LAW could have been expected to say so. I agree; and especially so in view of the reference to the lack of indication that the fracture originated at an inclusion on the fracture surface
- On 13 October Mr Rimlinger (Rowan) had written to Mr Cupstid (LeTourneau) on the subject of “Jacking System Discrepancies” enclosing a list of items which “must be addressed regarding the jacking system completion” adding “please be advised that the large amount of oil level items is most likely due to leaking oil sealing surfaces”. The list which was enclosed contained reference to 28 gear boxes described as “Oil level low”. The previous day the gearbox on E5 had been “raising Cain” and had been taken off line. Mr Ford was highly critical of the list doubting that those responsible for it appreciated the difficulties of checking the correct levels. Mr Niehaus said the same but also realistically acknowledged that filling the gearboxes was not a popular job. It was Mr Niehaus' evidence (which I accept) that the number of leaking gearboxes was probably about 8 to 10 rather than 28 because of the difficulties of reading the levels and the possibility that leaks could occur between jackings which would be addressed by topping up before the next jacking
- On 23 October Mr Butler reported to Mr Dowdy (among others at LeTourneau) in an internal memorandum on the “No 6 Gearbox Oil Leakage and Other items of Concern”. The memorandum included “information” that “the gearboxes have never been filled to the proper oil level prior to 15 October” and a reference to the report that 28 gearboxes were leaking with the comment that the accuracy of this information was “suspect”. The failure of E5 gearbox was found to be due to lack of lubrication. The seal was dented (as were seals on 6 other gearboxes). The “Conclusions” included a reference to the probable cause of leaking from the oil seals as installation damage and stated that “additional premature bearing and gear failures, similar to the one described in this report are likely due to the history of improper lubrication”. The “recommended repair procedure” had two provisions. First to replace the oil seals in the gearboxes that were found to be leaking with a procedure to establish which ones were leaking. The second provision was “remove motors from all gearboxes and inspect motor pinions for abnormal wear due to lack of lubrication. Rebuild each gearbox for which this wear is found ….” The first provision was carried out either by replacing the seal (in 4 cases) or the entire gearbox (in 6 cases) before the rig left Sabine. The second was not. The gearbox on E7 also failed on 23 October probably due to lack of lubrication. There were no other gearbox failures before 19 January and only one incident since in Canada in December 2000
- Mr Butler said in evidence (and I accept) that not only was his report influenced by the information from a mechanic that the gearboxes had never been filled to their correct level but that he would only have persisted with his second recommendation if it had been confirmed that all 72 gearboxes had been run with significantly low levels of oil. If it was only a small number, he would have been content with topping up and dealing with any failures as they occurred and perhaps checking with a stethoscope. He said whilst failures could be unpredictable and without warning (in the sense that the noise made by a failing gearbox would go unheard) it was not likely that more than one would fail at a time in such a way. What in fact was done, as Mr Neihaus said, was to change out all the leaking seals and inspect the motor pinions on those gearboxes and if there was any abnormal wear in them the gearbox itself was also changed out. Mr Dowdy accepted that if there was evidence of under-filling then it would have been prudent to inspect the gearboxes in question. But Mr Cupstid had investigated the matter and was satisfied that was not the case. I do accept Mr Ford’s evidence that the gearboxes were originally filled to the correct level and Mr Massey and Mr Snow's evidence that the oil levels were (or at least should have been) checked and topped up before each jacking. Mr Stone’s evidence, based on the design of the units, that the risk of severe spillage during the walk to the river (when there was a tilt of about 14() was limited to the gearboxes at the top in positions 7 and 8 on chords B, E and H and that no spillage could occur from those in positions 1 to 6 or on the other chords, was convincing and unchallenged. Thus only 6 gearboxes were even at risk from the tilt and, as Mr Stone also explained, the tilt was in any event in the “right” direction as it would take the remaining oil to where it was needed. Thus any risk of damage from low oil levels from the walk only arose in 6 gearboxes and only then after the walk was completed
- On 9 November 1998 Mr Sleightholme wrote to Mr Quinn and Mr Snow (with copies to Mr Cowie and Mr Hay) saying
"We understand that Rowan are carrying out an investigation into the gear/pinion failures which were experienced during jacking operations at Sabine Pass.
Since this is of great importance to the future reliability of the rig we would like to include it in our commissioning reports.
We would therefore appreciate a copy of the report that you receive, if you are able to release it."
- It is notable that this letter does not refer to any previous requests for a copy of any report nor complain about a failure to respond to them. Mr Sleightholme had also seen a number of jacking operations at Sabine Pass since the failures which had been carried out without a problem as the leg sections were fitted
- ADC's interim report (paragraph 122) on the Jacking System contained the following
i) Summary: "The jacking system, see comment, did cause concern during the team's early days at Sabine Pass, due to higher than expected failure rates of gears and pinions. The latest jacking operations have, however, been satisfactory. We have formally requested by memo to the Rig Manager (Mr Quinn) a copy of the report of Rowan's investigations into these failures."
ii) Action: "Inspect all pinion gears and racks for undue wear" showed Rowan's comment as "Done" and ADC's comment as "Status ex Sabine is satisfactory"; and Action: "Visually inspect storm locking mechanism. Function test at least once" with Rowan's comment "Done by ABS as part of Class Surveys" and a "tick" from ADC. (Mr Massey said he commissioned the storm locks at Sabine Pass and I have no doubt that he did).
iii) "Comment: At Sabine Pass there have been recurring problems, i.e. failure of gearbox output pinions, gearbox oil loss and bull gear failure. The oil leakage problem is currently being addressed by LeTourneau. Gear failure is also being investigated and is thought to be caused by over-stressing during the walkout at Vicksburg. This could be true as the rate of failure is now much less. The most recent jacking operations have been okay. ABS are taking a keen interest."
- Whilst Mr Sleightholme sought to say that ADC were not specialists in gear or jacking design and had no way of knowing prior to receiving any report whether there was still a serious problem or not, ADC's comments were clear and of course supplied to and not questioned by Amoco
- Mr Gillis' letter of 13 November (Paragraph 138) which enumerated Amoco's concerns about the RGV as it was about to leave Sabine Pass made no reference at all to the jacking system. Mr Gillis said that Amoco was satisfied with the fact that the rig had jacked up and down several times whilst the legs were fitted at Sabine Pass which had demonstrated that the system was adequate. He said it was not referred to in his letter because he did not expect it to develop into an issue. That reflects what I am quite satisfied was the view of everyone in any position of authority. The failures were not a concern; they were seen as commissioning problems; each failure had occurred when the pinion was potentially exposed to a massive overload; there was intelligent interest in the outcome of further researches but the events were water under the bridge. As Mr Massey put it, whilst he was not involved in the investigations into the causes of the pinion failures he was very confident in the safety of the rig or he would not have been on it at Sabine with a 400 foot air gap, 120 or so men under his command regularly jacking it. Mr Bowes said that ductility was not an issue in the design of the system nor a concern from an "application point of view" albeit Mr Benge wanted an investigation from a "scientific point of view" into the low values
- Mr Cowie wrote a "status report" on the RGV as at 15 November 1998 in which was included a note reading
"Broken Pinions. Amoco requested to be notified of the results into the inquiry into the broken pinions. LeTourneau have not yet responded with the results of the metallurgy tests on the broken pinions.
Gearbox failures. Rowan have been requested to provide information on progress regarding upgrading of gearbox seals."
- In the closing submissions Amoco sought to make much of the evidence of Mr McNease that he had assured Mr Gillis that pinion failures were not uncommon, the three failures had probably been caused by mis-wiring, and there was no question at all over the safety of the rig. On the basis that the evidence was (as it was) that three pinion failures on one rig were unusual and that Mr McNease's explanation that he was referring to gear train failures rather than specifically pinion failures was "wholly inadequate", it was submitted that this assurance was a lie. I unhesitatingly reject that submission. Mr Gillis himself did not even recall the conversation. His evidence was that the message he received from Amoco personnel at Sabine Pass or Mr Hosie was that the failures were no longer a concern and Rowan "understood what was happening and it was okay". That was the substance of everyone's accepted view at the time. It was the substance of what Mr McNease was saying. Gear trains which are overloaded will fail at the "weak" link. Mr McNease said, and I accept, that Mr Bowes had told him that in this sense the weak link in the 1000 Kip unit was the ¼ pitch pinion in contrast to the 375 Kip unit where it was the bull gear. Failure of the gear trains during construction and commissioning are on the evidence not that uncommon and I accept Mr McNease's evidence that he drew no distinction between the possible causes of failure
- Mr Yoder was one of two authors of a report dated 18 November, 1998 described as a failure analysis of Brake Heads on the RGV. The report records that during leg erection at Sabine Pass four brake heads were “determined faulty” by indication ground detector lights. The brake heads were replaced with parts from the RGVI and the failed ones returned to Longview for analysis. One was found not to be faulty. Three were found to have at least one overheated coil to the degree that the coil boxes had begun to melt. The surface of one inner magnet ring also showed signs of roughness due to rusting which could have prevented it from making positive contact with the outer ring. Mr Yoder’s conclusion was that rusting had taken place on the magnet rings because covers were unavailable at first during the construction phase and later were not always in place. If the brake head was not fully closed he said the coils would overheat especially during a prolonged continuous jacking
- In his evidence Mr Yoder himself and Mr Ford expressed doubts (with which Mr Trickey agreed) that rusting could cause the problem because the discs are likely to clean themselves when operated. Debris between the brake faces was a more probable cause. Mr Yoder also said that if you had to examine all the brake coils it would take about ½ an hour a brake to do so once it had been removed and placed on a test bench.
- A second report by LAW, dated 19 November 1998, was addressed to Mr Benge. It related to a visual examination of the failed E3 and E6 pinions. Law’s views were that
i) “The fracture surfaces of B8, E3 and E6 were similar in appearance and exhibited comparable characteristics .... Brittle fracture mode characteristics are also present in E3 and E6 pinions ...."
ii) The overall appearance of pinions B8 and E6 are nearly identical. Both possess a similar fracture surface and a torsional crack. The determined load was applied on the left side of the tooth. The appearance of the E3 pinion fracture surface is similar to the others, but is opposite in nature. The load appears to have been applied to the right side of the tooth.”
- When it is remembered that the gear trains were set in an opposed fashion with the even and odd numbers opposed to each other, the fact that E3 failed when loaded to the right side and B8 and E6 when loaded to the left side demonstrates that all three were reverse loaded. There is also no dispute that both E3 and E6 failed when mechanically overloaded above their specified tensile strength and that the failures initiated sub-surface at a location where there was no visible imperfection
- In the "Conclusion" to this report LAW stated that
"Analysis of the base material of these pinions would provide a better insight to the root cause of failure and data to determine if the pinion conforms to the ESP specification."
- LAW also faxed LeTourneau on 25 November the results of further mechanical testing of the B8 pinion. 6 sets of Charpy test results were provided all of which were below the value specified for the test coupons. The results from samples taken 1/8" from the tooth surface were 4, 7, 7 and 8, 8, 8 ft lbs. The specification for the coupon was 15 ft lbs. The results for elongation and reduction in area were also markedly below the specified coupon values. The results were indeed worse than those set out in the first LAW report (paragraph 262). On 2 and 3 December LAW provided LeTourneau with results from destructive testing of E3, E6 and the original test stand pinions. The Charpy values ranged from 2 ft lbs to 9 ft lbs. E6 had the lowest values. But the test coupon values all exceeded the specification, which was therefore complied with, and the values for the failed pinions were also of the same order as for the test stand pinions. There was, however, no discernible correlation between the average coupon values and the pinion values. Similarly the values for reduction of area and elongation of the coupons were all within the specification but the pinion values were markedly lower. The experts were agreed that the Charpy values on the pinions were low. They disagreed as to whether they were "remarkably" low and as to any significance to be attached to it. It is also, I think, the burden of the evidence that LeTourneau had expected a better correlation between the coupon and the pinion values than these figures, albeit Mr Bowes and Mr Dowdy had (rightly, as the experts agree) expected them to differ and the pinion values to be lower
- On receipt of the second Law report Mr Benge recorded (and notified Mr Bowes and Mr Dowdy) that an “in-depth evaluation” was to be carried out of cracking observed in a sample from B8 “to determine whether the cracking could have been detected when ultrasonic tested” by PSC. This was a reference to sub-surface cracks (to which alone ultrasonic testing applied) and so of very limited relevance to the present metallurgical issues. Amoco's only points about sub-surface flaws are that they may indicate the likely existence of surface flaws and that they required to be investigated at 19 January. In the event the results of the evaluation were not available until April 1999 and proved to be of no material significance. Mr Cupstid thought this further investigation was “eccentric and unnecessary” or a luxury. Mr Bowes said that after the second LAW Report the investigation was all "applied research" with which he was not concerned or involved at all. He knew in December that the flaws were not in a critical area and were smaller than the maximum size allowed by the testing. If anyone had thought this evaluation was urgent Dr Slater said it could have been completed in a "couple of days"
- The F7 Failure on 12 January occurred when Mr (Bryan) Quinn was operating the jacking system. Mr Price and Mr Jones of Amoco were on board at the time. The jacking had been twice interrupted to allow Racal to install a positioning instrument on the bow leg and to allow a wire to be released after it had become snagged. After about an hour an over-amperage warning light for F7 had come on and jacking was stopped again. Mr Quinn sent two crew members to investigate who reported that they could see nothing wrong. They were instructed to release the brake manually and disconnect the motor. There is evidence (including from Captain Broertjes), which I accept, that when that was done no one heard the torque spin out of the motor. Jacking was resumed and the pinion was immediately heard to break. On inspection it was found that one tooth of the pinion was cracked through but still engaged in the rack. It was decided to cut it out. Mr Quinn "bumped" the rig in the opposite direction to take the load off the pinion and it was only then that the torque was heard to spin out. It was for this reason when he was informed of it that Mr Quinn believed (and I accept) that the cause of failure was not (as he had at first thought) No 6 gearbox failure but that the brakes had bound either because of debris or because they had been wrongly set. In consequence when "released" the brakes had not in fact done so because of the build up of heat whereas they had cooled by the time the rig was bumped. There was no noise from the gearbox to suggest it had failed and if it had but the brake had not the torque would have spun out when the brake was first "released". F7 was not one of the gearboxes on Mr Rimlinger's low oil list (paragraph 266) nor one which could have suffered severe spillage during the walk (paragraph 268)
- BAO has been criticised because no examination was made of the brake but only the gearbox after the failure. That was perhaps unfortunate but not sinister. It would not necessarily have revealed the cause of failure which remains uncertain albeit corrosion or debris is the probable explanation. I accept Mr Quinn's evidence that a few days later he told Mr O'Brien in the presence of Mr Price that the cause was a brake problem and that, should it recur, it would be addressed in future by taking the entire brake head off if there was an over-amperage signal. Mr Quinn said, and I also accept, that Mr O'Brien was happy with this explanation and solution. The explanation was in effect repeated in Mr Marcom's letter sent on 18 January (paragraph 168 and Appendix B, 3)
- In the event two teeth of the pinion were cut out or, more accurately, burnt off by a welder. That took about 8 hours. F7 was left static and jacking was resumed. The RGV is rated to withstand 10 foot waves at the time the hull comes out of the water as it is jacked. On 12 January the seas at Halley North did reach over 10' in height but at the time when the rig had achieved a 15' air gap. Mr Quinn said he was not concerned at the sea level. Mr Gray said there was no emergency. Captain Pique said there were no safety concerns. So did Captain Broertjes. Both Mr Gray and Mr Quinn said that had it been really necessary to jack through the broken tooth then that could have been done provided the process was carefully monitored and controlled. But neither of them even thought of it at the time because there was no reason to do so and neither had ever knowingly done such a thing before. Mr Gray and Mr Quinn readily accepted jacking through would be very much a last resort and would involve the risk of further damage and unknowns but they were confident it could be managed. Having seen and heard them unshaken on this in cross-examination I am confident they were right
- It was not even put to Mr Quinn in cross-examination that if the failure of F7 had been replicated alongside the Arbroath platform rather than at Halley North there would have been any or greater concern. Mr Quinn said, again without challenge, if the RGV had been jacking up alongside the platform that would only have occurred in very good weather allowing time to cut out the teeth
- Captain Broertjes (who did not know of the previous pinion failures) said that although there was some concern on the rig as to the cause of the failure of F7 he was not at all worried that the safety of the rig had been in any way compromised nor that the failure would have prevented the rig from jacking alongside Arbroath. He pointed out that after the teeth had been burnt off the jacking system had worked without a problem at Halley North itself. Captain Pique (who had considerable experience) said
“I was not unduly concerned by the fact that one of the pinions in the rig had sheared. I have been on several rigs where pinions have broken during jacking operations. Pinion failures do occur from time to time.”
He said he was told at the time about two previous failures but that did not affect his view of the situation. It was not suggested to him in cross-examination that this view was unreasonable.
- Mr Price expressed no concern to anyone on the rig from Rowan/BAO or ADC about the F7 failure. Nor did Mr Jones. Mr Price's contemporaneous notes reflect no concern. Mr O'Brien and Mr Slater returned to the rig on 13 January. Mr Price handed over a list of his "observations/concerns". There was no mention of pinions or the jacking system. Mr Price's explanation in cross-examination that he hid his alarm so as not to create panic can hardly explain the matter of fact nature of his notes and the omission to include it in his list for Mr O'Brien. None of ADC's daily reports after 12 January mention the F7 failure
- Amoco's (Mr Gillis) fax sent on Thursday 14 January (Paragraph 162) referred to the pinion failure and the need to evaluate the risks arising from it. Mr Marcom received the fax that evening. He read it, as Mr Gillis intended, as suggesting a meeting would be arranged to conduct some kind of Hazop before jacking up the rig at the platform. The risk of a problem occurring at that time had already been addressed in June and November 1998 (paragraphs 114 and 121). The rig was being prepared for moving to Arbroath without reference to what is now said to be a compelling safety reason for not doing so. Mr Marcom did not receive Mr Denholm's letter faxed on the evening of Friday 15 January until Monday 18 January (Paragraphs 165 and 166). His reply on the jacking system was sent at 19.41 that evening. It was based on information obtained by telephone from Mr Rimlinger and Mr Quinn. In effect, unknown to Mr Marcom, by then termination was proceeding and I am quite satisfied that whatever he had said or written would have made no difference. Indeed what he did write was given scant, if any, attention. Mr Gillis was unable to explain or justify his lack of response. I should add that insofar as there was a conflict between the evidence of Mr Marcom and Mr Rimlinger on the matter I accept that at this time Mr Marcom was quite unaware and Mr Rimlinger did not tell him of the existence of any reports by LAW on the pinions
- On 18 January Mr Bowes (LeTourneau) wrote formally to Mr McNease (Rowan) about the failures of B8, E6 and E3. The letter referred to and enclosed the two LAW reports. It stated (rightly) that all the pinion teeth had sheared when reverse loaded. It expressed the conclusion that all three had sheared during the walk to the river but gone unnoticed at the time. B8 was said to have been mis-wired in a single-phase configuration and a fairly broad hint given that E3 and E6 had probably suffered from the same or reverse mis-wiring. The letter expressed LeTourneau’s opinion that the design, manufacture and installation of the system was “adequate for the normal operating and severe storm design events”.
