Maersk Oil UK Ltd (Formerly Kerr-McGee Oil (UK) Plc) v Dresser-Rand (UK) Ltd

During the 1990s, KMG was an independent oil company whose business included offshore oil exploration and production of oil and gas.  It was the operator of the Janice oilfield.  In 2005 KMG was acquired by Maersk Oil (UK) Ltd. (Maersk), who since that time have operated the oilfield.

Although Maersk is the nominal claimant, throughout the trial, and in witness statements, reports and submissions the parties have found it convenient to refer to the claimant as KMG.  I adopt this nomenclature and will refer to the claimant throughout as KMG.

Dresser-Rand (UK) Ltd (DR) is part of the worldwide group that supplies compression facilities to the oil and gas industry.

The Janice Field is located in the North Sea at a depth of 180 m, approximately 175 miles south east of Aberdeen.  The Janice oil is produced from subsea wells tied back to a floating production unit, which is a semi submersible vessel, known as Janice A.

Janice oil is exported from the floating production unit by pipeline to the Southern Y of J Block spur, and then through Norpipe for redelivery at Teeside.

Associated gas is also produced with the oil and exported from the platform.

DR sold KMG a compression facility for use on Janice A. Under the terms of the agreement DR undertook to provide a complete compression process and mechanical design, and to fabricate the package.  The original contract price agreed was £3,133,400.

KMG contends that the equipment supplied by DR was defective and dangerous, and once installed offshore, and in operation, the compressors caused excessive vibration, giving rise to component fatigue, gas escapes, and the production of damaging liquids in the compressor trains comprising the package.

KMG have formulated a claim of £7,927,487 by way of damages subject to the contractual limit of liability. DR counterclaim £166,000.

INTRODUCTION

The design, construction and installation of the equipment onto the Janice A platform was a fast track development.  The time between the award of the compression package contract and delivery was only eight months.

The development of the Janice Field by KMG and its joint venturers involved converting a six legged semi submersible accommodation unit into a floating production unit (FPU) forming the Janice Alpha platform which was also a semi-submersible. 

There were delays and commissioning did not occur until April of 1999.

The conversion inevitably gave rise to constraints of space for the process equipment erected on it, and constraints arising from the structural rigidity of the platform under the load conditions different from those appropriate for accommodation use.

The hull or level 1 of the topside structure is enclosed and houses the living quarters, galley, games room, sickbay, as well as the engine room, workshops and electrical switch rooms.  Above this and towards the rear is located the production deck, where all the production equipment, including the two gas compressor trains are located.  Forward of the production area, and protected by a blast wall are two levels of offices, with the bridge above. 

The main electrical generators, powered by turbines are located on the starboard side of the platform and are fuelled either by fuel gas or marine diesel.

The sub sea well heads are connected to the platform by flexible riser pipes.  These flexible pipes convey the well fluids, comprising oil, associated gas and produce water from the reservoir, into the production manifold, and thence into the production separator, known as the first stage separator at the rear of the platform. 

The production separator is designed for 60,000 barrels of oil per day and is a three phase separator.  Gas is separated from the oil and water liquid phases in the vapour space of the vessel whilst the oil and water separate by gravity separation in the central portion of the vessel.

There is also the facility to pass each well flow into a test separator, which is a smaller version of the production separator designed to separate the oil and gas and water phases.  This facility permits the flows from each well, to be measured periodically and enables KMG to analyse and manage the performance of the reservoir.

The stabilised crude oil is pumped from the production separator using two booster oil pumps.  The oil flow is metered for tax purposes and cooled before the oil pressure is further increased using two oil export pumps.  These are used to export the oil from the platform through a 14 inch diameter pipeline to its ultimate landfall at Teeside.

The produced water from production and test separators is routed by hydro-cyclones where residual oil is removed from the water and then discharged to the sea via a final degasser and separator.

The associated gas from the production and test separators is routed to gas compressor trains A and B. Each train comprises a pre-treatment section and a three stage reciprocating compressor.  The compressors are used to compress the gas for export, or as gas lift, where it is reduced to individual wells to enhance oil production.  The gas is exported via the 12 inch diameter gas pipeline to the Judy platform.

THE CONTRACT 

The Contract was made 19th of August 1997 and is entitled Janice Purchasing Contract PMD-C-0040.  It is made up of a series of documents.

Clause 40 of section 1 provides:-

The work undertaken by the defendant Contractor is defined in article 1 section 1:

“ Work shall mean the goods, equipment, plant and machinery, materials, documentation, drawing manuals work and/or services specified in the scope of work, including such particulars and details not expressly defined, but which are necessary and customarily provided for in the performance of the obligations and liabilities as described herein”

Clause 1.0 of section 2, states: 

“Introduction

Clause 11.1 section 1 Purchasing Contract provides:-

Clauses 11.2 and 11.3, of section 1 contain further warranties:-

These provisions are clearly complementary to one another, and part of the same scheme, once defective work or parts are replaced the two years warranty and guarantee starts again, in respect of the new parts or work.

In the event that the contractor does not comply with the obligations to make good the defects or work part 11.6 is triggered.

On 17^th April 1998 Amendment No 1 to Contract PMD-C-0040 Janice Project was agreed.

These are the relevant parts:

“NOW; it is hereby agreed that in consideration of the announcement of the forthcoming closure of the manufacturing operations of the Contractor at Wythenshawe, and in recognition of both the Contractor’s and the Company's desire to complete the compressor packages  in accordance with the Contract conditions the Contract shall be amended as set out below:

1(a) In recognition of the Company's commercial position, and the contractor’s inability to complete the contract in accordance with the original delivery date, as specified in the Contract the Company will accept delivery FCA Wythenshawe of the incomplete compressor packages on the 18th and 19th of April 1998 and transport them at the Company's cost to Aker McNulty Ltd’s Module yard. (The Module yard).  The Company and the Contractor shall prior  to the delivery of the compressor packages agree a list which shall set out the outstanding work under the Contract, that is to be performed at the Module yard, and all outstanding ancillary equipment, and the material to be supplied by the Contractor.

All other miscellaneous equipment and materials leaving Wythenshawe after the main consignments of the two compressor packages shall be transported C I F Module yard ( not offloaded) at the Contractor's expense.  The timing of the delivery of the miscellaneous equipment and materials to the Module yard shall be agreed between the Company and the Contractor, ancillary equipment and materials returned to Wythenshawe shall also be transported and insured at the Contractor's expense.

(b)

On a mutually agreed date after delivery of the compressor packages to the Company the Contractor shall dispatch to the Module yard the required supervisory personnel, who will assist and advise the Company to facilitate the completion of the compressor packages.  The Contractor shall advise the name of the leading supervisory personnel.  All expenses including wages travel, accommodation and subsistence associated to the provision of these personnel shall be paid by the Contractor.

(c), The Company shall provide personnel, plant and equipment, sufficient for the completion of the compressor packages at the Module yard and shall advise the Contractor of these personnel with job title and hourly rat together with the name of the prime contact and sample of the module yard typical worksheet. The Company's prime contact shall present on a timely basis time booked against work performed on the compressor packages to the Contractor’s leading supervisor and the Company's prime contact shall also provide on a timely basic basis details of any other costs incurred by the company in relation to this work.

(d)

Upon Mechanical Acceptance both parties shall document the agreed date of reaching this milestone, and at this time, the Contractor’s site personnel shall be released from the work”

Thereafter followed various payment provisions, which are relevant in relation to limitation of liability under clause 42 infra.

PRELIMINARIES

There are a number of preliminary contractual issues that must be addressed before general questions of liability can be addressed.

Extent of Remedies available

The first issue is whether the Claimant has any right to claim damages, other than such sums as it may be entitled to recover under the terms of clause 11.6 of the Purchasing Contract.

The resolution of this issue depends upon the true construction and meaning of the Purchasing Contract.

DR contends that it does not have such a right since clause 11.1 contains a warranty and guarantee that bites upon “the work” as defined in clause 11.1.

The first sentence of clause 11.2 contains an additional guarantee and warranty in respect of the engineering , workmanship, and all materials and equipment fabricated or provided by any of its subcontractors, whether it has had the benefit of approval of the designs or not.

This guarantee and warranty exists for a  period 24 months from the date of the Inspection Release Note  referred to in clause 3 3.1.

DR contends that the phrase in the second sentence of clause 11.2 “or failure by the Contractor to comply with any obligation, express or implied under contract” comprehends and applies to the guarantee and warranty referred to in clause 11.1.  Accordingly, in clause 11.5 reference to “the foregoing warranty” as being additional and not in substitution of warranties and other rights provided at law refers to those warranties set out in clauses 11.1 to 11.4 and cannot refer to other remedies, such as an implied warranty that the works would be reasonably fit for the purpose, or the general right at common law to damages for breach of contract, because clause 11.9 contains the rider disclaiming all other remedies expressed or implied.

The first part of clause 11.9 deals with very particular situations where equipment or parts are defective by reason of normal wear and tear or arise from a failure to resist erosive or corrosive action, or by reason of the Company's failure to properly store. In these events there is no liability upon the contractor.

DR further contends that clause 42 of the Purchasing Contract also excludes any general Common law right that KMG might otherwise have to sue DR for damages for breach of contract.  It is submitted that the scheme under article 11.6 exhaustively provides the remedies available for breach of warranty under clause 11 by DR and that KMG are deprived of the additional general right to claim damages for such breaches.

KMG relies upon passages in Lewinson The Interpretation of Contracts 4^th Edn in paragraphs 397-399 as accurately stating the approach to be followed by the Court.

“  Clear words are necessary before a court will hold that a contract has taken away any right or remedy which one of the parties would have had at common law” and further that “ where a contract does purport to exclude a remedy, that exclusion clause should be narrowly construed.”

I accept that this is the proper approach to construction in this case.

The wording of clause 11.5 by making reference to the foregoing warranty clearly refers to clause 11.2.and that far from limiting remedies purports to add to the menu of remedies by saving other warranties or rights provided by law.

This must be considered in the light of clause 42.  The remedies of the Company as set forth herein are exclusive “as stated”.   The qualifying words, “as stated” are crucial submit KMG, because, by way of example, the remedy of liquidated damages for delay is expressly stated to be an exclusive remedy of the Company and the sole liability of the Contractor.  Such is the rubric adopted in article 4 and referred to in article 42.

In my judgment, the interpretation contended for by KMG is correct. It gives sensible meaning to each of the clauses chosen by the parties, and it accords with commercial sense and a purposive construction of the contract as a whole.

I do not consider that the wording in clause 11.9 was intended by the parties to contradict the clear wording of clauses 11.5 and clause 42.  Rather, with reference to the subject of this clause, the equipment or parts furnished by the Contractor, it serves to emphasise the disclaimer in the first part of clause 11.9.

Notices

The next issue is whether it is necessary as a condition precedent to any entitlement by the Claimant to recovery, that the written notice provisions under the contract should have been complied with.

Clause 38 NOTICES provides that:-

The way in which KMG put its claim against DR appears in Paragraph 100 of the Amended Particulars of Claim: 

“The Claimant seeks to recover the above sums, £7,927,487.32.  Subject to the limit of liability under the contract

DR’s  case as to paragraphs 2 and 3 is that KMG has no such entitlement and is limited to any costs it can recover under and in accordance with the terms of article 11.6 of section 1 of the Purchasing  Contract,  is rejected.

As to the monies that KMG claim are payable under article 11.6 DR contends that in order for KMG to recover the total cost of its own personnel or of other contractors making good all defects caused by DR's breach of either or both of the guarantees or warranties set out in clauses 1.1 and the first sentence of clause 1.2, of section 1 of the Purchasing Contract, and which DR has been unable or unwilling to proceed within a reasonable time, KMG is required to give DR two kinds of notification

First, it must give the” requirement notification” under the terms of clause 11.1.  The relevant part of which refers to “work, repair replacement, making good defects….. attributable to the Contractor ….  apparent within the said 24 month period as may be required by the Company”.

Clause 11.6 similarly, refers to” the failure or unwillingness of the Contractor to proceed with corrective work in a reasonable time as aforesaid, required by the Company”.

Secondly, it must have given the written notification referred to in clause 11.6

“ if no or insufficient sums are available for full offset by the Company then the Contractor, upon receipt of the Company’s written notice of the Contractor's obligations hereunder shall promptly remit to the Company all sums due and owing pursuant to the terms of this clause”

Where there is insufficient available for offset the written notification of the Contractor's obligations is a necessary precondition to trigger the Contractor's obligation to make prompt payment.

It is desirable that any requirement under clause 11 is reduced to writing but there is no provision requiring that KMG must serve a notice complying with the provisions of article 38 or that they are bound to give notice in a particular way.

A requirement may be oral or evidenced in writing. Each claim based upon the two-year guarantee provisions of article 11 must be considered case-by-case to see whether any adequate requirement within article 11 was made and if so, when such requirement was made.  I reject the defendant's submission that such requirements had to be in writing or in any particular form.

This does not detract from the commonsense considerations that prudence would require some note or memorandum setting out the precise respects in which DR was in breach of any guarantee or warranty, stating what work, repair, rectification, replacement or making good of defects was required or caused by DR’s breaches, and the date and means by which KMG gave DR notification required by clause 11.2 together with a summary of its contents.

Such notification would also briefly refer to facts and matters relied upon by KMG in support of its contention, if appropriate, that DR was unwilling to proceed with corrective action within a reasonable time, following KMG's requirement and would contain particulars of the necessary corrective work that KMG personnel or others had carried out and the basis of costing.

Fitness for purpose/Suitable for purpose

By paragraph 27 of the Amended Particulars of Claim KMG  plead that DR’s contractual obligation under article 11.1 of the Purchasing Contract was provide a compressor, which was ”fit for its purpose”.

The expressions “fit” and “suitable” in the context of this case are interchangeable and are synonymous. Neither is qualified by adjectives such as “reasonably”

Warranty under Clause 8.3

Article 8.3, of section 1, provides that:-

This provision imports a warranty by DR that it was an experienced and competent contractor, and that it had or would obtain a full understanding and knowledge of the extent of the work.  As such, it is not controversial.

Conflicts between Purchasing Contract and Functional Specification

What is controversial between the parties are questions concerning the design obligation that the contract imposed upon the defendant.  The parties have identified an issue, which is relevant to this.

Was DR obliged to comply with the requirements of the functional specification, where such requirements would place it in breach of its obligations set out in article 11.1 of the Purchasing Contract that it shall act in accordance with good and sound industry practices, and/or would not allow the work can be performed in a first-class and workmanlike manner?

Article 40 of Section 1 sets out a clear order of precedence.  Whilst clause 1 of the Purchasing Contract states that:

…. this scope of supply relates to the Janice Field development project and is specific to the following documents with which the contractor shall fully comply

It is clear that where appropriate, DR was to comply with the requirements of the Functional Specification particularised in clause 1 of section 2, as well as the documents forming part of Part 6 of the Purchasing Contract, but the paramount obligation was that imposed by the warranty under article 11.1 of Section 1.

By way of example, if compliance with the Functional Specification or any standard referred to in it, gave rise to health and safety concerns  a design approach, which complied with health and safety requirements would be warranted even if this was at variance with such specification or standard.

Similarly, if the design approach prescribed in the Functional specification gave rise to a design which did not permit the works to be performed in a first-class workmanlike manner then the obligation would be to comply with the overriding obligations contained in the warranty in clause 11 of Section 1.

Clearly this may have cost consequences if a price is based upon requirements forming part of functional specification and compliance with the warranty requirements of section 1 is more expensive. 

Compliance with any standard referred to in the Functional Specification may nonetheless give rise to a breach of warranty under article 11. Whilst the court will strive to give a purposive construction to the contract as a whole, where there are clear and unambiguous terms chosen by the parties and an agreed order of precedence that provision should not be ignored in an attempt to reconcile the seemingly irreconcilable or to rescue a party from his unforeseen commercial disadvantage.

The warranty period-claims under clause 11.6

A number of issues arise in relation to KMG’s claims under this head.  First, when did the warranty period commence. Second, is there any prescribed mechanism for informing the contractor of a warranty claim?

The warranty period prescribed in articles 11.2 and 11.3 is a 24 month period.  Article 30.3.1 provides that an Inspection Release Note is issued to the Contractor by the Company, if an inspection takes place, or a  certified copy of the test result issued by the Contractor to the Company, where an inspection is not opted for,  shall be the commencement period of the warranty period

Clause 30.3.1.2 provides that the Contractor shall not commence preparation for transportation until the Company has issued the inspection release note.

The fabrication and assembly of the two compressor trains was envisaged by the parties only to have been carried out at the Wythenshawe factory of DR before transportation to the North Sea.

DR decided to close the factory before completion of fabrication and assembly and KMG agreed to accept delivery of the incomplete compressor packages on the 18^th/ 19th of April 1998 and agreed to transport them to the Aker McNulty, module yard at DR's expense for completion.

Prior to this, the parties had agreed a list of outstanding works to be performed at the module yard and of the outstanding ancillary equipment remaining to be supplied by the Contractor.

KMG agreed to provide personnel, plant and equipment, sufficient for the completion of the compressor packages under the supervision of staff provided by DR.

Clause 1(d) Amendment number 1/17 April 1998 provides that

“On mechanical acceptance, both parties shall document the agreed date of meeting this milestone, and at this time, the Contractor’s site personnel shall be released from the work.”

This provision reflected the changed circumstances and supersedes the article 30.3.1 trigger for the commencement of the warranty period.  The parties are in dispute as to the date on which the warranty period commenced.

KMG’s primary contention is that the warranty period commenced on the 10th of January 1999 when they issued DR with “a mechanical acceptance certificate” on that date.  There is no reference to such a certificate in either the Contract or the Amendment.

This document states that apart from a punch list containing items outside DR’s scope of supply, the machines were completed by the 20th of August of 1998.  This unilateral document does not support KMG’s contention.

Their secondary case is that the warranty period commenced on the 7th of March 1999 when a further “mechanical acceptance” certificate was issued by them.  This document records the mechanical completion of the compressors for commissioning on the Janice A platform following satisfactory mechanical erection, after their transportation to the platform.

Mechanical completion in the sense that the items of equipment removed and shipped from the Aker McNulty yard offshore had been satisfactorily reinstalled on the platform.

Neither of these documents came into being at the instance of the parties jointly intended to evidence mechanical acceptance for the purposes of clause 1.(d) of the Amendment.

DR’s case is that mechanical acceptance was reached on the 19th of June 1998.  By this date, the outstanding works, which were identified in the punch list, which accompanied the Inspection Release Notes of the 17th of April 1998 when the compressors left the defendant’s Wythenshawe factory, were completed.

It is evident that by the 19th of June 1998 all of DR’s personnel had been released, and the scope of their work was complete. The contemporaneous documentation relied upon by DR as evidencing mechanical acceptance for the purposes of Amendment 1(d) is equivocal.

Mr Horne DR’s project manager referred to a fax sent to KMG on the 30th of July 1998,   in which he asserted his belief that:-

“ … the DR equipment has been effectively mechanically  complete in an ex works complete status for  several weeks now”

Agreed minutes of a meeting held at Wythenshawe on 4^th August 1998 record that the compressor units were considered as essentially complete on-site by the 19th of June, with the exception of a punch list. Mr Horne said that the punch list attached to these minutes confirmed that all of the DR actions were either complete or passed on to commissioning activity with the exception of two items identified as DR/Aker McNulty joint actions.

On 9 July 1998 DR sent an invoice to KMG under a covering letter.  It is evident that the DR regarded mechanical acceptance as having been achieved but in their discussions with Aker McNulty it had come to light that there would be some delay in completing the paperwork to correctly record the completion of the milestone. The KMG recipients of that letter were not persuaded that mechanical acceptance was complete. 

The letter is endorsed ” have accepted mechanical completion by Dresser-when will the paperwork be complete - if they have not reached mechanical completion what is the list of work to complete to achieve”.

The invoice was not paid until 7th December 1998.

On 11th August 1998 KMG advised the Health and Safety Executive that the Janice A marine, utility and process systems would be mechanically complete before sailaway and that Janice A would sail from the Tyne during the week commencing 21st of August 1998.

By this date, I accept the “no load and nitrogen test” had been successfully accomplished, for which mechanical completion was a prerequisite.

I am satisfied that mechanical acceptance was completely achieved by 11th August 1998, and that the contemporaneous documentation emanating from both parties evidences their recognition of this as does the earlier release of the DR personnel from the Aker McNulty yard.

Cap on Damages

There is an issue between the parties as to the level of the cap on damages.

Clause 42 provides;-

“in any event, the total liability by the Contractor with respect to any claims under this contract, whether based in contract,  indemnity,  tort, ( including negligence and strict liability), or otherwise, shall not exceed the purchase price of the unit of equipment or part(s) upon which such liability is based.”

The purchase price, at the time of contracting was £3,133,400.00.

By virtue of the three variation orders and extra costs incurred by DR at Aker McNulty's yard set out in DR’s letter of the 29th of September 1998.  The total purchase price was increased to £3,259,263.00. DR were unable to complete the works on the compressors in their own factory at Wythenshawe, and the parties agreed that the work would be completed at the Aker McNulty yard in accordance with the provisions of Amendment number 1.

