Coal Pension Properties Ltd v Nu-Way Ltd

In this action the Claimant (‘CPP’) seeks to recover damages arising from a gas explosion which occurred in the boiler house of the BHS store at 252/258 Oxford Street shortly before midnight on 15 August 2001. The explosion occurred as a result of a gas leak from a GB3500 gas booster manufactured by the Defendant (‘Nu-Way’) in 1986.

CPP’s claim is for the costs of repairing damage to the building and, since it had no contractual relationship with Nu-Way, it makes its claim in the tort of negligence. In summary CPP claims that Nu-Way knew by August 1998 that there were inherent dangers in the design of the GB3500 booster, which meant that it could fail at any time, yet failed to issue any adequate warnings about the dangers.

The purpose of the gas booster was to raise the pressure of gas supplied to gas burners which in turn injected a gas flame into the boilers.

The GB 3500 gas booster consisted of a cast aluminium alloy casing (‘the fan casing’). Inside the fan casing was an impeller or fan made of the same material. The impeller was attached to one end of a horizontal spindle by a pair of grub screws. The other end of the spindle was attached to a pulley which was caused to rotate by a flat belt that was in turn driven by an electric motor. The booster was designed to cause the impeller to rotate at a speed of 11,800 r.p.m. The spindle was supported within the casing by a pair of bearings: one at the impeller end and the other at the driven (or pulley) end of the spindle.

The bearings were constructed of rings of hardened steel spheres which were set between two races and were contained within a cage to keep the balls in place between the races.

It was common ground between the experts that at some stage the impeller became loose from the drive spindle and rubbed against the casing, causing it to break apart. The resulting fragments of the impeller struck against and fractured the casing causing an escape of gas into the plant room. During the course of the trial it also came to be common ground that the cause of the impeller coming loose from the drive spindle was the loosening of the grub screws which attached the impeller to the spindle; and that this was caused by movement in the spindle, which was itself a result of the break-up of the bearings at the impeller end of the spindle.

In 1986 a new boiler system had been installed at the BHS store. This included two GB 3500 boosters which were designed to service the 8 gas boilers. The system was designed by Ronald Rose and Associates, who produced an Operations and Maintenance Manual (‘the OMM’). The OMM contained guidance under the heading ‘Routine Maintenance’.

Work Schedules

The sequence and frequency of routine maintenance should be carried out in accordance with the wall chart provided by Maintenance Department. It is important that the log provided is always completed immediately after the work has been carried out.

Inspection

All equipment and plant should be regularly inspected, in order to give early warning of any faults that might arise or become too serious and thus cause permanent and possibly expensive damage. Regular inspection should therefore become a matter of course, and in addition to any specified in the maintenance schedules.

The equipment should also be monitored for any changes in noise levels or new noise generation, variations in temperature, excessive vibration and discolouration of surfaces, since any of these symptoms could be an early indication of possible faults ...

Frequency of maintenance

The recommended intervals between the carrying out of maintenance on plant and equipment, are based on new plant in good condition. Consideration should therefore be given after a suitable period of operation, eg three years, to increasing the frequency in order to compensate for the usual ‘fair wear and tear’.

The GB 3500 booster was first designed and produced by Nu-Way in 1975. It has been in continuous production since then; and more than 2,200 have been manufactured. In 1976 a smaller gas booster model, the GB 3160, was introduced by Nu-Way and this too has remained in continuous production.

Over the course of time various design changes were introduced: some of these were the result of lessons learnt where boosters had been shown to have failed, others were the result of development to improve the product.

Nu-Way’s customers were original equipment manufacturers and CORGI registered engineers; and it kept its customers and gas heating engineers informed of product modifications and recommended changes in maintenance practice by updating the GB3500 Handbook and sending out Product Information Notes (‘PINs’).

In about May 1990 Nu-Way was notified of an incident involving a GB3500 booster owned by Marks & Spencer. A copy of a letter from Marks & Spencer dated 16 May 1990 which described what occurred was copied to Nu-way.

