FT Everard & Sons Ltd. v BP Marine Ltd.

The Claimant ( “Everard” ) claims that lubricating oil supplied by the Defendant ( “BP” ) was not reasonably fit for its purpose and caused damage to the pumps in the hydraulic systems of three of its vessels “Ability” , “Amenity” and “Authenticity” . Everard claims damages of some £725,000 for repair costs, loss of hire and expenses. BP denies the allegations, claims that the lubricant was suitable and that problems experienced by Everard with its pumps had a different cause or causes.

There is some agreement about the background history and the events leading to the crucial months in July, August and September 2002. There is however disagreement about the significance of those events and of the documents generated in the course of them, as has emerged in the evidence of the witnesses of fact. There is fundamental disagreement about the causes of the problems so I have heard evidence from two experts appointed by each side. There is also a dispute about the meaning of limitation and exclusion clauses upon which BP relies and the extent to which they can be shown to be reasonable under the Unfair Contract Terms Act. Before returning to these issues I will first set out the background, the facts which are agreed or not greatly in dispute and the boundaries of the disagreement about the more technical matters.

Everard operates a substantial fleet of vessels including the three small coastal tankers which are the subject of this action. The three vessels are in effect small oil tankers and were built in 1979 to make short voyages of no more than a few days at a time and, sometimes,of less than a day. Each vessel contains a hydraulic system powered by three bent axial piston pumps sometimes known as “aggregates” . Each pump is driven by a diesel motor which can be engaged and disengaged by a clutch. The pumps drive hydraulic fluid around the system to power several cargo pumps, the windlass, the capstan and a bow thruster (which powers adjustments to the vessel’s direction while manoeuvring). The hydraulic system is therefore often in use, generally when the vessel’s engines are off. The designer, manufacturer and supplier of the hydraulic systems was Frank Mohn AS operating under the trade name “FRAMO” . This company is based in Norway but has sites in the Netherlands and elsewhere. Hydromatik (Also known as Rexroth) manufactured and supplied the aggregate axial piston swash plate pumps used in the vessels’ hydraulic systems.

BP supplies lubricants and fuels to the marine industry including those required for the vessels. BP supplied lubricant under a written agreement dated 1st July 1980 ( “The Agreement” ) which I will mention in more detail later. From the start each vessel used BP Vanellus M-10W ( “M-10” ) lubricant in the hydraulic system. M10 is an engine oil for use in diesel and petrol engines, not a dedicated hydraulic oil. It is apparently not uncommon for engine oils to be used in hydraulic systems. BP do not however supply M10 for use in hydraulic systems in any other vessels. From 1995 onwards BP recommended to Everard that it change to a dedicated hydraulic oil, BP’s Bartran HV32. Such a change was financially unattractive as it would involve the expense of completely draining the systems and filling with new oil rather than just topping up. Everard decided to make some changes however after BP stopped supplying M10 in the size of barrel which it wanted. Everard considered a change to C3, another engine oil with a very slightly higher grade and performance than M10. After its advice was sought BP recommended C3. The specifications of the two oils and the chemistry of their additives were very similar. Everard contracted to buy C3 from BP and this product was introduced gradually, beginning from March 2000 although all three vessels were not fully charged with C3 until the spring of 2002. A complication, on which Everard placed increasing emphasis as the trial proceeded is that there were two formulations of C3. The first formulation was that made at Birkenhead. In early 2001 there was a minor change in the additive package after production switched to Stanlow and this became the second formulation. FRAMO did not approve the use of either M10 or C3 in its equipment.

In about 2000 Everard decided to replace the Dorman auxiliary diesel engines on each pump with Cummins engines. Amenity’s engines were changed in Poland in November 2000, Authenticity’s Number 3 engine was changed in October 2000 and the other two in April 2001. Ability’s engines were changed in Poland in August 2001. These changes led to varying degrees of misalignment which BP contends were likely to have been the principal cause of the problems that led to pump damage in each vessel, which came to a head in summer 2002. I will deal later with the problems which then ensued. As a result of the problems which came to light in the summer of 2002 Frank Mohn recommended that the oil be changed to Bartran HV32, a suggestion with which BP, without accepting that there was anything wrong with C3 concurred. After that change had taken place the vessels for the most part suffered no further relevant problems with their hydraulic systems and were in time sold on by Everard. Discussion of the events on each vessel requires some knowledge of how the hydraulic system works.

Each pump is, as I have mentioned, driven by a diesel engine to which it is connected through a clutch. In this way the pump can be engaged only when the hydraulic system is required. The pump barrel is driven round by the shaft connected to the pump end of the clutch. The flow from the pump, and thus the pressure, is regulated by a control system. The pump is designed to rotate at the same speed at all times and for its output in pressure to be varied by tilting the “swash” or “back” plate within the pump. This alters the pressure and rate of flow. Within the pump there is a stationary port plate with inward and outward ports. The barrel within the pump rotates against the port plate and the pistons in the valve move in and out. They suck fluid in from the suction inlet at the bottom and then pump it out through the delivery inlet at the top. The piston is attached to the swash plate at the back. The tilt of the swash plate determines how far the pistons are drawn in and out and thus how much fluid they deliver. The hydraulic system is fed by oil contained in a running tank holding some 3,000 litres of oil. The pumps run at a constant 1,500 rpm. There is a filter of 10 micron mesh which keeps oil returning to the running tank at the end of the cycle, free of debris. This short summary illustrated, I hope, by the sketch which is Attachment 1 to this Judgment is an oversimplification of a summary prepared by Mr Thompson, counsel for BP but it may assist comprehension of what follows. The position is complicated by the use of different terminology by some of those involved, particularly when working at speed on board. Some are straightforward for example, the swash plate is sometimes called the back plate and the piston cup ,the piston ball but others are more confusing. While many of the surfaces within the pump, such as the control plate, are steel, others such as the cylinder barrel face which rubs against it are made of bronze. There are three pumps on each vessel so there are about nine separate pieces of equipment.