The bulk of the completion works were carried out by KMG at a cost of £326,686.  Had they not done this work it would have been the obligation of DR to complete the works.

Payment of the purchase price was agreed by the parties could be in stages against certain milestones.  The payments were to be 10 % 30% 30% 25% and 5%.

As to the 25% payment it was agreed that:-  

In the event the fourth milestone payment was subject to the deductible sum of £326,686.00.

DR contend that the purchase price of £3,259,263.00 was reduced by this amount to £2,933,576.00, and thus for the purposes of article 42 the capped liability of DR is £2,932,576.00.

KMG contend that the purchase price remained the same but that in relation to the 25% instalment part was paid in kind, namely the agreed cost of the works undertaken by the KMG in DR’s stead.

I accept DR’s argument.  The purchase price and value given remained the same to KMG.  The manner of paying the 25% instalment of the purchase price was varied and reflects KMG’s having given monies worth of an agreed value as part of the purchase price.

DR further contend that the cap clause has the effect of ousting the court's discretion to award interest on any damages that may be awarded should such damages equal or exceed the cap.

I reject that argument.  The cap limits liabilities under the contract.  There is no contractual right to interest.  Interest on damages is statutory and discretionary.

Compromise

DR assert that KMG’s claim with respect to the modifications required to the condensate drain from the dehydrator inlet knockout drum MV 0804(Scott schedule 3.4) was settled on the 17th of September of 1999.

DR’s case is that significant vibration issues(Scott schedule 1 .0), matters relating to drainage from piping low points(Scott schedule 3.5), difficulties relating to the2/3 stage recycle valves(Scott schedule 4 .1), matters relating to parallel operation of trains A and B(Scott schedule 5.0), were also settled on  8th February 2000. 

DR contends that these matters, together with others not the subject of the present claims were settled by KMG agreeing to accept a credit of £37,500.

These claims, of course, fell within the warranty period and were matters notified to DR of which they took full cognizance. The value of the claims presently pursued, which DR contend were settled are valued in excess of £1 million.

It is evident that DR were aware of the particular concerns expressed by KMG as to vibration and liquid process issues.  To that end, DR personnel had been deployed offshore, whilst commissioning went on, and problems were being encountered.  They made charges for their services and attendance offshore. These charges were contentious.

On the 17th of September of 1999, there was a meeting at KMG's office in Aberdeen where current items of concern on the project were discussed.  The meeting at was in two parts, first a technical meeting attended by Mr Ian Sharp and Dr John Krska of KMG and DR personnel, including Mr Whittle.  A wide range of technical difficulties related to potential warranty claims were tabled and discussed.

The second meeting was a commercial meeting attended by Messrs Troakes, Dickinson and Mackay representing KMG, and Messrs Whittle, O'Connor and Greenhalgh for DR. It is evident that nothing was agreed at this meeting save an agreed approach to future discussions on the outstanding matters.

The minute of the first meeting recorded that the purpose of the first meeting was to discuss potential warranty issues arising under the contract “ in an open and frank discussion with a view to achieving a mutually satisfactory financial agreement in respect of the deficiencies” the minute went on to record that “….. a further meeting shall be convened to discuss the outstanding issues, which are awaiting the results of a technical study survey or the like.  For issues with an agreed resolution, DR to propose credit values and KM to calculate the value of charges. These shall be issued to the respective parties for review and comment.  KM advised in general, that problems with the Gas Compressor Package had prolonged the commissioning phase of the project”.

The date of the second meeting was 8 February 2000.  The agenda for that meeting was recorded in an e-mail from Mr Troakes of KMG to Mr Whittle of DR. Item 2 “DR to give response to meeting of 17th September on Offshore Man-hour reductions”.

There was no mention of response to any other identified matters.

It is clear that following the September meeting the parties hoped for some resolution of the potential warranty claims.  The agenda for the February meeting made no reference to responses to the substantive warranty concerns which were agreed by the parties, at the September meeting, to be subject to further technical appraisal.

I am satisfied, having heard the evidence of Mr Troakes, Mr Whittle, Mr Wilson and Mr O'Connor, that there was some measure of agreement achieved at the February meeting.  I reject the contention of DR that their intention to leave the February meeting having closed out all the issues raised by  KMG, including technical issues, warranty issues, various invoices and other sums due to DR from KMG, was fulfilled.  There may have been an earnest hope but the evidence considered objectively, does not support this contention by DR.

DR’s Mr Whittle attended the meetings in September, and in February.  He explained how the £37,500 sum agreed in February, was calculated by reference to DR’s claimed offshore man hours.  He accepted that the agenda of the meeting was set out in Mr Troakes’s e-mail.  Mr Swann of DR accepted that his recollection of the meeting 8 February, was limited, and that he took no part in the discussions prior to the meeting as to any specific outstanding issues.  The evidence of Mr O'Connor I did not find impressive.  In his statement, he wrongly claimed that Mr Greenhalgh and Mr Dickinson of KMG, attended the meeting 8 February.

Mr Troakes gave evidence which I accepted, that the purpose of the September meeting was primarily to discuss the problems that the KMG project team considered were the responsibility of DR. Hence they looked to the hours charged by DR for remedying the defects with the compressor which KMG considered they should not pay for.  Mr Troakes made the point, which has not been gainsaid by any other witness, that the considerable costs that KMG had incurred with IDEAS and IGL were never discussed at either meeting.

These third-party costs would of course have been highly material to any full and final settlement of warranty claims.

Both Mr Whittle and Mr O'Connor accepted that, as at 8 February 2000 they were not aware, whether or not, any of the appraisals and studies mentioned at the meeting on17th of September of 1999 had been carried out.

The contemporaneous documentation is instructive as to what was discussed in the February meeting.  First, the agenda indicated that DR's response would be in terms of offshore man hours; second Mr.O'Connor's single page manuscript note suggests that the £37,500 figure was referable to man hours.

Had there been a compromise, as now contended for by DR, it is inconceivable that when faced with a claim for remedial costs arising from vibration and parallel operation problems, that in their letter 14th of April 2000, they did not assert that these matters had been compromised in a binding agreement on 8 February 2000.

I reject DR’s submission that there was any such compromise as contended for.

I accept that at the 17th of September 1999 meeting the parties discussed the liquid recycle problems arising out of the sizing of components relating to the condensate recycle line.

Item 3 of the minute refers to:-

“liquid recycle problems-NRV,LCV and pipework sizing…….DR stated that according to design calculations this system should work as sized.  To enable the system to work valve internals were removed.  Nobody has a clear understanding of why there is a problem.  KMG agreed that this item could be removed from the list “(emphasis supplied).”

It is evident that the problem continued as recorded in the minute. DR contends that an estoppel by convention, arising from the fact that the minute states that the matter could be removed from the list of warranty issues.

There is no evidence that either party conducted its affairs on the assumption that this issue was resolved, or the claim abandoned by KMG. There is no evidence that both parties thereafter acted upon the assumption that this issue would never be pursued or that unfairness would result.  The fact that the parties agreed to remove the item from the list of warranty items is not unambiguous evidence of an accord or convention.  The basis of its removal was the understanding that there was a problem and no solution was then apparent in a context where further technical appraisals were being pursued by the parties.

I reject the submission that there was an estoppel by convention, arising from the agreed omission from the list, of this item.

INTRODUCTION PROCESS ISSUES

The Process Flow Equipment

In the pre-treatment sections, associated gas is produced in the production separator MV 0801 and test separators at pressures between 14 and 18 barg.  Each separator has a pressure control on its gas outlet, to enable the pressure to be controlled above the suction pressure required to each compressor.  Two gas streams from the separators co-mingle downstream of their pressure control valves and the flow rate is measured before the flow is split to feed the gas compression trains MS-0801A and B.

The design case flow rates for each compressor is between 21 and 24 millions of standard cubic feet per day (mmscfd).  The gas flow first enters the pre-treatment section of each compression train, which is designed to cool the gas to 30° centigrade to condense some liquids and remove them from the gas stream before the gas enters the first stage of compression.  Any excess gas over and above that which can be handled by the compressor is routed to the high-pressure flare. 

Thus, when a gas compressor is out of service gas can still pass through the pre-treatment section and be flared, enabling excess gas to be disposed of safely and thereby maintaining oil production, albeit at a reduced rate.

The gas entering each pre-treatment section at the beginning of each gas compressor train first passes through the compressor pre-coolers (MH-0801A) which cools the gas to 30° C.  This gas, with some liquid leaves the pre-coolers and enters the first stage suction scrubbers (MV-0801A) where the liquid is removed from the gas phase, as it passes through a mist pad or mat installed within the vessel. Liquid droplets coalesce and fall to the bottom of the vessel, where they are collected under level control before being pumped back to the production manifold using the condensate pump (MP-0801A).  In the event of an extra high level, occurring in the vessel there is an alarm and automatic trip of the compressor train.

The gas stream leaves the top of the first stage suction scrubber and passes to the suction of the two parallel first stage cylinders, numbers 4 and 6 on each compressor.  Before entering the suction manifold the gas stream co-mingles with the first stage recycle gas stream.  A recycle is required for reciprocating compressors such as these, so that when there is no forward flow from the discharge of compressor to the next stage or from the compressor outlet, gas can be recycled to the suction side of the compressor and thus keep a constant volumetric flow through each stage.  This enables the compressor to be started up and used operationally to deliver export flows which are smaller than the rated design capacity.

The first stage recycle, is controlled by pressure control valves and controls the suction pressure for he first stage of each compressor.  If the gas flow entering the pre- treatment section from the separators exceeds the volumetric capacity of the compressor then the excess gas is directed to the high-pressure flare.  The pressure controller, upstream of the pre-treatment sections on each compressor train is also designed to keep the suction pressure at about 12.4 barg by controlling the first stage recycle valves.

The first stage of compression

When the co-mingled inlet gas stream from the pre-treatment section and the first stage recycle stream enter the two parallel first stage cylinders the gas is compressed to approximately 30barg.  This compression causes the gas to heat, and so the discharge stream from the first stage is passed through the first stage discharge cooler (MH-0802A) where the gas is cooled to 30°C. 

In cooling, the gas will condense some liquids which are carried forward with the gas stream and into the second stage suction scrubber (MV-0802A) where the liquid is separated from the gas stream using a vane pack within the vessel.  The liquid collected in the bottom of the second stage discharge suction scrubber is returned under level control to the production manifold where an alarm and trip of the compressor train safeguards against the occurrence of an extra high level within the vessel.

For the second and third stages of compression there is a combined recycle stream which co-mingles with the discharge from the first stage of compression upstream of the first stage discharge cooler.

Downstream of the first stage discharge cooler there is a liquid condensate recycle line, where liquid collected from the discharge of the second stage of the cooling is returned to the process under level control.

The second stage of compression

The gas leaving the second stage suction scrubber flows to two parallel second stage cylinders in each compressor numbers 3 and 4 in MC-0801A. The compressed and thereby heated gas passes through the second stage discharge cooler MH-0804A and is cooled to 35°C.

The liquids formed on cooling the second stage discharge gas flow to 35°C, are separated from the gas stream in a Dehydrator Inlet Knockout drum MV-0804A.This vessel also known as the Second Stage Discharge Scrubber. It has an internal vane pack installed to entrain and coalesce the liquid, which collects on the bottom of the vessel.  The liquid from this vessel is discharged under level control and any excessive high level is prevented by an alarm and compressor train trip.

Further liquid gas separation may be achieved by passing the gas from the knockout drum directly into the filter coalescer, the Second Stage Discharge Coalescer (MV-0814A) where the filter cartridge is designed to remove ay liquid droplets exceeding 2 microns in size. The additional liquid removed by the filter is collected in the coalescer and drains under gravity to the Dehydrator Inlet Knockout drum below. 

The Filter Coalescer seeks to achieve a “liquid free” gas stream to the dehydration unit, which is fed from the discharge of the second stages from both compressors, trains A and B. A completely liquid free gas stream is operationally impossible to achieve but the Functional Specification prescribes what is acceptable for the gas stream to be regarded as such.

The Dehydration Unit

This unit was not supplied by DR. it was designed and manufactured by Latoka Engineering Limited.

The combined streams from the second stage discharge of compressors A and B co-mingle and then enter the dehydration package MS-1001.  The stream enters into the glycol contactor tower, towards the bottom and passes out through the tower against a flow of lean glycol, fed from the top of the tower.  On contact with the glycol the water content of the gas stream is removed and the gas stream leaves the top of the tower through a demister and is cooled in the gas/glycol heat exchange before being returned to the third stage suction of compressors A and B.

The glycol containing the absorbed water from the gas stream collects in the base of the glycol contactor.  It is then cooled in the heat exchangers before it enters the glycol flash drum (MV-1002) where any absorbed gas is flashed, with the gas being directed to the low pressure flare.  The remaining water laden glycol is routed to the glycol still, which regenerates the glycol by driving off the water vapour and leaving lean glycol for re-use in the glycol contactor tower.

The third stage of compression

The gas leaving the dehydration package is routed to the third stage suction of the compression trains A and B. Each compressor has two parallel third stage cylinders, numbers 1 and 2 where the gas is compressed from approximately76 barg up to 210barg.  The heat arising from the compression is removed by passing the gas stream through the third stage discharge cooler MH-0803A, where the gas is cooled to 40°C. 

The gas streams from the two compressor trains co-mingle and are routed as required to either the gas lift manifold or into the gas export pipeline.  The discharge pressure of the third stage is controlled by PIC-11001.  There is a second/third stage recycle line, which returns gas from the discharge of the third stage compression to the discharge of the first stage upstream of the first stage discharge coolerMH-0802A.

The flow rate of the recycled gas is controlled by the pressure control valves of each of the compressor trains and the second/third stage recycle effectively controls the pressure at the suction of the second stage, at approximately 30barg.

The overall pressure control

The suction pressure to the first stage of compressor A is controlled by the first stage recycle valve PCV-08153 using pressure indicating controller PIC-0 08115.  The controller receives a signal from PIC-08021C, which is designed to maintain a pressure of 12.4 barg.  If the suction pressure increases above this prescribed value then PIC-08021 starts to open the pressure control valves to the high pressure flare. 

When the suction pressure reaches 12.7barg, should the pressure continue to increase, the pressure control valves on the gas discharge from both separators begin to close in order to control and reduce the suction pressure.

The suction to the second stage of compression is controlled by the second/third stage recycle valve PCV-08153 on compressor A to approximately 30barg. 

The discharge from the third stage is controlled through PIC-11001 could normally set 210barg. If the discharge pressure increases to 215barg, then a signal is sent from that controller to the first stage recycle pressure controller, and is designed to reduce the suction pressure to the first stage, thereby reducing the discharge pressures of all the compression stages.

Compressor shutdown systems

To protect the compressors against damaging process deviations the package is equipped with process alarms and trips to cause an automatic shutdown of each compressor.  These comprise extra high level trips in the scrubber/knockout pots on the suction of each stage of compression, extra high pressure trips on the discharges of the first second and third stages of compression, extra high temperature trips on the discharges of the first second and third stages of compression, and extra low pressure trips on the suction of the first second and a third stages of compression.

Each compression package has a crankcase and cylinder lubrication system each of which incorporates shutdown mechanisms in the event of serious malfunction. Each package also has a separate control panel. In the event of the shutdown the compressor is designed so that the shutdown valves close at the suction of the first stage, the discharge of the third stage, and on the feed and return lines to the dehydration package.  Effectively, this isolates each compressor train from the process gas streams.  Similarly, shutdown valves on the liquid outlets from the scrubbers/knockout pots also close automatically giving rise to what is known as an emergency shutdown (ESD).

Following an ESD, each compressor is automatically depressurised to remove the potential hazard of high-pressure gas in the pipework, which has the potential to fuel either a gas release or a subsequent fire.  This decompression is effected by the opening of the blow down valves on each compressor train located at the discharge of the third stage of each compressor.

In the event of an ESD and the extra high-pressure trips failing to prevent any excess pressure within the compressor system, there is also a series of high pressure safety valves, designed to relieve any excess gas pressure to the high-pressure flare.

The compressors are only one part of the overall process design, the boundary conditions are important when considering the overall process design.  The gas lift and gas export system has nothing to do with the Dresser- Rand design or the HP flare.  The first stage separator, where the gas from the sub-sea wells is separated before it goes into the DR package, and the GDU unit designed and supplied by Latoka Ltd, were also outside the design responsibility of DR. Nevertheless, the performance of these vital parts of the process plant had to be quantified before sensible design could commence, let alone develop.

The boundary conditions are important when considering the interface responsibilities of DR under the contract and issues relating to any contractual obligation that DR had to produce process simulations for the overall process design.

Scott Schedule section 3.2 1^st Stage Recycle line

KMG complains of excessive liquid carry over on the first and second stage compressor suction.

“excessive water carry over and excessive condensate carry over on the first and second stage compressors, respectively were responsible for accelerated wear from the lube oil wash out, component damage, condensate leaks, and at least 3-off gas leaks through hydraulic locking.  The cause of the first stage problems was the failure to address the amount of water produced when cooler first stage recycle gas flows into the warmer suction line.  The cause of the second stage problems was a failure to address the reality of condensate carryover when in recycle mode, where the overlapping first stage and second/ third stage, recycle causes(on max recycle) a doubling of the gas flow(and velocity) in the second stage suction scrubber.  Increasing the lube oil rate to the compressor cylinders did not eliminate wear on both stages, was never a solution to hydraulic locking and involved excessive and costly consumption of lube oil.  Parallel compression required continuous recycle.  The only practical solution open was to reroute the first stage recycle to remove the possibility of water make after the first stage suction scrubber and to remove the recycle line overlap.  Change Request CR 2455 refers.

In Janice original design, the recycle for the first stage compressors, was taken from downstream of the second suction scrubber and returned immediately upstream of the first stage compressors.  The second and third stage compressors have common recycles, which take gas from the discharge of the third stage compressors to the inlet of the first stage discharge coolers.  The unusual configurations of the first stage recycle, overlapping the second third stage recycle gave rise to two particular problems.

First. The first stage recycle configuration created problems for the first stage compressors.  The first stage recycle gas was cool relative to the incoming gas from the first stage suction scrubbers.  This caused water to condense out… water thus condensed will enter the compressor cylinder, washing the lubricating oil off the cylinders in the first stage compressor and potentially causing hydraulic locking.  This is a problem encountered during recycle.  Much greater use was made of recycle when both trains were finally up and running, because, for as long as only one train was available it tended to be run flat out, with little or no recycle, and it was from experience gained after both trains could be run simultaneously that the full extent of the problem became apparent.

Second.  The flow rate through the first stage discharge coolers and the second stage suction scrubbers is increased while the compressors are recycling.  In the extreme case, when there is no forward flow of gas the flow rate through these components is doubled.  Neither of these components was designed to handle the recycle case and they were hence undersized.  Studies ultimately determined that the cooler could be operated satisfactorily but concern remained that the second stage suction scrubbers were undersized for the recycle case and that this would result in liquid carryover into the second stage compressors, washing off the lubricating oil from the compressor cylinders.

DR responded:-“The routing of the bypass line was imposed  by KMG  during the project's development phase and was necessitated by inconsistencies within the functional specification itself.  See section 10.2 of the Amended Defence and Counterclaim.”

Two distinguished process engineering experts gave evidence.  On behalf of KMG, Mr Roderic Sylvester-Evans a Chartered Engineer and Fellow of the Institution of Chemical Engineers with extensive academic and practical experience specialising process safety and risk management issues and accident investigation in the petro-chemical and transportation industries.

Professor Graham Davies gave evidence for DR. He was Head of UMIST Chemical Engineering Department for eight years, prior to which he was a Research Engineer. Throughout his professional career he had been involved in research in the field of separation of heterogeneous mixtures of gases and liquids and solids ranging from fundamentals of coalescence of droplets and the structure of porous membranes to the design of large-scale separation equipment.

There was large measure of agreement between the experts, whose evidence, well tested in cross examination, was of great assistance.  Their differences mainly arose out of their approaches to process design dictated by their views of the party’s respective contractual obligations.  Such views did not colour their technical evidence. 

Both experts agreed that so far as the process engineering was concerned there was a regrettable absence of thorough process simulation and analysis at all stages of the design development.

The original process simulation upon which the functional specification was based was undertaken by Kvaerner Oil and Gas Company Ltd. using a software programme known as Pro II. The programme HYSYS was later used by IGL, consultants employed by KMG to carry out investigation and remedial work.  It employs the same methodology as Pro II to which KMG had access and gave rise to unnecessary duplicated expenditure.

The Kvaerner data should have enabled the clear and precise definition of the process development to be undertaken. A process flow sheet was formulated in which the individual process stages were defined and set out.  This took into account the source, form and composition of the raw materials to be used as the input feed streams to the process and the form and composition of the products from the process.  Waste products were similarly defined.