- The letter was prepared for and delivered and discussed at the usual monthly meeting between LeTourneau and Rowan held on 19 January. Mr McNease and Mr Rimlinger were present for Rowan. Mr Bowes explained at the meeting that the B8, E3 and E6 pinions had been shown to be less ductile than expected and therefore may have failed in a more brittle manner when their yield strength was reached but the cause of failure was massive mechanical overload. He also said that it was believed that the failures had occurred or been initiated during the walk and were probably the consequence of mis-wirings. There is some uncertainty on the evidence as to when and to whom at Rowan the LAW reports were first provided. This is the first documented record. But Mr Rimlinger recalled receiving copies of the reports shortly after the rig left Sabine Pass. The probability, as I find, is that he alone did so and he alone was aware of Law's conclusions before this meeting. He was also assured by LeTourneau at the time (Mr Dowdy) that the conclusions were not a concern. Mr Dowdy told him the pinion was stronger than specified "but if it breaks it is brittle. It will not bend much, it is just going to break"
- LeTourneau's letter has been the subject of criticism by Amoco. I think, however, that it was both frank and straightforward. The reference to miswirings was, as I have held, probably right. So was a reference to the indication system acting as a warning as it had with F7. The changes made to the final version of the letter sent on 8 February did not, as Mr Bowes said, affect the substance of the letter, and were also accurate
- Amoco also submits that despite Rowan/BAO's assurances that reports on the pinion failures would be supplied to Amoco "it appears that this was never the intention". I reject that. I accept Mr Quinn's evidence that he did not have copies of any of the LAW Reports to give to anyone. Mr Rimlinger said he had not passed the first two reports on because they were incomplete and he never heard from Amoco again after the rig left America. I think, as stated, that insofar as the reports addressed the earlier failures the matter was thought to be water under the bridge and Amoco (and ADC) had in effect been told (and rightly told) that the cause was reverse overloading. That was, in my judgment the essential information together with the subsequent successful jacking operations, as Mr Marcom believed. After the failure of F7, and whatever criticisms may be made of the correspondence which followed, there was no realistic opportunity for Rowan/BAO to respond by supplying any reports before Amoco claimed to terminate the contract
- The ADC final report (paragraph 199) was prepared by Mr Sleightholme and Mr (Mel) Thomson. As Mr Sleightholme put it, he did not disagree with Mr Thomson's account recorded as ADC's "Final Comment" that "One pinion tooth broke when jacking up in North Sea, but breakage rate is acceptable," and "Copy of Rowan's investigation report has been requested". Mr Thomson's evidence was that he really knew little about the failure, it was not his area of expertise and he discussed it with no one although he (and not Mr Sleightholme) as ADC team leader was on the RGV from 12-19 January returning to it by helicopter on the 12th only shortly after the problem occurred. He said he probably should not have made the comment and Amoco's case is that it was wrong. But at the least I am quite satisfied that the comment reflected the lack of serious concern of all those on the rig at the time, including Mr Price and Mr Jones (paragraph 288). Mr O'Brien's view and my comment upon it are to be found in Paragraph 194(vi)
- LeTourneau held an internal meeting to review matters on 3 February 1999. In the course of the meeting Mr Benge’s Notes record Mr Eckerman asking Mr Benge the question
“Would you feel comfortable in being on RGV under adverse conditions with the pinions being operated?”
Mr Benge noted his own answer to this question as “No” but also an interjection from Mr Bowes that Mr Benge was unqualified to make “that decision” because he was not knowledgeable about “its redundant systems”. Mr Bowes did not remember referring to redundant systems. He did remember somewhat forcefully making the distinction between an applications engineer (himself) and a metallurgist (Mr Benge) who did not know the application, design loading or design criteria. The conclusion to the meeting recorded by Mr Benge was that LeTourneau would continue to work with PSC to determine the cause of the failure and the lack of correlation between the pinions and the test coupons. Mr Bowes said LeTourneau was not going to stop that.
- This document was one on which Amoco understandably placed considerable reliance. Mr Dowdy was not present at the meeting. He said the cause of failure was for him “open and shut” at the time: extreme overloading from a reverse direction. He also said the strength of the failed pinions was known to be adequate for proper functioning in normal operation. That was also the view of Mr Bowes
- A third report by LAW was dated 4 February 1999. It was addressed to Mr Benge. It summarised “the mechanical property data obtained from the samples removed from B8, E3, E6 and Test Stand” pinions. A draft of the report had been provided to Mr Benge on 18 January. Some of the test results had been provided earlier (see paragraph 281)
- Samples taken from the pinions had been mechanically tested to destruction by LAW. The samples were taken from 1/8 inch and 1 inch from the surface and at the centre of the tooth. The original API 8C test coupons were also taken 1 inch from the surface. Tensile strengths and Charpy V-notch tests at -40ºC were carried out on the samples. The results were also compared to the Mill Test Reports (MTR) of Pacific Steel. LAW’s “discussion of results” included the following
“…the material tested has a tensile strength which lies within a range of 122 KSI to 165 KSI. The averaged tensile strengths are higher than the Mill Test Report (MTR) with the exception of the test stand pinion. The elongation and reduction of area results for all of the pinions are substantially lower than those listed on each corresponding MTR”.
“The Charpy values are below 10 foot pounds. The shear percentage and lateral expansion values indicate that the material from which the specimens were removed is brittle at the test temperature of -40ºC. These values are consistently lower than the values listed on the MTRs.”
The Charpy results obtained from the test stand pinions were however similar to those from B8, E3 and E6.
- Mr Bowes said (and I accept) that LAW did not provide yield strength results for many of the samples despite LeTourneau's request because they were concerned that the high yield strengths and low reductions of area observed meant that their strain gauges might be damaged by destructive testing. An updated version of Mr Bowes 18 January letter (paragraph 290) was sent to Rowan (Mr McNease) ABS and DNV on 8 February enclosing all three LAW reports
- FTI Anamet, a testing laboratory, reported to PSC on the fracture face of the E6 tooth on 26 February. The clear conclusion was that failure was due to “gross overloading”. The report is, I think, ambiguous as to whether any casting flaws were seen on the surface of the tooth as opposed to the surface of the crack but clear both that there were internal casting flaws and that flaws (surface and internal) had “nil involvement” in the failure of the tooth. E6 was the tooth with the lowest (or “worst”) Charpy results. PSC sent this report to LeTourneau on 1 March
- In February Mr Mobbs went to the RGV in Dundee to carry out an inspection of the ½ pitch pinions to assess the extent of hub wear. He found a number of oil leaks at seals. He had reported some leaking seals in August 1998 which he said had been replaced although I think there is some real doubt whether they or at least all of them were. On 6 January a maintenance check had shown some 14 seals to be leaking and 5 thrust plates to be worn. Mr Mobbs' conclusions were set out in a report dated 9 February. 13 units were found to have 3/16 inch or more wear and to require “High Priority” hub and seal replacements. Of these 13, 4 were sufficiently worn for Mr Mobbs to recommend that “jacking be limited to minimal only” until they had been repaired. 8 units had ½ inch or more wear and were to be replaced as “Next Priority” and 7 were found to require only seal replacement
- In his evidence Mr Mobbs said that his caution to keep jacking “to minimal” until the four had been replaced was only because he did not want to cause further wear by jacking when, because the rig was stacked at Dundee, it was not necessary to do so. But had the rig been going to Arbroath the state of wear would not have inhibited or affected jacking in any way. He said the wear he saw would still have allowed another 20 to 30 hours (or 2 to 3 years) of normal jacking without a problem. I accept this evidence which was not substantially challenged. The problem of hub wear led to a minor redesign of the secondary geartrains on the RGVI and RGVII. The ½ pitch pinion seals, hubs and thrust plates were changed out by LeTourneau mechanics between March and July 1999
- LeTourneau and PSC continued at a leisurely pace to investigate and debate the reasons why test pieces removed from the four failed pinions had low ductility values. No satisfying answer emerged. Mr Dowdy said, however, that what was very clear from all the investigations was that PSC had produced pinions of very high tensile and yield strengths which were more than strong enough to perform their function and had only failed due to an extreme overload condition. None had failed because of a lack of toughness or ductility. Mr Neely of PSC wrote to Mr Dowdy on 9 July to that effect. LAW reported on F7 in September 1999 and their findings were essentially the same as for B8, E3 and E6
- Both ABS and DNV were informed of the results of Law's tests. Neither required any further testing of the pinions of any kind (metallurgical or otherwise) nor did they revise or withdraw the approvals they had given for the RGV. As stated (paragraph 202) DNV did seek confirmation of the "original theory" that the failures were due to overloading and not a manufacturing defect and were content when they received it in the form of Mr Neely's 9 July letter
Jacking History
- Mr Rimlinger calculated that between June 1998 and 12 January 1999 the system was jacked for a total of 9,124 feet for the great majority of which (8,343 feet) it was loaded with the full weight of the hull. In normal operations over a year, assuming 5 rig moves, the rig would be jacked only some 600 feet when loaded with the full weight of the hull. That gives an idea of the "proving" of the system prior to termination of the contract. The reason for this degree of jacking was the procedure adopted for fitting the leg sections to the rig (called "legging up"). The rig was jacked down to pick up each leg section, jacked up again so that section could be fitted to the top of the existing section, and jacked down again and so on section by section, leg by leg. That would of course be done with the system bearing the full weight of the rig and the equipment on it at the time about 600 Kips per pinion
- In early September 1998 the rig was jacked out of the water at Sabine Pass with partial preload which imposed pinion loads of about 800 Kips. On 10 September a static preload imposed loads of about 1100 Kips per pinion
- In November 1998, as part of some tests to check load sharing (of no relevance to the issues) the rig was jacked with preload on the starboard leg. Mr Bowes calculated that the load imposed on each pinion was 1260 Kips and the rig was jacked through one complete revolution. Thus each of the 7 teeth on each of the 24 pinions on the starboard leg was successfully loaded to that extent
- Although both Mr Trickey and Mr Hogg questioned the accuracy of the readings which were taken at the time, there is no doubt that in October 1999 in the course of a wet tow from Dundee to Sable Island (Nova Scotia) in the Atlantic Ocean the RGV experienced a severe storm when the legs were partially lowered with the tips of the spud cans over 150 feet below the keel. The jacking system was subjected to high pinion loads. Mr Dowdy calculated maximum loadings in excess of 1500 Kips on 34 pinions, 1600 Kips on 10 pinions and 1700 Kips on 4 pinions. Whilst those precise figures have to be treated with considerable caution, there is no real dispute that significant loadings would have been experienced and were withstood. The RGV sought shelter at Falmouth
- It is the unchallenged evidence of Mr Quinn that since the RGV left Dundee it first went to work on 29 March 2000 offshore Halifax, Nova Scotia, Canada. No unusual problems have occurred. Downtime has been limited. As at the time (13 March 2001) he gave evidence the RGV had been jacked up and down a total of 22 times at Dundee, Falmouth and at Halifax. On 9 occasions the jacking was next to live platforms. These operations have been carried out without incident or pinion failures, save that on one occasion in December 2000 an over amperage light came on for E1 as the legs were lowered on approaching a platform. That was successfully dealt with in about an hour by disabling the gearbox and completing the move to the platform. During the 9 static preload operations between May 1999 and December 2000 the average pinion load per leg exceeded 1200 Kips with maximum loads in excess of 1400 Kips on some pinions. Of course Amoco is right to submit that jacking undertaken since 19 January is irrelevant to how the jacking system should have been viewed at that date. But it does help to establish that if it was then judged to be fit for its purpose that judgment was right
Pinion Loads
- It was Mr Hogg’s calculation that the maximum vertical load on the pinions with the StormLOKs engaged which would be experienced at Arbroath in a 50 year storm was 1124 Kips. That evidence was unchallenged. LeTourneau’s calculation (made by Mr Mobbs) in December 1998 was 1327 Kips, later reduced to 1289 Kips (paragraph 230). Mr Hogg and Mr Trickey agreed that in the 50 year storm condition at Arbroath loads would exceed 1800 Kips unless the StormLOKs were engaged. But they could be engaged well before such loadings were experienced (paragraph 230). There was also agreement that to justify a design loading of 1800 Kips a measure of excess capacity is required of the order of 15 to 20% so that tests would, for example, have to establish a capacity up to 2115 Kips to justify the design loading of 1800 Kips. Mr Hogg’s calculations of maximum loadings in “Drilling Condition 1” were also lower than LeTourneau’s: 1444 Kips against 1842 Kips. Again those calculations were not really and certainly not convincingly challenged. Indeed Mr Hogg said they contained an element of conservatism by making no allowance for foundation “fixity” (resistance to leg bending moments at the foot of the rig) which Mr Trickey accepted would reduce the loadings further by some 10%, that is from 1444 Kips to 1300 Kips
Expert Opinions
- As will be apparent from Appendix D no less than 11 experts gave evidence on issues specific to the Jacking System. In addition, Mr Bartle and Mr Poss addressed the evidence from the standpoint of how a “reasonable and prudent operator” should have responded, and evidence was given by witnesses from both LeTourneau and PSC which involved considerable expertise. Captains Mallett and May considered the response of the Rig Mover and Warranty Surveyor to the facts
- To a great extent the volume of evidence resulted from Amoco’s developing case that the pinions were not or at least could reasonably have been thought not to be fit for their purpose because they were or might be liable to give way under loads they should have been able to sustain. That case whilst derived from the fact that there were 4 failures of pinion teeth (as well as the other matters referred to in the claim) was substantially based on what I shall refer to overall as the metallurgical issues. It is to those issues to which I shall first turn
- There are a number of specific expertises involved. Each expert’s opinion was to a considerable extent dependent on the views of others. The target was or at least became to establish the minimum size of surface flaw on a pinion which would create a risk of a tooth failing under a given loading and to address the question whether the inspection (MPI) carried out by the manufacturer (PSC) would have captured and removed a flaw of the deduced size. Fracture mechanics seeks to assess the critical size of flaw and stress which could lead to failure; the metallurgists address the properties of the pinions and the reasons for the known failures; the expert on MPI considers the process of detection of flaws and the “probability of detection” (POD) experts address just that: how probable is it that a flaw of a given size will be detected by MPI. Expertise on pinion loads addresses the question whether a given loading is probable at Arbroath and has already been summarised.
- I confess at once that after hearing the factual evidence and as an untutored mind I approached the materiality of these issues with a degree of scepticism. The skills involved are ones with which I was wholly unfamiliar and potentially a less than enthusiastic student. But the skills of the experts in explaining their subjects were such that not only did they make the subject interesting but one where the issues could be approached with reasonable confidence. However the result was that it served to confirm my scepticism. The thesis, admittedly stated crudely, advanced by Amoco in the reports on which it relies was that a surface crack at the most vulnerable point on a pinion of only 1.5mm in depth might not be detected and could lead to a tooth yielding under a load of 1300 Kips. After further analysis that figure became 1636 Kips at -6o C (the "maximum safe loading"). Both experts assumed a depth to length crack ratio of 1:5 for a crack of this depth
- I have seen a model of the pinions. I have seen one of the broken teeth. I have set out the testing required by ABS and carried out by PSC (paragraphs 234 to 236). The governing property in the design was strength (paragraph 237). The pinions have in fact been extensively used both before and after the failures (paragraphs 305 to 309). The 4 incidents of ¼ pitch pinion failures are ones in respect of two of which there is no dispute that the failures were caused by massive overloads. That means they did not fail at anything like 1300 or 1636 Kips but at a load well in excess of 1800 Kips. The overwhelming probability is that the same occurred with the other two failures; certainly there is no dispute that they too failed when reverse loaded. As for B8 I have considered the evidence in paragraphs 262 to 265. As for F7 Professor Miller’s evidence from examining photographs of the failed pinion was that it suffered a load in the region of 2600 Kips before failure. That evidence, granted it came late, was really unchallenged and certainly unshaken. Moreover there is nothing on my findings to suggest that a surface fault played any part in any of the failures. In the case of F7 there was nothing of the surface left to examine. The mode of failure was brittle fracture as all the experts would expect. But the cause of failure was overload and had nothing to do with surface cracks. Those findings mean that despite the low Charpy values the pinions were indeed stronger than specified despite and notwithstanding the low values or brittle character
- It struck me as extraordinary to suggest that an undetected crack of 1.5mm (let alone 1.2mm) in depth (and 7.5mm in length) on the surface of one of these massive pinions could cause it to fracture as Amoco’s metallurgical evidence suggested. Much more importantly Professor Miller said he had had just the same reaction when first asked to address the problem: “when I saw these dimensions quoted of 1.2 and 1.5mm my engineering judgment said ‘This cannot be true’”. Once Professor Miller had concluded his evidence, which, despite very skilful cross-examination, was something of a tour de force, I would doubt if anyone who heard it (and all the related evidence) could have failed to agree with him
Fracture Mechanics
- The fracture mechanics ‘contest’ was between Dr Manteghi and Professor Miller. In fact the contest was a somewhat unequal match. It needs to be borne in mind that reports and evidence were being produced and developed on these issues during the trial and seen in that context many of Amoco's criticisms of Professor Miller's evidence are I think wholly misplaced. Both experts changed and developed their opinions. There was a fundamental difference of approach between the two. Dr Manteghi believed that the low toughness and ductility of the pinions raised a question about their susceptibility to brittle fracture which required a fracture mechanics analysis. Professor Miller considered the correct approach was to analyse the cause of the actual failures which alone gave rise to any possible question about the adequacy of the pinions. He also considered that the prototype tests were far more valuable than theoretical analysis: theory is approximate, the experiment tells you the truth. It may be wondered how experts can produce figures of such precision as a flaw of 1 or 1.5mm depth. However there is no doubt that they can, but only by making choices about flaw shapes (gape), sizes and locations and crack and stress orientations. Moreover as the evidence proceeded it became apparent that the raw data which formed a vital part of the calculation (the other 'known' is the maximum principal stresses at the tooth root which is calculated by finite element analysis) was itself open to very considerable question. When it is explained how they are assessed (paragraph 250), the criticisms of Dr Slater and Professor Miller of Charpy values generally and the values actually available for the pinions were in my judgment well made. To describe it as a “primitive” measure is, if anything, an understatement. It explains why taking a single Charpy value rather that an average of a number of values is not good practice. That is not to doubt (least of all in the centenary year for Charpy) that the use of Charpy values in fracture mechanics calculations is an accepted and reasonable practice. It is. BS 7910 provides for it. But the limits of it must be acknowledged. The values need to go through a process of extrapolation and correlation to obtain a fracture toughness value. ABS did not require Charpy values. Yet it was the only raw data which was available for the calculations. Fracture toughness data obtained by direct measurement did not exist because it formed no part of the design or testing of the pinions and was not carried out for the purposes of these proceedings. The specification related to the coupons and it was known the pinions would have different and lower values
- Dr Manteghi maintained in his first report that giving the benefit of every possible doubt to LeTourneau “the situation was clear cut that these things (the pinions) with the definition of fitness for service that I had adopted were unfit for service.” The definition of fitness for service he had adopted was a flaw depth of 1mm or less and 5 to 10 mm in length because he did not believe that such flaws could be reliably detected by MPI and his calculations showed that at a loading of 1800 Kips and a temperature of -40ºC the maximum tolerable flaw size was less than that depth and length and at the same loading but a temperature of -6ºC (the lowest temperature likely to be experienced at Arbroath) the maximum tolerable flaw size was 1.2 to 1.4mm depth and 7.2 to 11.6mm long which were a size which also probably could not be detected.
- In his supplemental report Dr Manteghi revised the figures but still concluded that the maximum safe load at minus 6ºC was 1636 Kips assuming that a flaw with a depth of less than 1.5mm could not be reliably detected by MPI. Yet the only pinions ever to fail did not fail until they were loaded to an extent in excess of design and vastly in excess of Dr Manteghi’s figures
- I do not think it necessary to examine and set out the detailed analysis which led to Dr Manteghi’s results. Professor Miller’s analysis, also using established techniques, produced numbers which were of a different order to Dr Manteghi’s and, if reliable, established that the pinions were well able to withstand loads up to and in excess of specification and far greater than anything which could be expected to be experienced at Arbroath. Professor Miller's conclusion was that if the flaw was 1.5mm deep the tooth could withstand a loading at -6ºC of 1810 or 1831 Kips even without allowing for what he called the ‘U’ factor. But he also considered that it was wrong to ignore the ‘U’ factor which was an attempt to represent residual compressive stresses in the tooth and (less significantly) the “late crack opening effect” caused by the stresses in and extent of the constraint exercised by the material surrounding the notional crack and its tip which was being assumed for the analysis. Dr Slater was wholly convincing in explaining how the “mini-quench”, (a short water-cooling after the tempering cycle), the last process in the heat treatment of the ¼ pitch pinion, would create significant compressive stresses both at the surface and at a crack tip only 1.5mm below the surface. It is Professor Miller’s opinion that these factors have a marked effect in reducing the risk of fracture and must be accounted for. If they were his "maximum safe loading" rose to a range of 2410 to 2693 Kips
- I think it sufficient to state my overall conclusion on this evidence
i) The calculations involved are subject to a number of uncertainties and assumptions which can have a very marked effect on the outcome. There remains room for a considerable degree of judgment.
ii) There is no dispute that Dr Manteghi and Professor Miller both addressed what may be called the worst case in terms of the location (on the highest line of stress) and orientation (perpendicular to maximum principal stress) and shape (sharp tipped) of the postulated crack.
iii) Despite his claims to the contrary, I do not accept that Dr Manteghi has, even within the parameter of the “worst case” been in any way favourable to LeTourneau. In part that is because on matters with which other experts have dealt I prefer the views of BAO’s experts to those of Amoco’s experts on which Dr Manteghi understandably relied. But it is also because I much prefer the views of Dr Slater and Professor Miller which had a refreshing and compelling reality about them and because I do not think Dr Manteghi was able convincingly to maintain his position. In particular in rejecting any ‘U’ factor and in his selection of a Charpy value for use in his calculations I think Dr Manteghi inevitably produced a result which was extremely conservative.
iv) Even the result Dr Manteghi did finally produce was such as gave the pinions an ability to withstand a margin of 20% in excess of any loading (paragraph 310) likely to be experienced at Arbroath (1636 Kips against 1300 Kips) for a crack 1.5mm in depth. For a crack 1mm in depth Dr Manteghi's figure for maximum safe loading was 1992 Kips. Professor Miller's comparable figures are 1831 (1.5mm) and 2243 (1mm) Kips without any 'U' factor and at least 2410 and 2951 Kips if a 'U' factor is included.