There are three modes of lubrication, as Dr Stinton, one of Everard’s experts, explains, the two relevant to this case being hydrodynamic and boundary. In hydrodynamic lubrication the metal surfaces are kept completely apart by a film of oil. The thickness of this oil film depends on the viscosity of the oil, the relative speeds of the two surfaces and the load applied to them. Increasing the viscosity or the relative speed of the surfaces tends to increase the thickness of the oil film. Increasing the load on the surfaces or decreasing their relative sliding speeds tends to decrease the thickness of the oil film. In boundary lubrication there is no complete liquid oil film between the surfaces and metal to metal contact occurs at parts of the surface. Lubrication is maintained only if the oil contains extreme pressure additives. These additives are chemicals which modify the metal surfaces so preventing them from tearing and wearing away at an excessive rate.

The viscosity of an oil is its resistance to flow. High viscosity oil does not flow readily and looks thick. Viscosity is important to how easily oil flows but also to the thickness of oil films which are formed between rubbing surfaces. Higher viscosity oils give thicker oil films at a given temperature. If an oil film is thick it keeps the metal surfaces apart. As the oil film becomes thinner there comes a stage when the oil film breaks down and metal to metal contact occurs. At this stage tearing or ripping of metal can result unless extreme pressure additives are available for lubrication.

Additives are made by specialist companies and while their chemical compositions can increasingly be determined by analysis these remain confidential, full data not being supplied to oil companies like BP. The difference between first and second formula C3 lies in the change in the additive package which took place in early 2001. The view within BP, certainly as at 13th September 2002, was that the formulations of M10 and C3 were nearly the same and that both the additive package choices for C3 were from the same manufacturer using the same technology and would also have been used in M10. If there was anything significant in the differences in the characteristics of the packages used for the two C3 formulations, Everard has not produced any evidence to demonstrate this.

The documents and the experts’ reports are replete with descriptions of problems many of which, it is common ground, are irrelevant. The pleadings are imprecise about the nature of the problems and their cause, but in opening Mr Hancock QC for Everard said that those caused by BP’s oil were excessive wear, particularly on the face of the cylinder barrel, contamination with large metallic particles and scoring damage to the metal. Dr Stinton, Everard’s Tribological expert categorised the failures as scoring damage, unlikely to have been caused by the oil, overheating and wear of the control plate where it is rubbed by the bronze faced cylinder barrel . Mr Williams, Everard’s engineering expert, also identified overheating, cavitation, erosion and scoring damage to pistons, cylinder and control plate. By the end of the trial Everard had categorised four different types of damage. First, and to them overwhelmingly the most important, is overheating and wear at the barrel/port plate interface. The wear is said to be the consequence of metal to metal contact which accompanies overheating. This leads to loss of pressure. The overheating and wear-down damage at the barrel and port plate interface is the most important because it leads to the pumps becoming inefficient. That leads to a drop in pressure within the pump because of the fact that there is excessive leak-off of the oil within the pump. As a result there is too great a space for the oil so the pressure cannot be kept up. Too much oil leaks away rather than being used to push the other oil forward and to power the various elements of the system. Secondly there is cavitation, a term which, according Casey, a work cited by Dr Tomlinson, BP’s engineering expert, is the formation and collapse of bubbles within a liquid causing erosion and damage speckling the surface. Everard accept that this is not a major problem or significant in explaining the overall cause or causes of the damage. Thirdly there is overheating at the back (or swash) plate and piston cups which Everard says is consistent with the migration of heat from the overheating at the barrel and the plate. Everard accepts that it is not clear that, without the wear damage, these problems would have led to the need to replace the relevant parts of the pumps but say, without much opposition from BP, that it was sensible to replace the whole system. BP relies on this damage as being consistent with the diagnosis of over-pressurisation identified by their engineering expert Dr Tomlinson and, as being inconsistent with an Everard claim, based on Dr Stinton’s hypothesis, that the problems were caused by starvation of oil between the barrel and the port plate, at the time of start up of the pumps. Fourthly there is scoring damage which I need not describe in detail. BP claims with conviction that this is irrelevant and Everard concedes that it is not damage “which can definitely said to be due to oil; but equally it cannot be said that it was not” . It follows that discussion will be confined to the first and third categories, wear to the barrel/port plate interface and overheating and not upon the mass of material relating to other alleged damage.

Everard does not rely on any tribological evidence that the lubricating qualities of C3 were any worse than those of M10. Indeed it was BP who obtained an order for exchange of tribological evidence, opposed by Everard. In this case, unusually, it is the Defendant relying upon such evidence and the Claimant dismissing it as irrelevant rather than the other way around. Apart from Dr Stinton’s starvation theory which I will mention later Everard’s case depends solely on drawing inferences from the correlation of events.