The material and energy balance tables define the continuity of material and energy that must be satisfied across the entire process, and across each and every stage in the process.  Thus the mass flow of fluid into a cooler, must equate to the mass flow of fluid leaving the cooler and the total energy of the fluid stream entering the cooler must equate to the total energy leaving the cooler, plus the energy removed by the cooling medium and any energy lost or gained from the cooler to the surroundings.

Since the physical state of a fluid, whether it exists as gas and liquid or mixture of both phases is determined by composition, temperature and pressure, a series of complicated equations is undertaken, incorporating physico-chemical rules, thermodynamics and equilibrium thermodynamics.

When solved these should show the quantity and composition of gaseous and liquid phases in relation to various stages in the coolers and other process vessels.

It is upon such precise and lengthy computer analyses that the detailed design of the individual components in the process system and their relation to each other and other parts of the system should be based.

The Functional Specification in paragraph 2 provides:-

“The document PMD-PFD-2001 Rev 02- Overall Process Flow Diagram marked up in the BCM [Bid Confirmation Meeting] dated 23/24. 6.97 attached to the MOM [Minutes of Meeting] shall form the basis of the development of the Contractors P.& I.D…”

Each train shall be completely self-contained and shall  include, but not be limited to the following:-

…… Complete process design…

Paragraph 4.1 provides:-

“The Gas Compression Package is required to compress gas from the first stage separator be used as a fuel gas, lift gas, and for gas export.  The package comprises two 50% gas compression trains, including individual pre- treatment equipment.  The pre-treatment of the gas stream is required to maximise the liquids recovery in the field.  After the second stage of compression, the complete gas stream from both compression trains will be routed to the Gas Dehydration Package (supplied by others) for dehydration and the removal of a fuel gas stream.  The dehydrated gas stream, less fuel gas, will then be returned to the Gas Compression Package for further compression.

The location of the gas compression package within the processing flow scheme is detailed on the document PMD-PMF-2001-Overall Process Flow diagram.  Additionally, this details the major process inputs and outputs to the package.  The Contractor will be responsible for all the required controls within the package (not detailed on the process flow diagram) to ensure safe and stable operation of the unit within the operating range as defined below.

4.2

Process Requirements

This section defines the pressures, temperatures and flows required by or provided to the process boundaries of the gas compression package.

4.2.1

Summary of gas flows, pressures and temperatures at the package boundary

Refer to Heat and Material Balance tables PMD-PFD-2014.01/02/03Rev.04….

4.2.1.1 Gas flows

a)…….Refer to Heat and Material Balance Table PMD-PFD-20 14, Stream 21 for the gas composition of throttled gas to compression.

4.9

Other constraints

The gas fed from the compressed interstage system to the Gas Dehydration Package is required to be liquid free.  Of particular concern is the possibility of oil mist carryover.  In order to eliminate oil mist carryover in knockout vessel, incorporating a high efficiency or oil mist eliminator shall be installed in the feed line to the Package.

The reference in paragraph 2 of the Functional Specification  to” complete process design” is to the design of the process equipment within the DR. package.  It cannot relate to the complete process design since this would require information on the conditions and performance of the upstream equipment, namely the first stage separator, and would require detailed knowledge of the design conditions for parts of the package for which they had no responsibility.

Paragraph 2 contains the phrase “shall form the basis of the development of the Contractor’s P & ID”. This is a reference to the development of the piping and instrumentation within DR’s scope and not to the development of the flow sheet.

The tables “ Heat and Material Balance “ were produced, according to a non- marked up version of the Process Flow Diagram, which did not include or allow any gas recycle.  It did include a third stage suction scrubber.

These tables are part of the Functional Specification and serve to identify the purpose of the process plant, namely to receive and process specified material at the compression package boundary and to deliver processed hydrocarbon and gaseous material, according to specification, when it leaves the package boundary in such manner as the package process equipment remains efficient and serviceable for its projected life.

On the 3rd and 4th of September 1997 a meeting took place at Wythenshawe at which a discussion took place about changing the route of the first stage recycle line.  The recycle line originally contemplated, prevented the independent operation of the pre-treatment system.  According to the evidence of Mr Donald Ferns of KMG who described himself as “lead process engineer on the Janice development on the Project side”, a discussion took place as a result of which the route of the 1^st stage recycle line was changed.

KMG suggested the line of the re-routed 1^st stage recycle, DR had no objection and drawing 85900 S C- PG was amended in red manuscript.

Mr Horne, DR’s Project Manager for the Janice Project, said that he considered the impact of the drawing markup and was satisfied that it was a reasonable change to resolve the difficulty.  He confirmed that Mr Heaton of Latoka Engineering Limited, responsible for the process design and internals of the compression package scrubber vessels, was satisfied that the second stage discharge scrubber could cope with the change.

The contract had been signed on the 10th of August 1997.

No revised Heat and Materials data assessing the consequences of the changes were produced.

Professor Davies trenchantly expressed the view that the re-routing which refers to the introduction of the recycle stream into stream 23 suggested by Mr Ferns as “astonishing”.

“Mr Ferns was the Lead Process Engineer of the Janice development.  He would have reached this position after several years of experience in this type of process engineering.  Although I would not expect him to be an expert in thermodynamics and gas dynamics.  I would have expected him to have a working knowledge of these subjects sufficient to realise that if a gas is cooled below its dew point condensation of liquid will take place.  This is exactly what will happen at the mixing point of the two streams.  Stream 23 will be at its dew point having been in equilibrium with liquid in MV 0801 at a temperature of 35°C and the pressure of 12.7 Bara.  The recycle stream before throttling is at the same temperature, 35°C, but at 30.9 Bara.  When the recycle gas has expanded down to 12.7 Bara the temperature will have decreased.  The proposal was flawed from the outset.  Furthermore, to compound the mistake, as with the earlier proposal no calculations appear to have been carried out at the time to examine the implications on the process of the changes.  The Heat and Material balances should be revised…. to account for the changes. A simple calculation, reflecting the differences in composition would have given an estimate of the gas temperature after expansion and would have shown that there would be a problem of condensation in the pipe after mixing…… what is certain is that liquid will condense in the pipe. This pipe connects to the first stage gas compressor, and therefore the liquid produced by cooling of the feed gas below its dew point, will enter the compressor.

Mr Horne of DR was in a similar position to Mr Fearns.

Mr Sylvester-Evans agreed that the design of the 1st stage recycle was flawed.  He also agreed with Professor Davies that the second stage discharge scrubber was rendered inadequate.  The consequences of the change were not supported by any calculation or analysis.  Both experts agreed that it was elementary to carry out calculations and analysis quantifying the effects of change before embarking upon it.

The original Heat and Materials balance data was produced by KMG from Kvaerner.  On the 17th February 1998, a revised process simulation was handed to DR at a meeting by KMG. Mr Sylvester- Evans commented:- “ That particular one…. was not a complete reworking of the package, my Lord, I think it is important to note that that was just to examine part of the condensate recycle with respect to sizing on valves.” 

At no stage was any such data required from DR.

Paragraph 3 of the Functional Specification provides:-

“Upon Contract Award the revision status of all specifications, recommended practices, data sheets , basis of design documents design reports and other relevant documents will be frozen.”

DR submit that the effect of this provision was the parties should proceed on the basis of the original process flow diagram, thus there would be no design changes necessitating further revised process simulations, and in consequence no specific requirement could be placed on DR to produce its own process simulation model in order to deal with changes.

The parties clearly agreed that this provision should not operate.

Clause 8.11 of the Purchasing Contract in any event provided for change and design development.

“The Contractor shall submit to the Company for review and comment, all relevant documents including but not limited to sketches, drawings, calculations, reports, technical notes and recommendations.  These documents shall be submitted to the Company, in sufficient time to review and comment upon them without delaying the performance of the work by the Contractor, and in accordance with the master schedule.  The acceptance by the Company of any such relevant documents shall not relieve the Contractor of any obligations under the Contract or constitute the Company's assumption of responsibility for the work.

Mr Ferns proposed a solution he believed would enable KMG to operate the pre-treatment unit independently of the compressors as required in the Functional Specification. By so doing KMG resolved the apparent ambiguity in the requirements of the Functional Specification.

KMG submits that given that clarification it was incumbent upon DR to design and deliver a compressor package incorporating the process engineering requirements, and which was fit for purpose.  In so far as that required process simulation it was for DR to undertake it.

DR submits that the entire responsibility for the change, and hence its detrimental effect must lie with KMG.  It submits that KMG is estopped by convention from denying that DR was entitled to deviate from the Functional Specification to comply with KMG’s express requirements.

Estoppel by convention is described in Spencer Bower, Estoppel by Representation 4^th edition at VIII.2.1:-

“This form of estoppel is founded, not on a representation made by representor and believed by representee, but on an agreed statement of facts or law, the truth of which has been assumed, by the convention of the parties , as the basis of their relationship. When the parties have so acted in their relationship upon the agreed assumption that the given the state of facts or law is to be accepted between them as true, that it would be unfair on one for the other to resile from the agreed assumption, then he will be entitled to relief against the other…” 

There is no evidence that at the meeting there was any understanding or agreement reached to the effect that DR were in any way relieved of their design responsibility or released from their warranty in matters connected with the recycle line.

None of the witnesses who gave evidence before me as to that meeting mentioned any discussion to that effect.  There is no hint of any such discussion in the manuscript minutes or the typed minutes.

There is no unambiguous evidence to show that both parties acted upon the assumption that DR were relieved of design responsibility in relation to the first stage recycle line or of their warranty obligations in relation to it. The mere fact that KMG produced some process simulation data relating to valves does not assist DR. They had produced process simulation data before this meeting as part of the collaborative process under the contract.

There is no basis for the contention of DR that an estoppel arises.

It was for DR to design and deliver a compression package including the process requirements which was durable and efficient in operation.  To accomplish this it was both desirable and necessary at the preliminary design stage, and throughout the design process, to be armed with the appropriate process simulation data.

DR could have requested such data from KMG, or used  the services of a third-party such as Kvaerner or IGL, if it lacked the expertise and knowledge in-house.

The design to the 1^st stage recycle that DR used to build the package was defective.  The second stage scrubber MV0802, in consequence was overloaded.  It was necessary for the second stage cooler MH0802, and the second stage scrubber to be redesigned to take the higher flow rate.

It is clear that these difficulties were identified in the discussions between KMG and DR. The oral reassurance sought by Mr Horne from Mr Heaton of Lakota is but one example of DR’s awareness of the problem. 

I accept the evidence of Dr Krska, a careful and reliable witness, that DR twice were requested by the Janice project team to confirm that the 1st stage discharge cooler, and the 2^nd stage suction scrubber were designed for the higher flow rates that arise during recycle and that DR confirmed that they were adequate for this case. It was later confirmed to the project team that DR had not taken heed of the higher flow rate issue.  I accept that they and their subcontractor Latoka, later confirmed that these units were not designed to take the higher flow rate.

On the 16th of April 1999 DR in their customer activity report No16286 noted:- “during end of run  it was noted that there was a gas leak at the first stage number four cylinder outhead in the same area as the previous leak on number six cylinder approximately 6 o'clock position.  The first number four cylinder outhead, was removed and the gasket was found to be split at the bottom”

On 20th of June 2000 the OIM Janice E- mailed KMG with a Dresser-Rand Inspection report on B Compressor first second and third stage cylinders inspection: “ …note excessive wear measurement in the first stage cylinders, number 4 and 6 also piston ring clearances.”  It was noted that the opinion of  Dresser-Rand personnel on board, was “ that the machine was a non runner.”

On the 19th of April 1999.  The Janice OIM reported gas compression starts aborted due to A machine leaking condensate from the second stage cylinders.  It was stripped down to investigate condensate in the second stage and on 20th April 1999 there was a meeting on the Janice platform between Mr Smith and Mr Slater of DR and Mr Ferns of Kvaerner considering a discussion document giving three possible causes of the condensate carryover into the A machine.

During April May and June, consideration of the condensate problem continued.  There were meetings between the parties and exchange of e-mails.  On the fifth of May 1999.  There was a meeting at Wythenshawe, where it was noted that DR’s Field Service note how the compressor suction valve damage was found and that it was symptomatic of a liquid slug hitting the valve.

On second of June 1999, a compression system report noted:” condensate dripping from second stage discharge valve cover(number five cylinder)

A Compression System Report noted that “The A compressor  second stage, number five cylinder discharge valve cover was leaking condensate. Compressor was isolated and purged for” o” ring replacement.  The B compressor second stage ( number 3 cylinder) discharge 12 cover and lubricator leaking condensate.  Compressor isolated and purged from cover “o” ring replacement.

On 5^th September 1999 B compressor started but tripped on the second stage high discharge pressure.  The switch and others were blown and found to be full of condensate.

On 16th May 2000 Mr McGuinness of KMG e-mailed Mr Whittle of DR

“In my telephone conversation, I confirmed that as a result of mechanical problems with piston and cylinder wear on the A compressor we wish to explore the various avenues available to limit or eliminate the same.  It is evident that the compressor is handling condensate in the first and second stages, by the frequency of damaged valves and the lack of excess lubrication on cylinder walls when inspected.  The current visit inspection to carry out at 2000 hour routine on the” B” set  has highlighted cylinder bore wear of 1.1 mm and 2.18 mm on cylinders 4 and 6 respectively”

During July and August 2000 five DR reports record compressor damage to both A and B compressors and levels of cylinder wear on the first and second compressor stages, consistent with liquids removing the lubricant.

I am satisfied that DR knew of the difficulties being experienced by KMG in relation to liquid handling and condensates.  They were kept informed.  Their advice was sought, and their personnel were regularly on board the platform.  Their reports evidence that DR were fully live to these problems.

DR contends that significant damage was caused to the first stage scrubbers and compressor cylinders by the ingress of sand and water from a first stage separator into the DR compression package.  The Heat and Material Balance Table referred to in paragraph 4.2.1 of the Functional Specification prescribes what the process gas constituents should be at any stage in the process. Stream 21 indicates what the composition of the process gas as it leaves the first stage separator MV 0301 should be. 

A sample taken in April of 1999 contained a quantity of water in the hydrocarbon liquid.  Mr Sylvester-Evans regarded this in  the nature of a rogue sample.  There was a second set of analyses undertaken by Corelab and Oil Phase.  These are reported on an anhydrous basis and by the use of steam tables using Raoult’s Law the fraction of water within the overall mix it can be assessed.  Mr Sylvester-Evans accepted that assessment was approximate, but it was good enough for many cases.

Professor Davies had reservations as the quality of these results. They do not of course stand-alone. They should be considered in the context of the evidence of those on the platform who witnessed considerable quantities of condensate in all parts of the system.  Mr Colin Smith was amongst their number.

I accept Mr Sylvester-Evan’s conclusions that the sample showed, that which was introduced into the DR package, corresponded with the Functional Specification.

There is evidence that the pre-treatment separator collected quantities of sand.  Mr John McIntyre gave evidence deposits of sand in the first stage separator were low and monitored.  There is evidence that some minute particles found their way into the DR package.  Such material would be collected in the low points or depressions of the piping runs.  Some sand particles lower than 0.01mm in size were observed on metallic particles from a piston ring in the second stage. For it to come from the reservoir it would have to pass through the cooler on the first stage from the scrubber on the first stage, and the pulsation dampener and through the first stage of compression itself, and onwards through the discharge dampener and the cooler on the suction side of the first stage to the second scrubber on the suction side.

I am satisfied that no significant quantities of sand were introduced into the DR package, and that there is no basis for concluding that the damage was materially caused otherwise than by the condensates produced as a result of DR’s defective design.

KMG dealt with the problems arising out of the rerouting of the first stage recycle lines on A and B trains and the consequential undersizing of the second stage scrubber and cooler and claim £546,541.62, in respect of the cost of the remedial rerouting of the first stage recycle line and necessary ancillary work.  This claim figure was adjusted in a meeting between the quantum experts to £538,572.62.  Mr van Voorst, DR’s quantum expert, has agreed the sum of £516,363.20, leaving a balance of £22,209.34.

Mr Boyne, KMG’s quantum expert has given evidence attributing this balance to the remedial works undertaken.  I am satisfied that the sum of £4,844.34 relating to the manufacture of pipe spools is referable to the drawings relating to the first stage compressor recycle rerouting and that it is contemporaneous with the remedial work.  This sum is the subject of the Cansco Greig invoice and the claimants are entitled to it.

£4455 relating to Janice A Offshore subject of invoices rendered by IDEAS is not recoverable.

£12,910 relate to invoices from L&N(Scotland) Ltd and the provision of “Netherlock” MRL-L interlocks and related works.  Whilst there is no evidence that the previous works included interlocks I am satisfied that the remedial works necessarily included these items and that the cost is therefore recoverable.

£534,117.62 is recoverable in relation to this part of the claim.

Section 3.4 Scott Schedule Liquid removed from Knockout drums

This item relates to KMG’s allegation that the dehydrator inlet knockout drum in the 0804 was overloaded and in consequence there was inadequate liquid removal.  Complaint is also made that the level transmitters in the drums were of the displacer type, consisting of a metal float, which collapsed due to external pressure.

KMG maintain that the drainage arrangements provided by DR for the drums were inadequate and that the drain line, level control valve and outlet connections through the second stage condensate recycle from the dehydrator inlet knockout drum were undersized.

“this was exacerbated by condensate flashing within the liquid recycle line further impeding liquid flow. Incorrectly specified level transmitters, which suffered float collapsed.  This led to frequent high liquid level, shutdown trips from the dehydrator inlet knockout drum.  The second stage condensate recycle line components had to be increased in size, and the level control valve moved downstream to limit the effects of flashing off. The level transmitters had to be replaced with nucleonic or pressure differential devices”

Professor Davies in his report identifies the problems encountered in relation to MV0804 relating to liquid collection in the vessel and drainage from it.  The circulation of condensate around a loop from MV 0804 through MV 0802, the second stage compressor MV 0802 and cooler MH 0804 back to MV 0804 causes the liquid load onto NT 0804 to be significantly higher than shown for stream 28 and the outlet stream 44( in the Heat and Materials Balance Table in the Functional Specification.)

He states that

“..this increase in the liquid concentration on   the feed to MV 0804 may have profound effects on the condition of the gas leaving the second stage of this knockout drum, the filter condenser MV 0814.  The first stage of separation relies upon a vane separator in MV0804.  This was designed for much lower gas flow rates than will be achieved with gas recycle.  Again, my calculations suggest that the gas handling capacity over and above the original design ratio is modest.  The separation fluxes achievable under operating conditions are low, principally because of the low surface tension of the liquid at this stage in the process.  An increase in gas flow and liquid flow is likely to result in a greater than expected carryover of liquid in the gas entering into MV 0814.

Filter coalescer elements, as used in MV 0814 have a much smaller operating envelope in relation to liquid concentrations than all other ‘primary’ devices(vanes, meshes, axial flow cyclones, etc) and therefore these may well approach a ‘flooding condition’ at the higher gas and liquid flows experienced at high gas recycle ratio.”

Mr Sylvester-Evans agreed that the re-routing of the condensate from the dehydrator inlet knockout drum to the inlet of the second stage suction scrubber rather than the production separator late in the project had a profound and deleterious effect on the liquid accumulation within the dehydrator knockout drum.

The build up of liquid in this vessel led to frequent process trips due to the high liquid levels.  The liquid built up faster than it could be drained away.  I accept the evidence of Mr Sylvester-Evans that the work performed in August 1999 and thereafter was necessary and reasonable to remedy the problems encountered with drainage.  It is evident that one of the problems was that the 1 inch diameter nozzle acted as an hydraulic bottleneck in the drainage system.

Professor Davies agreed that he would not be surprised to see KMG carrying out works to modify the drainage arrangements from the dehydrator Knockout drum.

It is accepted by DR that the floats used in the level transmitter within MV 0804 were unsuitable. DR submits that they have no liability in respect of this matter, because they warned KMG that these floats would operate very close to their working limit and that capacitance level transmitters were being considered.  It is accepted by KMG that they originally expressed the view that capacitance level transmitters would not be suitable.  The design responsibility was DRs.  They decided to fit floats in the level transmitter.  They cannot now be heard to say that the decision was not theirs.

Subject to liability and general mitigation considerations, KMGs claim in the sum of £43,627.49 was agreed by the experts. KMG are entitled to this sum.

Section 3.3 Scott Schedule GDU

The claims in this section relate to the glycol contactor  column MV-1001.  It is also known as the TEG contactor which contains tri-ethylene glycol and its function (described above) is to operate as a Gas Dehydration unit (GDU).  It was designed and supplied by the Latoka Engineering Ltd.  It is the next vessel in the process scheme beyond MV 0814.

The GDU is intolerant to liquid contamination and so the gas feed to the GDU had to be liquid free.  It is impossible to produce a gas feed that entrains no liquid whatsoever.  Both experts agree that where the Functional Specification General Requirements section requires” liquid free” it means that the absolute carryover of liquid in the gas at that stage was to be   < 0.1usg/mmscf.

There is no direct evidence that at TP 05, the point where the gas stream leaves the DR compression package and before it is then piped upwards through the 20 metres of piping manifold designed and installed by others,  that it was not” liquid free”

The evidence relied upon by KMG is that liquid has been found in the GDU and damage has been caused.  There is no issue about this.