Metallurgy
- The respective experts were Dr Beardsley and Dr Slater. There was some measure of agreement between them. Essentially they were agreed that the actual failures were the result of reverse loading and were initiated sub-surface. They were agreed that the pinions were shown to have lower ductility than LeTourneau had expected. It was agreed that the lowest average toughness found was on E6 and near the point of failure. But there is no dispute that E6 failed only when overloaded beyond its design rating and not at a surface flaw
- Although she acknowledged that to an extent her conclusions took into account the views of Dr Manteghi and Dr Rudlin, Dr Beardsley concluded her first report with the statement
“I do not believe the pinions remaining on Gorilla V to be suitable for North Sea service as a consequence of both the poor properties, the likely presence of significant imperfections and the uncertain sources and magnitudes of loadings. These factors individually and in combination mean that their integrity cannot be relied upon”.
- In her Supplemental report this opinion appeared to be somewhat softened: “From a metallurgical viewpoint, I believe that the information regarding the properties and quality of the remaining pinions available at January 1999 meant that they could not be considered suitable for service without further evaluation….” But in evidence Dr Beardsley made clear that her opinion remained that she would not have any confidence that the pinions were fit for their purpose nor that further failures would not occur. Essentially that opinion was founded on what she described as “the remarkably low Charpy values" and susceptibility to imperfections whilst acknowledging that she (unlike the fracture mechanics experts) was unable to quantify the effect of those values. Impact loading, Dr Beardsley agreed in evidence, was not a concern in practice
- I should record that in my judgment this view is one of such an extreme nature (more extreme even than any other of Amoco’s experts) that it causes me serious concern not about Dr Beardsley’s objectivity or expertise (which were apparent) but as to the application of that expertise to objectively established facts. Dr Beardsley did acknowledge that her opinion, although expressed in broad terms, was limited to a metallurgical viewpoint. Nonetheless I think it represents a reaction to and academic analysis of the undoubted low toughness and ductility figures leading to an unsustainable overstatement of the reality of the risk addressed
- When subjected to analysis Dr Beardsley’s concerns were two related points: the “remarkably low” ductility and the variability of the properties of the pinions compared to the PSC specification. But it was agreed that if castings of the size of the pinions failed they will always fail in a brittle manner. Whether the properties Dr Beardsley was concerned about would contribute to a failure in the sense of cause a pinion to fail below its rated value is, as she acknowledged, dependent on the existence of a surface defect or imperfection in the pinion of a critical size to initiate a brittle fracture. That, and the likelihood of such a defect surviving PSC’s inspection processes, is, as she also acknowledged, a matter for the fracture mechanics and MPI/POD experts. And if the evidence of those experts showed that no critically sized defect was present then the metallurgical issue was only whether the pinions had adequate tensile strength which, it is agreed, they had. All the calculations of tensile strength show that the specification was in fact exceeded. Dr Beardsley expressly agreed in cross-examination that there would be no concerns about low toughness and ductility “if you know there are no critical defects there”. Thus her opinions were far more dependent on those of other experts and the conditions likely to be found in the North Sea than at times she appeared prepared to acknowledge. She accepted that what she meant by saying that she could not be confident that there would not be another pinion failure was that she could not be 100 percent certain no such failure would occur. That is a criterion which does not accord with ALARP. Nor, on the evidence, did any pinion ever fail below design loading because of a surface (or any other) imperfection. It also became apparent in the course of her cross-examination that Dr Beardsley had misread and so overstated the, in any event very limited, evidence that there were any surface flaws on the pinions
- Dr Slater considered that “the concept of ductility on a large component” like the pinions “really does not mean very much”. He pointed out that while the LeTourneau specification referred to Charpy values, in fact (and as was common practice at the time) no fracture mechanics analysis had been used in the design of the pinions and thus the specified values were not related to any such analysis. Manufacturers rely on the testing they carry out. The Charpy values are to ensure that good heat treatment is carried out. In the same way the manufacturer relies on non-destructive testing techniques such as MPI to detect any flaws in the material. Whilst agreeing that the Charpy values in the LAW Reports were low he said they “must not be considered in the abstract. The data must basically be considered in the available universe of information about these pinions”. That makes very good sense and in my judgment it is something Amoco’s experts (with the exception to some extent of Mr Trickey) failed to do or at least demonstrate
- Dr Slater accepted that the test block or coupon results for ductility and Charpy values did not reflect the results obtained on testing the pinions themselves (which were consistently low) and indeed that there was no clear correlation between the test block values and the values on a pinion. But what the test block values were showing was that if they were achieved (as they were) the strength data on the pinions was in excess of specification. When Mr Unwin put to him Amoco’s case that the four pinion failures coupled with the low values made it appropriate to investigate or evaluate the pinions further Dr Slater disagreed. He said
“The only information that is available is that we have some fractured pinions, we cannot find any flaws on the fracture surface. We know that these pinions have basically been reverse loaded. We know that the available load under those circumstances is huge. Therefore the obvious and reasonable assessment … is that these pinions failed under large overload, and the properties of the material, as far as fracture toughness interacting with flaws, was not an issue. You also then go to the various test stand data, where we have precisely the same type of Charpy values from the test stand pinions as we have from the pinions that have fractured and again we find that these pinions have performed – I hate to say it – but flawlessly. So under those circumstances it would seem to me that the information that you have to make a decision is precisely the opposite of what you are trying to put forward.
In other words there is no information to show that you should be going into a flaw-based evaluation, but rather that all the information you have is that aside from this aspect that the Charpy values are lower than the specified minima, everything else basically tells you that there is not a problem with these pinions; that what we have here are some one-off failures that have nothing to do with the way that these pinions are going to be used in service …. I think that is an entirely reasonable situation that they would say “These pinions are fine”.
- Dr Slater’s opinion was in my judgment really unanswerable. Certainly he was not in any way shaken in cross-examination. Mr Bowes' evidence to the same effect was equally impressive, adding that the inspection threshold for flaw sizes by PSC was for such a very small size that low ductility was of no concern at all
Magnetic Particle Inspection
- In the event, Dr Rudlin was the only expert to give evidence on the methodology and reliability of MPI as such. But the criterion of “reliability” and the approach to its assessment was based on Probability of Detection Curves. In that expertise the contest was between Dr Rudlin and Dr Wall. Again, in my judgment, the evidence and opinions of Dr Wall were very much to be preferred. He was restrained, objective and most impressive. His report convincingly demolished much of Dr Rudlin's analysis. It was Dr Rudlin’s conclusion that only surface breaking flaws in the ¼ pitch pinions of 1.5mm depth or greater would be reliably detected by MPI. Dr Wall believed flaws of a depth of 0.5 to 1mm and length of 1.5 to 3mm would be reliably detected
- The criterion addressed by both experts for “reliable” detection was “90% probability of detection at 95% confidence level”. That means more than 90 out of 100 flaws of a given dimension are detected 95% of the time or such a flaw has a 90% chance of being detected. It is, Dr Wall said, a very conservative approach. The practical problem with the methodology is that it requires at least 29 flaws per size group to give statistical confidence because the criterion is met (as a matter of arithmetic) on detection of 29 out of 29 defects or 45 out of 46. In cross-examination and through the evidence of Dr Wall I think the lack of data and the resulting arbitrary nature of some of Dr Rudlin’s calculations was readily exposed. For example, Dr Rudlin’s deployment of the data dictated an answer in excess of 1mm. He could not sustain his position that repetition of MPI made no statistical difference. Commonsense suggested and Dr Wall explained otherwise. The relevant international standards (both American and British) refer to flaws with major dimensions of 1/16 inch or 1.58mm (paragraph 241). It is not in issue that this is or includes a reference to the length of the flaw and not to depth (an MPI operator is unlikely to be able to see the depth of a flaw). Nor is it in issue that the aspect ratio of depth to length for such a flaw is at least 1 to 2 if not 1 to 5 (which was Dr Rudlin’s figure). Thus the depth of a flaw of 1.5mm in length would be 0.75mm or less. Equally a flaw of a depth of 1.5mm would have a length of 3 to 7.5mm. Yet the standards (based on many years of experience) assume and the evidence is that flaws of much less than those dimensions can be reliably detected by MPI. Nor is it in issue that the shape of the pinions makes an MPI inspection easy and that the PSC (Mr Costa) inspection was done to a very high standard by a highly trained and experienced employee (Paragraphs 254 and 255)
- Dr Wall was understandably concerned to stress that he believed it was appropriate to give a range of figures because of the sensitivities and uncertainties involved. But when asked by Mr Unwin he said he certainly believed the relevant depth was below 1mm but to go significantly below 0.5mm would be difficult. In re-examination, when pressed by Mr Siberry, he said “we could reasonably come down to a level of 0.6 to 0.7mm” and having demonstrated that the key benefit of repeat inspection was in the detection of very small defect sizes even below that level
- I am quite satisfied on the evidence that Dr Wall’s figures are to be preferred. It follows, of course, that the conclusions of Dr Manteghi’s fracture mechanics calculations proceeded on a false basis and produced a result which in any event considerably under-estimated the load-bearing capacity of the pinions. Indeed if Dr Wall's upper figure of 1mm was applied even Dr Manteghi's calculations show the pinions met their specified load ratings [paragraphs 225 and 321(iv)] albeit with only a 10% safety margin and those ratings were much greater than the loads calculated for Arbroath (paragraph 310)
Marine Operations
- There was a fair measure of agreement between Captain Mallett and Captain May. A Marine Warranty Surveyor is there for loss prevention and reports to underwriters. He will assess the condition and operability of the jacking system before approving any rig move and it is his duty to certify that the rig can operate on the proposed location and withstand the environmental forces imposed upon it. The two experts were agreed that the F7 pinion failure did not place the rig at risk because there was ample time to cut away the damaged pinion and resume jacking at Halley North before the weather deteriorated. The only circumstances in which the rig or the Arbroath platform might have been placed at risk would be if the jacking system was immobilised while the rig was alongside the platform in the transitional mode between floating (when it could if necessary be moved clear by the use of the moorings and tugs) and elevation above wave height and the weather and sea conditions deteriorated rapidly
- At full elevation the risk of jacking system failure would be insignificant assuming no metallurgical or engineering doubt about its capacity. Even in the case of immobilisation during the transitional stage any risk would only arise if the weather window required before any move alongside the platform would be permitted were to shorten unexpectedly. Weather forecasts for periods longer than 24 hours should be treated with caution and confidence in the accuracy of the last day of a 3 day forecast would at best be moderate
- The experts were also agreed that whilst “taken alone” the failure of F7 “might not be considered significant” “awareness of the previous failures would have created a need for an explanation before attempting to install the rig at Arbroath”. If the explanation involved engineering or metallurgical issues then they agreed suitable advice would be sought. They also agreed that the system was powerful enough to jack through an obstruction but such a course would only be adopted as a last resort. Where they disagreed was as to whether the information available on 19 January 1999 would have caused a Marine Warranty Surveyor to withhold approval for the move to the Arbroath platform without further investigation
- Captain Mallett’s thesis was that on the failure of F7 he would have expected to be told of the earlier pinion failures which would have triggered an investigation, including a metallurgical investigation, of much the same sort as has been carried out for this trial. That was because it was his opinion that “there was a higher than usual risk that the jacking system might fail”
- This opinion was based substantially on what Captain Mallett considered to be the surprising frequency of the failures, the failure to ascribe them to clearly identified causes and the “doubtful material quality” said to have been exposed by the LAW reports. He would not accept that commissioning failures on a new rig were different from operational failures, nor that LeTourneau’s view that the failures were by reverse overloading and the pinions safe would be sufficient to satisfy a Warranty Surveyor. Nor that ADC were satisfied. Nor that Amoco were not unduly concerned. Nor that Captain Pique, who was of course on board when F7 failed, said he would have been prepared to approve the final move to Arbroath without further investigation or testing. I prefer the views of Captains Pique (and others who were present) and May.
- Captain May said that the explanation for the F7 failure, which occurred when the leg was being lowered (not lifting the weight of the hull), had to be a gearbox or brake failure; the earlier failures were in the commissioning phase; if he had learnt of a metallurgical investigation he would have sought advice about it. He would have taken the opportunity to inspect the system at Halley North; you could “pop” the brakes and test the electrical circuits. That would in any event be done before any jacking
- If Captain May had heard about the gearboxes possibly running on low oil levels, he would question the crew and perhaps seek further assurances from LeTourneau or Rowan. Only if that did not satisfy him would he turn to his engineering department, in effect seeking the advice of a Mr Trickey or a Mr Stone
- As Captain May pointed out not only was the rig jacked up at Halley North after the two teeth of F7 were burnt off, but before arriving at the Arbroath platform the RGV would again be jacked down at Halley North, and jacked up again at the standby location. That, if, assuming as I think it would have been, it had been carried out without a problem, would have given Captain May considerable additional confidence in the system and he said he would certainly have given his approval to jacking up the rig alongside the platform in those circumstances. The positioning operation would normally be expected to be completed in 18 to 24 hours from the stand-by position. Full preloading would, itself, on the evidence, take some 18 hours. Regular weather updates would be obtained and monitored. Forecasts are available 24 hours a day. If the weather was to deteriorate or be forecast to do so there are stages at which the operation can be aborted. The approach takes some 3 hours. The legs are jacked down before preload begins. Preload can also be dumped and the rig jacked down. Full preload may not in any event be required for the expected conditions. Indeed at Arbroath even for a once-in-50-year storm only some 55 to 60% preload is required to ride out the storm. The existing and forecast weather conditions required for positioning next to a platform require winds less than 20 knots, seas of 5 feet or less and no significant swell. These are "better" than required for jacking at a stand-by or other location such as Halley North: Paragraph 285
- There are also other safeguards. If there is a gearbox failure or brake sticking it is likely to cause an over-amperage warning light to come on. It did before the smoking on B8 and the failure of F7 and on the occasion in December 2000 in Canada. There is uncertainty about E3 and E6, but the evidence suggests the devices may not have been calibrated at the time they failed. I do think by January 1999 the evidence is that a warning would probably have been given. The motor and brakehead can be disengaged to allow a ¼ pinion to freewheel (possibly even if it is damaged). That takes about an hour. The procedure was established following the F7 failure itself. The gearbox can readily be isolated from the gear train by cutting out a number of the small teeth on the one pitch pinion. That also takes about an hour. To cut out a tooth on a ¼ pitch pinion takes about 4 hours. That would only be necessary if it was jammed. Otherwise it could be left to freewheel. In the last resort a controlled jacking through could be attempted but as Mr Quinn, Mr Trickey and Mr Stone said the other available options meant there would be no need to consider it. It is very difficult to imagine how with all these possibilities the weather could deteriorate so rapidly and unexpectedly to give rise to a problem. In refusing to accept them, I think Captain Mallett was carrying theorising and improbability beyond the limits to which it can reasonably be taken. His view was an extreme one, particularly so in the context of my other findings, and one which I think would not have been and was not in fact adopted by anyone concerned at the time.
- The other issue which Captain Mallett and Captain May addressed was the likely waiting on weather time which could arise from delivery of the rig in December rather than October. It is convenient to comment on the issue here, but the question is only of real relevance if there was (which I have held there was not) a contractual obligation to mobilise the rig for delivery in October or possibly if (which it would not have done: paragraph 145) Amoco would have taken advantage of an earlier weather window had the rig arrived in Rotterdam fully prepared to drill. It is agreed that weather conditions in the North Sea generally turn for the worse from mid-September. The weather is also generally at its worst in December and January. But it became apparent in the course of his cross-examination that despite the terms of Captain Mallett’s report there were in fact no suitable weather windows in October 1998 in which it would have been acceptable to seek to position the RGV at Arbroath. If the rig had been ready to leave Rotterdam on 15 October 1998 (the end of the "anticipated" arrival window) it would have been 29 days before the move could in fact have begun. Captain Mallett’s criterion for the final approach to Arbroath was a full 24 hour forecast (Captain May preferred 36 hours). Captain Mallett accepted that there was a suitable weather window for the final approach between 9 and 12 January which the RGV could have used had it been able to leave Rotterdam in time to do so early on 8 January and it was agreed that provided the rig had been ready to leave Rotterdam no later than early morning of 16 December it could have done so albeit it would still have had to wait for 9 January before it could finally move to the platform as the weather would only have allowed it to reach the standby location. There is also a dispute between Captain Mallett and Captain May as to whether if the contract commenced on 25 December the weather conditions would in fact have allowed the RGV to leave Rotterdam in time to take advantage of the weather window between 9 and 12 January. Captain Mallett believed they were such that the rig could only have begun the move at about midnight on 8 January which would have allowed time possibly to get to the stand-by location but not alongside the platform. Captain May had much greater experience of rig movements from Rotterdam and in the North Sea. It was his evidence that the usual waiting time in the winter months (mid-October to mid-March) was 2-3 weeks to complete a rig move. It was also his evidence that the actual weather conditions would have allowed the RGV to leave Rotterdam both on 7 January and any time after 1800 hours on 8 January. I think, because of his greater knowledge (in particular of where the RGV was in Rotterdam and the conditions in which it would be safe to leave) Captain May’s evidence is to be preferred. Arbroath is some 350 miles from Rotterdam and the RGV would take a little less than 60 hours to get there. The weather conditions were such that departure on 7 January would have left sufficient time to go alongside the platform. Departure at or shortly after 1800 hours on 8 January would only have enabled the rig to reach the standby location at Arbroath. Thus the actual weather conditions in 1998 and 1999 would probably have resulted in less waiting on weather time for an arrival in mid-December than in mid-October and if delivery took place on 25 December the waiting time would probably only have been some 13 days. At the very best for Amoco, therefore, the actual weather conditions at the time do not assist the company’s case that delay caused it any loss or was of significance
Overview
- The experts who expressed an overview of the issues relating to the Jacking System were engineers experienced in offshore units and jacking systems. Both, Mr Trickey and Mr Stone, were impressive and knowledgeable witnesses.
- They agreed that B8, E3, E6 and F7 all failed when reverse loaded such as could occur from a mis-wired motor, failure of a brake to release or failure of any significant component in the gear train. They also agreed that the metallurgical properties of the pinions would be relevant to assessing the fitness for purpose of the pinions. Where they differed was in the need for and degree of investigation required by the low toughness and ductility of the pinions, the low oil levels found in the No 6 gearboxes, the leaks in the secondary gear train, the condition of the brakes and the damage found on one occasion to the H chord on a rack. In general terms Mr Trickey considered those matters gave rise to serious concerns which required investigation but Mr Stone did not.
- Inevitably and reasonably, Mr Trickey relied on the views of Drs Beardsley, Manteghi and Rudlin in reaching some of the conclusions he did. In his last report, dated 2 April 2001, relying on the latest reports of those experts, Mr Trickey took Dr Manteghi’s figure of 1636 Kips as the assumed load at which the pinions were at risk of failure. Allowing a suitable margin his opinion was that the maximum allowable load should be de-rated to 80% of that figure namely 1309 Kips. On the basis that LeTourneau had calculated the maximum pinion loads at Arbroath with StormLOKs engaged to be 1300 Kips (Mr Hogg’s figure is of course lower) Mr Trickey then revised his earlier reports as to the requirements he considered appropriate before jacking the RGV at the Arbroath platform. Those requirements were (or, so far as material, included)
i) A complete metallurgical investigation sufficient to achieve the figures presented by Dr Manteghi or figures sufficient to show that the pinions could meet the requirements for operations at Arbroath.
ii) An investigation into why F7 failed.
iii) A visual and MPI inspection of the roots of the rack on chord F in the area where F7 failed.
iv) “All brakes to be opened, examined and where necessary, cleaned and reset".
v) Inspection of all primary gearboxes not previously inspected.
vi) Opening of secondary geartrains for inspection on B8, E3, E6 and F7 and the side of rack H in order to check they were not damaged.
vii) Survey (as Mr Mobbs did in February 1999) of the ½ pitch pinions to investigate oil leaks, seal damage, shaft play and hub wear (paragraphs 301 to 302).
viii) Testing the StormLOKs and ensuring the crew understood how to fit them.
ix) Replacement of F7 and repair of any damage found to the primary gearboxes.
x) Performance of a jack-up test by undertaking a jacking cycle of 50ft with a load of 593 Kips a pinion followed by a preload jacking test raising the hull by 10ft with a load equivalent to 1200 Kips per pinion and then lowering the hull with a load of 1300 Kips per pinion (the difference is because of a different friction value between lifting and lowering the rig).