We recently had an incident on a Centrifugal Booster, fitted to increase the burner gas pressure on a blown gas boiler.

The impeller disintegrated, cracking the outer case scroll, causing a gas leak. We are arranging for the manufacturers to carry out a failure inspection and [to] let us have a report as soon as possible.

This was the first failure of the GB3500 booster which had been notified to Nu-Way in the 15 years since it had first gone into production. It was, and remains, unclear what caused the failure.

In November 1995 Nu-Way issued PIN No 60. This set out the procedure for replacing the impeller and bearing housing assembly and made clear:

This work should only be carried out by a Corgi registered engineer.

In about December 1995 Nu-Way was notified of another incident involving a GB3500 booster: this time at Brunswick Campus in Gloucester. Michael Yarker, then (as now) a Nu-Way Production Engineer, was one of the Nu-Way employees who examined the booster. Nu-Way’s initial report was contained in a letter dated 19 December 1995:

The Booster was examined by Messrs A Bishop and M Yarker. Their findings were, the impeller spindle had broken thus causing the impeller to strike the top casing resulting in fracture. It was noted that the impeller/bearing housing had been replaced at some time, on site.

It is suggested that out of balance forces acted on the impeller putting continuous bending loads on the spindle ...

Following this incident, Nu-Way referred the cause of the fractured spindle to Dr J Huddleston of Nottingham Trent University’s Department of Manufacturing Engineering. In his report dated 21 February 1996, Dr Huddleston concluded:

The visual examination, supported by optical and electron microscopy and by dimensional study indicates that the shaft failed by fatigue resulting from repeated application locally of excessive stresses. These stresses could be attributed to the combined effects of stress concentration from an insufficiently radiused change of cross section at the point of the final fracture adjacent to a bearing and to any imbalance in the system.

The reference to an insufficiently radiused change of cross section related to an inspection of the damaged shaft which appeared to have a radius at the change of cross section of 0.35mm, whereas the designed radius was an estimated 0.75mm. This pointed to a machining error.

Nu-way responded to Dr Huddleston’s report by a change in the design of the spindle. The diameter of the spindle was increased from 10mm to 11mm; and the radius at the change of cross section was also increased to take into Dr Huddleston’s observation. This new design went into production in October 1996.

In November 1996 Nu-Way added a reinforcing rib around the top half of the fan casing, in order to reduce the risk of fracture.

In June 1997 an explosion occurred at Kerrin Point, a block of flats in South London. Three Nu-Way gas boosters had been installed and commissioned a few weeks earlier. The incident was investigated by the Health and Safety Executive; and those responsible for the installation of the equipment, together with the London Borough of Lambeth, were prosecuted.

In October 1997 Nu-Way carried out testing in order to establish the damage that could result from a poorly balanced impeller or a weakened spindle. The spindle was tested to destruction but, although both the impeller and the casing were damaged, the casing did not fracture.

In February 1998 Nu-Way added a reinforcing rib around the bottom half of the fan casing, in order to reduce the risk of fracture.

In June 1998 a GB3500 booster was involved in an incident at the Nynex Centre, Manchester. The booster was examined on site and then returned to Nu-Way’s premises for further examination. This examination showed, among other things, that the booster had been fitted with a replacement bearing housing since its original manufacture, the spindle was of the ‘new’ design (ie 11mm diameter), the fan had broken causing the fan casing to break, grub screws were missing and a large amount of a tacky substance was found on the belt and pulleys and inside guard. It appeared to be belt dressing.

In an internal memo dated 21 July 1998 a number of questions were posed by Mr Yarker, including one which questioned why belt dressing had been applied to the fan belt, and another which asked who had fitted the replacement bearing housing and whether it had been fitted in accordance with PIN 60.

The memo concluded:

The majority of boosters still in service will have old style fan casing without reinforcing ribs.

In the event of a spindle failure there is a chance that the casing will be damaged. As boosters get older, so the likelihood of the bearings becoming life expired increases. Further failures must be expected.