When the two compressor trains are working in parallel two separate streams are mixed and then fed into the GDU.  Although the streams emanate from two identical compressor skids, slight temperature differences between the two gas streams, as low as half a percent centigrade according to Dr Krska , when mixed, could give rise to significant liquid formation in the GDU.  He opined that it was impractical to ensure that the upstream conditions of the two streams were identical.  Professor Davies agreed that it would be impractical to ensure that the temperature of the two streams at the mixing point were maintained within 0.5C.

The real issues are: first as to where the boundary of the package is: second what was the point at which liquid free gas had to be delivered.

KMG submits that the requirement was to supply a single gas compression package albeit on two separate skids, and from  the package should be supplied a single source of liquid free gas.  When there is parallel operation the comingled streams from each train should mix within the package and leave it liquid free.

The boundary of the package is at terminal point 05 on each skid.  The re-entry point on each skid is at terminal point us is at terminal point 06.

At 4.1 of the Functional Specification entitled” Package Description” is the following:-

“ the package comprises two complete 50% gas compression trains, including individual pre-treatment equipment…After the second stage of compression, the complete gas stream, from both compression trains, will be routed to the Gas Dehydration Package( supplied by others) for dehydration and the removal of fuel gas stream…”

There is no assistance to be derived from this wording as to whether the mixture of the two gas streams is to be within the package or without.

Reliance is placed upon a PFD diagram where the package limit is depicted by a broken line.  Within one of the two package areas shown, the gas stream from B train is marked as being fed into A train.  It is submitted that this is unambiguous evidence that there arises an obligation for DR to design and fabricate a mixing point to enable the co- mingled gas streams from A and B streams to emerge liquid free for delivery to the GDU outside the package.

The PFD relied upon, in fact shows the package divided into three parts.  There are two skids, for convenience diagrammatically depicted as one, and the separate GDU system.  The ‘as built’ system has the co- mingling point within the separate GDU system.

Paragraph 4.9 of the Functional Specification, entitled “Other Constraints” contains the provision that the gas fed from the compressor interstage system to the gas dehydration package is required to be liquid free”….

KMG in its Re-Amended Particulars of Claim avers-

“Further or in the alternative the defendant’s compression package did not deal adequately with the consequences of co-mingling gas from compression trains A and B immediately before the contactor.

68…… the defendant's design was defective because it failed to provide adequate liquid knockout measures for the comingled gas with the result that excessively wet gas found its way to the contactor.”

DR in reply answered:-

“147.2 The meeting point is within the piping leading from the compressor interchange system to the gas dehydration package, and accordingly DR had no responsibility to deal with the possibility of liquid knockout at this point.

DR had no obligation under the contract to design and provide a comingling point. The gas streams leaving the A and B trains respectively became the design responsibility of KMG.  There is no evidence that DR was given any information, let alone had any knowledge as to the extent conformation gauge or layout of the piping to be used by KMG to connect from the edge of the DR  package to the GDU package.

The 20 metres of piping manifold from TP 05 at the boundary edge of DR’s package to the GDU was for some time not lagged.  Surprisingly it was not trace heated either. Similarly the piping manifold leading from the GDU, back to the DR package at TP 06.

Professor Davies expressed the view that those responsible for designing these manifolds should have taken account of the need to reduce heat transfer.  In cross-examination he made position clear, the necessity was to ensure the gas did not fall below its dew point within the manifold before it reached the GDU.

“ If the lagging is 100% efficient, and it is a long line it still might drop its temperature.  It needs lagging and tracing.  Or trace heating and lagging, I should say, you put the trace heater on first.”

Neither Professor Davies or Mr Sylvester-Evans have produced any calculations as to the quantity of liquid that could have resulted from any defect in the design of the manifold.  The manifold slopes upwards to the GDU.  Liquid formed in the manifold would therefore drain to the lowest point at or below TP 05.

I am satisfied that the design and operation of the manifold within the GDU system caused the formation of some liquid, which may have contributed to the quantities of liquid within the GDU.

Whilst there is no direct evidence as to the composition of the gas flow at TP 05 I am compellingly persuaded that the gas flows leaving the DR package were not liquid free.

KMG originally intended to install an inlet scrubber to the GDU.  Mr Sylvester-Evans agreed that it was deleted as a cost saving measure.  Subsequently it was recognised that such a vessel was necessary and the GAPS vessel was designed and installed by KMG.

The GAPS vessel purchased by KMG was sized for three compressor trains whereas DR only provided two.  Clearly if a GAPS vessel is necessary by reason of DR’s breach the cost of a two train capacity vessel would be the appropriate measure.

Mr Boulding submits on behalf of DR that they initially intended to install a knockout vessel ( MV 0804) in the feed line to the package, which ensured that the absolute carryover of liquid in the gas at this stage was <.1usg/mmscf as required by paragraph 4 .9 the Functional Specification. DR sought advice from the Latoka Engineering Ltd and were advised that in order to achieve the removal of droplets greater than 2 µ a coalescer filter would be needed downstream of the knockout vessel  MV 0804.  Hence the installation of MV0814 A and B in each train containing a coalescer filter, because DR recognised the necessity of protecting the GDU. 

Glycol regeneration is intolerant to contamination from lubricating oil, and hydrocarbon condensate.  Lubricating oils cause the glycol to degrade and foul the filters and the glycol reboiler.  A hydrocarbon condensate, also causes pressure flow and temperature fluctuations within the regeneration package.  Both forms of contamination prevent the regeneration process from providing glycols sufficiently dry in order to allow the gas, water quality specification to be achieved.

KMG contend that the coalescer filters failed to work efficiently.  They blocked causing a differential pressure that sucked condensate back up from the return drain line which had to be removed manually and also gave rise to condensate entering the GDU.

Mr Sylvester- Evans expressed the view that there was a serious maintenance problem. The cartridges weighing some 2 kg could not be conveniently accessed for maintenance and cleaning.  It appears that KMG in laying out their equipment had not followed express advice of DR and left sufficient space for easy removal.  He accepted however, that had there been timely maintenance, the coalescers would have been effective.

Professor Davies reviewed the correspondence between DR and Latoka Engineering Ltd in relation to the design of the dehydration inlet knockout vessel MV 0804.  They recognised that primary separation alone, using a mesh demister or vane separator, would not meet the process requirements. To ensure compliance with the standard of entrainment in the fuel gas to the GDU of 0.1usg/mmscf a primary separator, followed by a fibre bed coalescer, or preconditioning stage, followed by axial flow site cyclones should be fitted.  He observed that the former was the method selected by the Latoka and concluded. “.. this is acceptable and has been and still is successfully used throughout the industry.”

Latoka Engineering Ltd designed and supplied the GDU whose input requirements governed the output requirements of MV 0804 and MV 0814.

Given that the drainage and level transmitters were remedied, the coalescers properly maintained, and that the manifolds to the GDU were properly lagged and trace heated, it follows that MV 0804 and MV 08 are able to produce liquid free gas when it leaves DR's package and is supplied to the GDU. There is no cogent evidence of the need for the GAPS vessel to ensure that the gas stream supplied to the DR package edge at TP 05 or the GDU is liquid free.  There may be other justifications that warrant this design decision. 

The design details of the gaps vessel have not been supplied to Professor Davies.  He was unable to draw any conclusion as to the intended benefits to be derived from fitting the GAPS vessel.  The costs of designing, supplying and fitting the vessel are not recoverable from DR.

Section 3.5 Scott Schedule Condensate Drainage

KMG complain that when the GDU was shut down, liquid  accumulated through condensation in the low points of the GDU feed and return lines.  This liquid would then travel to the third stage compressor in slug form causing extensive component damage, condensate leakage, and at least three of the reported gas leaks. A low drain originally fitted to the GDU return line was ineffective since batch draining could not be carried out when the system was pressurised. 

The solution was to fit drain lines both to the feed and return lines which could be used to batch drain under all operating conditions.

DR contends that the pipework referred to was not within its scope of supply.  It properly relies upon the fact that the points TP 05 and TP 06 mark the end of the DR package.  The return line from the GDU to TP 06 contains a splitting device installed by KMG to return the process gas streams back to the respective trains within the DR package. The pipework manifold, connecting TP 05 to the GDU sloped downwards to DR’s horizontal piping.

It is evident that a significant part of the problem arose from the design and layout of the pipework in the GDU packet.  Clearly, if that led to low points it was the obligation of KMG’s designers of the GDU package to incorporate drainage of the low points.  Mr Sylvester -Evans accepted this to be part of the design responsibility.

Insofar as there are undrained low points within the DR package that is their responsibility.  The accumulation of liquid in the low points gives rise to the formation of liquid in slug form entering the third stage compressor. Originally a third stage scrubber was planned for, in the design.  It was removed at the request of KMG for cost or weight considerations.

Professor Davies said that the removal of the suction scrubber from the design at the instance of KMG was indefensible.  Cross-examined, he accepted it was reasonable for KMG without the drum in front of the first stage compression to subsequently install low point drains at these points

“ Well, you would definitely have to drain it and I hope you are not going to suggest to me that these low point drains, and these low point drains, and being able to drain them, are going to satisfy the performance of or the duty of a third stage suction drum, are you“?

The removal of the third stage suction drum in the design put a premium upon the necessity of having effective drainage of the low points.

At the Aker McNulty yard the need for a drain in the vicinity of TP 06 was identified and one was installed by KMG, which was inadequate for its purpose.  KMG cannot now be heard to complain that it was inadequate having assumed the responsibility of putting it in at a time when they undertook to complete work on the compressor package in the stead of DR and for which there was a reduction in price on account of work done in lieu.

The minute of the meeting 15 April 1998 relate to glycol carryover from the GDU and the acceptance by KMG that they were to provide protection in the low points of piping from GDU to the gas compression package.

On the 20th of April 1999, condensate was found in the suction and discharge valve posts, on investigating a leak of gas from the third stage and number one cylinder.  On the 30th of April 1999 a third stage compressor suction and discharge valve was found to be damaged it was thought by a liquid slug.  In May DR confirmed that to be the case.  On 24th June 1999 a meeting was held between KMG and DR’s senior management at Le Havre where many of the technical problems recently experienced were discussed and as it is recorded, a full picture of the various issues being dealt with, was given.

The cause of liquid formation, and importantly its presence within the DR package after the gas stream had gone through the GDU does not result from any breach by DR.

Scott Schedule, section 3.8 Condensate pump and valve seals

The LP condensate pump returns condensate from the first stage suction scrubber to the first stage separator.  The pump seals failed frequently.  DR contends that these were items subject to normal wear and tear within the terms of clause 11.9 of the Purchasing Contract. I agree. 

In any event no breach has been demonstrated.

Scott Schedule, section 4.0

This item has been dealt with under vibration above

CONTROL ISSUES

Compressor Control Shortcomings

Paragraph 4.8 of the Functional Specification provides:

“The overall control system for the compressors is within the package supply.  The Contractor is responsible for the design and supply of all necessary instrumentation and control systems for the safe operation, start up and shutdown of the gas compression package.  This includes the ability to start and stop one compression train without at affecting the operation of the other train.” (emphasis provided)

Since the compressor package was itself an independent component of a larger process system, the other component parts of which was supplied by other contractors, it is important to consider the interface requirements.  These are contained in the contractor Administration Instructions 7.0

KMG raise two specific complaints with respect to the control interface mismatch and parallel start-up, and one general complaint with respect to parallel operation.

Scott Schedule 5.2 control interface

The first stage recycle controller adjusts the recycle flow rate in order to prevent low first stage suction pressure.  When operating automatically it adjusts the flow rate, based upon signals received from the pre -treatment pressure controller and the third stage discharge pressure controller.

However, it could not operate automatically, because the signal from the pre -treatment pressure controller (outside the DR package) drove the recycle controller’s  set point the wrong way.  There was a mismatch between the compressor controller system supplied by DR and the external control system supplied by Silvertech Ltd.  As a result, the automatic operation of the controller gave rise to compressor shut downs.

Mr Sylvester-Evans accepted that this was something that should have been an obvious to both parties. I agree. It is not contended that DR’s control system was at fault. Merely that it was incompatible with Silvertech’s.  There is no allegation, let alone evidence that there was any failure to consult or to cooperate with KMG’s other contractors. There is no merit in this allegation.

Scott Schedule 5.4 Parallel operation pressure trips

KMG allege that when the second compressor was brought on line there was a transient pressure reduction in the first compressor, which resulted in it experiencing low suction pressure trips.  This is attributed to the mechanism used by DR to prevent excessive pressure differentials between the suction and discharge ends of the compression package. DR chose to measure the suction pressure, and the discharge pressure was inferred from the release valve setting.  No harmful differential pressures were created across the compressor, because the discharge pressure was also drawn down when the second compressor came online.

DR assumed that the provision of stable suction pressure to the gas compression package was KMG's responsibility.  And so it was in normal circumstances, save when the second train was also operated.  The operation of the second train as part of the complete package was a matter that should have been taken account of and a differential pressure switch measuring the differential in pressure directly should have been provided for. 

Scott Schedule 5.6 Parallel operation Compressor control

KMG contend that they encountered numerous control problems in attempting to achieve parallel operation of the two compressors.  This included cascade tripping, where a trip on one compressor would cause the other compressor to trip.  Mr Sylvester- Evans expressed the view, which I accept that this was because the normal operating conditions of the compressor has been very close to the design limits for which DR were responsible. The gas compression package was not tolerant to relatively small process deviations. The system was not tolerant to sensitive high suction pressure with small variations causing a first stage high discharge pressure trip.

These problems required diagnostic work and resolution by prolonged iteration.  Performance improvements carried out numerous and necessary investigations to deal with these problems.

DR did not participate in finding solutions.  They regarded it as a KMGs problem since DR submitted, the resolution of these problems required knowledge uniquely possessed by KMG such as gas lift rates, export composition and interface conditions with other parts of the process.  I have no doubt that this is the case.  There is no evidence however that this information was asked for, let alone sought by DR and denied by KMG.

In relation to the installation of differential pressure switches to eliminate low pressure trips during parallel operation KMG claim £8,363.  This claim is proved.

In relation to Scott schedule 5.6, KMG can properly recover £34,649.64, the cost of the investigations and modifications to address the problem of stable compressor control.

ELECTRIC MOTORS

Scott Schedule section 6

This section relates to complaints of electric motor failure and cooling fan failure with accelerated wear. The electric motors drove the compressors.

KMG contend that: “… starting in October of fan clearly showed the failure mode to be induced by torsional vibration.  The result was a series of emergency process shutdowns, purchase of replacement fans, the hire of an external blower and air compressor investigative work and the  redesign (originating from DEI) and manufacture of a stiffer fan.  Each compressor motor also suffered rotor core failures(breaking up) in the latter half of 2001.  The internal cooling fans had also 1999 both compressors had suffered at least eight failures of the external air cooling fans.  An inspection of the partially failed suffered from some cracked fins. The failures were due to high torsional stresses.  The results were  unplanned removal and repair of both motors, each compressor out of service for considerable time, torsional vibration surveys, (DEI) and investigative modelling carried out.(Peebles) a spare motor also had to be purchased to ensure that KMG could continue to use the compressors during periods when problems with the supplied notice were being resolved.”

DR comment-“ KMG omitted to inform/or discuss the fan failures with D R for a considerable period after they occurred (25th October 1999 -13 February  2001).  Even when DR were made aware of the fan problem the defect was not reported, in accordance with the contract requirements. DR produced an updated torsional analysis utilising data refined by Peebles and discussed it fully with KMG and Peebles.  It must be commented that there was to close correlation between the initial and the latter DR torsional studies, DR offered to do additional work, but did not receive a request to do so. Prior to rotor failures the motors had been in operation for approximately 9000 hours.  DR will accordingly rely on clause 11 .9 of the Purchasing Contract, which  states “ no equipment or part furnished by the contractor shall be deemed to be defective by reason of normal wear and tear or the Company's failure to properly operate or maintain the equipment in accordance with good industry practice or the specific recommendations of the contractor if, which is denied the motors were defective and/or were not suitable for their purpose then this was due to default on the part 20 defendant….”

DR has chosen not to plead its positive case, in accordance with CPR 16.5, of fair wear and tear or of  failure to use or maintain otherwise than in accordance with good industry practice save asserting that the engines lasted 9000 hours, and by implication, were fit for purpose, and that the failures did not amount to a breaches of the term that the equipment would have a 15 year life span as set out in paragraph 4 .11 of the Functional Specification.

Damage to External Fans

There is no evidence of any failure by KMG to maintain or use otherwise in accordance with good industry practice.  Professor Garvey the joint expert was not even asked to consider this possibility.  The failures of the external fans were caused by torsional vibration at or slightly below the fourth and fifth resonant frequencies. I am satisfied that the occurrence of frequency droops was not the cause of torsional excitation.  In order for this to have causal significance, such droops must occur on a large-scale, in order to give rise to the repeated application of cyclic stress.  There was no evidence of any such pattern or intensity of frequency drooping.  At certain harmonics it is likely that the occurrence of drooping, would result in a reduction of the forcing effect of the motor and thus reduce this mode of vibration.

I was not assisted by the documentation referred to in the file references B 1.6, B 1.7, B1.8, B1.9,B1.10, C1.2, C1.3, E1.2, E1.6, J 1.3,L 1.1, and L1.3.

DR submitted that unreliability of the diesel generators was a material cause of drooping.  Professor Garvey accepted that diesel generator unreliability could cause frequency drooping but there is no cogent evidence as to the extent and degree of any such alleged unreliability or drooping. It seems that the problems with the diesel generators were addressed in 1999 and the failure of electric motor A occurred in the autumn of 2001 and motor B in January 2002.

Damage to Internal Fans

As to the internal fans there was some uncertainty as to the precise cause of damage.  Professor Garvey's view was that the primary likelihood was  when some other part of the system fell apart harmonics would arise causing damage to the internal fans.  This damage caused by the failure of the external fans or the rotor shaft connection would change the functioning of the system in such a way as to produce damage to the internal fins.

Weld Cracking Rotor Spider and Core

In relation to the cracking of the welds between the arm of the rotor spider and the inside of the rotor core Professor Garvey was certain that this was caused by torsional vibration associated with the first resonant frequency.  The source of this torsional vibration was excitation produced at the harmonics of the running speed of the motor.  Repeated direct online (DOL), starting of the induction machine causing torsional transients was canvassed as a possible cause.  Professor Garvey accepted this as a theoretical possibility. In any event, the Functional Specification required the rate online starting.

Bearing Pitting

The clear evidence of Professor Garvey, was that the most likely cause of damage to the surface of the steel race of the bearing, termed as’ bearing pitting’, was that stray magnetic fields  inside the motor, induced voltages across the bearing, caused current flows to breakdown the oil film, causing mechanical damage to the bearing.  These bearing currents can be prevented by bridging the rotating and stationary parts of the bearing so that the currents do not have to flow through the rotor.  Such bridging is accomplished by brushes fitted to contact the shaft.

Stator

A supporting block was dislodged from the stator end of the winding support.  Professor Garvey described this as relatively minor damage.  The copper electrical windings of the stator are supported by steel rings or bracing, and sometimes by blocks as in this case.  The supporting block that was used was unable to withstand the very substantial forces from the magnetic fields caused by the inrush currents arising from the direct online starts of the induction machine. This is a breach of the requirement that the motors should be suitable for direct online starts.

Quantum electrical

DR submitted that the damage to the motors was caused by fair wear and tear, and that in to KMG failed to properly maintain the equipment. In particular, by failing to lubricate the cylinders.  I reject this submission.  There is no reliable evidence of a failure to routinely monitor and maintain.  DR refer to” the possibility” of a routine failure to lubricate.  There was one event of failure to top up a Manzell reservoir.  I am satisfied that this was an isolated omission.

The motors were defective in so far that after 9000 hours serious defects appear at when the life contractually required was 15 years.  The defects were not of sacrificial components but of vital components which should have had a long and reliable life.

The defects became apparent in November of 1999 but DR were not contacted until February of 2001.  It appears that when contacted they did not act expeditiously.  Seven months later, on the 14th of September 2001.  Mr Sharp of KMG wrote:-

“Despite numerous requests and exchanges of correspondence between our companyies interspersed some time ago, with communications to Parsons Peebles the issue of Janice compressor motor fan failures remains unresolved.

While efforts expended by Dresser- Rand to date are acknowledged, Kerr McGee patience is finite and cannot permit the current state to continue.  Our records indicate the last contact made between Dresser- Rand and Parsons Peebles being back in June of this year, with neither evidence of subsequent progress, nor action.”

DR replied on the 21st of September 2001 :-

“ I agree that progress is slow and unacceptable, and as such, we will inject a priority “

DR blamed Peebles for the delay.  There was a dispute between DR and Peebles over the supply of drawing of the existing shaft, and then further discussions leading at to the instruction to carry out repairs and rebuilt work in December 2001.  DR did not act with any degree of expedition.  Its subcontractor was dilatory.