BP relies upon the expert evidence of Professor Malcolm Fox, Research Professor in the Institute of Tribology at the University of Leeds and Emeritus Professor at De Montfort University. He was until his retirement in July 2004 Professor of Lubricant Technology at De Montfort. He is the author of over 150 papers and of a standard text book. He has held numerous posts of distinction in the area of lubricant research as well as acting as expert witness in other cases. Professor Fox is clearly of great distinction in his field. He arranged tests of M10, first formulation C3 when produced at Birkenhead, and second formulation C3 produced at Stanlow refinery with a slightly different additive package, and Bartran HV32. Four tests were of lubrication performance and one of water separation. Professor Fox concluded from these tests that it was his “firm view” that the results “show that there is no material difference in tribological wear performance between… M10 and either of… the C3s” . He also concluded that there was no evidence that water held in suspension in relevant conditions significantly reduces the wear performance or efficiency of C3. He also further concludes that it is his firm view that “ on the basis of the tribological evidence and the other material referred to in this report, there is no evidence of a causal relationship to be drawn between the use of… C3 in the vessels and the reported pump damage. Indeed the tribological evidence and the other material referred to in this report in my opinion dispel any such causal relationship” . Professor Fox found the formulations tested satisfactory notwithstanding the fact that FRAMO had not approved the use of M10 or C3. His opinion was that both oils “clearly met and indeed exceeded those requirements”, confirming Mr Waldrum’s evidence that they met and exceeded equipment manufacturer’s specifications. He also examined the used oil analysis results accumulated by consultants on Everard’s behalf over the years and concluded that the wear elements were considerably lower than he would have expected if there had been a relevant lack of lubrication.

None of these conclusions was seriously challenged by Dr Horace Stinton, Everard’s expert. Dr Stinton has degrees including that of DPhil from Imperial College. He worked for some time for a firm of consulting engineers specialising in engine design, becoming that company’s chief chemist and managing amongst other units the lubricants laboratory. He became an independent consultant in 1986 and works for a range of clients in the motor petroleum and chemical industries. Dr Stinton carried out no tests himself and his position after his first report and a joint memorandum following a meeting of the experts was that Professor Fox’s tests were insufficient. Although Dr Stinton accepted that all the standard tests had been conducted he pointed out that these involved steel on steel and not steel on bronze and further testing was essential. He did not explain why this would make a material difference. He did not devise in detail or carry out the alternative tests which he proposed in general terms.

Everard contended that Professor Fox’s tests were not relevant to the particular problems identified. Two tests, the four-ball wear and DEFSTAN were designed to measure the ability of oil in place to protect against rubbing, not a condition apparent in this case. The other two tests, four-ball EP and the FZG scuffing test examined extreme pressure performance of the oils when lubricating two steel surfaces. The relevant part of the oil for these purposes is the additive pack which provides a chemical reaction upon the surface of the relevant piece of steel. Everard submits that this test does not help with a bronze surface. In re-examination Professor Fox suggested that the additives would protect the bronze in the same way as the steel surfaces but did not go into detail.

Dr Stinton put forward no reason why C3 might be distinct from M10 which had lubricated a bronze and steel contact for 20 years. He did not explain what practical difference the presence of brass would make. Everard, although not Dr Stinton personally as he was instructed relatively late, had every opportunity to carry out the additional tests they say were required. Everard were unenthusiastic about carrying out testing as the solicitors’ correspondence shows. A direction for tribological evidence was made by the court on 25th May 2005. Everard knew of the proposed tests by 10th June. By 2nd June oil had been blended available for Everard’s expert to test. Dr Stinton was instructed towards the end of July and carried out no tests. In these circumstances I conclude that if Everard or its advisers seriously believed that further tests could or might erode Professor Fox’s conclusions they would have carried these out. Professor Fox is, as I have said, very eminent in his field and was a wholly convincing witness. While keen to develop his points he was measured and very careful to confine his testimony to matters within his expertise of tribology. He would not be drawn into subjects where he was not expert nor into speculation. Dr Stinton is an able and qualified consultant, well versed in the subject of tribology, but not to the same degree as Professor Fox. He was frank and realistic in his answers in that area and well aware of his duties to the court. I deal separately with a hypothesis put forward for the first time in Dr Stinton’s Supplemental Report about starvation of oil at the critical surfaces. This hypothesis turns more on questions of viscosity and consideration of the chronology and is not a purely tribological matter.

Everard has failed to show that there is any difference in the tribological qualities of the oils in question. The evidence suggests the reverse. This does not mean, as BP claims, that the Claimant must fail on this ground alone. This lack of difference in the tribological qualities, and absence of evidence of anything wrong in the makeup of the product, is nevertheless an unpromising start to a claim like this.

Everard says, essentially, that it is an irresistible inference from what it alleges is the close correlation between use of C3 (and in particular the second formulation) and the breakdown of the pumps, that the quality of the oil was responsible. Everard also attributes the problems to Dr Stinton’s theory that C3 caused starvation at the pump surfaces. BP claims that the chronology is more consistent with a correlation between misalignment and the damage. BP also says that this correlation is a meaningless exercise without what it describes as a “failure mechanism” which identifies how the alleged cause actually led to the damage. Everard also relies upon what it says are clear conclusions to be drawn from testing one pump on the Amenity in September 2002. BP disputes both the facts surrounding that testing and the conclusions drawn by Everard. BP also relies on Dr Tomlinson’s expert evidence that over pressurisation caused the problems.

Just as there is no agreed glossary of technical terms in this case there is no agreed chronology either. Mr Williams, BP’s engineering expert, produced a chronology with the original report in April 2003. Mr Ferguson of BP also produced a chronology exhibited by Dr Tomlinson. There are other summaries about, each seeking to be helpful by identifying what the writer sees as pertinent. Despite what I was told about the tendency of marine engineers to take photographs incessantly there are very few in this case and none regarded by the experts as of acceptable quality.

Many of the documents are understandably informal in nature, prepared on board or at the yard, and compiled at a time before the issues in this case had surfaced or seemed significant. The meaning of some of the terms used is open to legitimate debate and few of the authors gave evidence. As well as Everard’s own records emphasis has been placed on those generated by Frank Mohn, Highland Hydraulics (a company overhauling some of Everard’s hydraulic plants in Inverness), Frank Harlington, Marine Engineer and Micron Hydraulic (engineers based in Bradford who inspected and reported on the problems with the systems). The meaning and significance of many entries in these documents have been extensively canvassed in evidence, often inconclusively.