- It was Mr Trickey’s estimate that granted availability of equipment and personnel and whatever time might be required to move the rig to Dundee and to agree the scope of work, those requirements would take approximately 6 weeks from the time that the work was put in hand
- Mr Trickey agreed in cross-examination that he believed that the RGV would probably have passed his jack-up test. He also of course agreed without question that if Drs Beardsley, Manteghi and Rudlin were wrong then his figures would have to be re-calculated and that his view that the RGV's loading capacity had to be de-rated was dependant on their views. By way of illustration, if the minimum reliably detectable depth of crack size was changed to 0.75mm (not Dr Rudlin’s 1.5mm) but otherwise Drs Beardsley and Manteghi were right the failure load of a pinion would be 2314 Kips or 1851 Kips at 80%. The comparable figure for a crack of 1mm depth are 1992 Kips and 1594 Kips at 80% [paragraph 321(iv)]. If, of course, Professor Miller’s calculations were used the answer would be even more favourable to BAO
- In cross-examination, Mr Trickey accepted, as again he had to as his evidence followed much of the metallurgical evidence, that it was “very unlikely” the failure of B8 had anything to do with the properties of the pinion. He also agreed that E3 and E6 failed when overloaded in the sense of loaded above design. The logic of those opinions is, as stated, that probably all three (and certainly E3 and E6) pinions in fact performed better in loading strength terms than specified and their Charpy values had no effect on their performance
- The F7 failure, Mr Trickey agreed, was caused either by a locked or dragging brake or a faulty gearbox. The gearbox was found to be locked after the event but that could have been either an effect of the failure or its cause. He also agreed that if the torque did not spin out on release of the brake then a brake defect was the cause. Again it follows, as Professor Miller said, that the cause of the pinion failure was in all probability that it was overloaded. Although Mr Trickey was reluctant to accept it I think it follows that there was no basis at all in any of the four failures for thinking that there was any problem with the material properties of the pinions. If anything, the reverse was true. Indeed if it had not been for the LAW reports I doubt if Amoco or this court would ever have heard about the metallurgical issues. The thrust of Mr Trickey’s evidence on those issues was that he would have been concerned once he was aware of the low Charpy values (especially on E6) on the failed pinions, which were lower than the design criterion (albeit that related to the coupons all of which were within specification). That he said would have “horrified” him and led him to seek the advice of an outside expert
- Mr Trickey fairly accepted that he should have referred in his reports to the degree of comfort or confidence which could be derived from the jacking operations in Sabine and was also unaware even when he gave evidence of the “satisfactory” report by ADC on the jacking system (paragraph 294)
- When asked whether he now accepted that, even on his analysis, it would be reasonable to ignore metallurgical investigation and go straight to his preload test Mr Trickey still thought not as he would need some guidance on the maximum limit to which he could test. But that, as stated, was dependent on the views of Drs Beardsley, Manteghi and Rudlin which I have rejected
- The other matters on which Mr Trickey relied (Paragraph 346) can be addressed shortly
Item (ii) The reason for the failure of F7 was, on the evidence, a brake defect which could have been avoided if Mr Quinn had been told that the torque did not spin out when the brakes were "released" (see Paragraph 283). The existing information at 19 January readily justified that conclusion.
Item (iii) MPI inspection of F chord. There was no evidence of any rack damage caused by any of the pinion failures. They were each inspected at the time. Mr Stone said an MPI inspection would only be required if a visual inspection suggested the rack had been damaged. In any event Mr Trickey agreed such an inspection would only take about an hour.
Item (iv) Brakes The evidence of brake failures is to be found only in Mr Yoder’s report in November 1998 (paragraph 276) and the circumstances of the F7 pinion failure (paragraphs 283 to 284). That is a very thin basis for Mr Trickey’s opinion. Mr Stone said that the brake failure rate was a reasonable one “and almost to be expected” bearing in mind how hard the brakes had been worked during the legging-up process. The basic design was no different from the 375 Kip units which had an established successful performance record. It is also a standard routine to "pop" the brakes to test them before any jacking. In any event to check the brakes would only involve seeing that the coil was undamaged and the magnetic rings collapsed together. A megger test on each leg would prove the electrical circuits and establish that the coils and motors were good. Only if there was a low reading would it be necessary to inspect further.
Item (v) (See Paragraphs 266 to 267). Only 2 primary gearboxes were changed out before LeTourneau delivered the rig to Rowan, E5 and E7, both of which were replaced because they were making a loud noise during jacking. There is some evidence both were affected by lack of lubrication. Gears could also fail because of bad bearings, blown seals or debris inside them. Mr Trickey fairly accepted that Mr Rimlinger’s list of 28 gearboxes said to have been found with low oil levels could have been substantially overstated due to a misunderstanding as to the correct oil level. He also agreed that low levels could have been the result of leaking output pinion seals, and, provided it was done thoroughly, the problem would have been addressed by the check on oil levels which was carried out at Sabine before every jacking operation. Mr Trickey was not prepared to accept that the response to the supposed low oil levels (see Paragraph 268) was sufficient to avoid the need to inspect all the gearboxes or at least a sufficient number (50%) to satisfy the inspector there was no problem. In part that seemed to be based on his opinion that the F7 failure could have been caused by a gearbox problem (which I think most unlikely on the evidence). Again all the gearboxes had worked without a problem after mid-October from when there is no dispute that they were regularly topped up. Mr Stone said, and I accept, that if there had been a problem it would have been manifest in that time and indeed almost at once. The extent of the jacking involved in the "legging up" has already been stated. Whilst Mr Trickey is no doubt right that a gearbox may deteriorate gradually the evidence also is that in such a case it will usually make a discernible noise to give warning before it does fail. It may also cause an over-amperage warning. This was an instance in which I think Mr Stone was plainly right that LeTourneau’s response at the time was appropriate and sufficient and Mr Trickey failed to satisfy me that his approach was a reasonable one.
Item (vi) Secondary Gearboxes The evidence shows that B8, E3, E6 and F7 were inspected at the time of the pinion failures. As to the evidence of damage found to an H rack tooth in June 1998 during the walk to the river, in the light of Mr Dowdy’s unchallenged evidence that it was decided in discussions which included ABS that it was so minor as not to require repair and the subsequent jacking I found Mr Trickey’s continuing reluctance completely to accept he should delete it from his requirements unfortunate. The point was rightly not even pursued with Mr Stone in cross-examination.
Item (vii) ½ pitch pinions Mr Trickey’s point was that Mr Mobbs investigation should have been carried out before and not after the rig went to Arbroath. Whilst I accept Mr Trickey’s approach as a reasonable one ADC did not follow it, Amoco did not ask for it and I also accept that it was rightly not seen to be a matter which required urgent attention at the time or which created any risks to going on location or to the operational capacity of the rig. Mr Stone agreed that “in an ideal situation” LeTourneau would have dealt with the problem before delivery of the rig but the oil was always topped up before jacking and no secondary gear train ever failed. Even the worst wear reported by Mr Mobbs (5/8 inch) was such that a failure would only occur after a further inch of wear. Mr Mobbs inspection was completed in one day.
(viii) StormLOKs Mr Massey said, and I accept, that the StormLOKs were commissioned; ADC acknowledged that: see Paragraph 271(ii). ABS witnessed it and were satisfied. The StormLOKS were not complicated. The evidence is that instructions for their use were present on board. It would take at most two to three hours for the crew to practice fitting them and that could have been done at any location.
(ix) Replacement of F7 was, on the evidence, unnecessary in view of the considerable redundancy in the system. As Mr Stone said, it could simply be taken off line.
- Mr Trickey’s estimate of how long it would take to carry out his requirements was based on the metallurgical investigation being the critical path item which might take up to 6 weeks and the preload test would then follow. The rest of the work could be done at the same time and “could probably be done in a couple of weeks”. Mr Trickey acknowledged that the estimate for the metallurgical investigation was a matter better left to others. Dr Manteghi thought an analysis such as the one he had done would take some 2½ to 3 weeks but actual destructive fracture toughness tests would take 6 weeks (the source of Mr Trickey’s estimate). Mr Bowes said fracture toughness tests could have been done in about 2½ weeks; Professor Miller thought 12, 24 hour days
- Mr Stone did not consider any of Mr Trickey’s requirements were necessary before jacking up proceeded at Arbroath. He also said that if it had been thought necessary to test the pinions it could have been done in stages by using the weight of the hull (that is without or with only very little preload) and testing half the pinions on each leg at a time which would have involved lifting loads of 1200 Kips. Any safety concerns could be met by the ability to engage the other half of the pinions at the push of a button. That he said could be done without going into a safe harbour and would take some 5 to 7 days
- The basis of Mr Stone’s opinion (accepting that he too would have left any metallurgical issues to the advice of others) was that he considered the three pinion failures at Sabine Pass were “pretty much explained” and not that unusual for a shipyard that had been mothballed for some three years where the labour force was dealing with the first opposed gear train. It was his belief that B8 was caused by mis-wiring and either the same applied to E3 and E6 or it was caused by a brake failure but one which was momentary as it only lasted for part of a rotation damaging one tooth and nothing else. A pinion takes 3.78 minutes to complete a rotation. B8, he thought was no mystery. It ultimately failed because the brake was re-set without being wired up again, hence the torque was heard by Mr Snow to spin out when the brake was released, and, without power, the warning light would not function. The key point for Mr Stone was that the explanation for each failure had to be human error. Mis-wirings would inevitably cease to be a problem as they would become apparent on first operation. The brakes were very simple devices and again would be proved by “popping” them as a long established routine every time before jacking. As for F7 Mr Stone said something plainly was dragging and he explained how it was that with the rig lowering the leg the motor should be “regenerating” and the over-amperage light would not come on until in effect whatever is causing it to drag also causes the motor to go to forward power level and to exceed the over-current level. That could result (as happened on the evidence) in the over-amperage warning light operating only at the last stage. But his opinion was that the immediate cause of failure was human error in not then ensuring that the brake was released and because no one informed Mr Quinn that when they thought they had released it the torque had not spun out. He also said that on examination of the brake, provided the coils are intact, it is improbable that the cause of it dragging would be revealed. A coil failure would cause the earth warning signal to operate, which it did not. In his opinion there was nothing in the history of events, leaving aside only metallurgical issues, which indicated a need for any jacking test as he did not think there should have been any question with regard to the strength of the pinions
Operator Issues
- The expert role of Mr Bartle and Mr Poss in the proceedings was to report on the approach that a reasonable and prudent Operator of a rig would have taken given the condition of the RGV at 19 January 1999
- To a very great extent therefore their opinions were dependant on the other evidence both factual and, particularly in the case of Mr Bartle, expert
- The very limited measure of agreement between them and their divergent views are apparent from a Joint Memorandum dated 18 January 2001. As regards the Jacking System they disagreed whether there was any risk of immobilisation of the system but agreed that if that potential reasonably existed then they would have sought specialist advice. They agreed that the response from BAO in the fax letter dated 18 January (Paragraph 168) regarding the pinion failures “was inadequate for Amoco to make an informed decision regarding the safety of jacking the rig alongside the Arbroath platform”. Mr Poss added that it was his opinion that “Amoco’s move to terminate without further investigation of the matter was wholly inadequate and not as a reasonable operator would have done.
- The essence of the different view on the totality of the issues was captured in the last two paragraphs of the Joint Memorandum
Mr Poss believed that the [RGV] would have been sufficiently ready on 19 January to move alongside the Arbroath platform and for operations to commence.
Mr Bartle continued to believe that the operability and safety of the jacking system and the well control system were questionable and the rig floor equipment had not been completely commissioned at contract termination. He believed that the option of manual operation was unacceptable for both operational and safety reasons”.
- Whilst the expert role which he was fulfilling may to some extent account for it, in the course of his evidence Mr Bartle was, I think, frequently prepared to act more as advocate than expert. He also appeared to have difficulty in discounting complaints about the rig floor equipment which had been abandoned and in appreciating that he was addressing the conduct of a reasonable operator which was not necessarily the same as an operator who was seeking grounds to terminate or re-negotiate the contract. I regret to say that I found much of his evidence tendentious and unbalanced. Mr Poss, on the other hand, was both reasoned and objective and quite willing to criticise BAO when he considered it right to do so
- Mr Bartle readily acknowledged that his views on the jacking system were largely based on the reports of Mr Trickey, Dr Beardsley, Dr Manteghi and Captain Mallett. He had expressed those views before reading any of BAO’s expert reports. He said when he did read those reports he did not find it necessary to change any of his opinions
- Mr Bartle's conclusion in his report was that
"Due to the condition of the jacking system on 19 January, and the fact that insufficient information had been provided regarding the pinion failures and that no other information was forthcoming that other problems had been dealt with, I would not have approved the move of the drilling unit alongside the production platform. I would not have agreed to the rig move until I was fully confident that failures, in particular those of the 1/4 pitch pinions, were unlikely to re-occur and this risk was as low as reasonably practicable.
- Amoco places considerable reliance on Mr Poss' agreement in cross-examination to this conclusion of Mr Bartle. However the real difference between the two experts (as Mr Poss' evidence made clear) was as to what would or should have been required to provide the necessary confidence. Mr Poss (like Mr Stone) considered the information available at the time to be sufficient; Mr Bartle, like Mr Trickey did not. Amoco also submit that BAO's expressed willingness (had the contract not been terminated) to provide information should be "viewed with the greatest scepticism". I do not agree. I see nothing in the history of events to support a case that BAO ever sought to keep any information from Amoco (or ADC). Amoco and ADC had representatives on board the RGV throughout the events. Their reports demonstrate that they were aware of all material events and discussed them with Rowan/BAO. Mr McDonald had an entirely open and frank relationship with Amoco onshore. It is, as I have stated, in my judgment quite wrong to characterise the correspondence which immediately preceded termination as supporting Amoco's case. To the contrary
- When questioned about Mr Marcom’s fax letter of 18 January (paragraphs 168 and 289) Mr Bartle said that the essence of the complaint about its contents was that it gave insufficient information to enable an operator to make a decision “either way”. When challenged that the logic of that response had to be to seek further information, which of course Amoco did not do and Mr Poss said a reasonable operator would have done, Mr Bartle’s first response was in effect to say that BAO appeared to be shutting the door by denying there was any reason for concern, a matter of fact which derives no justification from the documents and which I reject on the evidence. The reality, as I find it, was that the boot was on the other foot. Amoco shut the door and had no intention of seeking any further information. Amoco decided that day to terminate the contract and did so the next day without giving any real consideration to Mr Marcom’s letter. Moreover there was no expression of concern by those on board at the time, Mr Jones, Mr Price or ADC. The extent of Mr O’Brien’s concerns are apparent from his reports (paragraphs 189 to 194). Mr Bartle would not have been satisfied by LeTourneau’s report to Rowan on 19 January on the earlier failures (paragraphs 291 to 292). He would not have been satisfied without Mr Trickey’s postulated investigations because the precise cause of the failures had not been identified
- Mr Poss said that most operators would not even know that a pinion had failed but agreed the 12 January failure would cause concern in the context of the earlier failures and that Amoco was right and justified in seeking further information. He would have expected the concern to be raised orally at once (not by a letter dated 2 days later) and he would have expected to have seen all the information available to LeTourneau. There should have been a meeting to achieve a joint evaluation of any problem and to seek to satisfy an operator it was safe to proceed. Whilst Mr Poss did not resile from his view that Mr Marcom’s fax letter of 18 January did not provide sufficient information for an operator to be satisfied, he did say the situation was not one which would normally give rise to letter writing
- I think it is quite apparent in the context in which the letter exchanges were taking place that Amoco was not in fact in the least concerned to seek further information or interested in making "an informed decision" about the jacking system but was rather maintaining a case on which it had embarked before 12 January on the basis that the 15 January was a key date for the operation of Clause 28 of the contract [paragraphs 158(ii) and (iii)]. It would hardly be surprising if that was to affect the nature of any response. Even so, and despite the views of both Mr Bartle and Mr Poss, I do not find the 18 January letter either uninformative or unhelpful and I unhesitatingly accept that Mr Marcom wrote what he genuinely believed to be the views of Rowan/BAO on the causes of the pinion failures. The key information (confirming "advises previously given to Amoco's representatives on board") was that the three failures at Sabine were in effect due to "overstress" (E3 and E6) and the other to "a wiring fault", and that the F7 failure was due to a brake fault for which a procedure had been established to prevent any recurrence. The letter also referred to the jacking system being satisfactorily load tested and certified. All those statements, as I find, were accurate. The blind were not leading the blind in providing this information as Amoco submitted in its closing submissions. Moreover the substance of the information was the same as provided by LeTourneau to Rowan: paragraph 290. LeTourneau's opinion was to be respected. Had Amoco been concerned to seek information it would have pursued the matter. Indeed it would have learnt that ADC and Mr O'Brien were content to proceed, assuming Amoco was not in fact aware of that at the time. The same was true of the Complex OIM (Mark Jones) and Captain Piqué. There was also no basis or justification for assuming that the 18 January letter was or could reasonably be taken to be BAO's last word on the matter
- Mr Poss said that the probability was that had a reasonable operator sought further information it would have been satisfied by LeTourneau’s advice justified as he thought it was by all the testing of the system and the explanation of the pinion failures. Mr Poss “seriously doubted” that anything more including any fracture mechanics analysis would ever have been sought or done. On all the evidence I have now heard I have no doubt at all he was right about that and I think there was more than sufficient information available at 19 January for the same judgment to have been properly and soundly reached then, and a reasonable operator would have taken that view. In any event even any lingering doubts could and on my findings would have been readily dispelled by any of the physical preload tests suggested by the experts. Moreover if, as I think on the evidence would have been the case, such a test would have been done in a sheltered area (such as Dundee) it would still probably have been achieved in time for the rig to meet a suitable weather window between 30 January and 2 February to return to Arbroath and go to work
Conclusions
- Amoco's case on the facts fails. At 19 January, with a minimum of further enquiry, it would readily and rightly have been concluded that the jacking system met its specification and so and by definition was more than fit for use at Arbroath. The essential information was that the four pinion failures were explained by human error leading to reverse overloading, the system had more than proved itself and a procedure had been established to address the immediate circumstances (brake binding) which caused the F7 failure. The metallurgical properties of the pinions were such that they were stronger but more brittle than the values specified for the API test coupons. That should have been of no concern
BOPS/WELL CONTROL SYSTEM
Description
- A very basic description of the well control equipment is to be found in Paragraphs 15 to 17. The BOP Control System (or Koomey Unit or Accumulator Unit), is designed to provide hydraulic power generation, storage of pressurised fluid, and distribution of the fluid to the various functions of the BOP stack and diverter system, together with a remote control capability from the driller's panel and the toolpusher's panel as well as from the control manifold itself. It provides a volume of fluid sufficient to perform the various operations on stored fluid only (i.e. without the pumps operating) so that the well can still be shut in even in the event of a loss of power
- The key components of the entire system and their component or item (on a schematic) numbers, where relevant, are as follows
- The Hydraulic Power Unit ("HPU") Tank is a stainless steel hydraulic fluid reservoir tank containing the volume of fluid (3600 litres) that is to be pressurised and stored. Connected to the tank by short lengths of piping are two pumps (10.1 and 10.2) each of which is capable of supplying enough fluid for the system. The pumps are fitted with stainless steel filters on both the suction (between the tank and the pumps) and discharge sides (11.1 and 11.2, 12.1 and 12.2 respectively) and automatic pressure switches to stop and start the pumps on use (very approximately) of about 10% of the stored pressure. The HPU tank is located in an equipment room on the mezzanine deck, some 200 feet from the drill floor. The filters on the discharge side of the pumps (12.1 and 12.2) are referred to as the main or in-line filters. They were designed to remove even very fine material of 20 microns or more from the fluid (a micron is one millionth of a metre: the diameter of a hair is about 70 microns)
- The Accumulator Bank consists of a bank of 56 54-litre cylindrical pressure storage vessels ("accumulators") for the 3000 psi fluid generated by the HPU pumps. Cylindrical rubber bladders in the accumulators are filled with pressurised nitrogen gas which is compressed by the hydraulic fluid pumped into the accumulators. When a function is operated the high pressure fluid is forced out by the pressurised bladders. The accumulators are situated in the same equipment room as the HPU tank a few feet away from it.