It is DR’s case that the hire of blower the cool the motors would have been avoided had KMG acted more promptly.  Prof Garvey stated the obvious when he said that that some of the later remedial action was potentially avoidable.  He referred to the hire of the blowers.

There is no real basis on the evidence for concluding that had DR been notified earlier then an earlier solution would have been found.

The B motor failed on the 16th of November 1999.  The claim for the hire of these blowers spans the period November 1999/31st of July 2000.

I am satisfied that the blowers were hired to provide cooling air, when the fans an integral part of the electric motors that drove the compressors, failed.

I do not accept Mr van Voorst’s evidence that there is insufficient evidence connecting the hire of the blowers from Whitaker Engineering over this period.  There is no evidence that such equipment had any other purpose on the platform and was not wholly referable to fan failures.

KMG claim £126,270 the costs of purchasing an extra standby motor, in case of further motor failures.

The decision to purchase it was made in November of 2001 when Parsons Peebles, the motor fabricators were repairing and rebuilding the other motor.

The repair of motor A was completed in December 2001 and motor B was still working.  KMG were confident that the problem with the motors had been ascertained, and that motor A would be properly repaired.  The third motor purchased was of the same design as motors A and B.

In my judgment the purchase price of the third motor is not recoverable.  It was not reasonable or proportionate to recover both the costs of repair, and/or rebuilding of the other two motors, and the cost of the third motor, which may be used and brought into service when the third compressor is brought into commission.

The work carried out on each of the motors was reasonable and proportionate and flowed from DR’s breaches.  The inspection work on motor B in October and November 2001 was appropriate and necessary.  The cost of this, £16,213, is properly recoverable.

I find that the monitoring work wrongfully characterised by Mr van Voorst, as condition monitoring was reasonable and necessary.  They should be no problems with vibration or the necessity for a vibration survey by Parsons Peebles of Rosyth.  These costs are clearly the result of the motor failure through vibration, which would not have been incurred but for DR’s proven breaches.

Quantum

KMG claim a total of £244,179. Save for the cost of the 3^rd motor (£126,270) they have proved their claim in the sum of £117,909.02.

Scott Schedule 3.1 Liquid Handling General

VIBRATION ISSUES

Scott Schedule Section 1.1 vibration.

KMG summarised its case in the Scott schedule. “ Both compressor packages suffered from excessive vibration problems to the extent that numerous component failures occurred individually or in combination through one or more of all of the following: the resonant vibration/pulsation vibration/acoustic vibration/fluid flow turbulence/vortex shedding/torsional vibration.  It is the vibration related failures which frequently exposed poor design and manufacturing defects.  The results were numerous gas leaks, frequent process shutdowns and significant additional operating expense through required modifications, ongoing inspection activity, adding piping supports and reengineering or modification of many pipework components. DEI and IDEAS were employed to measure and analyse these problems.  The main vibration problems were due to mechanical and acoustic vibration”

DR commented,” that as they were not actively involved in the resolution of any vibration issue, an orderly approach to problem-solving was not adopted.  There was a tremendous increase from the original 5 locations identified in the tests by DEI, and KMG is put to prove that all of these modifications were required due to default on DR's part.  Further and in any event, claims in this action were compromised and KMG is now barred from seeking to recover sums in relation to the same in this action”.

By clause 11.1 DR warranted that:-

“ work…. shall be free from defects, and that the works or materials goods and equipment incorporated into the work in all parts therefore shall be of prime quality…suitable for the purpose and use to which they are intended and shall perform in accordance with the requirements and conditions of the contract.”

By clause 11.2 DR warranted that:-

“ Irrespective of whether any designs, data or information prepared by the Contractor have been approved by the Company, the Contractor undertakes and guarantees and warrants the engineering workmanship, and all materials and equipment fabricated or provided by it or any of its subcontractors against any and all defects for a period of 24 months from the date of the Inspection Release note.

The scope of supply refers to specific documentation that the Contractor must fully comply with. The DR quotation, 16 June 1997 contains the following:-

“DR designs its pulsation bottles to limit peak to peak pulsation to 5% or less as a standard and guarantees that there will be no harmful pulsations.

The scope of supply requirements include:-

“ Pulsation dampeners are required to maintain pneumatic vibrations within safe limits

The Contractor shall provide a low vibration package”

The natural frequency of the skid with equipment mounted is designed such   that excitation is not encountered during operation.

DR seek to defend the claim for breach of warranty on the ground of the package was built to API 11P, which is referred to in the scope of supply under the heading “normative references “.

It seems that DR relies upon this compliance as complying with the warranties at clause 11 of the Contract and of the specific requirements that there be no harmful pulsation and a low vibration design. 

That cannot be the case, where clause 11 takes precedence over the functional specification, by virtue of the provisions of clause 40.

Equipment supplied under the Functional Specification was to conform with the nominated codes, which included API 11P only where applicable.

The two compressor trains designed and fabricated by DR did not comprise the complete process system installed by KMG on Janice A.

The compressor package was one component part in a complex system chosen by KMG and fabricated and designed by others. It was nonetheless, a component designed to operate within that process system, and not in isolation from it. 

When considering issues of fitness of purpose it is vital to bear in mind, the extent of the express contractual obligations imposed on DR under the contract for the performance of the DR compressor package and any effects caused to, or by, the equipment supplied by the other suppliers.

It was common ground between at the vibration experts, Mr Don Smith and Mr. Campbell that ideally, more rigorous computer simulation testing as to the effects of both mechanical and acoustic vibration should have been carried out.  The functional specification referred to the specification for packaged reciprocating compressors for oil and gas production services to be found in the American Petroleum Industry Specification 11P (API 11P (SPEC 11P).

Mr. Campbell expressed the view that a system design for the environment in which that package was to be deployed should include such work as is necessary to prove that system was fit for service or purpose. 

That work would include as a minimum, an acoustic pulsation study to API 618 Design Approach 3.  Mr Don Smith, in his report agreed that the system should have been designed according to API 618 Design Approach 3. “EDI always recommends that, when practical, the compressor/piping system should be designed to obtain pulsation levels at or below the levels recommended in Design Approach 3”.

In his cross-examination on day 19.  He said “So I think the whole thing is, what we are trying to say is that, unfortunately, whoever made the decision to go with API 11, 11P, they should have gone with API…… I mean, API design approach 3 should have been for the analysis of the vibration and pulsation part of it.”

Later in the course of his cross-examination, Mr Smith, in his homely fashion encapsulated the question that the court has to decide. At page 2210 day 19:-

DR in this case chose ‘to roll the dice’.

The experts who dealt with these issues were Mr Robert Campbell, for KMG and Mr Don Smith for DR.

Mr Campbell is the Managing Director of International Design and Engineering Services Ltd.  His area of specialisation is static pressurised plant integrity, and his experience covers the use in applications of most piping and pressure vessel design codes and standards, particularly European and American.  The integrity work that he has been involved in includes most damage mechanisms including vibration, acoustic, noise, fluid/gas flow, flow transients, pulsation, corrosion, erosion, motion, settlement and creep.  His career spans 34 years and includes hands-on experience, through apprenticeship to distinguished academic qualification and a consultancy dealing with practical in design problems over many years.  He has particular experience of North Sea oil and gas extraction and investigation of catastrophic events such as Piper Alpha.  I found his evidence impressive, following the rigorous testing by the searching and skilful cross-examination by Mr Boulding.

Mr Donald Smith is a Senior Staff Engineer and currently the President of Engineering Dynamics Inc. He has been involved with vibration and pulsation problems associated with rotating and reciprocating equipment, for more than 40 years.  His academic qualifications are impressive, and include the research and testing of rotating and reciprocating equipment.  Both experts have considerable experience with computer and analogue modelling. Both have been instructed by major oil and gas companies all over the world.

Both experts may be described as eminent in their fields.

They have met together in order to come to agreement where they could.  They have reported at length on the issues where they cannot agree.  Sadly, there is discernible a degree of antagonism in their dealings with each other, Mr Smith putting in question Mr. Campbell's good faith.

Mr Smith had a presentational style whilst giving evidence, which can be described as homely and robust.  Mr Campbell presents as a more careful and cautious witness.

The differences between them as to the strict technical and scientific aspects of the evidence are not great.  Their approach to the design and fabrication of reciprocating compressors, and the ancillary vessels and pipe work is markedly different.  Mr Campbell emphasises the need for rigorous investigation and testing using computer simulation, finite element analysis and other sophisticated techniques leaving nothing, that may be predicted, to chance, where it could affect the performance and long life integrity of pipes and vessels used in the compressor package.  He equates fitness for purpose with fitness for service and has considerations of safety uppermost in his mind.

Mr Smith is properly informed by commercial practicalities and has a robust attitude to compressor design, which may be described as “ you get what you pay for, a Rolls Royce specification does not come for a Ford price “

The Expert’s points of agreement state:-

Because compressors have moving parts which cause pulsation it is impossible to have a reciprocating compressor, which is not the cause of some vibration in any piping, small bore branches (SBB) and equipment fitted to it.  Such piping, SBB and equipment, have their own mechanical natural frequencies.  Acceptable low amplitude pulsations produced by the compressor, predictably may cause any piping, SBB, and equipment vibration to be amplified when coincident with its mechanical natural frequency.

The amplification factors for piping mechanical natural frequency, depends upon the available damping in the system.  When the piping systems have comprehensive damping, and low amplification, any piping vibration will disappear quickly after the piping is impacted, and the risk of high cycle fatigue failure is minimised.

In July 1999 KMG commissioned a report from DEI.  Initially, on 5^th July, five items were identified as giving rise to concern after strain measurements had been taken. Mr Smith said in evidence…” these are items that are of concern.  When you have high cycle fatigue failures, they can occur in a matter of hours or days, and so they have to watch them 24 hours a day.”

On the 21st of July of 1999 the DEI report showed that the number of items, giving concern on the SBB system, requiring non destructive testing and monitoring on a 24-hour basis had risen to 25.

The experts agreed that the leaks within the DR work scope were a cause for serious concern to a responsible operator, and warranted intensive investigation.

DEI had been contracted by KMG, to perform a vibration survey of the export gas compressor MS - 0801B on the Janice platform. The initial scope of the survey focused on 60 pipework items specified by KMG and Kvaerner Oil and Gas.

On completion of the initial survey further strain measurements were performed on other elements of the pipework, compressor frame, compressor cylinders, compressors skid and supporting structure.

The report concluded that the survey performed on the gas compressor has shown that machine was producing acceptable levels of vibration for this type of reciprocating compressor. The vibration was found to be of an acceptable amplitude with no evidence of unacceptable characteristics. The anti- vibration mounts were found to have an acceptable performance, and the vibration levels transmitted to the deck at the areas of support were also within acceptable limits.

The conclusions of the DEI report, given the extent of the investigation undertaken, gave the compressor, the skid and the anti- vibration mounts a clean bill of health in terms of the levels of vibration.

Mr Smith regarded this data as very important, because it meant that to significantly reduce vibration levels modifications would have to be added to the piping to reduce piping vibration.

Mr Campbell accepted that the initial tests appear to have been performed properly.  Both experts agreed that further investigation was warranted.

The essential difference between the approach adopted by Mr Campbell, and that of Mr Smith was the nature of that process of investigation.  Mr Campbell said that in order to assess the fitness for purpose of the compression package, and particularly the attached pipework, SBB and any attached equipment, it was necessary to quantify its performance. Further, in order to ascertain whether it would have a guaranteed life of 15 years, some proper measurement of its performance and assessment of the future durability of the components was necessary.  An approach that was reactive to incidents that were potentially dangerous or life-threatening was not appropriate.

Mr Smith took a more robust view favouring an ad hoc approach. That of dealing with difficulties as they arose, and forestalling future harm by monitoring component, piping and SBB vibration using strain gauges where appropriate to measure its potential for causing damage.  It was a stance wholly consistent with his understanding of the contractual bargain struck by the parties.

Mr Smith’s approach to his evidence was informed throughout, by his view that KMG had “rolled the dice” and by making reference to a less onerous standard in the Functional Specification (API 11)” have got what they bargained for” overlooking the primary obligation of DR to design and supply a compression package fit for purpose for installation and use on a semi-submersible oil platform in the North Sea, where safety and reliability considerations are paramount.

I found the evidence of Mr Campbell impressive, and reliable.  Criticism was made of his independence as an expert witness, on the grounds that IDEAS, the company of which he was the managing director, had done investigatory and remedial work on behalf of KMG.  I am satisfied that he approached his task as an expert witness objectively and independently. He was clearly able to evaluate work done by IDEAS in a dispassionate and objective way and where he disagreed with it’s utility or value he said so.

Both distinguished experts come from a very small world wide pool of people equipped with the necessary expert technical expertise and practical experience. Inevitably, they have done work for all of the major international oil and gas companies.

The DEI reports are significant, in so far as they illustrate the extent of the problems relating to vibration on Janice A. and the depth of concern.

I accept that they demonstrate that the compressors were free from defect as the expert witnesses agreed.

I do not accept, however, that their findings justify the conclusion that the vibration levels in the pipework, SBB system and equipment attached to the compressors comprising the DR compression package were also by and large, satisfactory.

I accept Mr Campbell's evidence that DEI’s report was a useful survey, competently performed, which produced useful measurement data, which could be used as the basis of further investigation.

DEI first carried out a visual survey following KMG and Kvaerner’s initial instructions relating to the 60 sites.

The second stage was an attempt to measure stress at the perceived vulnerable sites, using strain gauges.  These were frictional gauges measuring strain which had to be converted to  stress by multiplying the measured strain by the modulus of elasticity for the appropriate material. This data was then compared with BS 7608, which in part deals with the allowable stress in various classified types of weld.

Mr Campbell described the difficulties in taking measurements on the small bore branches(SBB), as in locations with welding, where maximum stress occurs, and the physical conformation of the piping, SBB and attached valves is such that it is not practicable to get a close reading.  Many of the readings were taken on tapered surfaces and even on bracing used to support valves, piping and SBB.

In cross-examination, he said that the stress at a relevant point, properly adjusted, could be 50 times the measured stress.

I accept that  in order to obtain realistic fatigue predictions it was necessary to carry out Finite Element Analysis (FEA) of Small Bore Branch piping (SBB) to achieve the appropriate adjustment and enable a valid analysis to be made.  This was undertaken by IDEAS.

It is evident that DEI did not have the necessary stress analysis skills and understanding of fracture fatigue to predict failure with the requisite degree of accuracy, for safety and durability assessment.

Between June 1999 and November 2003 DEI made 111 reports of which 106 were directly related to piping and small bore branch vibration.  These reports comprised vibration and stress surveys, technical notes and site reports.  They evidence that piping and small bore branch vibration problems were ongoing and extensive.

The state of affairs that had arisen in relation to piping and small bore branch vibration was serious and warranted the deployment of specialist assistance such as that, in fact supplied by the Consultants instructed by KMG, and which may have been available from DR.

A schedule of incidents was served by KMG, and some of these were put to Mr Smith on day 19.  It was not wholly satisfactory, because he did not have sufficient notice of the detail. DR  served a counter schedule, which I have considered in the context of the evidence that has been received in the trial.

It is clear that there were several gas leaks from within the package.  There were of course, escapes of gas from the other process piping and equipment not supplied by DR off skid and in presentation there was an over emphasis as the extent of the risk to safety caused by DR.

Mr Smith accepted that some failures looked as though they were repetitive in nature, such as the tubing breaks on the Pressure Safety Valves (PSVs), and many of the events were of the same type, which warranted intensive investigation to find the source of the problem.

It is clear that that DR were responsible for a significant proportion of the incidents contained in the schedule with the exceptions of items 1,8, 9, 10,13,14,15, 20,25, 26,27,29,38,39,40,41 and 55.

I do not accept Mr Smith's evidence that the nature and the extent of these incidents were such as normally occur during the start up and commissioning process of offshore reciprocating equipment in the North Sea.

The excepted items principally relate to PSVs.

Quantum Scott Schedule 1.1

KMG claim £172,319 for remedial work under this head of claim:

“The remedial work carried out in relation to excessive vibration generally comprised mobilisation of personnel, investigation surveyors, measurements, modelling, finite element analysis and recommendations for modifications.”

KMG contends that it was put to expense as damages for breach of contract.

Clearly in consequence of DR’s breaches KMG was entitled to seek the assistance of third party consultants and advisors. Insofar as specific problems were addressed and specific resources put to the investigation of it, it is not difficult to identify what was done, by whom, when and to what purpose. In order to charge for these services the service provider has to justify to the customer the charges in terms of labour, materials and associated expense.

KMG rely upon a series of invoices from IDEAS and DEI. These do not clearly identify the works carried out.

Mr Smith, whose evidence I did not accept in relation to the necessity of these investigations and studies was of the opinion that most were not required, or if justified were performed in such a way that they were of no value. His evidence informed of the careful approach of Mr van Voorst DR’s quantum expert.

Insofar as he took account of Mr Smith’s evidence as to the utility of the investigations and apportioned the costs attributable to DR’s breaches I do not accept that evidence. Insofar as he rigorously sought attribution I prefer his evidence to that of Mr Boyne.

Mr van Voorst before factoring in the Smith view arrived at a figure of £127,879.19 (£79,362 plus £48,516).

There are five invoices from DEI totalling £24,054.12 which the parties pragmatically “split the difference” to avoid the disproportionate costs of investigating them in court.

There are four DEI invoices which remain controversial. One for £14,709.23 bears the invoice inscription “vibration surveyors of compressor and pipework under various combinations of train and process conditions. Stress management as/when required. Ad hoc work on WI skid and deluge pumps.”

It is evident that whilst on the platform for work wholly referable to DR breaches, KMG took the opportunity to use DEI to do collateral work not attributable to DR’s breaches.

There is no evidence as to the exact apportionment of costs attributable to either cause. I adopt a crude apportionment (see Laing Infra) which may have the effect of penalising KMG but which is a consequence of their failure to properly account. £7,350 is recoverable under this head.

There are three DEI invoices totalling £5,676.70 with the legend “condition monitoring service”. This on the face of it could refer to routine maintenance and thus not be recoverable. I am satisfied however that the references to compressor bridle pipework investigation indicates that these relate to work to the end of remedying the consequences of DR breaches.

The total recoverable under Scott Schedule Section 1.1 is therefore £152,932.95.

Scott Schedule section 1.3 Pressure Safety Valves

There is an issue between the parties as to whether the excessive vibrations caused early lifting of the pressure safety valves, PSVs 08116/08216 and the failure of the pilot impulse lines, and the early lifting of pressure safety valves 08130/08320.

Mr Hornee of DR said that the PSVs were supplied as individual pieces of equipment to be installed by KMG. DR were not responsible for determining the location or mounting of the PSVs.

The PSVs were pilot operated valves manufactured and supplied  by Greenwood Crosby. Instead of a spring, process pressure was used to keep the seat disc of the PSV closed at pressures below the set pressure. The pilot controlled the pressure on the top side of the unbalanced moving member steel tubing and was connected to the main valve, located some distance from the compressors in the off skid piping, with stainless steel tubing.

The first stage PSVs 08116 and 08216 would lift at a pressure of 28.5 barg when the set pressure was 33.5 barg.  The second stage PSVs 08130 and 08230 were recorded as lifting at pressures approximately 15 barg below the set pressure.

I am satisfied that vibration caused the early lifting problems since the set pressure adjustment is by a screw on a spring, which can easily be affected by high vibration.  Excessive vibration levels can also result in fatigue failures of the tube connecting the pilot valve to the main valve.

KMG used Kvaerner to install the PSVs.

Mr Brian Fitzpatrick of Kvaerner, carried out a reciprocating compressor pipe support survey between the 14th -19th April 1999.  He found that there were spans of relief pipework, which were unsupported and at the limit for carbon steel pipe in normal service.  Mr Fitzpatrick advised that additional piping restraint should be installed to reduce the piping vibration levels at the PSVs.

Between 4^th and 11th of May 1999 Mr Austen, of DR, found that the valves were subject to excessive vibration levels.  They were vibrating at multiples of the compressor speed, with maximum amplitudes at 44 Hz or three times running speed.  The pulsation levels measured in the top and bottom of the PSV pilot valves showed that the pressure pulsation levels were low, indicating that the excessive vibration levels were mechanically excited.  Had the pulsation levels been excessive, that would have caused the PSVs to continuously trip.  There is no evidence of this, although it is recorded that the same PSVs have failed on more than one occasion.

The first stage PSVs had been installed on a long line, which had too few supports and clamps. There were also a number of elbows immediately upstream of the PSVs, points where pulsation can easily couple to the piping.

I am satisfied that compressor pulsation inevitably contributed to the vibration and PSV lifting. The acceptable levels of compressor pulsation and consequential vibration have to be satisfactorily managed in design terms for the associated piping SBB and equipment. In relation to the PSV lifting and injurious vibration it has not been. 

DR is not liable for this failure. Some of the piping and equipment affected was off skid and clearly not in the compressor package contracted to be supplied by DR. The siting and installation of the PSVs in any event was not the responsibility of DR it having been entrusted to Kvaerner unknown to DR. DR had no contractual obligation to monitor or supervise the siting or installation of the affected PSVs.