Contemporaneous evidence about what happened to the pumps was limited. Mr Leonard Ralph, senior superintendent engineer at Everard had helpful evidence to give about the background, maintenance of the vessels, the weekly log sheets and defect and incident reports which were prepared and the practices of oil sampling and analysis. As to the detail Mr Ralph saw only the Authenticity and then only for the period around 23rd September 2002 when the pumps were open. He was not present for the further activity on 30th September, which I shall mention, or at any other time. Mr John Pratt was at the relevant time assistant engineer superintendent for Everard. He had close dealings with what happened on both the Ability and the Amenity. Unfortunately Mr Pratt did not have an opportunity before giving evidence to remind himself of the events or to read through and consider the documents. As he frankly accepted he made a number of basic errors including at one point even confusing two different vessels. He had almost no detailed, reliable recollection. Both Mr Ralph and Mr Pratt were completely straightforward and honest but of very limited assistance to the matters at issue. The court did have the evidence of Mr Michael Eckles of LGSA Marine, surveyors and consultants who conducted an investigation and provided final reports on each vessel. Mr Eckles was frank, straightforward and well-informed. He was as well qualified, if not more so, than Mr Williams though not of course not giving expert evidence as such. He was commendably cautious in some of his professional judgments.

For reasons I shall come to BP had almost no opportunity to participate in the evaluation or decision taking which led up to the final repairs and closing up of the pumps. There is no directly relevant evidence from BP witnesses although Mr Ferguson and Mr Waldrum gave helpful evidence about other matters. Samples of the oil which was actually in the systems when it allegedly caused such damage might have been highly significant. A sample was apparently taken from Amenity in September 2002 but lost by Everard.

With these considerations in mind I turn to the chronology for each vessel and in the interests of brevity concentrate on the features which each party submits to be important. I am cautious about reaching firm conclusions about what actually happened within the pumps on these vessels.

According to LGSA, the hydraulic system had operated in a generally satisfactory manner from the time that it was built until shortly after the new engines were fitted in August 2001. At the same time the system was drained and refilled with C3. Ability received top ups of C3 from April 2003 onwards. There were problems with the Number 3 pump in November 2001 but it is unclear whether that was related to a problem which Everard attribute to the oil. In July 2002 Number 1 pump and clutches 1 and 3 were overhauled. Number 3 engine was realigned with its clutch and pump, but 1 and 2 were not realigned until September. Number 1 pump had failed to maintain pressure. Excessive wear was found at that stage attributed to misalignment. In August metal shavings were found in Number 3 pump and that was removed to Micron with similar damage being seen on Number 1. Micron found that reworking was needed and that the housing was showing excessive wear. In late September Ability docked at Hull. There was apparently a hydraulic system failure due to a problem with number one clutch. Pumps one and two were inspected at Micron and when opened up damage was found which Micron later attributed to lack of lubrication, a view informed by experience on the Amenity. Number 2 pump required a new barrel piston and port plate assembly as a result, apparently, of wear. At the end of September the pumps and clutches were repaired and overhauled, the hydraulic system was emptied, flushed through and re-filled with Bartran HV32 on 4th October. After the system had been run with the new oil no damage was found and Ability returned to service on 10th October. There were some control problems in number 1 pump and minor further issues in November.

Everard points out that while the August 2002 inspection referred to misalignment Micron’s later report concluded that the problem was due to overheating as a result of lack of lubrication (albeit in a composite report prepared in April 2003 by Mr Kon the workshop manager). Everard emphasises that there was no scoring damage or record of steel particles being found in the hydraulic system attributable to causes other than oil. Everard say the damage was wear-down and cavitation with signs of blackening and bluing on the photos consistent with those of overheating.

BP points to the continuing clutch and pump control problems apparently seen in October and November 2002 and submits that there is no correlation in timing between the introduction of C3 and the pump failures. They are more closely connected to the engine misalignment. BP points to the lengthy period after the introduction of C3 but before misalignment when there were no problems. It also points to there being no period when C3 was in use, the engines were properly aligned with no clutch problems yet pump failures occurred. Failures did not start until June 2002 but there had been C3 in use since a top up in March 2000 and a complete charge in August 2001. BP says that it is striking that Ability suffered less than the other vessels and that misalignment may be the cause but C3 cannot be. BP submits that it is significant that Ability had C3 earlier than the other two vessels and yet suffered the least damage.

In March 2000 an unspecified amount of C3 went into the hydraulic system. That system was apparently under “some stress” with pump problems before the new oil was introduced. In November 2000 all three engines were replaced. In January 2001 there were control problems with Number 2 pump, vibrations reported in Number 1 were believed to be a bearing problem. The control problem apparently caused surges in the system consistent, BP says, with misalignment. In early February 2002 pumps one and three had pressure problems. Highland Hydraulics opened number three pump on 20th February finding scoring which led them to overhaul the pump and replace bearings, barrel and port plate. There were further pressure problems with all three pumps between March and June, metallic debris was found in the return filters, clutch problems and vibration continued and the vessel was withdrawn from service from 26th May until 9th June. The system was refilled with C3. The pumps were opened up from 1st July and steel and bronze particles found in all three pumps. There was damage to those parts including the swash plate. Everard suggests that bearing damage and other mechanical damage would have obliterated any sign of overheating, if this had occurred. Pumps 1 and 3 were sent to Highland Hydraulics and inspected and overhauled on 3rd July. Apart from scoring being found number three’s cylinder case and port plate were “blued” consistent with overheating. Repairs were carried out and completed on 18th July including steps to correct what was considered to be a “considerable misalignment” of the engines.