- The Hydraulic control manifold (or "main control panel") distributes the control fluid from the accumulators and pumps via supply lines to the Valve Module (situated under the drill floor close to the BOP stack) and on to the functions of the BOP and also via other lines directly to the choke and kill valves and the valves on the diverter. The connections to the Valve Module come up through the main deck above the control manifold and first terminate at the Deck Cluster. These connections are in hard or "shipboard" piping. There are 53 lines of hard piping each between 45 and 49 feet long and having 5 90° turns. From the Deck Cluster to the Valve Module and on to the various functions the connections are made by flexible hosing running on trolleys up and down the side of the cantilever section of the rig (known as "the catenary"). The hosing to the Valve Module is about 300 feet long. The control unit also provides hydraulic command signals ("pilot" signals) to the Valve Module. The operation of a pilot line from the control unit opens (or closes) the chosen valve so that pressurised hydraulic fluid from the supply line is directed to the relevant function
- The 3000 psi available from the pumps and accumulators can and will usually be adjusted to a lower pressure to operate the BOP stack and diverter system. Normal system pressure to the Ram BOPs is 1500 psi and for the annular preventer 500 - 1500 psi. The adjustment is made by regulators (32.1 and 32.2 and 33.1, 33.2 and 33.3). The regulators are themselves controlled from the BOP control unit by increase/decrease manifold blocks (item 66 for the Annular and BOP stack and item 67 for the diverter). The pilot supply for both items 66 and 67 comes via a 50 micron pilot in-line filter (item 15/16)
- Item 66 contained two valves, one leading to the annular regulator (Item 32.1) and on to the main annular supply line and the other to the manifold regulator (Item 32.2) and on to the main supply line to the rams. Item 32.1 regulates the pressure applied to the annular via the annular preventer control line; item 32.2 regulates the pressure applied to the BOP rams via the control lines to each pair of Rams. Between item 66 and items 32.1 and 32.2 there are throttle valves which can be used to regulate the flow and "mini" accumulator bottles pre-charged both to provide immediate pressure and to control the impact of any sudden increases in pressure. The throttle valves were items 62.1 and 62.2, the mini accumulators were item 40.1 for the annular (intended to be pre-set at 25 bar) and item 40.2 for the BOP rams (intended to be pre-set at 55 bar). 25 bar is about 360 psi. 55 bar is about 750 psi. The purpose of the throttle valves is somewhat uncertain. Indeed Mr Ellis was not familiar with their use on such systems, said it was not easy to tell what position (or setting) they were in and did not see the need for them. He suggested they were there just to fine tune the system. Mr Marymee thought they might be intended for use to avoid or control sudden increases in pressure
- Item 67 leads to the regulators for the diverter system. There are 3 regulators: item 33.1 for the flowline open/close valve, item 33.2 for the overshot packer open/close valve and item 33.3 the regulator for the diverter packer itself. The configuration was the same as item 66. There were throttle valves (items 62.3, 62.4 and 62.5) and 3 mini-accumulators (items 40.3, 40.4 and 40.5) intended to be pre-charged respectively to 10 bar, 10 bar and 20 bar
- The manifold regulator (32.2) system can be by-passed via the manifold by-pass (item 73.1). If the by-pass is in operation full accumulator pressure can be applied directly to the selected BOP function. The purpose is to enable instant full pressure to be applied in an emergency or as a fail-safe in the event of a fault with the manifold regulator
- The functions which had dedicated control lines running to them direct from the Control Panel included the Choke/Kill valves, the Diverter System Overboard Valve, Flowline Valve and the 4" Block Valve. These lines were formed of ½" steel piping with ¼" bore and terminated at the controlled function. The 4" Block valve itself is so situated that access to it to hook it up to the hoses can only be achieved by use of a riding belt or cradle to reach the diverter housing. It was in fact not required at all for drilling at Arbroath. There was also a separate pilot line to operate the Shear Assist by opening a valve (item 76). The purpose of the Shear Assist is to guarantee sufficient pressure from a separate bank of accumulators to ensure that the blind shear ram will shear any pipe in use.
The Contract
- The contractual provisions on which Amoco relies are Clauses 3.1. (paragraphs 35 and 36); 6.1.2 (paragraphs 59 and 60), and 6.1.5 (paragraphs 62 to 65) which itself refers to Appendix 6. The introduction to Appendix 6 to the Contract (Appendix A p 62) refers to "the well control acceptance standards" to be used by Amoco "to determine if the Drilling Unit is acceptable to Amoco" for the drilling programme. It also refers to "additional checks that will be made by Amoco well control equipment auditors" and to the possibility that the audit might be completed by independent specialists "including the close out of all corrective action requests raised from the audit"
- One feature of the various tests which are set out in Appendix 6 which needs to be highlighted here is to be found in section 19.0 which refers to "BOPs & Control Systems Function Tests" and requires a function test to be conducted at normal 1500 psi operating pressure "to record function times and fluid volumes for each function, maximum permissible response times will be as per API RP 53 -API RP 16e" (my underlining)
- "API" stands for the American Petroleum Institute. "RP" is short for "Recommended Practices". The Third Edition of API RP 53 sets out (at March 1997) the recommended practices for BOP Equipment Systems. The First Edition of API RP 16E (October 1990) sets out the recommended practices for design of well control systems. Recommended Practices are expressed to be "guides" but they are also said to be "based on extensive and wide ranging industry experience", and to reflect "proven, sound engineering and operating practices"
- Paragraphs 12.3.2 and 12.3.3 of API RP 53 provided that
"12.3.2. BOP systems should have sufficient usable hydraulic fluid volume (with pumps inoperative) to close one annular-type preventer, all ram-type preventers from a full open position, and open one HCR valve against zero wellbore pressure. After closing one annular preventer, all ram-type preventers and opening one HCR valve, the remaining pressure shall be 200 psi ...or more above the minimum recommended precharge pressure. (An HCR Valve is a choke or kill valve).
12.3.3. Accumulator Response Time. Response Time between activation and complete operation of a function is based on BOP or valve closure and seal off. For surface installations, the BOP control system should be capable of closing each ram BOP within 30 seconds. Closing time should not exceed 30 seconds for annular BOPs smaller than 18¾ inches ... nominal bore and 45 seconds for annular preventers of 18¾ inches nominal bore and larger. Response time for choke and kill valves (either open or close) should not exceed the minimum observed ram close response time. Measurement of closing response time begins at pushing the button ... to operate the function and ends when the BOP or valve is closed effecting a seal. A BOP is considered closed when the regulated operating pressure has recovered to its nominal setting ...."
- API RP 16E was in the same terms as those quoted from Paragraph 12.3.3. The annular preventer on the RGV was larger than 18¾ inches nominal bore
- API RP 53 also provided for pressure testing of the equipment. Paragraph 17.3.6 referred to Pressure Gauges
"Pressure gauges and chart recorders should be used and all testing results recorded. Pressure measurements should be made at not less than 25 percent nor more than 75 percent of the full pressure span of the gauge"
- 17.3.7 provided that
"The results of all BOP equipment pressure and function tests shall be documented and include, as a minimum, the testing sequence, the low and high test pressures, the duration of each test, and the results of the respective component tests.
Pressure tests shall be performed with a pressure chart recorder or equivalent data acquisition system and signed by pump operator, contractor's tool pusher, and operating company representative ...."
- Amoco does not allege in respect of the well control equipment that there were breaches of Clause 3.1 as regards conformity with all relevant laws nor Clause 11.7.1 (compliance with the Safety Case or Safety Management System) nor any of the provisions of Appendix 7A (Appendix A at 7A page 70 and following)
Amoco's Case
- The allegations made in paragraph 4A of the Claim are as follows
"(1) The well control equipment did not meet the requirements of Appendix 6, for the reasons set out in paragraphs (2)-(4A) below.
(2) The blow-out prevention ("BOP") accumulator unit contained contaminated hydraulic fluid affected by construction debris and corrosion. By the Termination Date:
(a) Although some of the accumulator unit components had been replaced, the unit as a whole was still not operational, because
(i) there remained a risk of additional component failure because of the contaminated fluid, construction debris and corrosion which had previously been demonstrated to have affected the operability and reliability of the unit. The BOP and accumulator unit therefore needed to be stripped down for (at least) partial inspection to ascertain the extent of further damage. Such component failure did in fact occur subsequently, as pleaded at paragraph 4(4A) below;
(ii) the in-line filters had not been replaced and the flow meters were not functional.
(b) The BOP had still not been fully pressure tested. In particular,
(i) the BOP system as a whole had not been tested;
(ii) the relevant certificates had not been supplied to the Plaintiff;
(iii) the rams had not been low pressure tested;
(iv) the valves on the BOP had not been tested;
(v) the replacement choke and kill lines had not been tested in situ.
(2A) As at 15 December 1998 the BOP had not yet been function tested so as to achieve the closing times required by Appendix 6 and by API RP 53, and it was a requirement of DNV's and ABS's interim certificate that such tests had to be successfully completed. On 24 December 1998 a function test of the BOP was carried out which failed to meet the specified closing times. Nevertheless by letters dated 24 December 1998 the Defendant (a) informed DNV and ABS that the BOP had been "satisfactorily tested as per applicable API standards 16E and DNV MODU Rules" when in fact it had not met the relevant standards, and (b) confirmed to the Plaintiff that all certification requirements had been fully satisfied when in fact (because of the failure of the test on that date ...) they had not been.
(2B) (a) The design and/or construction of the hydraulic and electrical catenaries was defective such that when skidding the cantilever the hoses to the BOP were likely to be kinked or crushed and/or the electrical control cables damaged. This defect was not rectified by 19th January 1999.
(b) As a result of the defect in sub-paragraph (a) above a number of the armoured hoses to the BOP were kinked or crushed and were in a condition that compromised the reliability of the well control system and was a safety hazard to personnel. The damaged hoses were neither pressure tested nor replaced by 19th January 1999.
(3) the BOP accumulator unit did not have an independent back up power supply, as required by the specification in Appendix 6.
(4) The diverter had not been fully assembled and tested.
(4A) Cameron Controls (the manufacturers and suppliers of the BOP) continued to carry out rectification and/or commissioning work to the BOP systems and/or their controls on 1 - 3 February 1999, 12 - 13 April 1999, 20 April 1999, 22 - 23 July 1999, 26 - 28 July 1999 and 5 August 1999. This included work on the valves and manifolds which had been contaminated by the debris pleaded in paragraph 4(2)(a) above and on the remote flow meters referred to in paragraph 4(2)(a)(ii). The remote flow meter on the driller's panel was not made functional until 23 July 1999. Cameron Controls carried out a full commissioning procedure on the BOP for the first time on 5 August 1999."
The Evidence
- Once the system had been built, Cameron (Houston) carried out its own factory acceptance tests. ABS witnessed the tests. The tests took place on 24 October. ABS confirmed acceptance subject only to points which are not suggested to be significant. The interim Class certificates had noted that the function and timing tests for the system were outstanding: see paragraph 140
- The system was also commissioned by Cameron (Houston) at Sabine Pass between 4 and 12 November 1998, but a small number of components needed to be replaced to complete the commissioning and Mr Rimlinger arranged with Cameron for them to be installed when the RGV arrived in Rotterdam.
- Mr Leaman (Cameron) went out to the rig in Rotterdam arriving on 19 December and departing on Christmas Eve. His Service Report (counter-signed by Mr Snow) is a full record of the work he did. It included, on 20 December, his finding a loose Allen screw in the manifold regulator (item 32.2) and, on 21 December, seal debris as well as "a lot of debris" on the filters items 12.1 and 12.2. "Due to large amounts of contamination in system" he arranged to filter the control fluid in the HPU tank through a home-made flushing rig leaving it running all night on 22 December. Mr Leaman referred to "a brown-coloured scum" in the tank sight glass with some particles "finer than sand" in it. Mr Leaman also serviced and re-fitted the regulator
- Mr Leaman believed the system was in working order by 23 December. But when he tested it on the morning of 24 December the annular would not close. The reason was found to be that the control valve operated by the pilot line had an internal snap ring which had popped out of its grooving jamming the piston internally. The "close" pilot line to the control valve was also found to have a pellet of welding slag in it at the deck cluster. Once the snap ring was rectified the annular worked. Removal of the pellet had not had that result. I should note that this account of the event reflects the evidence (which I accept) of Mr Leaman, Mr (Bryan) Quinn and Mr Snow. Mr (Mel) Thomson's evidence was materially different including timing the event to 23 not 24 December. I think Mr Thomson's evidence unreliable and probably connected to the events later on 24 December: see paragraphs 410 and 411
- ADC's Interim Report on the BOPs had noted in respect of low pressure tests that they were "outstanding". Rowan's comment was "in progress". The annular tests and fluid function tests of the control unit were shown by ADC as "function test only" and "final function test in Rotterdam" respectively. Both, however, were ticked and marked up as "satisfactory 24/12/98" by Mr Sleightholme on a copy of the report. The references to "record opening/closing times" and "record volume for opening/closing" and “record time for opening/closing from regulator pressure gauge on the unit" were also marked by Mr Sleightholme with a tick and the date 27/12/98, "in hand 26/12/98", and also a tick and the date 27/12/98 respectively
- There are two extant documents which contain records of times and pressures of tests carried out on 24 December both of which are included in Appendix E. The first is signed by Mr (Bryan) Quinn and Mr Snow. As Mr Quinn put it, and I do not doubt, he recorded what he saw. The times recorded did not meet those set out in the API test in 3 respects. The Annular closing time was 52 seconds (API 45 seconds) and the middle and lower rams closing times were 35 seconds (API 30 seconds). The end pressure did match the API provision of 1200 psi after one round of closings and openings. The second record is in Mr Sheen's writing. It contains the same figures set out in a different (and in fact mistaken) form. The document itself was included in ADC's interim acceptance book. It also has noted on it (as does the first document) different times of 45 seconds to open and 32 seconds to close for the middle rams.
- ADC prepared a Daily Report for 24 December. The report on "Hydraulics/BOP" carries Mr Sheen’s name but Mr Sleightholme thought he had written it on the basis of what he understood Mr Sheen to have told him. It recorded (by reference to the Sections in ADC's Interim Report)
"Section 3 - Deficiencies additional to Interim Report.
The fault on the BOP control unit hydraulics has been found to be a loose spool in a valve. This has been repaired by Cameron and the system now works satisfactorily.
Section 4 - "R" items demonstrated by Rowan.
The BOP has been functioned from the main unit and all stations. Functioning times are satisfactory. Accumulator test carried out. Checked for air or fluid leaks throughout the system - all satisfactory."
- Mr Walker (Amoco) who was also on the rig wrote in his report of 24 December: "... perform functions tests. Closing times satisfactory and ADC do not foresee problems with BOP Acceptance." Mr Walker said he got his information from Mr Sheen. Mr Leaman said everyone was happy with the tests otherwise he would not have been allowed to leave the rig as he did. Mr Welton said the same
- It was the tests carried out on 24 December which led both to the issue of the ABS and DNV Class Certificate for the RGV dated that day and to BAO's (Mr McDonald) letter to Amoco of the same date confirming that all certification requirements had been fully satisfied and that charges were to commence under the contract (see paragraphs 147 to 150). As stated there, the circumstances in which ABS and DNV expressed satisfaction with the BOP and Accumulator tests have been the subject of criticism by Amoco, based on the undoubted fact that the timings achieved on 24 December did not meet the stated API requirements. Amoco does not submit that these events have any particular consequence as regards the claims but does rely on them to allege that Mr Haggard and Mr McNease were guilty of deception of ABS and DNV and so are matters material to their credit
- In the afternoon of 24 December Mr Haggard was telephoned by Mr Quinn from the rig and told that everyone including ADC and Cameron was satisfied with the tests. Mr Haggard said Mr Quinn read out the results to him and he wrote them down. Mr Haggard telephoned Mr Wesselingh and told him the accumulator test had been satisfactory and asked him to draft a letter to be sent to both ABS and DNV. Mr Wesselingh’s first draft stated that the BOP control unit "was satisfactorily tested as per applicable API standards 16E and DNV MODU Rules. Please find enclosed the test specifics in Attachment 1 for easy reference”
- Mr Wesselingh said that when he prepared this draft he was hoping to have the actual test results to send with the letter. But during a further conversation with Mr Haggard he was asked if that was necessary and reported back that it was not after checking the ABS letter of 22 December (paragraph 148)
- In the event the draft letters were altered to read that the BOP control unit
"was satisfactorily tested as per applicable API Standards 16E and DNV MODU Rules.
It is requested that the item be deleted from Outstanding Recommendations for the Rowan Gorilla V.
I trust this information is sufficient ....
Should you have any further queries please do not hesitate to contact the undersigned."
- I do think that BAO's approach to ABS and DNV was unfortunate. I agree with Mr Ellis who said he would have read the re-drafted letter as saying that the API timings had been met. But I do not think on the evidence that the letter was intended to mislead. First, I unhesitatingly believe Mr (Bryan) Quinn when he says he, and everyone else on the rig, believed the tests had been not only successful but sufficient. Despite attempts by Mr Sheen and Mr Thomson to say otherwise ADC's records speak, and I find speak accurately, for themselves (paragraphs 393 and 395). Mr Quinn had no qualms about recording the times he did and when compared to the stated API times the discrepancies are apparent. There was no attempt to hide them. They were given to ADC. Cameron's records are to the same effect. So are Amoco's (Mr Walker). Further I am quite sure that Mr Wesselingh believed what he wrote was true and that whatever standards should have been met had been met
- Mr Haggard readily agreed in cross-examination that had he not believed (as he said he did) on 24 December that the problems with the BOP system had been resolved he would not have regarded the RGV as ready for delivery to Amoco and that it could not be delivered before it was certified
- Mr Haggard said he believed that those on board would know what was required and he was confident all was well. He nonetheless telephoned from his home to Mr McNease in Houston and read out to Mr McNease the times he had been given by Mr Quinn. Mr McNease (who was familiar with API standards) told him that the closing time for the annular was outside the API standards (with which Mr Haggard was not familiar) but they were only guidelines and if ADC and the people on the rig were satisfied then the results were acceptable and the certifiers could be so informed. When he went to the office in Yarmouth and met Mr Wesselingh there Mr Haggard did not have the times with him and he did not want to return home on Christmas Eve to get them. Hence his conversation with Mr Wesselingh about the need to refer to them and the changes to the letter to ABS and DNV. I accept this evidence and am satisfied neither Mr Haggard nor Mr McNease acted in any underhand let alone dishonest way in what was said to ABS and DNV. They did believe the tests were acceptable and reasonably so and neither Mr Haggard nor Mr Wesselingh thought the letter was misleading. Mr Haggard was wholly convincing in saying that Rowan’s relationship with ABS was based upon a degree of trust and it would have been crazy to mislead them. He agreed in hindsight that he probably should have referred specifically to the closing time of the annular. Mr McNease was equally convincing in saying it would never even have crossed his mind to mislead ABS or DNV.
- Further, whilst conscious that their evidence was admitted only in written form and so not subjected to the cross-examination to which it was plainly exposed, BAO has adduced statements from Mr McKay (DNV), Mr de Quelerij (DNV), and Mr Van Rooten (ABS), which go a little way to putting what happened in a more favourable light. Mr de Quelerij says that the timings are not to be viewed strictly but allow some room for flexibility and that the actual results, now he had seen them, did not give him any reason for concern. Mr Van Rooten says much the same and accepts that he had not asked to see the actual times. Mr Snow and Mr Poss also said, and (despite Mr Marymee's doubts) it seems to be commonsense, that in any actual well control event decisions to shut in a well are not usually taken in a hurry and even if they had to be in a sudden emergency the likelihood is that the BOP Rams (rather than the annular) would be the first port(s) of call. In other words, a few seconds one way or another was neither here nor there. Nonetheless I think the contract itself was clear that times were "maximum permissible response times": see Paragraph 381. API RP 53 (paragraph 383) is I think to the same effect. It also remains difficult to see what objective standard the certifiers could have applied to justify this approach. Certainly they were not given the information on which to make any judgment at the time
- Mr Walker noted in a report dated 26 December
"function test BOPs and times for closing rams and hydrill OK".