When a casing is damaged there is no protection against gas leak unless the unit is fitted with a slam-shut valve, gas leak detectors or other safety device, none of which are mandatory. It is unlikely that booster units are therefore protected from failure causing serious gas leak.

In view of the serious nature of a possible accident it is most important that Nu-Way address the problem immediately.

Mr Cadwell, Nu-Way’s Research & Development Manager, had been asked to prepare a technical report on the Nynex failure. In a draft of that report, which bore the same date as the final version, he wrote:

Conclusions

…

This is not the first time a booster has failed creating a potential disaster. A letter from Marks & Spencer, dated 16 May 1990 describes the failure of a booster casing.

There are three failures that have caused fracturing of the casing due to fragmentation of the impeller. The impeller moving as described, the shaft breaking and failure of the bearings.

It must be stressed that these failures are probable [sic] the most serious that can occur in the heating industry.

The shaft and impeller were modified in March 1996 and redesigned in October 1996 following the recommendations in a report, dated February 1996 by Nottingham Trent University after the failure of a booster spindle. The casing was strengthened in November 1996 and further strengthened, following another failure, in February 1998.

Actions Required

The spindle should be redesigned to stop axial movement of the impeller. The bearing assembly should be redesigned to prevent the impeller touching the housing in case of bearing failure.

Due to the extreme seriousness of this failure mode, and bearing in mind the Kerrin Point investigation is still ongoing, the actions required for boosters produced before February 1998 must be decided at Director level.

In the event the more critical passages in Mr Cadwell’s draft were omitted from the final version: first, the passage which began, ‘This is not the first time ...’ and ended ‘... the most serious that can occur in the heating industry’; and secondly, the passage which began, ‘Due to the extreme seriousness...’ and ended ‘... must be decided at Director level.’ A new sentence was added,

There was evidence that there had been incorrect maintenance,

and there was an additional recommendation that bearing assemblies should in future be supplied as a complete unit.

A meeting of Nu-Way’s senior management took place on 24 July 1998. This identified 13 action points to be ‘implemented immediately’ (‘the July 1998 meeting’). These points included:

1.

All fan casing castings without the reinforcing rib all around the casing… must be purged from stock and scrapped.

2.

Any GB3160 and GB3500 boosters in stock must be inspected and any without the reinforcing rib all around the casing must be removed and re-worked with the modified casing ...

4.

No further bearing assemblies to be sold as a spares item. When a bearing assembly fails it will be replaced by new booster fan casing assembly complete with pipe-work to connect to pressure switches. The replacement casing assembly should not be offered for boosters supplied before January 1989. We would either recommend factory reconditioning the existing booster or a complete new unit. Replacement bearing assemblies will be available for all boosters manufactured from September 1998…

6.

PINs will be prepared giving fitting instructions for both replacement fan casing assemblies and replacement bearing assemblies…

8.

The gas booster handbook and data sheet to be updated and printed. Old copies to be destroyed when amended copies are available.

9.

Method of securing fan impeller to spindle to be investigated and improved…

Under point 4 there was a reference in manuscript to PINs 100 and 101, as being attached.

The evidence of Mr Yarker, who had attended the meeting, was that it was known that the Nynex bearing assembly had been replaced in February 1997, and it appeared that belt dressing had been applied to the pulleys. Importantly in the light of later events, the grub screws had become loose, causing the impeller to move along the spindle. Nu-way could not establish a definite cause of the failure; but there was a sense that the source of the problem may have been mechanical damage caused to the spindle at some time prior to installation in February 1997. It was for this reason that Nu-Way decided that in future all housings would be supplied complete; and that, in order to guard against the risks inherent in poor maintenance practices, bearing assemblies (i.e. the impeller, bearings and spindle) would no longer be supplied as separate parts. Instead, the entire bearing housing assembly would all be supplied as a single unit, fitted within the newly designed casing.

As Mr Yarker expressed it in a later memorandum dated in February 2002, after the BHS incident,

If all future housings were supplied fitted complete in the fan casing and properly tested at Nu-Way it was felt that no further problems of this nature would arise. Furthermore, when boosters with the old-style casing came up for renewal they would automatically be equipped with the additional security of a reinforced fan casing.