Where pilot impulse line valve failures occurred a solution was found by relocating the pilot valves and fitting flexible impulse lines isolating the pilot valves from the vibration.

Scott Schedule 1.4 Cracking of Pipe /SBB

The claimant contends that the excessive vibration also gave rise to fatigue failure and overstressing on the small bore connections to the pulsation bottles on both A and B trains.

I am satisfied that this was the case and that DR is in breach of the Purchasing Contract as alleged. 

The defendant contends that if there was a breach, nonetheless, there should be a finding that the excessive vibrations were not caused by the compressor.

Such a finding would be at odds with the experts’ evidence recorded in their points of agreement.  Whilst DEI found an acceptable level of pulsation in the compressors, that nonetheless, predictably contributed to the vibration in the piping, small bore branches and equipment attached within the DR compressor package, and which caused fatigue failure gas leaks, process shutdowns and some additional operating expense, whilst remedial work was undertaken, which involved considerable investigation, by way of measurement and analysis, and the need to fit supports and bracing to piping, small bore branches and valves.

Scott Schedule 1.4: Quantum

Under this head KMG claim for the cost of the remedial work carried out comprising the investigation survey, re-tightening, mobilisation of men off-shore, designs for and work to support stiffened clamp and brace pipework and to install flow and restriction orifices in the pipework and replacement of the small bore instrument valves with lighter designs to mitigate vibration problems. The cost of this was £472,839. (RAPC per 85 D)

Mr van Voorst in the experts joint agreement of June 2006 agreed a figure of £391,667 leaving in issue £70,329. Three invoices from Alco Hi-Tek in the sum of £2,759 have been subsequently agreed.

The balance remaining in contention comprises £66,084 of which £40,668 comprises the following DEI invoices:

A balance of £25,415.96 remains in dispute. There is a claim for £8,078 representing a deduction by Mr van Voorst from invoices presented by IDEAS in respect of excessive hours. This is properly claimed.

There is a sum of £1,632 which relates to four invoices from NECE. I hold that these are not recoverable. There is insufficient evidence to justify this claim.

A sum of £15,289 representing two invoices from Whitaker Engineering is claimed for pipe support. The purchase order refers to specific drawings and the supports are for DEI items number 6, 23, 24, 46, 47, 21, 1, 33 and 34.

The job number 1722 clearly relates to vibration issues and the job numbers in my judgment are sufficient evidence of attribution. I am satisfied that the locations identified are all within the DR package and the appropriate skid. The sum of £15,289 is properly recoverable.

An invoice from Caledonian Petroleum Services in the sum of £415 for a pipe support manufactured and supplied in accordance with drawing 8140-312-P-PS-66B is clearly referable to job 1722 and is properly attributed to the remedial works consequent upon DR’s breach and is properly recoverable. No further sums are recoverable.

Scott Schedule section 1.5 Dampener Crack Failures

At section 1.5 of the Scott schedule complaint is made of dampener crack failures and design anomalies.  On 7th April 1999 there was a weld fracture on the first stage dampener which gave rise to a gas escape.  Defects were found in compressor B in a 3/4 fitting attachment, and on compressor A on an N2 fitting on the pulsation dampener of the first stage suction. 

On the 20th of August 1999 it was reported that a temporary wooden support had been fitted to the second stage discharge bottle on compressor B to reduce vibration and piping stress.  A similar support had been fitted on compressor A.

There then followed a static dynamic and pulsation report by IDEAS which reported that the dampener vessels analysed were found to be excited by mechanical and acoustic sources.  The dampener crack failures were apparent by June 1999.

IDEAS controversially concluded that there had been a shell mode of vibration and recommended a mechanical clamp.  In fact a welded clamp was fitted to the dampeners at modest cost.  KMG claim the costs of the expensive investigation conducted by IDEAS in addition to the modest welding costs.  The welding work had to be done onshore.

The issue between the parties is whether it was appropriate to carry out a Static Analysis of the piping system.

Item 6 of the experts’ Points of Disagreement records that Robert Campbell believes that the static analysis of piping carried out by IDEAS was necessary to quantify the status of the piping system for the Static Loads and basic Fitness For Service (FFS)

Mr Smith disputed that saying there had been no reports of problems due to excessive thermal loads such as load cycle fatigue failures.

The thermal analysis was carried out on most of the pipework with the exception of the SBB.  This was not done because it is difficult to model the actual shape of the small, often hand fabricated tubes, for a pipe stress analysis.  Many of the fatigue failures occurred in the SBB.  Such failure as occurred on the PSV tubing was caused by high cycle fatigue.

Mr Campbell accepted that thermal analysis of itself, would not reveal anything about high cycle fatigue.  The utility of this work in relation PSV tubing is not immediately evident.  Mr Smith says it has no relevance.  Mr Campbell says it gives one necessary data as a prelude to making a value judgment about fatigue life.

I accept that to be the case.  However, since much of the PSV tubing was not the responsibility of DR and the thermal analysis had no relevance to SBB's it is difficult to know how much of the cost of this analysis is attributable to matters for which DR may be held liable.

A great deal of time was occupied during the trial in investigating whether analyses should have been carried out by DR at the design stage.  In an ideal world and were the project not a fast track project to be completed in eight months, some of the work carried out by DEI and IDEAS after installation on the Janice A platform could usefully have been done at the design stage.  Clearly this has cost implications, and DR, in Mr Smith's words ”rolled the dice” and took the risk, that absent quantified analysis, the compressor package as designed, would not be fit for purpose for installation and use in the North Sea.

At paragraph 1.4 of the Scott schedule the defendant comments: “DR was in a competitive bid situation when making this proposal and their price estimate, like those of its competitors was based on the requirements of the functional specification.”  In any event the lead time necessary for the fabrication of the dampener vessels, significantly shortened the time available for analysis as to their design capacity.

In my judgment there were analyses that should have been conducted by DR at the design stage despite lack of prescription by KMG in the functional specification.  The failure to carry them out, gave rise to the design and fabrication of a compressor package that was not fit for purpose.

Quantum

KMG claim £87,704. Mr van Voorst believes that IDEAS hours were excessive. They are not demonstrated to have been excessive. He valued the claim at £79,214 disallowing half of the hours charged for by IDEAS.

KMG are entitled to recover £87,704.

Scott Schedule section 1.6 Thermowell Failures

Paragraph 1.6 of the Scott schedule states that:- “ two thermowells failed on the second stage discharge dampeners causing gas leaks. The well in each case breadth of the friendship area through a high cycle fatigue mechanism.  KMG carried out intensive nondestructive testing of all compressor thermowells and replace them with interim standards before fitting re-engineered and redesigned thermowells.  The original thermowells had not been wake frequency checked for the possibility of vortex shedding which may have given rise to a risk of high cycle fatigue failure.  The thermowell design was to DR in-house standard.  The material was confirmed as bar stock, and no vibration calculations were carried out on behalf of  DR.”

Investigation of the failure included frequency spectrum analysis and radiography.  The Materials Engineering reports record that both failed from fatigue cracking. Mr Campbell said that, in his view the fatigue could be wake induced resulting from the flow drag of the process gas in compression.  He canvassed a further possibility; that there was a basic excitation, as a result of the state of the vibrating branch on which the thermowell sat.

Whilst both experts agree, that is whatever the exact cause of failure the two thermowells were not able to withstand conditions in the service it is necessary to determine the probable cause of failure in view of the cost of investigation and replacements claimed by KMG.

In my judgment the more likely cause of failure was wake induced fatigue failure as a result of the flow of gas in compression. I accept the evidence of Mr Campbell as to this cause being the likelihood.

These failures cannot be considered in isolation.  The first thermowell failure was part of a sequence of vibration related incidents some involving gas releases, all of which gave rise to serious concern.

The second thermowell failure also on the second stage discharge dampener served to heighten that concern.

Two courses were open to KMG.  To investigate conditions at the sites of the thermowells, with studies to determine the suitability of the thermowells to withstand those conditions or to replace the thermowells with wave checked thermowells that KMG felt confident would not fail.  Both courses would involve expense.

KMG, in fact replaced both of the failed thermowells, and all other thermowells in the pulsation dampeners of the other stages.  Additionally, they commissioned an interim stock of carbon steel thermowells whilst waiting for permanent replacements made in steel.

The KMG solution was not a quantified solution.  It was to fabricate shorter replacement thermowells.  Mr Campbell accepted that the effect of the shorter thermowell could even have exacerbated the problem because the mechanical natural frequency might thereby have been raised and become coincident with the excitation frequency.

The KMG solution was overkill and they are only entitled to the cost of replacing the two failed thermowells. The cost of these replacements and fitting is agreed in the sum £1,758.77.

Scott Schedule Section 1.7 Cooler Crack Failures

At paragraph 1.7 Scott schedule KMG epitomises its case in relation to cooler crack failures:-

”During a survey of small bore items DEI discovered a gas leak from the first stage cooler end cap on B train.  The baffle plate inside the cooler had suffered a fatigue crack, starting at the end of the baffle to outer shell weld, which had propagated along the weld and through the outer shell.  Pressure pulsations and flowed turbulence were the likely excitation forces acting on the baffle plate.  The result was an emergency shut down, a long time with the compressor out of service, pulsation surveys, reengineering of all cooler end caps with the involvement of IDEAS,DEI and Glapwell Contracting Services.  A further similar failure on the first stage cooler on A train occurred before it had been upgraded.”

DR responded: “this explanatory note refers to an alleged gas leaks from the Channels (a.k.a. Bonnets, End Caps ) of the Shell and Tube Heat Exchangers…(reference is made to)… anticipated problems with tubes that were subsequently resolved without modification to the coolers.  No communication took place between DR and KMG on the issue of cracking of channels prior to disclosure.  KMG has failed to mitigate its losses in that extravagant methods of problem resolution were adopted by KMG, involving five different suppliers, none of which was the equipment manufacturer OEM a renowned specialist in this field.  Failure to notify DR of this alleged defect did not allow DR to reduce potential losses by involving Cooler OEM, either on a warranty basis (at no cost to either KMG or DR ) or alternatively at considerably lower cost…”

There are four gas coolers to each compressor package.  Pre –Cooler(Suction Cooler) MH 0801, the 1st Stage Discharge Cooler, MH 0802, the Final Discharge Cooler(3^rd Stage Discharge Cooler)MH 0803, and the 2^nd stage Discharge Cooler)MH 0804.

Only the 1st Stage Discharge Cooler on unit B failed(MH 0208B).  There was a gas leak, due to a gas leak through the vessel wall. Cracks were discovered in the vessel wall and the partition plate.  The crack occurred in the removable bonnet section of the cooler.  The bonnet sections from coolers MH 0801A, MH 0801B,MH 0802A,MH 0804A, and MH 0804B, were taken to inspection by a surveyor from Engineering and Technology Group of Lloyd's register to see if the vessels complied with the requirements of ASME Div 1 1998 part of the Functional  Specification.  Despite the design of the coolers, having been approved and inspected by Lloyd's register there were several items that did not meet the code specifications.

A detailed metallurgical investigation of the coolers on both trains indicated that the crack on MH 0802 B, was associated with a stop/ start crater caused by stopping then starting a new weld.  Such a crater could lead to high stress concentration and was some evidence as to why the crack occurred where it did.

The quality of the welding on the cooler was considered to be generally good.  However, the weld connecting the baffle plate to the shell had an uneven leg length and a concave profile and the joint where the crack occurred showed the stop/ start defect.  There is a suggestion that the design of the joint was open to question, where the baffle plate thickness exceeded the shell plate thickness.  In relation to the cooler end box MH 0801A some surface cracking was detected when a dye penetrant was used but no failure was associated with this cracking.

On the 28 June 2000 KMG raised with DR(UK) the question of DR at Broken Arrow US.being involved in checking the design of the 1st stage Discharge Cooler. DR’s report 19^th August 2000 relating to the Janice A compressor makes reference to the failure of MH 0802B. It is evident that the cooler manufacturers were contacted. There is no evidence however, of any warranty claim being pursued by DR.

On 5th July 2000 IDEAS performed a preliminary analysis on the failed cooler and recommended a short-term solution involving stiffening, using plate and web techniques with the expectation that the key life would  be increased to just over two years, assuming pulsations inside the cooler at 10%.  The assumption was necessary because the pulsations had not been quantified.

On 21st July 2000 IDEAS issued a Finite Element Analysis report using assumed pulsation levels. They concluded that all of the cooler boxes except MH 0801 were liable to fail and recommended the re-design of all boxes.

Criticism is made that the assumptions made were far too conservative.  In particular that the compressors would shut down and start up each day, when the operating expectation was that compressors should run for several months without shut down.

In principle it seems that Mr Smith was prepared to accept  that an FEA was justifiable, albeit reluctantly: “… But I do not think you should immediately jump in and start doing an FEA without contacting at least the manufacturer”(T 19)

The real issue between the parties is as to the assumptions made for the purposes of the analysis.

Mr Boulding submits there was a straightforward engineering solution to the problem suggested by MEL who undertook the metallurgical examination, namely that upgrading the welding of the failed cooler box would suffice.

Mr Campbell discounted MEL’s recommendation on the grounds that …”they did not have the background or the understanding of design and fracture mechanics to make any value judgment like that”.

Mr Boulding went on to submit that Mr Campbell was forced to dismiss these recommendations out of self-interest to support his company’s unnecessary and expensive investigation by way of FEA.  I reject these submissions.

It is clear that a more sophisticated investigation by FEA was warranted.  The conservative assumptions made for the purposes of the analysis I am satisfied were appropriate.  The failure data properly interpreted ranks failure in order of priority and is intended to enable the degree of risk of failure to be assessed and not to predict when any particular failure will in fact occur.

Modifications were carried out to all coolers.  In relation to MH 0802 and MH 0804, a central stiffener was added under the petition plate in the Outlook section to reduce the localised stress.  A full penetration weld was used on the partition plate to reduce its susceptibility to the fatigue and to eliminate the area of discontinuity at the original crack initiation.  The dome end and barrel wall thickness of the gas boxes for MH 0801 and MH 0802 was increased to 12.5mm. 

In relation to MH 0803, the semicircular flange separating the inlet and outlet areas of the gas box was re-placed with a circular flange ring.  All coolers were subjected to full nondestructive testing of all welds before they were put into service. 

Mr Smith, in his original report said that the modifications to MH 0801 MH 0802 and MH 0804 were primarily to reduce the stress on and improve the strength of the welds, thus, by simply improving the welding afforded a cheaper and quicker solution.

In view of the nature of the failure and the potential risk to safety and prejudice to long-term continuous running it was reasonable in my judgment, to undertake a quantified analysis.  This did not arise out of an isolated event.  It was part of a growing history of failures, and it was inappropriate to deal with it on the basis of a pragmatic sticking plaster solution. 

I am satisfied that excessive vibration was the dominant contributory cause of the failure of cooler, MH 0802B and that the modifications were appropriate. 

I accept the evidence of both Mr Campbell and Mr Smith that it was appropriate on discovery of the failure for KMG to go back to DR in case they, or their supplier had data, which quantified the performance of the boxes and was relevant to their designed fitness for purpose, and sufficiently rigorous for a 15 year life on the North Sea.

DR clearly had that opportunity.  The matter was specifically brought to their attention, as was acknowledged in the report of the 19th of August 2000.  Doubtless, had they such material, or indeed, their suppliers, it would have been put before the court.  They would not have sat on it.

Scott Schedule Section 1.7: Quantum

KMG claims £164,387.37. £156,243.37 is agreed by Mr van Voorst.

The dispute relates to a single NECE invoice in the sum of £8,144.

The invoice is one of a number relating to “inspection of gas compressor, cooler/tube bundle”. The timing is consistent with work in respect of cool end camp failures but there appears to be a mismatch between the purchase order, which refers to off-shore work, and this invoice, which refers to on-shore work.

It is evident that the nature of the work carried out involving preparing the work pack, attending meetings and providing technical assistance was done both on-shore and off-shore. Mr van Voorst in evidence showed caution as to attribution because the Cost Code shown was not consistent with the KMG job number. I am satisfied that this was because it would not be a KMG job number, rather the job number of NECE who carried out the services which I am satisfied were part of the remedial works attributable to DR’s breaches. I hold that this sum of £8,144 is recoverable and thus the total claim by KMG for £164,387.37 is proved.

Scott Schedule Section 1.8 Pressure and Temperature Switches

Scott schedule, paragraph 1 .8 relates to pressure and temperature switches.  KMG assert that:- “ the early switching the pressure and temperature switches was common due to excessive and mechanical vibration.” this caused many trip  shutdowns and, on at least one occasion, caused a gas leak.  This required the following switches, to be relocated off the skid to escape the vibration input.( TSHH08143 et seq... and the equivalent switches for unit B. ) Reference Change Request 2014.  Subsequently, pressure differential switches PDSH-08132 and PDSH-08232 required to be located under Change Request number CR 3112.”

DR responded “ From the evidence produced... it is clear is that KMG made no effort to advise DR of their dissatisfaction with the referenced switches, and that DR cannot therefore be held responsible for costs associated with in this head of claim. DR can find no indication that KMG asked the advice of DR, or more importantly, the instrument suppliers regarding the reported problems.”

It is accepted by DR that the compressors experienced problems due to early switching of pressure and temperature switches due to excessive vibration levels. 

It is submitted that during the initial commissioning of the compressor and the associated equipment are several other pressure and temperature switches that were mounted on the ends of the small bore branches on the compressor pulsation dampeners, failed due to excessive vibration of the small bore branches. In consequence the failures of these instruments resulted in many trips due to incorrect readings which were taken by pressure and temperature switches. 

The cause of the problem was the fact that the vibration levels on instrumentation mounted on the skid, such as pressure and temperature switches, can be amplified by a factor of 50 when the mechanical natural frequency of the support bracket is coincident with a multiple of the compressor running speed.  Hence, excessive vibration can occur even when the vibration levels on the skid compressor cylinders and the suction and discharge dampeners are acceptable.

The Functional Specification provides that the overall control system for the compressors is within the package supply.  “The Contractor DR is responsible for the design and supply of all necessary instrumentation and control systems for the safe operation startup and shutdown of the gas compression package.  This includes the ability to start and stop one compression train without affecting the operation of the other train.  These provisions include the overall compression system controls, as well as the individual machine capacity controls.

The inlet pressure at the gas compression package will be controlled by the company.  The contractor is to provide suitable controls so that the package produces the discharge pressure over the full range of operating conditions i.e. the contractor is to provide a fixed discharge pressure control system.

The controls for the package are to include all necessary alarms, trips, relief valves and bursting discs required to the safe operation of the equipment.  Also all required shutdown valves etc. are to be included from the safe startup and shutdown of the package along with the required control logic.  Additionally, the following external signals are to be sent by the package to the company's external control system for each compressor :- -running, stopped.

In addition the following general package signals are also required:- common alarm, common trip, unit alarm, unit trip….”

The switches were clearly within DR’s workscope.

Mr Ronald Curley, the offshore operations supervisor, and later offshore installation manager of Janice A between April of 1999 and February of 2001, described how he was concerned at various levels, with the commissioning and how Janice A took years to achieve steady state reliability.  He described how on the platform KMG first experienced hydrocarbon releases in small bore piping and pilot lines, and then saw escapes from thin walled pipes of 2 inch bore and greater.  “We saw at least one failure of the 4 inch pipework.”

The failure of the large bore pipe is even more troubling than the failure of the smaller bore pipework, because the larger pipes carry more gas and the potential consequences of the gas release are more serious.  The next type of failure, we saw was the failure of the thermowells.  This is extremely worrying, because this was on the high-pressure side of the compression package.  In addition to gas releases was suffered instrument failures to vibration-pressure switches producing spurious instructions to shutdown-and valve control problems.

Mr Curley was in an unrivalled position to evidence the difficulties on board the platform of which he was effectively the Captain and charged with statutory and regulatory responsibility for personnel and plant to the Health and Safety Executive.  He described the compression system reports which were compiled by IGL process engineers on board, who were’ babysitting’ the compression package.

“The IGL engineers were required, as things were so bad with the gas compression system that the Operations Supervisor could not cope with the package, and everything else, without assistance of dedicated process engineers. IGL were recruited to monitor on a 24 hour seven days a week basis, the compressors.  The reason we went for professionally qualified engineers, is that the problems were so technically challenging that we needed dedicated hands-on personnel for our own criteria of safety, and in deed, the use of IGL was endorsed by the HSE.”

Mr Curley played tribute to the DR personnel on board:-

“DR personnel were good at the mechanical side of the compression package, i.e. they had a good knowledge of the mechanical side, with the problems went beyond mechanical problems and, while I respect their ability to do with maintenance and servicing of the compressors, they were unable to deal with the vibration and process problems we were experiencing.”

Mr Curley confirmed that the problems experienced with the switches described at section 1 .8 in the Scott schedule were due to excessive vibration.  The switches were visibly affected by vibrations, and when relocated this put an end to the problem of early switching. 