- Mr Walker said this entry was derived from the ADC Report for 26 December, prepared (as I find) by Mr Sheen that read
"1900 hrs. Repeated function tests of BOP from rig floor panel in an attempt to obtain flow readings. Flow meter found to be not working. Told that there is a spare on board.... BOP functions timed by using annular and manifold pressure gauges. Timings satisfactory. Accumulator pressure dropped to 1000 psi and pumping back up to 3000 psi took 6 minutes 32 seconds, i.e. more than satisfactory. Both pumps cutting in and out at correct pressures."
- In fact, judged strictly against the stated API timings, the closing times of the 3 BOP Rams were still not "satisfactory". Mr Sheen had no other acceptable explanation for what he wrote than it meaning what it says. Mr Sheen recorded the figures achieved on 26 December and his record also appears in Appendix E
- Mr (Mel) Thomson was the author of ADC's Daily Report on "Hydraulics/BOP" for 2 January. He noted
"Function testing of diverter valves - 4" valve on fill-up line functioned open - closed - satisfactory. Flowline valve would not operate in either direction, hydraulic lines disconnected and blockage caused by welding slag/debris found. This was cleared and lines flushed. Valve now operating satisfactorily." (The valve had been installed on 1 January).
- Mr Sleightholme, Mr Thomson, Mr Cowie and Mr Price met Mr O'Brien and Mr Slater on 3 January at the conclusion of Mr O’Brien and Mr Slater’s first visit to the rig. The record of the meeting contains no reference to any problems with the BOP system
- Mr Cowie said ( and I accept) that after he returned to the rig on 2 January he reviewed ADC’s report and came across the actual test results for 24 December recorded on two documents in ADC’s papers. Mr Sheen had left the rig on 31 December. Mr Cowie noticed that the results did not meet the API timings. He was extremely angry and thought that Amoco had been deceived by Rowan/BAO as the certification requirements were the basis for BAO’s 24 December letter putting the rig “on contract”. He took the matter up with Mr (Bryan) Quinn in strong terms. There is no dispute that Mr Quinn’s immediate response was that Rowan would do the test again. Having heard both Mr Cowie and Mr Quinn I have no doubt that Mr Cowie genuinely believed that Amoco had been deceived but also no doubt that in fact he was wrong in the sense that Mr Quinn, and ADC, had made an innocent mistake. I accept Mr Quinn’s evidence that he did not know the precise times required. On the other hand, I think Mr Cowie’s undoubted anger and their embarrassment may go some way to explaining Mr Sheen and Mr (Mel) Thomson’s evidence (including, in Mr Sheen's case, a fax sent to Mr Hay on 3 January and his tally book) seeking, despite the clear documentary evidence, to distance themselves from and indeed to contradict the events and records of the tests which did take place on 24 and 26 December. I regret to say that I found that evidence to be untruthful
- In particular I cannot accept that Mr Sleightholme misunderstood or misrecorded what Mr Sheen told him nor that Mr Sheen and Mr Thomson simply chanced upon the test being carried out on 24 December. Nor can I accept that they did no more than watch it. Mr Walker was aware of the intention to carry out the test beforehand and I unhesitatingly accept the evidence of Mr Snow and am sure that Mr Sheen and Mr Thomson were also aware the tests were to be carried out and were present to witness them. Rowan/BAO’s target was to achieve acceptance and Mr Mel Thomson was aware (as was Mr Leaman) that Mr Leaman would not have been allowed to leave the rig for Christmas before the system had been satisfactorily tested. I also accept Mr Quinn’s evidence that he discussed the tests afterwards with Mr Sheen and everyone was happy that the system had performed very well. It is apparent that until Mr Cowie raised the matter on 2 January everyone was proceeding on the basis that the test had been passed and was satisfactory. Moreover I think the reason why the tests were repeated on 26 December was that ADC’s report form (but not API RP 53) required further information than Mr Sheen had recorded on 24 December but which he did record on that date (see paragraph 406)
- On 3 January Mr Cowie, Mr Price and Mr Quigley noted that an attempt had been made to record opening and closing times again for the rams and annular but the accumulator unit was malfunctioning. Decreasing pressure in the manifold regulator produced an increase in pressure on the annular and flowline seal gauges
- Mr Leaman returned to the rig in the early morning of 4 January. His Daily Report Sheet for 4 January shows that the problem with the accumulator unit was to be dealt with, after advice from Mr Lehmker, by fitting a check valve in the return line. The report also records that the pressure on the manifold regulator would not decrease and that Mr Leaman removed the shear assist pilot valve (item 76) and found "the internal parts were all blocked with some sort of debris (metallic or fibrous)" and that the internal filter screen had split and been pushed back into the valve body causing internal damage. A new valve was ordered and the filter repaired.
- The next day Mr Leaman's report records that the timings of the BOP operations were raised with him by Mr (Bryan) Quinn. Closing times were said to be too slow. This led to agreement that the control fluid would be replaced with a less viscous fluid which was ordered along with flushing equipment.
- The Amoco people on board the rig were kept fully informed about the problems as appears, for example, from their report of 4 January which noted
"[Mr Leaman] is of the opinion that the whole system requires flushing to ISO standards, this would involve stripping the control unit, disconnecting the lines to the drill floor and the operating lines to the stack. Obviously this would take some time and we will await Rowan's decision on the way forward, but it confirms that the fundamental problem could be debris in the system. The question of operating times will have to be addressed when the unit is up and running."
- This is a different exercise than stripping down the whole system as it has been referred to during the trial and in Amoco's case, but more than Mr Leaman says he said. It is not difficult to see how there might be confusion. The flushing exercise itself involves disconnecting all the lines to the valve module and from the module to the BOP stack. It would also necessarily require repair or replacement of components which were found to be blocked when the system was functioned after the flushing was completed. There is no dispute that the flushing and replacement fluid was intended to address both the debris in the system and the closing times. Nor does BAO suggest that the problems did not need to be addressed. In the event I see no need to try to resolve the various conflicting accounts of what Mr Leaman may have said. I accept his evidence that he did not believe that a strip down of the whole system was required at this time let alone after the flushing exercise had been completed
- On 6 January Mr Leaman recorded that "4 armoured hoses had been crushed when the derrick had been skidded back". One hose was said to be crushed almost flat and the condition of all 4 was said to be indeterminable because "we cannot get access through the armoured sheath to inspect internal damage". Mr Leaman recommended to Mr Quinn that he inspect them. He also sought an extra man from Cameron to help to flush the system
- On 7 January Mr (Brian) Thomson arrived on the rig to assist Mr Leaman. The tanks and accumulators were drained and Mr Leaman's report noted: "Reservoir inspected and found large deposits of metal and other unidentifiable materials. Sample provided for further inspection". In another report of the same events Mr Leaman referred to "large deposits of welding slag and other substances"
- Mr Leaman and Mr Thomson began to flush the entire hydraulic system on 8 January. The lines were disconnected and flushed through. Mr Leaman also redressed the valve (item 76) removed on 4 January (paragraph 413) and re-fitted and tested it. Flushing of the system continued throughout 9 and 10 January. On 9 January Mr Leaman referred to "large rust deposits seen coming from hoses", and on 10 January to the fluid being clean to NAS, Class 10. There is no other report of the flushing exercise revealing or evidencing any other debris in the system.
- The specified standard was NAS Class 10. That is a reference to National Aerospace Standards which state maximum contamination limits based on a 100 ml sample for fluids in hydraulic systems. The lower the number of the class the cleaner the fluid. The Class 10 maximum limits were 256,000 particle sizes of 5 to 15 microns rising up to 250 particle sizes of 250 microns. Class 9 required 50% of those figures and Class 8 50% of those for Class 9. The classes ranged from 1 to 12 each changing by a factor of 50%
- On 11 January the reservoir tank was filled with the replacement, less viscous fluid. Mr Leaman and Mr (Brian) Thomson's note dated 11 January, which they confirmed in evidence, records
"Function tested BOP rams and timed them opening and closing. Timings found to be within Rig requirements.
Waiting on Rig Manager to pass system timings."
- On 12 January Mr Leaman and Mr Thomson's daily report sheet records that they fitted two check valves in the return line as advised by Mr Lehmker (paragraph 413). The report continues
Carried out BOP test and function valves on BOP stack. All ram valves closed within specified time and the Annular closed within 40 seconds. Carried out Accumulator Test again. Shear assist found not to be working. Shear assist pilot valve removed stripped down cleaned and examined as far as possible. Found possible problem with internal check valve being stuck and not allowing fluid to pass. New valve ordered ....
- It was noted that, save for the shear assist, the system was fully functioned during the evening. Mr Gray's tally book suggests that in fact one closing time for a BOP Ram was 1 second in excess of API RP 53. Although the evidence is at best confusing I do think it improbable that the shear assist had been successfully tested on either 11 or 12 January but failed shortly afterwards on the occasion recorded in this report. The report itself suggests a lack of surprise at what happened. The valve had been found to be blocked by debris on 4 January (paragraph 413). Mr Leaman had to reinstall it to carry out the flushing exercise on 8 January (the replacement had not arrived and blanking caps were not strong enough to withstand the flushing) and as he was unaware of its internal configuration I think the probability is that contrary to his belief it still contained debris which he failed to remove but which had been present on 4 January before the flushing exercise. The valve was unique to the system in incorporating two small opposing internal check valves: See also paragraphs 433 and 435
- Mr Price's "observations/concerns" provided to Mr O'Brien on 13 January included a reference to the well control system
"The BOP control system was found to be out of specification. Further investigation pointed out that the system was contaminated with solids and emulsified hydraulic fluid. This required the Cameron engineer to insist that the entire system be flushed and displaced with a less viscous fluid. This was accomplished during the tow to the standby location. A third party inspector will now verify the system.
- That is an express and uncritical reference to what it was upon which Cameron had insisted. There is no reference to any, let alone a complete, strip down assuming that was understood to be different from flushing the entire system. It also accords with Mr Leaman's evidence and what was in fact done
- The daily report for 13 January records that the system was fully functional on the evening of 12 January but that at the beginning of the shift on 13 January Mr Leaman and Mr Thomson had been called because of a problem with the manifold and annular regulators (32.2 and 32.1) on which pressure could not be increased or decreased. The report records that when the system was charged up pressure rose rapidly up to 500 psi and then stopped and failed to go any higher. Mr Leaman and Mr Thomson noted that there was "good pressure and good flow" after the in-line filter and on the inlet supply to the increase/decrease manifold (item 66). But pressure at the regulator was 500 psi and would not increase and the same applied at each component after the manifold in the line
- Mr Leaman and Mr Thomson concluded that this indicated there was "a possible blockage somewhere inside the manifold block". It (item 66) was stripped down, cleaned, examined, had compressed air blown through all ports and valves, reassembled and refitted but the problem persisted. The regulator was also stripped but nothing was found to be wrong with it. The problem was discussed on the telephone with Mr Coull and it was suggested that the increase/decrease manifold should be checked again as a filter might be blocked. Mr Leaman and Mr Thomson therefore re-stripped item 66 "into smaller individual components" but found nothing and the problem remained when it was re-fitted and the system tested again. A replacement was ordered.
- The next day (14 January) the diverter regulator was found to be passing fluid. One of the seals under a seal plate was found to be damaged and an Allen screw was "floating around in the body of the valve". The regulator was rebuilt and the seal replaced
- Item 66 was stripped down "even further" once again. Nothing was found. Further pressure tests were carried out along the system. Inlet and outlet pressures on item 66 were at 3000 psi. That could only be so if item 66 itself was not the cause of any problem. The daily report notes that when the pressure was tested "below the regulator throttle valve (item 62.5)" 0-250 psi was achieved very quickly but from 250 psi "the pressure increase was very slow" (The reference to item 62.5 should have been to 62.1). The report also noted that when the pilot filter (item 15/16) was removed and the system retested "inlet pressure was more stable", and that, over an hour's testing, "the operation of the regulator was OK". However in his "Closure Report" Mr Leaman wrote that after removing item 15/16 "the response was a little better but still not working correctly". Mr (Brian) Thomson also said the regulator was still not working properly when Mr Lehmker arrived on 15 January
- Mr Coull reported to Mr Duensing (Mr Lehmker’s superior) on 14 January that
"The system was drained of control fluid and flushed/purged with clean fluid, but from experience any system being subjected to this degree of contamination without a major strip down and overhaul of all components, further failures of components will occur."
- Amoco understandably relies on this report as support for its case that the system required to be stripped down and indeed that Mr Leaman said as much. Amoco's letter asserting that this was Cameron's view was sent the next day (paragraph 166). But the evidence of Mr Coull was that the statement in his report to Mr Duensing had been derived from his experience which was obtained on much finer small bore hydraulic systems and that Mr Duensing's response had been that a strip down was not required and the matter should be left to Mr Lehmker who was going to the rig on 15 January. Despite the attack on this evidence, I accept it as truthful. Mr Coull was an impressive and straightforward witness and his evidence derives some support from a later report (see paragraph 452). Mr Lehmker was in fact unaware that the system had been flushed until 15 January, and once the flushing had been completed and subsequent function tests carried out no one was suggesting that any or any further form of strip down was required in any comparable sense
- On 15 January Mr Leaman and Mr Thomson took samples of fluid in the HPU tank and the Accumulator Unit and found they were NAS Class 10 and Class 9 respectively.
- Mr Lehmker arrived from Germany on 15 January to assist Mr Leaman and Mr Thomson. Mr Lehmker gave his evidence with the assistance on occasion of a German interpreter. The daily report noted that they had carried out a function test on the Shear Assist which failed because the pilot valve (item 76) failed. The report continued
"Removed valve and inspected found that the check valve was full of debris
- The regulators for both the BOPs and the annular were also found not be working
- Mr Lehmker had stripped item 76 down further than Mr Leaman had done. He wrote a note recording the work he performed on 15 (and 16) January. In relation to item 76 his note records
there was a lot of dirt and debris inside the valve housing causing the hydraulic supply fluid to be blocked at the inlet side so that no hydraulic outlet signal could pass through this valve.
- In relation to the regulators, Mr Lehmker noted that
"It was found out that the ... pilot valve block (item 66) ... created the problems. This valves are installed to ensure either remote or local pilot operation of the ... BOP - Manifold Regulator (item 32.2) ... and of the ...Annular Regulator (item 32.1) ..... It was decided to replace the ... pilot valve block because damage was assumed to the internal parts during the operation time of the BOP Control System with heav(il)y contaminated fluid. This was also (supported) by the inspection of the inlet fluid supply filter element (item 16) of this valve that was totally clogged with debris and dirt...."
- Item 66 was changed out on 15 January being replaced by one from the RGVI. The shear assist valve was changed out on 16 January (paragraph 441) and has caused no problems since
- On 16 January Mr Leaman continued to flush the HPU tank until the fluid was cleaned to NAS Class 8.
- Mr Leaman's daily report continues
"Spares arrived on helicopter. Shear Assist valve and Increase and Decrease manifold replaced and tested ... system worked as it should at the correct pressure. API test carried out on BOP testing timings and function. Test passed and signed off by Rig Manager and ADC examiner. "
- There is a test sheet showing timings and pressures dated 16 January signed by Mr (Bryan) Quinn and Mr (Mel) Thomson, a copy of which is included in Appendix E. In his "Closure Report" Mr Leaman also recorded and Mr Thomson said that after the test had been passed functional checks were successfully carried out several more times and that whilst diverter functions had not been carried out because the diverter was not fitted "it had tested OK and without problems previously"
- Mr Lehmker's notes for 16 January also refer to fitting a new Item 76 and then Item 66 and continue
"With this new installed parts it was possible to start an official customer witnessed BOP-Stack function test. This test includes the operation of the integrated Shear Assist system and also the local/remote pilot operation of the ... BOP-Manifold Regulator and the ... Annular Regulator. During the test actual BOP operation times were measured and recorded in the attached test sheet. The Accumulator Volumetric Capacity Check was also carried out and the results were recorded in the attached test sheet. The system charge time from precharge pressure to maximum working pressure is also recorded in the attached test sheet.
The achieved test results were totally in accordance to the customer expectations without any objections, so that the attached test sheet was signed by the rig manager and the ADC inspector.
- "The attached test sheet" was the same as that referred to by Mr Leaman (Appendix E). Mr (Mark) Jones recorded in his tally book "BOP function tested to the satisfaction of ADC and BP Amoco". Mr Price also said the test was satisfactory and that apart from two flowmeters everything was working on the BOP control unit
- On 17 January Mr Leaman checked the equipment and noted "no problems seen, system working fine". He left the rig on 18 January. He said that when he left he was confident that the system would function 100% and he did not think or say that it was necessary to strip down the accumulator unit or inspect the chambers of the BOPs. In his witness statement Mr Leaman said
“The steps which I took with Brian Thomson to flush the unit and the parts which were replaced amounted in my view to a stripping down of the unit. We had isolated the tank, the pilot and control lines and various other components of the system, including valves and regulators, and each was flushed thoroughly. We had also stripped down, cleaned and repaired all parts which were found to be affected by contamination. I have no doubt that the exercise was appropriate and successful in dealing with the problems that had arisen. There was a possibility that, as a result of contamination having been in the system prior to the flushing, or because there was a very small amount of debris left in the system, one or two components may need to be changed out later. The need to do so could be minimised by regular maintenance, such as changing filters, sampling and flushing, and any parts which did need to be changed out could be changed quickly, using the spares onboard”
I accept this evidence as truthful. Mr Lehmker said much the same.
- Mr Mel Thomson's daily report for 19 January referred to a function test on the diverter system. It noted that the system was "in good working condition" but with "deficiencies" stated to be
Regulators for flowline seals, overshot packer, and diverter packer element could not be operated from the drilling panel, or the office panel, trip tank pump also could not be operated from drilling shack on start/stop function.
- The problem (described by Mr Gray in evidence) was similar to the problem on the Annular and BOP system. Pressure on the diverter functions would rise to 500 psi but go no further. Moreover all three diverter functions suffered from the same problem: the flow line seals, the overshot packer and the diverter packer. The problem was attributed to the diverter increase/decrease manifold block (item 67). Neither Mr Marymee nor any other witness could explain how debris in the inlet line could possibly cause the pressure to go up to 500 psi and then stop for one let alone all three valves. Nonetheless at the time, as with item 66, the problem was thought to be caused by debris in the valves
- The status of the system at 19 January has also to be seen in the light of Mr O'Brien's reports (Paragraphs 193-194) and "Omissions Profile", (Paragraph 190). So far as material as regards the Well Control System it was recorded that the problems with the manifold pressure regulator (item 66) and shear assist pilot valve (item 76) required replacements which had been fitted on 17 January and the control system was “hydraulically operational”. Mr O'Brien could hardly have thought that successful acceptance tests could be carried out in 5-7 days if he really believed there was a serious contamination problem requiring a strip-down of the entire system. Moreover there is no dispute that BAO had agreed to Mr O'Brien's acceptance tests which Mr O'Brien agreed would have exposed any remaining problems in the system. Mr O'Brien did say he was confident the system would fail his test but that is not what his report says
- ADC's final report on the Well Control Equipment included a "Summary", written by Mr Mel Thomson, which stated that
Although a certain amount of BOP testing has been carried out, some remains to be completed. For details see the checklists in the following section. Acceptable closure times for the BOP rams and annular were only achieved after changing the hydraulic fluid to a lower viscosity.