In addition advice was taken from Dr Daizhong Su of Nottingham Trent University to advise on the design of more certain means of securing the spindle to the impeller than the use of grub screws. As an interim measure, Nu-Way applied ‘Loctite’ glue to improve the fixing of the grub screws.

PIN 100 entitled ‘Nu-Way Gas Booster, Instructions for fitting replacement Fan Casing Unit’ was issued on 24 July 1998. It stressed the importance of ensuring that such work was undertaken only by CORGI registered engineers. After giving advice about the fitting of replacement fan casing units, it stated,

Nu-Way gas boosters are lubricated for life but require regular inspection to ensure fan unit bearings and drive belt are in good condition. The belt should be inspected for frayed strands and cracks. The pulleys should be free from dirt or damage and the fan bearings should not be slack and should run freely and silently when spun by hand with the belt removed. Belt dressing must not be used.

PIN 101 had been intended to give instructions for fitting a replacement fan and bearing housing assembly. However in the light of the decision not to offer bearing housing assemblies as a spares item, PIN 101 was never issued.

PIN 102, headed ‘Re: Gas Booster’ was issued on 6 August 1998. This noted:

Also, with immediate effect any bearing assembly failure will be replaced with a fan casing unit complete with bearing assembly (see PIN 100 attached). This will be handled by our Parts and Components Division.

Enclosed is a copy of our latest gas booster handbook and data sheet which have now been combined. Please ensure that all previous copies are withdrawn. Further copies are available on request.

The handbook referred to in PIN 102 was a revised manual, ‘Handbook and Technical Data’ (issue 8/98). This stated that replacement bearings would no longer to be made available. It also advised,

Operation

The booster control, electrically linked with the burner and appliance, enables it to work automatically to meet pressure requirements.

Where a standby booster is fitted, it should be operated occasionally, or used in rotation.

Maintenance

Periodical Examination

The booster should be examined and maintained in good working order in accordance with good engineering practice.

Motor

The units are fitted with life sealed bearings: maintenance is not required until the bearings need replacing.

On 15 March 1999 Dr Su issued a report recommending the use of a key and locking nut to fix the impeller to the spindle. The impeller was also lightened. Following this advice Nu-Way tooled-up with newly designed machinery; and built a prototype which was tested over a period of seven months. This new design went into production in September 2000.

CPP criticise Nu-way’s response to the Nynex incident as inadequate, since it failed to address the recognised risks inherent in the unmodified bearing assemblies in current use.

On 16 May 2001 an explosion occurred at a BASF factory on Merseyside. There was an escape of gas from a Nu-Way GB3500 gas booster. Mr Yarker inspected the booster for Nu-Way; and found that the booster had been fitted with replacement flanged pulleys (instead of crowned pulleys), replacement non-standard bearings and had been run when the bearings were in a worn condition.

The immediate cause of the explosion at BHS has always been recognised to have been an escape of gas from booster No.1. As already noted, it is common ground that the ultimate cause of the fracture of the casing was the breakdown and disintegration of the impeller end bearing. The cage had disintegrated and the hardened steel balls were found to be deformed. The disintegration of the bearing had caused the grub screws to become loose and the impeller to become detached from the spindle.

There were a number of features in the arrangements in the BHS boiler house prior to the explosion, which were the subject of investigation during the course of the trial.

No explanation has been given for these changes.

A feature of the BHS incident was that, at the time of the explosion, booster No.1 had been in operation despite the fact that boosters were only designed to operate when the boilers were operating. They would not normally be operating during the summer months (from the end of March to the beginning of November) or at any other time, apart from 7.00 am to 3.00 pm.

These features of the boiler room and the operation of the boosters led to an investigation of the maintenance practices of BHS and their maintenance contractors which, Nu-way contends, were of a particularly low standard.