In August Highland Hydraulics went to the vessel at Hull because of an absence of flow or pressure in number three pump. Metal particles were found in pumps 1, 2 and all three were taken away by Highland Hydraulics for inspection. That inspection revealed wear, the presence of steel particles, scoring, some cavitation and “blueing” , consistent with overheating. The pumps were returned on 5th September and tested the following day. Metal particles were then found in all three pumps. There was a further inspection by Micron of Number 1. This found scoring and marking as well as extensive overheating damage later appraised by LSA as damage to the pump barrel and port plate consistent with a lubrication problem. Subsequent inspection of 2 and 3 on 12 September showed bad scoring and evidence of heating in the bearings. BP contrasts damage consistent with the presence of external metal particles, contamination which Everard was trying to eliminate during this period on this pump, with particles produced by metal to metal contact of the kind which occurred on 17th September. After further overhaul the pumps were returned to the vessel and a further test on number 1 pump was run on 17th September. The work was supervised by LGSA and Frank Mohn. They found a copper layer on the cylinder barrel badly damaged, two shiny circles on top of the barrel, overheating to the port plate and damage to the swash plate. No metal particles or scoring damage were found. Overheating was detected after only one or one-and-a-half hours.

Everard sees a link between the first full charge of new formulation C3 on 22nd November 2001 and problems being reported in February 2002. They also put particular emphasis on the final test of Amenity. It says that by this time all other parameters had been resolved and other potential causes eliminated, and that after only an hour-and-a-half serious overheating and contact damage was found without scoring damage. The oil was accordingly changed and the problem did not then recur. It follows that the oil caused the damage.

BP points to the fact that there were no problems between March and November 2000 when the system was full of C3. There was a fresh charge in June 2002 after the problems began in February. Numbers 1 and 3 suffered serious misalignment, but number 2 did not. BP emphasises that the system was badly contaminated with metallic debris as at the beginning of September 2002. The testing in early September occurred while there were repeated attempts to get rid of that contamination. By the 12th September inspection it appeared that the debris was external and had come not from the pumps as a result of wear and the system was still not “clean”. On BP’s case by 12th September the contamination remained. BP argues that there was no damage on 12th September like that which caused such immediate effect on the 17th September test. BP argues that the problem on 17th September could not have been the oil because the damage did not occur at the earlier test even though the oil was still in place. BP also suggests that the breakdown on 17th September may have been caused by insufficient clearance at the crucial surfaces. I will deal with the parties’ suggested explanations for what happened on 17th September when reviewing the testimony of Dr Stinton and Dr Tomlinson in this area.

According to Mr Williams’ review of the history component failures in the pumps and other rotating machines did occur in Authenticity between July 1997 and March 2000 before the first C3 delivery on 22nd of March which was used as a system top up. The Number 3 engine was replaced in October 2000, the Number 3 clutch removed for overhaul in November and the other two engines replaced in April 2001. After this the Number 1 clutch suffered a number of problems attributed to misalignment. That clutch was overhauled and realigned in July 2001. A problem with a shaft seal continued and the seal was renewed again in September 2001. The system was refilled with C3 in December 2001. Problems arose with pumps 1 and 3 in February 2002. When Number 3 was opened up 2mm of wear were found to the bronze face of the cylinder valve. There were further clutch problems in February, April and June 2002. On 12 ^th or14th September 2002 the vessel left service for a routine refit. The pumps showed signs of overheating and were found to be misaligned and adjustments were carried out. On 30th September there was an inspection at Micron which LGSA attended. This revealed that Pump 3 had worn down about 2mm with evidence of overheating, the control parts being affected by general wear. Number 2 pump barrel and port plate had no scoring but were blackened around the mating surfaces and the piston cups of the back plate. Number 1 pump barrel had wear and cavitation and the centre pin was loose consistent, according to Everard, with overheating and other parts had suffered wear. This was consistent with what Frank Mohn had found when the vessel docked at Sunderland. All three pumps were overhauled, the system was drained clean and flushed through and then refilled with Bartran HV32. By this time similar treatment had appeared to resolve the problems of Amenity in Hull. Although the pumps showed no problem on the test further debris was detected in the system in November 2002 and there were some further problems between December 2002 and April 2003.

Everard suggests a correlation between the damage suffered by Authenticity and the introduction of second formulation C3. Everard suggests that the first full charge of this product was on 16th August 2001 and that pressure problems were reported on 2nd February 2002 together with the wear on number three to which I have referred.

BP says that the correlation is greater between the pump problems and misalignment than with the use of C3. BP points to those problems arising before C3 was introduced and to an extent after it was replaced, and claims that serious problems coincided with misalignment. BP submits that there is a remarkable correlation between the replacement of the engines in October 2000 and April 2001 and the onset of problems with both clutches and pumps. Problems occurred first with Pump 3 and only with Pumps 1 and 2 after the engines had been changed.

As Mr Ralph points out in his evidence, by September 2002 it was apparent that the pumps on all three vessels were suffering similar problems at more or less the same time. He and Mr Pratt investigated the possibility of a common cause and retained LGSA instructed by Everard’s insurance broker arm. LGSA produced a separate report for each vessel. As to Ability the conclusion was that the damage to the clutch units could have occurred as a result of misalignment and the damage to the parts “may have occurred as a result of the operation of the pumps with an incorrect grade of oil” . LGSA recorded that the owners had formed the opinion that C3 was the cause. LGSA, and probably later Mr Williams, recommended that samples of C3 be analysed by a tribologist but Everard did not take this up. Similar recommendations and conclusions were reached for Authenticity and for Amenity, Mr Eckles attaching no significance to slight differences of wording in the three conclusions. The view of Mr Eckles in his witness statement was this:-

On 12th September Frank Mohn contacted BP in Norway about the oil on all three vessels with which problems were being experienced. There are internal BP emails about the issue and there were relatively brief discussions including a telephone conversation between Captain Lyons of Everard and Mr Waldrum of BP. Although the exchange attracted some controversy at the trial it does not contribute much to resolving the issues as they now stand. Everard were clearly moving towards identifying the oil as the culprit, BP rejected this but in the circumstances recommended a change to Bartran HV32 as they had done three years previously.