Mr Campbell gave evidence that the configuration of the switches made them highly susceptible to vibration, because the pressure and temperature switch locations typically had longer stand outs and large unsupported masses.

It is naive to conclude that DR personnel were not aware of the problems.  In relation to temperature transmitters and solutions requested on the first of September of 1998 by Mr Cashin of KMG , Mr Greenhalgh of DR responded, stating that the possible vibration problem will be addressed when the instrument performance is assessed during the imminent unloaded run.

The issue was raised by DR personnel again in a fax from Mr Colin Smith on 21st of August of 1998 when he stated on the half of himself and Mr Perry a colleague:-

“ we feel very strongly that should the transmitters remain in their present condition during the operation of the units there is no doubt that the failures will occur, which could result in prolonged shutdown of the units and problems to DR.

the only long-term solution that we can see is replacing the existing instruments and repositioning off the compressor pipework.

We realise that to implement this modification is major, but given past experience this may well be the only effective long-term solution.”

Recorded instances of high vibration levels at switch locations were noted by DEI. The initial vibration surveys carried by DEI and detailed in their first interim report CM.  204/01, indicated unacceptable levels of vibration, on five of the pressure switch locations on the each skid, judged against DR’s own velocity limiting criteria.

Quantum was agreed in the sum of £79,176.12.subject to mitigation.

Scott Schedule Section 4

2^nd/3^rd Stage Recycle Valve

KMG complain that the second/ third stage, recycle valves, had poor control and position stability.  It is explained that in the second and third stage recycle valves, recycle the gas around both the second and third stage compressor in one step.” this is an arduous duty for a valve as it takes a  particularly high-pressure drop.  The valve actuators could not hold  the valve position correctly, due to vibration.  The vibration was found to be of a different character and frequency from their calls to buy the compressor pulsation this vibration was induced by the valve itself.  The valve was found to be incorrectly fitted; when this was rectified the situation improved her whether further investigation determined that a larger, more sophisticated valve was required.”

DR’s response was that it could not find any evidence that a larger, more sophisticated valve was required and maintained that the originally supplied valves following correction in August of 1999 were fit for their purpose. “ The expenditure by KMG of £118,000 possibly improved the overall performance of the recycle system but this was an improvement, and not a necessity and/or expenditure, which was caused by any default on the part of DR.”

It is noted that there are no specific indications of problems with the original valves between September 1999 after their correct orientation, and the fitting of the replacement valves in July 2000. DR also noted that no record exists within the Scott Schedule of  repairs to replacement valves to the value of £29,000 which occurred between October 2000 and April 2001.

It is common ground that the valves were installed by Fisher controls the wrong way round, with the result that the gas flow went through the valve is the wrong way.  The valves were fitted in such a fashion, because the flow direction arrow had been fitted by Fisher controls in the wrong direction.  The valves were fitted correctly, in August 1999 the work being complete on 12th August.  They were fitted with a Whisper Trim 111, an anti noise device designed for noise reduction.

In July 2000 KMG replaced the valves with larger and more sophisticated CCI valves, which were more expensive.

There are two issues.  Firstly were the original valves designed with a sufficient capacity and secondly, when refitted, did they work adequately.  The resolution of the second issue will determine the first.

Mr Curley whose evidence I found impressive and impartial, in his supplemental witness statement said :-

“My recollection is that these valves were particularly problematic even after they had been oriented in the correct position.  The vibration was so bad that it caused a number of failures on the valve positioner feedback control arms, which wore away the metal on the arm.  In an attempt to stop this vibration and improve the operation of the recycle valves I remember large amounts of bracing being fitted to the valves.  Eventually the valves were changed out in an attempt to improve reliability.  I recalled that once new valves were fitted reliability significantly improved.”

Mr Curley's evidence was corroborated in part by the measurements recorded by DEI in their report 21st of December of 1999 six months after the valves were reorientated.

Mr Smith expressed the view that the vibration disappeared when the valve was reorientated.  That is not wholly borne out by the evidence. I am satisfied that it may have been necessary to consider replacing the Fisher valves to deal with the unacceptable degree of vibration. The chosen replacement CCI valves however were not free from difficulty. 

When  operating in recycle some tripping was experienced on high first stage discharge pressure, requiring modification.

Mr Sylvester-Evans, KMG’s process engineering expert accepted that there were tripping problems in 2003 “…. when you are operating in recycle more often, then this was a problem, yes.” He accepted “that the tripping problem was attributable to the valve being oversized “… but it was not causing the vibration issues.”

Mr Sylvester-Evans had only seen the preliminary design information for the CCI valves.  He had not seen the final design of the trims, or sizing of the internals.  He could only say that the CCI valve was a more sophisticated valve( than the Fisher valve), which was designed to deal with and limit the  degree of vibration due to flow, and that four months after its installation in July 2000 it had to be sent away for repair.

The cost of replacing the original Fisher valves is agreed at £79,176.12.

I am satisfied that the wrong fitting of the Fisher valves gave rise to excessive vibration.  I am not satisfied on the evidence before me that the fitting of the CCI valves was an appropriate or economic way of limiting such vibration as persisted after the refitting of the Fisher valves.  Furthermore, it is evident that the more sophisticated replacements constituted a betterment so far as process flow is concerned, as opposed to the vibration issues.

There is no cogent evidence before me as to what element of the agreed £79,176.12 could be apportioned to represent the degree of vibration limitation achieved by fitting the replacement valves. A nominal £1 will be awarded under this head.

Scott Schedule Section 2 Weld Quality

The Scott schedule, at paragraph 2 epitomises the claimant's case in relation to welding issues.

“ Pipes and vessels within the compression package contained highly flammable gas.  The gas was at pressure and present in large quantities.  Uncontrolled release of gas present a risk of fire and explosion. 

If the pressure retain the integrity of types of vessels and joints in the system has been compromised leaks will result producing uncontrolled releases of gas. Leaks of high-pressure gas in large quantities can be catastrophic…….

the excessive vibration within BR's package together with failures due to poor world quality lead KMG to carry out work to satisfy itself of the system’s integrity.  This work was performed in order that KMG it could be satisfied that the system supplied by the DR could be operated without risk of death injury or damage to property…… KMG rely on the vibration damage and leaks described in section 1, as well as the failures and defects listed below.

Without limitation of this case, KMG also contend that the weld quality on the large diameter spool pieces became a very high profile issue following the failure of the 4 inch first stage recycle line on B compressor in January of 2000. The weld quality of the affected spool was found to be very poor.  In addition, the stalled peace was found to be in thinner wall 10s rather than the required 40s. 

Due to the background of numerous weld with failures through excessive vibration, and the potential enormity of main process lines containing serious welder defects a further programme of radiographic inspection by NECE was carried out.  This uncovered a least two other seriously defective welds in a 10” intersection of suction piping.  A review by NECE of Dresser Rand’s records revealed missing records and inconsistencies. Spools considered to have a weld quality were replaced under Change Request number CR1722.”

DR responds: “DR’s alleged defaults(which are denied) are not the dominant cause of the costs allegedly incurred by KMG.  With the key document relied upon by KMG appears to be Rev 3 of NECE document NECE/PRJ/1375.  At section 2 .1 NECE, stated said that five systems were highlighted for  Inspection following a risk assessment carried out in 1998.  Only one of these systems, namely produced gas, was within DR’s work scope.  At the date when NEC determined the systems, which required testing( i.e. September/October 1999) there had been one only failure (of a1 inch purge connection) on DR supplied material. 

The evidence indicates that only two DR welds in fact failed (one1” one 4”).  Both welds were on piping, which was subjected to 10% radiography only, and on spools, which were successfully hydrotested by DR. DR puts KMG to prove that any piping failed as a result of defaults on DR ‘s part rather than due to other matters such as the flexibility of the vessel (or lack of the same ).

DR maintains that the cost attributed to this head of loss ( £459,142.31, of which £433,575.62, were costs associated with the additional nondestructive examination ) was not caused by any default on the part of DR and is in any event, wholly disproportionate.

The starting point is KMG’s Functional Specification included in the Purchase Contract.  It states that ASME/ANSI B 31.3 - Chemical Plant and Petroleum Refinery Piping- was to apply for use in the package design.

The Functional Specification did not impose any specific requirements for nondestructive examination of pipe welds. DR advised KMG in its Inspection Test plan that the standard being adopted was DR's in-house standard.  It is contended that this accords with good and sound industry practice.

DR’s case is that this is good enough even if, as happens, a weld fails, provided there is compliance with the standard.

Reliance is placed upon the recognition by Mr Campbell that defects are inherent in welding, and the acceptance by Mr Sharp of KMG that the welding shop at Wythenshawe doubtless was a fully qualified and certified ASME U-stamp facility and ISO 9000 approved involving quality audits are welding shop performance.

KMG's case is that once it is demonstrated that there are a number of serious weld failures, the integrity of the pressurised system is compromised and for it to continue to operate safely and efficiently in the North Sea  systematic nondestructive testing must be carried out in relation to entire system.

It cannot be that any weld fracture qualifies as a justifying an open-ended investigation.  Clearly that  investigation would have to be proportionate to safety considerations and expense and take account of all contributory causes.

The first fracture occurred 7 April 1999.  A crack in A compressors, first stage suction dampener (MV 0805) was observed in the vent connection.  It had been supplied with a blind flange only. It had been used by KMG as a purge connection and a number of heavy valves added to it according to Mr Hartley of DR the only witness, who gave first-hand evidence about it.  The evidence stood uncontradicted and unchallenged.

The inference is that the vibration, which may have caused the weld to fail was itself caused by loading inadequately supported pipework with a valve in a position not designed to receive such equipment.

KMG’s Mr Sharp accepted such a scenario was likely to cause a weld leak.

The second weld fracture occurred on the first stage recycle line, where the 1 inch connection of PIT 8215 branches off from the 4 inch recycle line.  It occurred on 4th June 1999 and was a very serious failure with potentially catastrophic consequences.  There is no evidence of any unforeseen use or abuse of that part of the piping system.  It failed under normal conditions, and the failure was wholly attributable to the quality of the weld and the occurrence of excessive vibration in the compressor package.

In January 2000, a weld on the 4 inch line upstream of the first stage recycle valve PIC 08215 fractured.  This too was a very serious failure with potentially catastrophic consequences.  Again, there is no evidence of unforeseen use or abuse of that part of the piping system.  The inference is clear, namely that it failed  under normal operating conditions in the North Sea at a very early stage in the compressor package’s warranted life of 15 years.

It is not surprising that KMG put in hand a systematic expert examination of the welding in the package not only by reason of the third failure admitted by DR but also in the context of the cumulative, emerging picture.

KMG is criticised because it did not seek the assistance of DR in carrying out either the repair work or the testing of the system. DR may well have replied that the notification of defect was not in the proper form, and in any event, the welding shop was properly certified and registered as proficient and able to carry out welding to a high standard.

Some vibration emanating from the compressors was inevitable. The consequential effect of it had to be managed and designed for, since it affected the remainder of the package which included the piping and small bore branches. One effect is vibration damage caused by fatigue. It affects the materials comprising the equipment and the welds.  The welding issues do not stand alone as isolated matters.

This head of claim as quantified by KMG has undergone change.  By July of 2005 it had grown to £459,142.31.  Mr Boyne, KMG's Quantum expert has adjusted this value to £111,591.14.  It is based upon an invoice review, prepared by NECE Ltd.  The review exercise was conducted by Mr Donald Smith NECE’s Managing Director.  This hearsay document was not the subject of any notice.  It records that those who wrote the report were of the opinion that:-

“ it was highly likely that the work charge for would have included gas compression inspection work and that a percentage    could be used to cover gas compression work “

Clearly a great deal of other work was performed by NECE and charged for, hence the original claim for £459,142.31.

No witness was called from NECE.

Mr van Voorst having rigorously considered attribution values this part of the claim at £37,202.94.

On this part of the claim I prefer the approach and reasoning of Mr van Voorst. It would not be appropriate to apportion on any other basis. Under this head I value KMG’s claim at £37,202.94.

QUANTUM

There are a number of quantum issues of general application that must be considered before the remaining detailed claims are dealt with.

Limitation of Liability

See paragraph 109 above

Mitigation

The cost of measuring investigating problems and carrying out work in order to put the compressor package in a condition where it conformed to the contractual warranties are all expenses that flow naturally from the breach of contract and are recoverable under the first limb of Hadley & Baxendale (1854) 9 Ex. 341.

The right to recover such expenses is not unqualified.  As Viscount Haldane L.C. said in British Westinghouse Co v Underground Railway (1912) AC 673, at 689:

“The fundamental basis is as compensation, the pecuniary loss  naturally flowing from the breach; that the first principle is qualified by a second, which imposes on the claimant the duty of taking all reasonable steps to mitigate the loss consequent on the breach, and bars him from claiming any part of the damage, which is due to his neglect to take such steps “

DR contends that KMG failed to mitigate in three important ways arising out of the use of third parties.  First, the resource of expertise of DR was not deployed. Second, that failure caused delay and additional expenses thereby.  Third, unnecessary and thereby unreasonable work was performed by third parties.

DR's case as to the use of third parties to investigate and advise as to remedial work is succinctly put in para 802 page 514 of the closing submission:-

“It is quite clear… that KMG, simply threw time and money at the difficulties it allegedly experienced with the compressors with either providing DR with any proper opportunity to carry out rectification works or adopting a reasonable and cost-effective approach.  There was simply no intention to allow DR to get on with commissioning or warranty works from the very start.”

DR submitted that the proper approach when reliance is placed on third parties is illustrated in Governors of the Hospitals of Sick Children v McLaughlin & Harvey Plc (1987) 19 Con LR.  25  where it was held that the that a Claimant can generally recover costs incurred in pursuance of the course of action, which is being recommended by a professional adviser even if the course of action eventually increases the claimant’s loss.

DR relies upon the passage in the judgment of HHJ Newey QC at page 96:-

“However reasonably the plaintiff acts he can only recover in respect of loss actually caused by the defendant.  If, therefore, part of the plaintiff’s claim does not arise out of the defendant's wrongdoing, but due to some independent cause the plaintiff cannot recover in respect of that part… The event may take the form of negligent advice upon which the plaintiff has acted.  Another way of expressing the matter might be that the defendant could not reasonably have foreseen that the plaintive would act on negligent advice.”

There is no evidence in relation to the investigations undertaken by IDEAS, DEI, IGL or any other of the third-party contractors, who undertook remedial works, of negligence, of aggravation or that the investigation undertaken was of no value.

No such case was expressly pleaded by DR, perhaps because of the unsatisfactory state of disclosed quantum evidence, although as the case progressed the approach of IDEAS was impugned by Mr Smith, DR’s vibration expert, and the value of that evidence thoroughly tested.

KMG’s decisions to seek third-party advice, was in Mr Boulding’s submission  ““unreasonable, and the costs arising from it were entirely unforeseeable and accordingly are irrecoverable, whether or not, they were incurred following professional advice”The passage cited above from the Governors of the Hospital for Sick Children case is not authority for that proposition.  It is an example of a break in the chain of causation.

The real issue is as to causation, as is demonstrated in Webb v Barclay’s Bank Plc 2000 PIQR p 8, a personal injuries case, where Barclays were liable to Mrs Webb for an accident at work in which she injured her knee.  A surgeon advised her to have an amputation of the leg below the knee.  The advice was negligent.  Mrs. Webb would not have lost the leg if properly advised.

In the judgement of the Court of Appeal at para 52 :-

“ The question here is whether, when an employee, who is injured in the service, and by the negligence of her employer, his liability to her is terminated by the intervening negligence of a doctor brought in to treat the original injury but who in fact made it worse.”

The Court of Appeal about the relevant law said:-

“53.

Unsurprisingly, there is no general rule on this question.  As Laws LJ said in Rahman v Arearose Ltd…’ It does not seem to me to be established as a rule of law that later negligence always extinguishes the causative potency of an earlier tort.  Nor should it be.  The law is that in every tortfeasor should compensate the injured claimant in respect of that loss and damage for which he should justly be held responsible.’

The same question was considered in the High Court of Australia in Mahoney v Kruschick (Demolitions) Pty Ltd.(1985), 156 CLR 522…We get from the headnote :

‘Held, that  the employer was liable in damages to the workmen, it might be able to prove, that if the doctor had been sued by the workmen, the doctor would be liable to some of the damages recovered by the workmen, and in that event, the employer would have been entitled to an order from contribution under s. 5(1)(c) of the Act.

Whether a tortfeasor can avoid liability for subsequent injury tortiously inflicted by a second tortfeasor depends on whether the subsequent tort and its consequences are themselves foreseeable consequences of the first tortfeasor's negligence. Per Curiam. Where any injury is exacerbated by medical treatment, the exacerbation may easily be regarded as a foreseeable consequence, for which the first tortfeasor is liable. If the plaintiff acts reasonably in seeking or accepting the treatment, negligence in the administration of the treatment need not be regarded as a novus actus interveniens, which relieves the first tortfeasor of liability for the plaintiff's subsequent condition.  The original injury can be regarded as carrying some risk that medical treatment might be negligently given.”

Finally, we agree with the editors of Clerk & Lindsell on Torts, when they say: Moreover, it is submitted that only medical treatment so grossly negligent as to be a completely inappropriate response to the injury inflicted by the defendant should operate to break the chain of causation.(18th edition, 2-55)”

The test is whether the advice of IDEAS and the other consultants as to the extent of investigation and diagnosis of the causes of vibration was such a completely inappropriate response as would break the chain of causation.

The costs of the following are clearly costs that are recoverable under the Webb test. (1) IDEAS’s Static Dynamic and Pulsation Report. (2) Finite Stress Analysis Report. (3) Work to assess nozzle loads. (5) Piping flexibility analysis and piping stress analysis.

The utility of the BOS fluid programme is less clear.  Mr Campbell, KMG’s vibration expert accepted that the BOS fluids work led nowhere and was abandoned. In cross examination, he agreed that the work was “spurious but sensible”.  In re-examination, he explained that he would not have carried out this piece of work but had allowed it to go on, because it was a relatively small part of the overall work, and there was a possibility that it might produce something. 

Mr Smith said that it was work of no value.  I hold that the cost of the BOS work is properly recoverable.  It was sensibly undertaken by a competent professional.  The outcome was spurious but the line of inquiry a proper one to follow.  It was investigation by analogy.

Mitigation: the burden of proof.

If damages are to be reduced on the grounds of failure to mitigate the defendant must also prove by how much the loss would have been reduced by any alternative course of action.

In Standard Chartered Bank v Pakistan National Shipping Corporation and others (2000) EWCA  55 (C.A.) it was argued that once unreasonable conduct had been established causation was broken and the burden of proof on the question of what loss would have been avoided shifted to the claimant.  This argument was rejected.  The burden of proving by how much the loss would have been reduced remains fairly and squarely with the defendant throughout.  In the judgment at para 38:-

“ It is trite law that the onus of proof on the issue of mitigation lies on the defendant; see Roper v Johnson(1873) LRCP 8 167, confirmed by the House of Lords in Ganac Grain Company v Faure & Fairclough (1968) AC 1130 at 1140.  The interrelation of that issue with the general obligation upon the plaintiff to prove his damage has been clearly and succinctly stated by Sir Owen Dixon CJ in the High Court of Australia Watts v Rake (1960) 108 CLR 158 as follows:

‘ The law of course places upon the plaintiff who sues in tort for liquidated damages the burden of satisfying, the tribunal of fact of  the damages he has suffered both special and general and of the quantification in money that should be adopted in the sum awarded.  That is the legal burden of proof, which rests upon him throughout. Only in one respect is the burden of proof upon the defendant and this is when he sets up matters in mitigation of damages.  If it appears satisfactorily, that damage in a particular form or to a particular degree has been suffered by the plaintiff as a result of the wrong, but the defendant maintains that the plaintiff might have avoided or mitigated that consequence by adopting some course, which it was reasonable to him to take, it seems clear enough that the law places upon the defendant the burden of proof upon the question whether by the course suggested the damage could have been so mitigated and upon the reasonableness of pursuing that course…’(emphasis added)

The burden of proof that any loss or part of it could have been avoided must be discharged by DR.

DR’s expert evidence on quantum does not attempt to do so.

DR contends that KMG, unreasonably failed to give notice in particular form (see above).  Full written notice of defects was not given within the warranty period in relation to many of the defects proved.  There is some evidence that DR had administrative arrangements to deal with warranty claims, separate from the field teams who were clearly on the Janice platform for a great deal of time undertaking commissioning work and instructed additional works. 

Nonetheless, there is cogent evidence that through the field teams and directly with DR senior personnel, KMG made clear to DR the nature and extent of the difficulties being encountered by KMG.

DR’s defence of compromise as to warranty claims in respect of compressor vibration issues (£1,000,913.25), weld quality (£110,285), parts of the claims in respect of liquid handling (£1,253,900), second and third stage re- cycle valves (£79,157), and simultaneous use of trains A and B (£63,628), presuppose sufficient and timely knowledge of the complaints within the warranty period.