- The checklists included "Record opening/closing times" and the "final ADC comment" "Satisfactory, See final results below". They also referred to certain low pressure tests as "outstanding"
- The "final results" were those which had been obtained on the test carried out on 16 January, the record of which had been signed by Mr Quinn and Mr Mel Thomson at the time (paragraph 440 and Appendix E)
- ADC’s final report on the diverter recorded that everything on ADC’s requirements was “satisfactory”. Mr Gray and Mr Perkins said the diverter was tested in the early hours of 19 January in a manner suggested by Mr Mel Thomson. The test was witnessed by Mr Thomson and apart from the pressure rising to and remaining at 500 psi everything was satisfactory. Item 67 was changed out a few days later for one from the RGVI
- The damaged hoses were the subject of a CAR (No 55) which "suggested" "at least a pressure re-test or replacement of the sections". On 19 January this CAR on which Mr (Mel) Thomson had written "completed" was signed off by Mr Sleightholme
- Mr Leaman and Mr Coull wrote a "Close-out Report" dated 21 January. It was Mr Coull who wrote that the system had been left functional but, because of the contamination there could be "no guarantees that another failure would not occur". The Report did not, however, make any reference to a strip down of the system. Mr Coull said in evidence the system was ready for use and if there was a failure it would be a matter for maintenance. He said you can have a failure in any system and that is why the systems build in dual redundancies and are not single point failure. It would of course have been wholly irresponsible and commercially disastrous for Cameron to have left the rig at a time when it was believed to be going to Arbroath if they had thought their system was in any way unsafe or unfit for use
- ADC's final report also included a copy of CAR 56, one of the 3 CARs which were issued "late". Mr Sleightholme's evidence was that the form used was not the same as the form he used for CARs up to the time when he left the RGV for the last time on 5 January. It was therefore his conclusion that Mr Mel Thomson, who succeeded him as team leader from 5 January, must have issued it after that date. Mr Thomson wrote the number “56” on the form. Indeed from its wording it must have been prepared after the flushing exercise was completed. The evidence of Mr Price and Mr Quinn suggests that it was prepared between 16 and 18 January
- The wording, under the printed heading "Non-Conformance", is
Due to constant problems with the B.O.P. control system, and control hoses, the system filters showed a considerable amount of foreign particles. There were numerous operations of the BOP Rams before the contaminated fluid was displaced with clean fluid. Therefore in the interest of good engineering practices it would be strongly advised to open at least one of the operating chambers for inspection.
- Mr Sleightholme said he discussed this with Mr Thomson when they were preparing the ADC final report. That is borne out by Mr Sleightholme's hand-written comment on the document "Not done at 19/1/99"
- I would make the following comments
i) The only strong "advice" is to open one BOP chamber for inspection. The evidence is that this would enable a visual inspection to be made of the piston, the cylinder bore and the seals to see if there was any scoring. That is very different from a strip down of the whole system as Mr Thomson agreed.
ii) Mr Price accepted that the contents of this CAR addressed the concern which he had about debris in the system and which he had expressed to Mr Quinn. Indeed Mr Price said, and I accept, that the CAR was issued at his request.
iii) As the "contaminated fluid" was replaced over a period of 4 days ending on 11 January the CAR must have been prepared by Mr Thomson after that date. His evidence to the contrary is mistaken and I think it unlikely it was prepared before 16 January.
iv) The language used is of good engineering practice. There is nothing to suggest that the matter was seen as one involving serious safety considerations.
v) The CAR is the only documentary evidence of any "concern" about contamination apart from the 15 January letter (Paragraph 166). But its terms are much more consistent with the 20 January letter than that letter (Paragraph 172). The 20 January letter acknowledged that the accumulator unit was "now working" save for some flow meters (which it is agreed were not important: paragraph 477).
- Mr Quinn said he thought it was not only unnecessary to open up a chamber but positively dangerous as it could lead to damage to the unit. He said he would not have done it unless Amoco had insisted and accepted responsibility for it. He said (and I accept) that when Mr Thomson handed the CAR to him, Mr Thomson said he had been instructed by Amoco to give it to Mr Quinn but Mr Thomson himself did not believe that it was necessary. Mr McNease said that had Amoco insisted it would have been done but he too thought it unnecessary. So did Mr Marcom. Their judgment was in fact right: see paragraph 465. I also have no doubt at all that it was an honest and reasonable judgment at the time supported as it was by the successful tests. Rowan and BAO would not risk the safety of their personnel nor indeed what was a major asset of Rowan. But I also have no doubt that had Amoco insisted BAO would have done as asked.
- It is an important part of the reasoning of Amoco's experts that further failures or malfunctions occurred in the hydraulic control system after 19 January, and so after the comprehensive flushing of the system, which were caused by or at least attributed to contamination remaining in the system
- There is no doubt (or dispute) that later failures or malfunctions were attributed to contamination. It follows and the evidence is that such failures were thought to be the sort of matters which could be caused by contamination even after flushing. But there is doubt and dispute as to whether (save in one case) the attribution to the earlier contamination was correct. It is I think material to note that Cameron were wary of Rowan blaming Cameron for problems with the system and whilst I do not mean by that to suggest that the views expressed were anything other than genuinely held both at the time and indeed for long afterwards, I do think that contamination had come to create something of a mind-fix which can now be seen to have been overstated
- Mr Leaman was sent out to the RGV again on 1 February to investigate a fault with some flowmeters. It is agreed that the fault itself was not a matter of any material significance (Paragraph 477). But whilst investigating it Mr Leaman recorded that another fault appeared with the replacement manifold block for the diverter (Item 67) which was passing fluid. The valve was removed and stripped down. Mr Leaman noted on 3 February that "small deposits of debris was found in the internals these were cleaned up and valve tested and found to be working correctly". It was then reinstalled and worked correctly. Amoco submits that this was the first date on which it could be said that the well control system tested satisfactorily
- In April 1999 there was a further problem with Item 67. On this occasion Mr McPhail was sent to the rig by Cameron. He found erratic pressure readings and fluid passing. He arranged removal of both items 66 and 67 for testing and repair where required. Whilst there Mr McPhail checked the cleanliness of the hydraulic fluid and some filters reporting that the fluid was in reasonable condition and that there was no excessive contamination in the filters
- On 13 May Mr Coull sent Mr (Bryan) Quinn a technical report and proposal to "repair and function test" the RGV BOP system. The conclusion of the report was expressed to be
"Both the manifold valves (Nos 66 & 67) have become faulty due to the initial ingress of contamination in the system and will have to be removed and replaced with new units."
- In fact Mr Coull was, I am satisfied, wrong in attributing the matters dealt with by Mr McPhail in April to debris. Mr McPhail did not do so and Mr Ellis convincingly explained the likely cause of what Mr McPhail found which had nothing to do with debris
- On 22 and 23 July Mr (Brian) Thomson returned to the RGV. He replaced items 66 and 67 with new valves. But when tested it was found that the same pressure control problems appeared as before. He also noted that one (item 40.3) of the mini-accumulators was set too high (at 35 not 10 bar) and also referred to adjusting the throttle valve. 35 bar is about 520 psi. After adjustment he tied the throttle valves into position. On 27 July it was noted that the new valves were leaking at a rate of about half a litre a minute. On 28 (or 29) July 3 small pieces of welding slag were found in the hose leading to the 4" block valve (Paragraph 379) at the bottom of the fitting at the deck cluster. Tests were carried out after removal of the slag which were wholly satisfactory. Amoco nonetheless says this incident is symptomatic of the risk of finding debris and debris causing malfunctions even after flushing and apparently successful functioning of the part. This is considered further in paragraph 482
- Since 19 January the BOP system has been function tested regularly including function and pressure testing in Dundee and Canada. The annular was stripped and inspected in Dundee in September 1999. No debris or scoring was found. The system underwent a successful pre-drilling test in Canada. The BOP ram covers were removed on 17 February 2000 to install cold weather elastomers required for Canadian conditions. No debris or scoring was noted as being present in the operating chambers
Expert Opinions
- Whilst as the factual evidence developed there were understandable divergences between Mr Marymee and Mr Ellis as to the number and nature of the incidents of debris being found in the system and their significance, the nature of their disagreement and the measure of any agreement is reflected in a Joint Memorandum dated 5 March 2001
- For Mr Ellis it was significant that such contamination as was found was in the pilot signal lines not the main supply lines, that it was not unexpected to find it in filters (which exist to catch it) and the HPU tank and the flushing and partial strip down or change-out of parts found to be defective was both effective to remove it and a proper and usual response. Mr Ellis also made the point that there was a difference between the debris such as the Allen screw and snap ring which were contained within the parts and debris which could affect the whole system. The former were not symptomatic of more widespread contamination
- Mr Marymee's emphasis was on the accepted safety-critical nature of the well control equipment, and the risk of further contamination with unpredictable consequences. It was his opinion, despite the limited terms of CAR 56, that the control system should have been completely stripped down prior to drilling and that it was necessary to strip down and clean all mechanical components in contact with the fluid. He also considered that all carbon steel piping should be "pickled" (a process of placing acid in the pipe) to remove rust. That was because of the reference to rust being seen on 9 January (Paragraph 419)
- The experts were also in disagreement about pressure testing and as to the extent to which it had in fact occurred. They agreed that the whole system would be pressure-tested prior to the start of drilling operations and at least fortnightly thereafter as a matter of routine
- They were agreed that the response times achieved on 24 December did not satisfy API RP 53, but that the 16 January test did satisfy all applicable standards
- They were agreed that the recommended practice was for each of the control system pumps to have an independent power supply. Mr Ellis thought the design met this criterion because each pump could be supplied from either the main or the emergency power supply and the only common component was a heavy copper busbar in the switchboard which, in any event, could have been dealt with by 6 - 8 hours of re-wiring if anyone had really thought it necessary. Mr Marymee disagreed whilst acknowledging that failure of the switchboard was improbable and could be dealt with by re-wiring
- They were agreed that the diverter system had been fully assembled by 19 January. Mr Ellis considered there was ample time to pressure test it as an installed system in the days before it would be required. The experts disagreed on the importance of the diverter manifold block (item 67) which could not be adjusted from 500 psi. Mr Ellis also considered there was ample time to obtain a replacement
- The damaged hoses could have been replaced with spares on board if there were doubts about their integrity
- In summary the conflicting positions were set out in answer to the question
"Was the BOP Control System fit for its purpose at 19 January 1999?"
- Mr Ellis' response was
"Yes, the BOP Control System was capable of controlling all functions of the BOP and Diverter on 19 January 1999. Minor troubleshooting and repair (including repair or replacement of the manifold block Item 67) could and would have easily been completed prior to the need to deploy the diverter or BOP stack."
- Mr Marymee's response was
"In my opinion, the Rowan Gorilla V BOP System was not fit for purpose on 19 January 1999 for reasons including, but not limited to, the following: 1) The Diverter System Increase/Decrease Manifold Block (Item 67) was not properly controlling the pressure regulators; 2) there remained a risk of further contamination caused equipment failures (as occurred with the 4" Block Valve 29 July 1999); 3) the Driller's and Toolpusher's Control Panel flowmeter displays were not functional, these flowmeter displays were a requirement of the certification standard DNV Drill (N); 4) the two BOP Control System pumps were not supplied by independent power sources; and 5) some of the control hoses had damaged armour and the integrity of the underlying hoses should have been verified. Additionally, the BOP System had not completed acceptance testing specified by Contract Appendix 6 ....”
- I should add that the flowmeter displays were agreed to be "desirable" but in no sense essential. The flowmeters on the main Control panel were working and there was no relevant contract or class requirement for them. DNV Drill (N) was applicable only to drilling in Norwegian waters. The only contractual requirement as regards the function tests was for "maximum permissible response times" to be as per API RP 53 - API RP 16e: see Appendix A at page 68
- The evidence in relation to "independent" power supplies is overwhelmingly that it was of no significance. It did not trouble DNV or ABS or HSE. It was not required by the contract. It could readily be dealt with if anyone had required it. The position in relation to the hoses is equally clear. I accept Mr (Bryan) Quinn's and Mr Snow's evidence that the integrity of the underlying rubber hoses was established by inspecting and testing them; only the outer sheath was damaged not the rubber hose inside. Moreover various tests were in fact later carried out without a problem which must have involved functioning the hoses. It would have taken only some 3-4 hours to fit new hoses (spares were on board). ADC were content (paragraph 451). The same hoses are still in use on the RGV. Mr Marymee would have done himself more credit if he had conceded the point instead of seeking to maintain a position which I think was exposed as untenable. These matters really have no place at all in a dispute the parameters of which are set by Clause 28 and the law of repudiatory breach of contract
- The allegations about pressure-testing or the lack of it are also of no substantive significance. Not only is it recognised that all relevant tests would routinely be carried out just prior to the start of drilling (and at least fortnightly thereafter) I accept the evidence of Mr Perkins, Mr Wesselingh and Mr Snow that in fact low pressure tests were carried out at Sabine Pass as they described and no one has even suggested any reason why the tests would or might have failed had they been carried out or repeated. It is also agreed by Mr Bartle and Mr Poss that any further testing would not have affected "the critical path" for operation of the RGV because there would be ample time to carry out tests in the 8 days or so which would elapse between the RGV arriving at the Arbroath platform and the commencement of drilling
- If there is any substance to be found in Amoco's complaint under this head in my judgment it has to be found and found only in the contamination issue. That case is really founded on what is said to be an unacceptable uncertainty created by the risk of continuing contamination. But in my judgment Mr Marymee carried theoretical risks far beyond reality or what any reasonable inference from the actual facts could sustain, and beyond any requirement which could be required by ALARP. I think Mr Ellis' evidence was much to be preferred as a realistic and measured assessment of the matter
- The people best placed to judge the actual risks were those on the rig, perhaps particularly those who would be on it if a well control incident occurred. All of Rowan, Cameron and ADC and Amoco personnel who were party to the events involving the BOP System were experienced people well aware and respectful of the safety considerations involved. Their views as documented are all of a kind. Their judgment at 19 January was that the system was satisfactory. It is now known that they were right because it has proved itself in Canada in the sense that it has had no problems when tested frequently (weekly, on the evidence) and when the operating chambers of the BOP were in fact opened up to change seals (to meet Canadian conditions) no damage was found to them. Mr Marymee, at least at first, nonetheless stood by his opinion that all parts of the system in contact with the fluid should be stripped down and that it should still be done even today. He later modified that view to suggest it would be sufficient now to do only "half" a strip-down if that revealed no problem. In my judgment that is to ignore reality. Indeed I think that was a characteristic of some of Mr Marymee's evidence as well as an inability to accept as fact what had plainly been established to be so
- There is some agreement that flushing is the usual response to debris in a hydraulic system and there can be no dispute that the flushing carried out by Mr Leaman and Mr (Brian) Thomson was very thorough and achieved a standard of fluid cleanliness better than that called for by the specification. There are, as I find, satisfying explanations for all the incidents which might otherwise be characterised as illustrations of debris remaining after flushing. Item 76 (the shear assist valve) is a complex valve and the only one of its configuration on the rig: it has two internal opposing check valves so that flushing might well not remove debris lodged there. The probability is that the debris was present before the flushing and was not removed by it (paragraph 423). Despite the evidence (albeit uncertain) of Mr Snow and Mr Thomson that they believed the 4" block valve of the diverter system had been successfully function-tested the day before the occasion in late July when it was found to have debris in it and some confusion in the documents about the date and number of tests I think that unlikely. I can think of no good reason why BAO should want to test the system a day after a successful test and so I think it most unlikely that the valve was tested successfully on 28 July but unsuccessfully on 29 July as Amoco contends and some of the documents suggest. Had that happened, however, I think it would probably have been expressly recalled if not recorded, which it was not. In my judgment the probability, especially in view of its redundancy, the likelihood that it would otherwise have been manifest earlier and its location, is that this was the first occasion the valve itself had been connected to the hoses since the flushing and so also the first time the whole line had been tested (see paragraph 464). Thus the debris, found in effect at the end of a self-contained dead-end line, had been there at least since the flushing exercise. Mr Ellis explained how if the hoses had been tested when blanked off debris would hardly move in the line but when they were connected to the valve the pressurised fluid could move into the valve and force the debris to the deck cluster for the first time. In other words it is an illustration of debris remaining after flushing but also of it revealing itself when first tested. Even if I was wrong in this conclusion I do not think the discovery of debris in one redundant line even after a successful prior test would begin to justify concern that it was symptomatic of a wider problem. Moreover, albeit a very late BAO entry in the suggested causes, I also think the probability is Mr Lehmker is right that the problems with the manifold blocks (items 66 and 67) were nothing to do with debris or contamination. Mr Leaman's frustration at stripping down item 66 on three occasions and finding no contamination and nothing wrong with it was wholly genuine. I agree with Mr Lehmker and Mr Ellis that no sensible explanation attributable to debris can be or has been provided for one let alone both regulators and 5 control valves rising to a set pressure and remaining there. Everyone agrees the problems with items 66 and 67 probably had the same cause. Why both? Why all 3 diverter functions? Why to the same pressure? Debris is most unlikely to cause consistent problems. Mr Lehmker said you could carry out experiments for 100 years and not achieve such a result. Even if it requires a measure of human incompetence, and cannot explain fully all the matters recorded in the documents, it is much more probable (or, as expressed in evidence, the least improbable theory) that the noted pressure was derived from the mini-accumulator bottles (albeit wrongly set) and that the throttles were in effect set so as to be nearly closed thus preventing the system pressure reaching the regulators. The possibility of this occurring derives some support from the events in July 1999 (see paragraph 464). Mr Marymee was not prepared to offer an alternative "least improbable theory" of his own. As for item 67, the replacement fitted in January came from the RGVI and the debris later found in it may well have come from the RGVI or been the result of installation. That of course is not conclusive because there can be no doubt that in January 1999 Cameron did believe that contamination was the cause of the problems with items 66 and 67 and the proper reaction to it has to be considered in that context. As Mr Coull put it, Cameron was chasing ghosts with items 66 and 67 but the question remains whether and how they needed to be chased at 19 January
- Nonetheless it remains in my judgment an unrealistic overreaction to advocate a strip down of the system. It was not thought to be required at the time by Cameron (paragraph 443) nor by ADC (paragraphs 447 to 448 and 454 to 456) nor, indeed, by Amoco itself as the change in the terms of the letter of 20 January from those of the letter of 15 January (Paragraph 172), the records and evidence of Mr Jones and Mr Price (paragraph 442), the fact that all on board on 19 January believed the rig was to go to Arbroath, and Mr O'Brien's report of 20 January (paragraph 446) demonstrate. Mr Marymee agreed that he had never heard of a strip down being required before. His point was that he had never come across a system as contaminated as this one. I think that is to exaggerate the problem. I prefer the view of Mr Ellis and those present at the time. Mr Ellis said the level of contamination was not unusual. He had seen far worse. The flushing achieved a level of cleanliness in the fluid considerably better than specified; repeated operation and testing of the system could be expected to reveal any continuing problems before the system was required for potential use in earnest; and changing out of those parts which had been shown to be affected (or which might have been) had taken place (item 76) or was planned. The likelihood of debris being trapped in the system could be considered and addressed in the case of items such as item 76. But as Mr Leaman put it, supported by Mr Ellis, "these valves are not very forgiving. If there is a damaged valve it will tell you it is damaged". That is why repeated successful functioning of the system justifiably gives confidence in its integrity. It is also, I think, material that both Mr Ellis and Mr Poss (and Mr Quinn) said (and I think it is commonsense) that stripping down could itself cause risks in re-assembly; the design of the RGV was such that it was possible to bypass the manifold regulator and apply pressure direct to the rams and that there were two sets of rams (in addition to the Annular) available for any given size of pipe. That is not to say that it would be reasonable let alone consistent with ALARP to start drilling with a belief or reasonable suspicion that an item of the well control equipment was deficient or that there was contamination in the system
- The need to "pickle" piping to which Mr Marymee refers also needs to be seen in context. The purpose is to remove rust from the inside of piping. The only piping on the system which could be subject to rusting is carbon steel (or shipboard) piping. There was not a great deal of such piping included in the system (Paragraph 374). Moreover such evidence as there is of rusting is to be found in Mr Leaman's note of 9 January (paragraph 419). When he gave evidence Mr Leaman acknowledged that it was no more than an assumption on his part that it was rust because "it looked like a rust deposit. It was brown in colour and had a gritty texture to it". Mr Leaman said, and I accept, that Mr Quigley was mistaken to state that rust was also found in the suction filter when it was removed: "it did not contain rust. It contained the type of debris it is put there to catch". There is no other document which refers to rust and not only do I think Mr Leaman would have recorded it if it had been found but I think his recollection was confident and truthful. That is a thin basis for concluding that the shipboard piping required to be pickled. Moreover Mr Lehmker said flushing was the normal way to get contaminants out of the system and there is no evidence at all of any rust being found in the fluid after the flushing and replacement of the fluid. Indeed such evidence as there is shows that it remained "clean" (paragraph 461). In my judgment pickling was not reasonably required and Mr Marymee has moved too far from reality to theory in saying that it was
Operator Issues
- As regards the well control system the views of Mr Bartle and Mr Poss were again divergent. Mr Bartle considered that the continuing risk of contamination even after the flushing meant “the complete operating system” had to be dismantled and inspected. Mr Poss considered that the flushing was a sufficient response. They did agree that the closing times were adequate at 19 January. They did not agree that an operator would willingly have accepted times a few seconds outside the API recommended practice. Mr Poss said he would; Mr Bartle said he would not
- Again, and understandably, Mr Bartle had relied on the views of Mr Marymee in expressing his opinions on the reliability of the well control system and what if any further work was required to make it fit for purpose. He agreed that at 19 January he, as a reasonable and prudent operator, would have taken into account the views of ADC, Mr Jones, Mr Price and Mr O’Brien and those of Cameron and Rowan/BAO. But when reminded of their views he said he could not and would not have accepted them and in some cases expressed doubts about their expertise and competence. It should not need to be said that all of those companies and people had the same concerns about safety as the Operator. But Mr Bartle was quite unable rationally to explain why a reasonable operator would ever have got to the position of consulting an expert such as Mr Marymee. The fact is that Amoco did not do so. Insofar as views were expressed on the contamination issue they reflected CAR 56 and the letter of 20 January, as Mr Bartle came to acknowledge. I should record that I found Mr Bartle’s evidence on the contamination issue particularly unsatisfactory. Mr Poss realistically and sensibly said that to make a judgment you would have to be there at the time. Those who were did so. They were well placed to do so. And in fact on my findings their judgment was right
- Mr Bartle accepted that none of the other matters in relation to the well control system (pressure testing, independent power supply, flowmeters and damaged hoses) would have been a reason for not allowing the RGV to proceed to the Arbroath platform. The reason being that if they did need attention what was required would take little time and could be done at Arbroath: as Mr Bartle put it, they “could have been sorted out in a short space of time”. He had, however, described them as “major problems” in his report
Conclusions
- In my judgment, the evidence establishes beyond any serious question that if Amoco had not determined to terminate the contract, the RGV would have gone to Arbroath and if there were any continuing problems with the well control system they would have been revealed before drilling either by Mr O'Brien's acceptance tests or by the normal process of testing and would have been resolved by the repair or replacement of parts as required in time for drilling to commence when the rig was otherwise ready to do so. Moreover in fact, as I find, the problems were not even of the extent they appeared to be
- No or no further strip down would have been sought. No problems which would have justified such an approach would have been revealed. Mr Ellis' opinion was that the combination of flushing and the replacement of two or three components thereafter was a satisfactory and appropriate way to address the contamination issue. I agree. The fluid was shown to be cleaner than specified and the system had been repeatedly functioned without problems
- I should perhaps add, as the issue was fully ventilated in the evidence, that even if, contrary to my findings, some further form of "strip down" had been required I am satisfied it would readily have established that the system was fit for its purpose without the necessity of including more than a limited number of sample items in the process. Mr Ellis thought it would take some 10 days to carry out the 50% procedure advocated by Mr Marymee. To strip the whole system he said was "a huge job" it could have been done but "I think you could have built a new system in less time". That is the measure of what was put forward by Amoco on this issue (despite the correspondence and documents) albeit the highest it is put in its Closing Submissions is that "if BAO had been prepared to acknowledge and address the problem, a hazop and 3-6 weeks work would have seen the end of it"
- It is an important part of Amoco's case that "BAO's attitude" to the contamination of the BOP system was itself repudiatory. I reject that totally. As at 19 January the only outstanding (and very recent) request from Amoco was to be found in CAR 56. That also, as I find, reflected the views of Mr Price expressed to Mr Quinn. It is also reflected in the 20 January letter (paragraph 172). Mr Mel Thomson of ADC in fact thought even that unnecessary (paragraph 457). It is framed as a request not an instruction. BAO was right to consider a strip down of the whole or any part of the system to be unnecessary. If Amoco had insisted on CAR 56 (or even a complete strip down) BAO would in fact have done it albeit whilst maintaining it was unnecessary (paragraph 457). Had BAO itself or Cameron thought it necessary or desirable I have no doubt BAO would have done it in any event. What BAO wrote about the well control system on 18 January (paragraphs 169 and 170) and 19 February (paragraph 201) was in my judgment both accurate and reasonable. Whilst Amoco rightly points out that both letters were written on the basis that the contract subsisted that does not mean that the reality of the context should be ignored, nor is it repudiatory to assert your case, but with or without being read in that context the terms of the letters cannot sensibly be characterised as Amoco seeks to characterise them now which is also in some contrast to how Amoco characterised the first letter at the time (paragraph 172)
PRS/TRAINING
Description
- A sufficient description of the PRS appears in Paragraph 20. The contract (Clause 6.2.1) required BAO to furnish properly trained, qualified and competent personnel in the numbers and classification set out in Appendix 3
Amoco's case
- Most of the allegations made by Amoco under the general heading of drill floor equipment were deleted in the course of the trial. Although Mr Barnes understandably presented the deletions as a consequence only of the need to concentrate on the areas of greatest importance and strength in Amoco's case, in reality they reflected the overwhelming burden of compelling evidence already given and evidence yet to come from some 15 further witnesses who, in the event, were not required to give evidence on the basis that their witnesses statements were accepted by Amoco. It also reflected the expert opinion of Mr Wentzell instructed on behalf of BAO
- The remaining allegations (Paragraph 4B(8) and (10) of the claim) made by Amoco are only that
"(8) (a) The automated pipe racking system could not access all the available rows.