On 3rd October, after some further exchanges, Everard put BP on notice of claim and invited them to make a joint inspection if they wished. By this date the systems on two of the three vessels were closed up. On 14th October BP responded with a list of questions. Everard Insurance Brokers prevented the release of documents to BP without Everard’s consent. By 14th October the pumps on all three vessels had been closed up so there was nothing to inspect. On 30th May 2003 Everard sent a claim letter to BP which led eventually to this trial. It was no doubt in connection with this that on 14th April 2003 Micron produced an “inspection report” . Mr Kon, the workshop manager reported on what had been found on the records of the various inspections of the pumps in the previous year. The problem was “most probably” a “lubrication problem most likely caused by the type of oil being used”. Micron record that all pumps had excessive wear and damage with signs of overheating of the bow and valve plate and other areas. Mr Kon encloses earlier quotes and reports and says that these should be read, for what they say about possible causes of the damage, as having been written without the knowledge of the type of oil being used. Mr Kon concludes that “overall” the major problem suffered by the pumps allowing for probably not inconsiderable service life in four out of five cases, was due to inadequate lubrication from the oil being used.

On the previous day, 14th April 2003 Mr Williams, Everard’s engineering expert had delivered his review undertaken “to determine if the system failures had a relationship with the period during which… C3… had been supplied and used” . I will turn to Mr Williams and the other expert witnesses after first summarising briefly the position of each party thus far, since each argues that it prevails on this side of the case without needing to rely on its expert evidence.

Everard says that wear and overheating damage is generally associated with a failure in lubrication. Those having investigated it, Mr Ralph, Mr Pratt and Mr Williams, all concluded that the damage was due to this reason. The oil was a hydraulic not an engine oil. The oil was not approved by system manufacturers and this was known to BP. The acid test was the 17th September test on Amenity. Once the oil was changed the problem did not recur and the vessels traded for at least a further two years. The problems are closely related to the introduction of the new formula C3. There are no other explanations consistent with the evidence other than one involving a failure on the part of the oil. The reason why the oil failed to provide proper lubrication is irrelevant. The fact that it did not do so means that it was unfit for its purpose.

BP says that there is no simple explanation for all the relevant damage. The correlation process suggests that it is more likely to be misalignment of the engines than the use of C3. Everard’s case does not explain the full range of damage but focuses only on part of it and part of the relevant period. Correlation is irrelevant without identifying a failure mechanism and this approach depends upon overlooking the tribological evidence which reveals nothing wrong with the oil.

Mr Williams has very considerable marine engineering experience at sea and long service as an engineer superintendent working for a large oil company. He has been a consultant since 1999. He is IENG and AMIMarEST. He was retained by Everard in April 2003 to review the systems and to report on failures. His initial review which forms his first report concludes that the clutch problems were caused by misalignment, that there is no absolute correlation between the use of unapproved oil and the sum total of hydraulic system damage experienced but “evidence that oil contributed to a proportion of the aggregate failures, particularly in the case of Amenity and the damage found latterly on Authenticity and Ability is present” . He sees C3 as the most likely cause of the problems but “failures of this nature are complex and usually manifest as a combination of factors so one cannot exclude the possibility that one or more of the other matters referred to … were contributing factors” . Mr Williams has compiled some helpful summaries and schedules. His approach is one of correlation between the introduction of C3 and his assessment of the damage and has been adopted in the case put forward by Everard I have outlined above. He suggests that ageing is probably not the cause of the breakdown as he has not detected signs of this when monitoring maintenance records until 2004. Mr Williams produced a supplemental witness report replying to the evidence of Dr Tomlinson at the end of which he concludes that, on balance, the piston pump damage that occurred on the vessels most likely resulted from a lubrication problem related to the use of C3. While Mr Williams’ views are a helpful addition to the debate about correlation and his evidence assisted the court in understanding some of the terms being used, his considerable practical experience was not matched by any training or insight into the scientific side of the problem. He understandably had no academic training in this extremely complex area and his expertise contributed little to discussion of why the correlation theories might be right. His expertise did nothing to explain what might actually have been wrong with the oil to cause it to damage the pumps. Mr Williams and Dr Tomlinson pass each other on different sides of the road so far as engineering expertise is concerned. Mr Williams was unable to match Dr Tomlinson’s understanding of the science but, as I mention below, his practical experience did help to demonstrate shortcomings in Dr Tomlinson’s approach.

When referring to the test on 17th September on the Number 1 pump on Amenity I mentioned BP’s suggestion that there was some other cause of extreme pressure which explained what happened. That suggested cause lay in the small clearance between the valve and the port plate referred to in a fax from LGSA dated 3rd October. The clearance for Number 1 pump was given as a 0.5mm, the minimum clearance according to the manufacturer should be 0.45mm. The clearance for the other two pumps was 1.2mm and 0.95mm. During the final rebuilding of the number 1 pump the clearance was increased to be between 0.8mm and 0.9mm. The possibility of extreme pressure caused by low clearance (or high pre-tension) was cautiously acknowledged by Mr Eckles and accepted by Dr Stinton in cross-examination. BP’s suggestion is that the clearance was adjusted tighter when the pump was overhauled so the surfaces would have been more prone to the risk of overheating. Once that clearance was eased the problem did not recur. In cross examination Mr Williams suggested that in fact the clearance had been adjusted in the opposite direction making it even tighter and the pre-tension greater. This appeared to be speculation on his part which contradicted the report and the evidence of Mr Eckles, and Mr Williams accepted that this would not have been a sensible thing to do after the failure on 17th September. Mr Williams has not given evidence before as an expert. He occasionally attributed clear inconsistencies in his testimony to his use of English. This was unconvincing. While clearly truthful about matters of fact, his opinions occasionally gave the impression of being influenced by an awareness of which side was calling him as a witness. Thus, to give an anodyne example, what started as “a maximum pressure of 180 bars” in the overview to his report was corrected to “normal operating pressure” and the working figure increased to 220 bars, the manufacturer’s maximum and then increased by a further 10%.