There is no cogent evidence that had the KMG given such written notice as contended for by DR that they would have acted to remedy the defects. The factory at Wythenshawe had closed.  The task of completing the fabrication work, and the removal of the compressor package was delegated to KMG.  Neither was any evidence led by DR that what they might have done would have been cheaper than that which KMG achieved.

DR further submit that where KMG carried out work without giving DR’ proper notification’ and/or without allowing DR any opportunity to rectify, KMG's recovery is limited to the sum which it would have cost DR to carry out the relevant element of rectification works. In support of this contention they rely, by analogy, upon Pearce and High v Baxter and Baxter (1999) BLR 101 (CA). 

This case concerned a defects liability clause in the JCT Minor Works Form.  The employer became aware of defects within the defects liability period, but did not inform the contractor of them before pleading them as part of his claim to damages for breach of contract.  The Court of Appeal held that the contractor was not liable for the full cost of the repairs, and that the employer could not recover more than the amount which would have cost the contractor himself to remedy the defects.  Evans LJ page 104 column 2 said:-

“ There remains, however, the assessment of damages which the employer is entitled to recover if the contractor does repair the defects, then no loss will be suffered apart possibly from consequential losses, which both parties agree are not barred by clause 2 .5.  If he does not, then the measure of loss will be the cost of the employer of having the defect repaired, unless in special circumstances the diminution in value of the property in question is appropriate.  The cost of employing a third-party repairer is likely to be higher than the cost to the contractor of doing the work himself would have been.  So the right to return in order to repair the defect is valuable to him.  The question arises whether, if he is denied that right the employer is entitled to employ another party and to recover the full cost of doing so as damages for the contractor's original breach.

In my judgment, the contractor is not liable for the full cost of repairs in those circumstances.  The employer cannot recover more than the amount which it would have cost the contractor himself to remedy the defects.  Thus the employer's failure to comply with clause 2.5, whether by refusing to allow the contractor to carry out the repair or by failing to give him notice of the defects, limits the amount of damages which he is entitled to recover.  This result is achieved as a matter of legal analysis by permitting the contractor to set off against the employer's damages claim the amount by which he the contractor has been disadvantaged by not being able or permitted to carry out the repairs himself or more simply, by reference to the employer's duty to mitigate his loss.”

The decision in Pearce is predicated on the basis said the contractor had a contractual right to return to remedy the defects.  No such right arises in this case.  In any event there was no evidence led by DR as to what cost savings would have arisen had DR assumed responsibility for remedying the defects.

A failure to notify DR may be relevant to mitigation where there is clear evidence that DR had a particular resource such as a software programme that had been offered and had it been used would have shortened time expended in investigation thereby reducing cost.  There is evidence that DR had a suitable pulsation software programme.  There is no evidence that it was offered to KMG.  There is no evidence as to whether there would have been any cost saving.

The diesel claim

The claim for the cost of diesel is by far the largest head of claim.

KMG’s pleaded claim is as set out in section 7 .6 of the Scott schedule:-

“One of the purposes of the compression package, mentioned in the contract-and therefore contemplated by the parties-was to enable associated gas to be burned in the solar turbines used on Janice for electrical power generation. Without compression the gas could not be used to power the turbines.  Whenever the compressors were inoperative KMG was forced to use an alternative fuel to power the turbines-marine gas oil (diesel).  Although a certain amount of diesel would have been consumed in the turbines even, had there been no defects in the compressors (because some downtime is inevitable the amount of additional downtime caused by DR's breaches of contract was considerable, and the consumption of diesel fuel disproportionate.

DR’s breaches of contract caused KMG to consume (and therefore to buy) additional diesel fuel.  Over the period to which this head of loss relates (1stJuly 1999 to 31st of December 2002) the additional diesel consumed in the turbines was over 23,000 tonnes and its cost £3,767,359.  KMG refer to appendix 4 for calculation of this loss.”

Fuel gas was one of the commodities produced on the Janice A platform.  It was flared off, exported, or used to power the Janice A turbines.  When exported it would produce profit.  When used to power the Janice A turbines it conferred an economic benefit on KMG.

If there was compressor downtime attributable to any DR breach a direct consequence would be the loss of fuel gas either to fuel the turbines or for export.

DR submits that this is a claim for consequential loss and is barred by clause 42 of the Purchasing Contract.  I do not accept that clause 42 has this effect as can be seen in that part of the judgment dealing with preliminary contractual issues.

KMG submit that it is wrong to characterise the loss as consequential loss a loss in respect of that claim.

Clearly there was loss. In my judgement it was consequential loss which falls within the provisions of clause 19.7 of the Purchasing Contract.

“Notwithstanding any other provisions in the Contract to the contrary, the Company and Joint Venture Partners and the Contractor shall each indemnify and hold harmless, and the other against and from any liability for, loss of production, loss of business, or any other indirect losses or consequential damages arising during and/ or as a result of the performance or non-performance of the Contract regardless of the cause thereof, including but not limited to that caused or contributed  to by the negligence and/ or other legal fault of the party seeking to allow this provision.”

DR further submit that even were this claim not barred by the provisions of clause 19 .7 of the Purchasing Contract it cannot succeed because it is a global claim in which the loss alleged derives from a number of causative events only some of which were the responsibility of DR.

KMG submits that the global claim is susceptible of apportionment.  It relies upon the evidence summarised in the Scott schedule and the evidence of its quantum expert, Mr Boyne.

Section 7.6 Scott schedule:

“Diesel was supplied to the Janice platform by two suppliers, namely ASCO Ltd and Seaforth Maritime Ltd (Seaforth).  The total volume and cost of diesel discharged to Janice during the period from the 1 July 1999 to 31 December 2002 was calculated from information contained in the relevant ASCO and Seaforth invoices.  From 2003 onwards, diesel consumption dropped to more normal levels as a consequence of improved reliability of Janice Gas Compression Package following the successful completion of the various remedial work programmes.  In order to calculate the volume of additional diesel consumed as a consequence of compressor downtime it was first necessary to establish a baseline diesel consumption level, i.e. the volume that would have been confessed, consumed in any event.  For the purposes of this exercise, the level of diesel consumption recorded in calendar  years 2003 and 2004 was used as an estimate of baseline diesel consumption.  The results of the exercises in terms of cost and volume of diesel consumed are presented in Table 1(see appendices)”  (emphasis provided)

The calculation is based upon the validity of the” baseline diesel consumption level” which assumes a standard consumption of 171 tonnes per month, costing in total £1,251,151.

The total cost of diesel consumed for the period in question (1st July 1999 to 31st December 2002), was £5,011,898, which net of the assumed consumption gives the sum of £3,770,746, to which is added shipping costs of £15,612, giving the claimed total of £3,776,359.

Mr Boyne, KMG’s Quantum expert, at its paragraph 3 .18 of his first report stated:-

“I shall reserve position regarding the validity of the baseline allowance, pending further investigation.  In order to validate the baseline allowance it will be necessary to consider other factors other than compressor downtime, which may affect the fuel gas availability, and therefore diesel consumption… factors to be considered shall include well up time, on well gas production and the use of gas for gas lift and export purposes over the entire period.”

In his addendum report, he identified a significant anomaly in the period February/August 2000, because there were prolonged periods during which gas was exported but no fuel gas used.  He calculated the cost of diesel used during this anomaly period at £797,129.  No evidence was advanced by him to support the validity of this calculation.  He was forced to abandon this assessment in cross examination as being intellectually insupportable.

KMG brought this substantial claim, presumably possessed of the necessary documentation to support it.  They were given every opportunity, by way of adjournment and indulgence to fully particularise this aspect of the case. Mr Boyne was given leave to serve an addendum report in which he signally failed to validate his approach.  The purpose of his chosen “baseline” approach was set out in the addendum report, page 9 para 6.2:-

“The purpose of using a baseline allowance was to avoid the onerous exercise of identifying and quantifying each and every event that would have affected diesel consumption over the claim period. Such an exercise would entail the production and examination of detailed operational records in order to ascertain the volume of diesel consumed each day, and to identify the precise reasons for such consumption in order to establish whether or not, it could be attributed to compressor unavailability.”

Asked about it in cross examination, he said:-

There are many reasons for downtime in compressor operation, having nothing to do with the breaches proved by KMG.  There was no evidence adduced by KMG as to the occasions or periods when a compressor was down or the reason why, when either of them were down.  Trips may occur in the early stages of production, initiated by the emergency shutdown system.  The restart sequence then commences, and if the reason for the ESD is uncertain, and a subsequent trip not unlikely, the gas turbines would probably be kept online with diesel fuel. 

Evidence of trips included those caused by the solar turbines, platform power failures and difficulties with the dehydration package none of these having anything to do with DR.

The evidence of Mr van Voorst ,KMG’s quantum expert, as to the appropriate way to assess KMG's claimed entitlement to diesel costs, was to analyse the daily production logs so that the relevant value could be attributed to diesel consumed on the platform due to compressor issues.  I found the evidence of Mr van Voorst reliable and well researched.  I was impressed with his objective approach and willingness where appropriate to point out any underestimate in DR’s assessment. 

Mr van Voorst expressed the opinion that are on the data presently made available by KMG it was impossible to establish the average baseline consumption, based on an average diesel consumption of later years.  He said:-

“ Whenever the facilities are down the  Average operator will concentrate on re-establishing gas and export lift to maximise revenue, rather than re-establishing fuel gas oil or production generates an income per barrel produced, which would mean that if the facility was to be producing 10,000 barrels of oil per day the income at an average of USD 40 per barrel would be USD 400,000 per day.  The sum expended on diesel oil is insignificant compared to this income stream which necessarily takes priority.”

There can be no doubt that downtime caused by DR’s breaches did gave rise to some enhanced diesel fuel user on the Janice A platform.

There is no evidential basis upon which this court can determine the proportion of diesel cost expended upon the purchase and transport of diesel attributable to DR's proven breaches during the relevant period.

For the court, to attempt to do so would be arbitrary.  Mr Boyne attempted to do so, and in so doing, undermined his authority as an independent quantum expert.

The claim presented is in the sum of £3,776,359.  It includes significant elements not caused by DR breaches and is a global claim. 

Useful guidance in relation to such claims is to be found in Laing Management (Scotland) Ltd v John Doyle Construction Ltd ]2004] BLR 295

John Doyle were works contractors employed by management contractors Laing under an Amended Scottish Works Contract Form. The project was the construction of new corporate headquarters for Scottish Widows. John Doyle claimed a 22-week extension of time and £4.8 million for loss and expense caused by delay and disruption. Laing argued that since John Doyle had made a global claim, if any of the loss and expense claimed was not caused by them then the whole claim must fail.

It was held that provided a contractor is able to establish that all of the events on which he relies are the responsibility of the employer, in law, it is not necessary for him to prove causal links between individual events and particular heads of loss.

Where the loss was caused both by matters for which the employer is responsible, and also by matters for which he is not responsible, then for the claim to succeed the contractor must establish that those events for which the employer is responsible, are the proximate or dominant cause of the loss. The proximate or dominant cause established by proven fact and commonsense analysis will be treated as the operative cause and the person responsible for it will be responsible for the whole loss.

In the event that the contractor cannot identify the dominant cause of the loss, it may be possible to apportion the loss between the causes for which the employer is responsible and other causes.  In such a case it is obviously necessary that the event or events for which the employer is responsible should be a material cause of the loss.  Provided that condition is met, the apportionment of loss between the different causes is possible in an appropriate case.

At page 303 column 1 Lord Maclean said:-

“Such a procedure may be appropriate in a case where the causes of the loss are truly concurrent in the sense that both operate together at the same time to produce a single consequence…….

Apportionment in this way on a time basis is relatively straightforward in cases that involve any delay.  Where disruption to contractors work is involved matters become more complex.  Nevertheless, we are of the opinion that apportionment will frequently be possible in such cases, according to the relative importance of the various causative events in producing loss.  Whether it is possible will clearly depend on the assessment made by the judge or arbiter, who must of course approach it on a wholly objective basis.  It may be said that such an approach produces a somewhat rough and ready result.  This procedure does not however seemed to us to be fundamentally different in nature from that used in relation to contributory negligence or contribution among the joint wrongdoers.  Moreover, the alternative to such an approach is the strict view that, if a contractor sustained a loss, caused partly by events for which the employer is responsible and partly by other events he cannot recover anything because he cannot demonstrate that the whole of the lost is the responsibility of the employer.  That would deny him a remedy even if the conduct of the employer or the architect is plainly culpable… It seems to us that in such cases the contractor should be able to recover for part of his loss and expense, and we are not persuaded that the practical difficulties of carrying out the exercise should prevent him from doing so.

An apportionment procedure of this nature has been used with apparent success in the United States in cases before the Court of Claims.  Thus in Lichter v Mellon-Stewart Company the plaintiff’s total cost claim on one contract was rejected on the ground that a substantial amount of their loss was the consequence of factors other than breaches of contract by the defendants.  Nevertheless the Court of Claims allowed a claim based on another contract between the same parties to succeed in part and its decision was upheld about United States Court of Appeals for the Third circuit. The Court of Claims had held that part of the Plaintiff’s extra cost on this contract was attributable to the fault of the defendant and part was attributable to the other non-compensable factors……

The important point that emerge from this decision are, first, that the Federal Courts in the United States are willing to undertake an apportionment exercise and, secondly, that any such apportionment must be based on the evidence and carried out on the basis that is reasonable in all the circumstances.  In our opinion, a similar procedure should be available in Scots law.  We stress, however, that the allocation must be based on the evidence, and that under Scottish procedure the evidence must be based on a foundation in the pleadings.”

If an apportionment, based on the evidence is possible, albeit difficult it would be manifestly unjust to deny a remedy, where there are plain contractual breaches by the defendant.

It is not possible on the basis of the evidence KMG has chosen to lead to identify a dominant cause for the cost of the diesel for the period relied upon.

The need to use the quantity of diesel paid for doing this period was caused by a combination of factors some of which were KMG's responsibility, and some of which were DR’s.

On the basis of the evidence led by KMG it is not possible to attempt to apportion the loss.

This globally presented claim fails in its entirety.

Supplementary services and spares

KMG claim reimbursement of money paid to DR for offshore services between 1st June of 1999 and 31st December of 2002, amounting to £495,000, and for spares, £359,916.

KMG put their case:-

“In attempting to resolve the problems in the compression package caused by the defects that form the subject of this action the claimant used, and therefore paid for the services of others, including DR.  These costs would not have been incurred had the package conformed to the contract, and been free from defect.  This head of claim relates only to the additional sums paid to DR.  These payments represent a benefit received by DR as a result of their own breaches of contract.”

The method of quantifying the loss is also set out in section 7 .1 of the Scott schedule.  It proceeds on the basis that commissioning would have been completed by the 31 May 1999, had compression package conformed to the requirements of the contract.

The services supplied and the spares fitted were pursuant to instructions under a Supplementary Services Agreement dated 19th May of 1998.

The relevant terms:-

B.The Company requires that the Contractor shall provide supervisory and technical personnel at the Company's site and/or offshore installation for the purpose of overseeing the installation and all commissioning of the goods.

1.1 The Contractor shall provide the Company with technical and all supervisory personnel to carry out services at onshore site and/ or offshore at the request of the Company issued in accordance with the administration instructions.

There were provisions for monthly invoices, and for the price to be paid subject to justification and audit within 24 months subsequent to payment of the final invoice.  There was also a provision for the Company's claims to be presented within such 24 month period.

These claims do not arise under the supplemental agreement.  They arise as heads of loss for breaches of the main agreement.

By way of defence to this claim DR relies upon an estoppel, maintaining that KMG never told DR that it reserved the right to claw back any monies paid for work or services under the Supplementary Agreement and that both parties proceeded on the basis  that DR were entitled to the money for services rendered, and thereafter DR used it in the normal course of their business, and it would be unfair for them to be required to return it.

No estoppel by convention arises.  There is no evidence that  KMG represented that it had abandoned any of its rights under the Purchasing Contract or the Supplementary Contract, still less of that the parties proceeded on such a common basis.

Furthermore clause 39 of the Purchasing Contract provides:

“Failure by the company to insist upon strict performance of any terms or conditions of the contract or failure or delay to exercise any rights or remedies provided here in or bylaw or failure to properly notify the Contractor in the event of breach or the acceptance of all payment for any work or the review or failure to review designs, shall not release the Contractor from any of the warranties or obligations of the Contract and shall not be deemed a waiver of any right to the Company to insist upon strict performance hereof or of any of its rights or remedies as to any prior or subsequent default thereunder, nor shall any termination of Work under the Contract by the Company operate as a waiver of any of the terms hereof.”

In the alternative DR contends that this is a global claim and that the costs were caused by defects for which DR were not responsible and for aspects of routine maintenance that would have been necessary in any event.

Mr Sharp of KMG gave evidence of the need for the services rendered under the Supplementary Service Agreement in order to deal with the remedying of defects.  He stated:

” Initially, at least during the early years, if the compressors were to be internally maintained then it will be done by or in the presence of Dresser- Rand personnel.  I can say by reference to the defect categories contained within the Scott schedule that the defects damage and failures, including those resulting from liquid handling issues coupled with extensive strip- downs and rebuilds of the packages where piping or pressure vessels had to be removed to repair vibration induced failure, all resulted in the need for Dresser- Rand, to be in attendance.  In addition, their presence was necessary, to witness compressor start-up and initial runs after any major works such as the repairs and replacement of compressor motors.”

Both quantum experts accept that the attribution of particular invoices to particular defects is impossible. 

Furthermore, to be taken into account  are not only the services described by Mr Sharp, but routine maintenance and the resolution of problems arising out of the maintenance lapses by KMG personnel relating to their failures to lubricate and replace vital filters giving rise to DR attendance offshore.  These matters were described in the unchallenged evidence of Mr Colin Smith.

KMG submit nonetheless that the causes of the expenditure were concurrent, and that an apportionment should be made identifying  that proportion of the expenditure caused by DR breaches and that incurred by reason of routine maintenance and other causes.

There is no evidential basis upon which such an apportionment properly should be made.

I reject KMG’s claim to recover the costs of services supplied by DR under the Supplemental Services Agreement.

Money paid by KMG to DR for spare parts

KMG claims £367,730 in respect of sums paid to DR for spare parts that would not have been required but for the defects with the compression package. Mr Boyne its quantum expert reduced the claim figure to £359,916.  The sum actually spent on spare parts during the relevant period was £559,916.  Mr Boyne sought to apportion this figure as between spare parts routinely required, and those made necessary in consequence of DR breaches.  He deducted from the total expenditure ‘the reasonable sum’ that would have been spent over this period for routine replacement on the basis of anticipated service intervals of 8000 miles.  He attributes £200,000 for these routine costs. 

Mr van Voorst has agreed the total expenditure figure for spare parts at £508,366 deducting £51,000 that he attributes to the use of one-off expenditure associated with commissioning.

Mr Sharp of KMG gave evidence:

“Furthermore, the defects described in the Scott schedule, in particular those associated with liquid handling resulted in increased wear of sacrificial components such as piston rings, and this in itself increased routine inspection frequencies, far beyond what we would have considered a normal maintenance cycle of 8000 hours.

…as to the actual interval of 3000 hours, as described in the Scott schedule I understand that this is the service interval current at the time of the preparation of the Scott schedule.  The 3000 hours is derived from forecasts of wear on for example, piston rings.  This in turn is derived from physical evidence gathered at previous inspections. In my view, this illustrates in a simple way the extent of the unusual wear in the compressors- at one stage we were down to inspection frequencies of 1000 hours. This is compounded by the need for unplanned interventions, where for example glands or valve covers had to be maintained due to a gas release. Furthermore unplanned interventions were made due to failures detailed throughout this statement.  Every time we opened up a compressor for inspection and maintenance or repair, typically in the presence of Dresser- Rand personnel consumables (spare parts) such as gaskets O rings, oil seals, nuts and piston rings were automatically replaced.”

DR contends that an estoppel arises in relation to the payments for these spares rendering the cost irrecoverable. I reject that contention. Neither party operated on the basis of such convention.

DR also maintains that the claim is a global claim which is not susceptible of apportionment and it should be rejected it in its entirety. It may not be susceptible of scientifically accurate apportionment but that does not mean that no award should be made since part of the claim cannot be demonstrated to have been caused by the proven breaches. 

If a significant element of the claim by the evidence is shown to have been caused by breach, an award of damages may be justified, provided it is based upon a very cautious assessment of competing causation, and which safeguards the defendant. Such an approach will certainly penalise the Claimant because some of his loss is irrecoverable.

DR have not chosen to lead any evidence as to the expected consumption of spares or plead a positive case in reply to the Amended Particulars of Claim of July 2005.

Under this head, KMG are entitled to £208,366.69.  From the expenditure figure of £508,366.69 based on Mr van Voorst's evidence has been deducted the sum of £300,000, representing a very cautious estimate of what would have been expended on spare parts for routine maintenance.

Counterclaim

SUMMARY OF DAMAGES

The Defendant invites me to make an issue based costs order. Costs and interest therefore will be dealt with at a further hearing.