(b) The pipe racking system also suffered from control problems. On 25 December 1998 the PRS ran back to the end of the track and crashed against the derrick crossbrace. At the Termination Date the PRS still required manual intervention to enable it to pick up pipe. Personnel from Varco (the supplier of the pipe racking system) continued to work on PRS control problems both before and after the Termination Date. The Plaintiff serves herewith at Appendix B a schedule setting out the work carried out on the PRS by Varco personnel between 22nd December 1998 and 28th June 1999 as indicative of the state of unreadiness and/or unfitness of the PRS.
(10) The rig crew had not been properly trained on the PRS as installed on the Drilling Unit or in using such equipment as part of an integrated system."
- Appendix B consisted of a list of extracts from service reports relating to the PRS together with the comments of Amoco's expert (Mr Ulliott) upon them
- However on 6 June 2001 (having informed Freshfields Bruckhaus Deringer on 18 May) Herbert Smith informed the Court that Amoco did not intend to call Mr Ulliott to give evidence and did not wish to cross-examine Mr Wentzell. On 15 June Herbert Smith wrote further to the Court to explain "how this affects our case under Paragraph 4(8)(b) of the Statement of Claim". The explanation was that
"we now accept that these (control issues) were minor and were resolved by 7th January (as stated in paragraph 7.16 of Wentzell's report). We do not rely on any problems after that date. Control problems remain in the case only for their effect on training. We would add (if it is not already clear) that this is now the main focus of the PRS case as a whole."
- What that comes to is that control problems with the PRS are only relied upon insofar as they can be shown to have made the PRS unavailable for training the crews and the crews were not in fact adequately trained to be able to use it. Amoco also accepts that what is left of the case under this heading cannot be relied upon to support any case under Clause 28 nor can it amount to a repudiatory breach of the contract. It is relied upon only, so it is submitted, as adding weight to the other issues in the latter context
The Evidence
- It was the evidence of Mr (Bryan) Quinn and really all those who had been in senior roles on the rig that whilst "minor glitches" remained in the PRS after 19 January there was nothing which would have prevented the rig from drilling. The evidence was wholly compelling. Belatedly Amoco have in effect accepted it, albeit doing so involved a departure from the burden of the song of at least some of the evidence adduced by Amoco and from what was said by Amoco in correspondence at the time, for example in the letters of 4 and 8 January (Paragraphs 154 and 156). In the context of training, the evidence is unchallenged (from Rowan/BAO and Varco) that Varco personnel were to be present on the rig when it first went to work drilling at Arbroath. The purpose was two-fold: to ensure that any further glitches could be addressed thoroughly and to assist the crews in the operation of the PRS should they need it. Mr Gray (and others), supported by the documents and the evidence of Mr Cowie, described the hands-on training given to the crews at Sabine Pass and instruction on a simulator at Varco's "school" attended by 7 crew members (5 of whom were drillers or assistant drillers) before the rig left America. Some of the crews were also involved in the commissioning and installation of the PRS at Sabine. Mr Snow was confident about using it in a drilling situation; Mr Gray was comfortable with the way the equipment was run and so were those who were to run it; primarily the Assistant Drillers. Mr Lanier (an experienced driller) described (rightly on the evidence) the crews as experienced and "hand-picked" who were excited at the prospect of getting started on something new. Mr Sykes in particular said much the same. Operation of the PRS was easy to learn; it was performing automatically functions with which the operators were wholly familiar and were basic to any drilling operation. The PRS was not rocket science; it was not complicated; it was repetitive; its main purpose was simply "tripping pipe", that is to take stands of pipe and hand them over to the well centre. As Mr Edwards put it, the PRS “is a tool assisting (the crew) to carry out what previously was a manual operation” done by them on every rig
- Mr Morton (another experienced driller) was also confident that he could operate the PRS properly after doing so whilst the rig was in Rotterdam. He said it was working correctly after it was repaired following the collision with the derrick on Christmas day. Mr Perkins and Mr Devaney said the same: "sure we had room to improve and we was going to improve". Mr Barker (an Assistant Driller) was 100% confident he could use the PRS safely and efficiently whilst acknowledging that speed would come with time and experience. Mr Walker (Amoco) is on record at 24 December as acknowledging at least the basic competence of the Drillers and Assistant Drillers to use the equipment.
- It was on 2 January, after the “diving board” had been levelled up, that it was found that the 5 slots on the 6 5/8" pipe racks furthest from the diving board were too narrow to allow pipe held by the PRS to reach them and that all 11 slots in row one of the 5" pipe racks (also the farthest from the diving board) could not be reached even with the maximum available arm extension. The diving board, as its name perhaps implies, is a walkway high up in the derrick running between the two sets of racks with a space on either side to permit access to the racks for pipe held either by the PRS or manually. Mr Quinn pointed out that the problem had not been put right to this day but the rig had successfully drilled a well and several abandonments in Canada. The unchallenged evidence is that the slots were in any event not necessary to enable the rig to drill in Arbroath because there was more than sufficient capacity (144 and 198 slots respectively each holding 90 feet of pipe) without them to store the maximum length of pipe required. Further, even if they, or at least the 11 5” slots, had been needed, all that was required was to send a man up the derrick to operate them manually
- Much of Amoco's case on training derived its focus from an internal record made by Varco of a meeting with Mr Snow and others on behalf of Rowan held on 2 February 1999 when the RGV was at Dundee. It is recorded against the title "Training" that
"The need to utilise the time available in the shipyard to ensure the correct operation of the equipment and the competence of the personnel was discussed. All present at the meeting recognised the need to implement a structured training programme for technical and operational personnel. Almost none of the present crews have previous experience of operating or maintaining semi-automated drilling systems. Various options were put up for discussion, the general feeling being that the mechanical and electrical personnel should be scheduled to attend Varco Systems courses for maintenance personnel held at Varco Montrose. Operational personnel would also attend a theoretical classroom based course and then undergo practical training and assessment on the safe operation of the equipment. Upon successful completion of the operational assessment the operators would be issued with a certificate of competency in safe operation of the equipment (driving licence). Varco will present a schedule and costing for training as outlined above."
- Mr Quinn's response to this document was typically restrained. Others were more forthright. Mr Quinn agreed that the crews had more to learn, but the 4 Assistant Drillers had each been sufficiently trained to drill and the best way to learn was from experience in using the PRS to drill. He said that after the contract had been terminated the rig was idle in Dundee, Varco was based nearby at Montrose and Rowan/BAO took the opportunity to train "everybody" including roughnecks, mechanics and electricians. Mr Marcom thought it was good business for Varco. The "driving licence" was a new scheme being introduced by Varco
- Mr McKenzie (Varco) said that at 19 January the Assistant Drillers were definitely capable of drilling a well using the PRS with Varco present on board. That is supported by an internal report he made on that day. Mr Fenton, also of Varco, said that whilst he would not say the Assistant Drillers were fully trained they were competent enough to drive the PRS and start drilling a well. Mr Hewitt confirmed the same. He said the training after 19 January was mainly for mechanical and electrical maintenance
- The evidence to which I have referred was wholly unshaken in cross-examination. It leaves me in no doubt at all that those whose job it would have been to operate the PRS had the RGV been allowed to drill at Arbroath at any time in January 1999 would have been comfortable and competent in doing so. The presence of Varco personnel would have provided added comfort. The intelligence and ability of the personnel was apparent. Their confidence was compelling and transparently honest if not understated
- There are other material considerations which should be recalled. First the contract (Appendix 10) contained specific provisions for a variance in the rig rate if the PRS did not operate or operate properly. In the sole case of the PRS the reduction was to be $5000 (in the other cases it was $15000) : see Paragraph 95. That is some measure of its significance in the context of the overall performance of the RGV
- On the evidence, the level of reduction reflects the fact that even without the PRS the crew on the rig were wholly familiar with and capable of moving drill pipe to where it was intended to go, and torquing it up or untorquing it as required. The main advantages of automation were repeatability and (to a limited extent) safety on the drill floor rather than any time saving in drilling
- Even in the context of escalating disputes about the rig rate and delay and the complaints about the drill floor equipment made generally in the letters of 26 December (paragraph 152) and 4 January (paragraph 154) and made generally about the PRS in the letter of 8 January (paragraph 156) Amoco authorised the rig move to Arbroath on 7 January. Belatedly (paragraph 496) Amoco has also accepted that issues about the operability of the PRS were minor and in any event had been resolved by 7 January. Despite Mr Gillis' protestations that it would have been unsafe to train or to complete the training of the rig crews whilst the rig was jacked up alongside the Arbroath platform the fact is that he authorised the rig to go to Arbroath, weather permitting, when he knew or certainly had the means of knowing the state of the crew training at the time
Expert Opinions
- There was not, in view of Amoco’s concessions, much room for expert evidence on this subject. The burden of Mr Munter’s evidence was that because of the access problems, which were due to installation errors, the PRS had not been “commissioned” at 19 January because it was not mechanically complete. I found that neither relevant nor realistic. The experts were agreed that in terms of time there was little to choose between manual and automated operation. Nor was there any challenge to the evidence that for Arbroath there was no need to use the 16 slots which could not be accessed by automated operation. Mr Munter’s concerns about the safety of manual operation and generally were, in my judgment, greatly exaggerated and Mr Edwards' views much more realistic and to be preferred. Mr Bartle and Mr Poss formally agreed that the access problem “was not a significant issue”. They were plainly right on the evidence to do so. Yet Mr Munter would not agree because “you have short-changed the man that is renting the rig” and he stood by the opinion expressed in his second report that the rig was not fit for drilling operations on 19 January because of the access problems alone. That, I think, was to seek to sustain the obviously unsustainable. Both Mr Bartle and Mr Poss said they would probably have accepted Varco's views about the PRS and training. It was not even suggested to relevant witnesses that the access problem had any effect on training. Indeed both Mr Quigley and Mr Bartle accepted that it did not. Mr Bartle did suggest that because the HSE would have required a documented record or demonstration of the training so would he. But the COSC (approved by the HSE, paragraph 119), simply provided that Rowan would ensure the crews were trained by the vendors and by hands-on-training as well as by their presence during installation and commissioning and running the equipment. Mr Bartle was adopting a fastidious approach which the COSC did not. There is no suggestion in any document or evidence that the HSE had any concern about Rowan/BAO's ability to assess crew competence
Conclusion
- In my judgment there is no substance at all to Amoco's case concerning the PRS and Training
OVERALL CONCLUSIONS
- It is Amoco's basic case that at 19 January the RGV was unsafe and unusable because of two "show stoppers"
i) The history of events concerning the jacking system which "called for a full investigation in order to determine the integrity of the system and the risks of jacking alongside the platform" and BAO's response to and explanations of the events which it is alleged was inadequate and inaccurate insofar as it was communicated to Amoco; and
ii) The well control system which was contaminated and could not be relied upon to prevent a blow-out that could have led to disaster. It should have been stripped down and BAO refused to do so.
- It is also Amoco's case that the delays and lack of crew training are part of the background which made the show stoppers "all the more serious in their effects, and contributed to Amoco's loss of confidence in BAO"
- Amoco's case fails on the facts at every point. But for termination, the jacking system would readily have been established to be (as it was) both safe for Arbroath and indeed capable of meeting its performance criteria had that been thought material. In all probability the jacking system would have been proved by performance. So also the well control system, at least when first required for use, would have been shown to be fully and safely functional. Indeed that was the case on 19 January save that the pressure control problems manifested at items 66 and 67 needed to be resolved "prior to spud". The rig would have passed Mr O'Brien's tests or the tests would have revealed any continuing problems which could readily have been addressed. In no relevant sense did BAO behave "intransigently" nor refuse to carry out work nor provide inaccurate information. Whilst Amoco's case of loss of confidence is I think contrived it was certainly not justified. The complaints about delay and crew training have no contractual relevance nor do they have any place in the context of repudiation or otherwise. Even had I accepted Amoco's factual and expert case in its entirety it would have been reasonably arguable that no repudiation had occurred. That case at its highest is that some 6 weeks to 3 months would have been required to put the matters now complained of right and to establish that the RGV was (as it was) safe and fit for drilling at Arbroath. Moreover there was good reason to believe that such would be the outcome of any further work. The contamination would have been addressed as Amoco claim it should have been. The jacking system would have been expected to pass and would have passed any reasonable test. In the context of a 12 month contractual commitment but a drilling opportunity limited to some 7½ months it must I think be very questionable whether even such circumstances would meet the test of repudiation. Be that as it may on the facts as I find them to be such a case is of course hopeless
- There was in addition no right to terminate the contract under Clauses 23.1.1 and 28.1.d) at least because
i) As regards the well control system BAO did not fail to conduct its operations in a diligent skilful and workmanlike manner nor could Amoco reasonably have concluded that it had;
ii) No adequate notice was given by Amoco of any such failures now relied upon;
iii) BAO did not in any event fail or refuse to remedy any failure or to proceed diligently to do so within the meaning of the Clauses.
- There was no right to terminate under Clause 28.1.b) at least because
i) The 30 day period had not expired;
ii) No relevant breakdown had occurred;
iii) In any event the well control equipment was not broken down at 19 January following the successful test on 16 January.
REMEDIES
- As I have found that Amoco was not entitled to terminate the contract it is accepted that there is no defence to the counterclaims for the second half of the mobilisation fee and the costs of mobilisation from Rotterdam to Arbroath. I have also rejected Amoco's claim to apply the "no remuneration" provision following the "failure to test" of the BOP on 24 December as a matter of construction of Clause 5.4 (Paragraphs 48 to 50). The tests which Amoco alleges "failed" did not result in the suspension of operations. Amoco also submits that "if it is held that the rig was not in a fit state on and from arrival in Rotterdam but that (for some reason) Amoco was not entitled to terminate" it should not have to pay any day rate for "the period until the rig was ready.
- The basis for this unusual if not remarkable submission is the decision in Sonat v Amerada Hess [1988] 1 Lloyds Rep 145, the fact that the rig rate was expressed to be payable for services (Clause 5, paragraph 42) and the asserted consequence that if the rig was not fit to perform those services then the rate should not be paid. The submission ignores the fact that the parties agreed a detailed and comprehensive contract to govern their relationship which provides specifically for the abatement of the rig rate (including in effect a nil rate) and for Amoco to terminate the contract which for the purposes of the submission must be assumed not to assist Amoco. It also ignores the fact that Amoco does not claim to have suffered any damage as a result of any breach of contract (let alone negligence) by BAO on the basis (as must again be assumed for this purpose) that it was not entitled to terminate the contract. Both the circumstances and the relevant contract terms were very different in Sonat, and in my judgment substantively so. In any event I have in effect held that at all relevant times BAO was able to provide and perform the substantial services required of it under the contract. The rig was never "out of action" in the sense relied upon in Amoco's submission or which occurred in Sonat as a result of an explosion
- Amoco does not contend (rightly in my judgment) that BAO is otherwise not entitled to the "standby" rate for which it claims for (in effect) the period of one year. I shall leave it to the parties on receipt of this judgment in draft to consider the terms of the precise orders to which it gives rise and, if they cannot be agreed, will consider the issues later when the judgment is delivered