Dr Stephen Tomlinson is s senior technical consultant with a consultancy advising commercial and military concerns. He has a Phd in fluid power systems, is a Chartered Engineer, a Member of the Institute of Electrical Engineers and an Associate Fellow of the Institute of Mathematics and its Applications. He has long experience in the fields of hydraulics in a variety of applications and has designed hydraulic pump control systems. He has practical experience in the sense that he has been to sea on naval ships when carrying out or implementing his professional advice. In contrast to Mr Williams he has no practical experience with the operation of the vessels such as those which are the subject of this case. Dr Tomlinson’s report contains extremely helpful information about the scientific background and a series of diagrams which help clarify the technical complexities. He has regard to the fact that the engineering side needs to be seen with the tribology. He identifies a number of causes for pump failures including misalignment, contamination, poor maintenance, aeration and cavitation. Dr Tomlinson put forward a theory to account for the problem of over-pressurisation. Taking the existence of misalignment between engine and pump he says that there would have been consequent oscillation in the shaft speed, that this would have led to variations in the pressure within the system which in turn would have caused contact damage and a wear at the valve end. Dr Tomlinson’s confidence in this theory is reflected in his assessment of it as being 95% likely to be the reason for the damage. This theory was comprehensively demolished in cross-examination and ran up against the additional difficulties that it was a phenomenon that Dr Tomlinson had never come across (although he believed colleagues had done). It was outside the experience also of Everard witnesses and Mr Williams who had been at sea with engines of this kind for a very long time. Dr Tomlinson turned then to the concept of “dieseling” a phenomenon leading to over-pressurisation at the piston ends and cups. Dr Tomlinson also put forward the theory of clearance/pretension and did so with such enthusiasm and commitment as to undermine confidence in the rest of his evidence. Dr Tomlinson had a dogmatic and complete certainty in the correctness of his views. While I have no doubt that this was a genuine belief on his part that he was clearly right, and not at all a wish to assist the party retaining him, I had no confidence in his conclusions although I was grateful for the considerable value of his background knowledge. His first report contains an admirable explanation of the technical issues and of how the pumps work. As Mr Thompson for BP put it “he did not give his evidence with a composure that one would have expected or hoped for from this side of the court… it may be that he over-egged his opinion slightly, in particular in searching for a single explanation for these phenomena but, of course, that is precisely what the Claimant is trying to do as well” . Indeed BP suggests that this expert’s testimony showed the flawed nature of any exercise to look for a single cause of failure.

After meeting Professor Fox and agreeing to the joint memorandum Dr Stinton produced a supplemental report developing the new theory that when the pumps start up the second C3 formulation does not reach the surfaces because of its viscosity. As a result overheating and damage occurred. The mechanism is well summarised by Mr Hancock QC in his closing submissions as follows:

These pumps go from standstill to full speed very rapidly. The shaft is turning at full revs; and the pump is then brought into action when the clutch is engaged.

The cylinder barrel and valve plate are pressed together, because the system is pretensioned. The effect of this is that when the system is at rest, the oil which might otherwise sit between the two surfaces is forced out.

The combination of the above two factors means that at start up the cylinder barrel and valve plate move against one another very rapidly without initial lubrication.

How rapidly lubrication starts depends on how rapidly the oil within the system sets up an oil wedge sufficient to force the two surfaces apart. This in turn will depend on the viscosity of the oil. A higher viscosity oil will take longer to begin lubrication.

Here, the second C3 formulation had a higher viscosity than any of the other oils which have been used in the system. So it is more likely that it will take longer to provide lubrication at start up; and hence more likely that there will be excessive contact for a period at start-up.

This excessive contact will lead to wear damage at the cylinder barrel face, which is bronze and softer than the steel valve plate. It will also lead to overheating, due to the friction caused by metal to metal contact.

The pattern of the damage found, and the evidence of overheating seen, is consonant with this theory.

Mr Thompson, for BP, was highly critical of this theory. It was a hypothesis unsupported by evidence or empirical study, vague and undeveloped when calculating time and distance and supported by an inapposite and distant analogy with a turbocharger. There was also a number of detailed criticisms put forward, the force of which Dr Stinton to some extent accepted. The suggested effect would only last a matter of seconds but would still need to cause damage of severity and do so on every start up. Numerous oils complying with manufacturer’s advice would necessarily give rise to some failure. If the viscosity of new formula C3 had this effect then it would do so with other oil including the original C3 and M10, if the start up occurred in temperatures a few degrees lower as it often would in colder months and in more northern waters. This theory too seemed to me to be demolished by cross examination. Dr Stinton seemed to accept this in a calm and fair-minded way. He accepted that the theory was “possible” and not “likely” . This was retrieved in re-examination so that the theory was still said to be the most likely explanation leading Mr Thompson to characterise Dr Stinton’s view of his own theory not wholly unfairly as being the best of a bad bunch.

There is no tribological evidence that the oil is defective. That testimony tends to indicate that C3 has no relevant characteristics not also present in M10 or indeed Bartran. It was open to Everard to conduct further or different tests to those carried out by Professor Fox if it claimed otherwise.

Dr Stinton’s theory is wholly unconvincing for the reasons given above. Dr Tomlinson’s theory is equally unconvincing. The inability of the experts to produce tenable reasons based on a single cause for why the damage occurs suggests that there may be no clear or single answer.

In my judgment the absence of a “failure mechanism” is not of itself fatal to Everard’s claim. Where however there is also nothing to show the respects in which the oil was of poor quality the circumstantial evidence must be clear and compelling for the Claimant to be able to prove its case. The quality of the evidence about what happened when these failures developed is not high. None of the theories put forward by the experts are convincing. The case based on correlating the introduction of C3 with damage is unsatisfactory. Generally the broader and the further back one looks the less convincing Everard’s claim becomes. Many of the points about timing depend on inadequate evidence. There is even a difference between Dr Stinton and Mr Williams about the timing of the damage to which correlation is to be applied. Those companies and individuals who blamed the oil in 2002 and 2003 may have acted more on intuition than analysis and naturally would not have pointed blame towards themselves. The overall pattern seems to me at least as consistent with misalignment as with C3 and I am left with the impression that while problems on all three vessels came together at about the same time this may well have been for a variety of different related and unrelated reasons. No reliable conclusions about causation can be drawn from this correlation exercise. While civil courts prefer not to decide cases on the burden of proof that is a pertinent factor in a situation such as this.

Everard says that the outcome of the 17th September test is so stark and clear that the evidential burden shifts at that stage to BP to come up with an explanation as to why there was a failure in lubrication in general terms which did not involve the oil. I cannot draw broad conclusions from the particular facts of the 17th September test on the basis of one pump on one vessel. The test was a working operation designed to get the system in service, not a properly designed or monitored exercise. For the reasons given by Mr Thompson on behalf of BP I do not accept the picture put forward by Everard of the 17th September test being carried out in conditions when all other possibilities were eliminated. There may well have been a special factor applying on that day to bring damage when the previous tests had not. It may have been clearance. It may have been something not identified. While lack of lubrication may well have been one of the factors causing or contributing to particular aspects of the damage suffered by these pumps the situation remains obscure, particularly when nothing is identified about C3 which Everard can show caused it to act differently to M10 or Bartran in relevant respects. In the end Mr Hancock said that Everard’s case was “quite simply, BP’s oil was not reasonably fit for the relevant purpose because it damaged the vessels’ pumps” . I am not satisfied that the oil did damage the vessels’ pumps. I remind myself that this stage in the exercise is not an evaluation of whether a product with an identified defect has caused damage. The question is whether the connection between the oil and the damage is so clear that the product must have been defective. Everard does not come near to showing this. Everard’s case therefore fails.

BP contends that if Everard’s claim succeeds on its merits it is nonetheless invalidated by clause 10(c) which provides as follows:-

notice of the claim, supported by a full statement of the facts and evidence on which it is based, is not given to BP and BP’s local representative at the port in question within 30 days of the delivery; or

the Buyer shall not give to BP a sample for testing purposes and permit BP to make such other tests and inspections as BP may reasonably require; or

the Marine Lubricants in question have been stored or transported in movable containers provided by the Buyer or altered, contaminated or adulterated in any way.”

BP says that no notice was given in time so the claim is invalidated under (i), that no sample was proffered and that BP was not given the tests and inspections it reasonably required and so falls foul of (ii).

Everard says that the clause does not apply since it is not a “quality” claim within Clause 10. Everard argues that the clause relates to a claim that a product is not specifically what was contracted for, not to a claim that the oil is generally unsuitable for use in a system. Everard says that the kind of claim contemplated by the clause is one that will become apparent within a short time after delivery. It relates to the qualities and description of oil before it is used, hence sub clause (iii).

Everard further argues that if the clause does apply it is not in breach. Further, the clause fails to satisfy the requirements of reasonableness imposed by Section 3 of the Unfair Contract Terms Act 1977. It is for BP not Everard to show that the statutory requirements are met.

BP responds, as did I initially, that a provision dealing with a defect in quality must cover a claim under Section 14(3) of the Sale of Goods Act 1979. BP seeks to substantiate the claim that there was a breach and then argues by reference to the statutory guidelines and factors invoked in some of the cases that the term satisfies the requirements of Section 3.

It is not necessary for me to decide this issue in view of the other conclusions I have reached. I will therefore address the point only briefly, Clause 10(c) needs to be seen in its context both within the contract as a whole and as a part of Clause 10 dealing with “measurement of quantities and claims” in a standard form contract for commercial sale of marine lubricants in substantial quantities and in a wide variety of grades. The claim envisaged by Clause 10(c) is one which if not settled by agreement is referred to a referee under Clause 15(a). That referee is, in the absence of agreement, nominated by the President of the Institute of Petroleum. Matters of a broader nature are to go to arbitration. These all indicate a focus on measured quantity, quality and specification. It seems to me that Mr Hancock is right and that the clause is indeed addressed to quality in the limited sense of BP’s obligation to supply the product ordered to the appropriate specification quality and quantity and uncontaminated. It deals with defects identified in the oil before it is used not to broader claims of the kind in this case concerning the fitness of oil for a particular purpose. That construction gives sense not only to (iii) but to the timing in (i) and to the sample requirement in (ii). Moreover, when read in that way, the provision is clearly a reasonable one, both in commercial terms and also should one apply the statutory factors. That approach is also consistent with construing the provision for what it is, an exemption clause drafted by the party seeking to exclude liability. It follows that if Everard had been successful on the merits Clause 10 would not have invalidated its claim.

As there will therefore be judgment for BP I believe it is unnecessary for me to address any other matters raised by the parties, but I will of course consider any views to the contrary. I shall be grateful for corrections of the usual kind and for a draft order. Finally, I express my thanks to counsel and to solicitors on both sides for the very able way in which this case was presented, prepared and organised.