In a letter of 15 October 2001 Mr Cadwell wrote to Mr Kerswill of BHS indicating a reasonable life for the bearings of 3-5 years. On 19 November, Mr Kerswill replied that he had never previously heard that the bearings might have such a short life; and suggested that it might be more useful to have an estimated life expressed in terms of hours run. Mr Yarker took up this point with the manufacturers SKF and was told that the bearing life was 74,400 hours. In the light of this figure, which was the equivalent of 8½ years of continuous running, he considered a life of 3-5 years to be reasonable.

On 5 April 2002 Nu-Way wrote to all Non Domestic CORGI installers enclosing PIN 108. The covering letter explained:

We have identified that there is a risk of serious failure, particularly with Gas Boosters manufactured and installed prior to November 1996, which are incorrectly maintained and not regularly inspected.

The terms of PIN 108 were as follows:

Maintenance Advice for Series GB3160 and GB3500 Gas Boosters

Experience has shown that maintenance of booster units can be overlooked or undertaken in an inadequate manner. Such practice can lead to the catastrophic failure of the bearing and fan units. The fan bearings and gas seal are of particular importance and have an estimated life of three to five years.

It is essential that all units are subject to regular inspection to ensure that no wear or damage has occurred to the bearings, fan, seals and drive elements.

The documentary evidence about the maintenance of the boosters is exiguous. From 1990-1999 Reliance Environmental Services Ltd carried out the maintenance on behalf of BHS. Their engineers recorded the work which they carried out on the labels attached to the control panels of the two boosters. These showed that maintenance work was carried out on both boosters in 1993 and 1995; and on booster No.2 in 1997 and 1999. It is likely that work was also carried out on booster No.1 in 1997 and 1999; however, the label is not easy to read. On 27 January 1999, the words, ‘Checked changed duty’ were written on the label for booster No.2, indicating that the boosters were used in rotation; although it is not clear which booster would have been used after 27 January 1999.

The other item of written evidence is an undated document with the title, ‘Planned Preventative Maintenance of Mechanical Services, Sprinklers & Extinguishers: September 2000 to August 2003’, to which BHS and a company named Integral UK Ltd were parties. Integral’s obligations appear to have included annual servicing of the heating equipment at all BHS stores, including the cleaning and checking of the operation of the gas boosters. In addition Integral undertook to prepare an Asset Register:

An updated asset register will be submitted by 31 January 2001 for all items of HVAC equipment under Contract on each site and this shall include a summary on the condition/life expectancy of each item of equipment.

It seems that no such asset register was prepared in respect of the gas boosters at the BHS Oxford Street store.

Mr Kerswill, who was BHS technical services manager with responsibility for identifying BHS’s need for future plant and services from 1997 to 2005, gave evidence of his belief that the gas boosters were inspected and maintained annually. However, it seems that any log books, service records or maintenance histories that were in the possession of Reliance were handed over to Integral; and Integral say that all relevant documentation was lost by them in the course of an office move. It also appears that BHS now has no relevant inspection and maintenance documents.

For CPP, Mr Edwards-Stuart QC submitted that in August 1998 Nu-Way knew that there was a problem with the unmodified booster casings, which had been shown to fracture. The problem had been identified and put in writing by both Mr Yarker and Mr Cadwell. The response in the July 1998 Meeting was adequate as far as it went; but PINs 100 and 102, which were the product of the meeting, failed to draw attention to the susceptibility to fracture in the event of damage to the impeller or spindle caused by excessive vibration from excessively worn bearings. He submitted that Nu-Way failed to address the problem of old and un-reinforced boosters and that if a proper warning had been sent out, BHS would have acted upon it and would have checked all boosters, and replaced the casings. He submitted that Nu-Way’s failures were negligent and were causative of the loss. To the extent that the bearings were in a seriously deteriorated condition prior to the explosion it was not negligent of BHS or its contractors to have failed to discover their condition; and, in any event, any act or omission by BHS or its contractors was not such as to break the chain of causation.

Mr Moody submitted that Nu-Way did not know what the cause of the Nynex incident was in July 1998, but were justified in the view that defective maintenance and/or damage in transit was involved. The steps taken by Nu-Way in the July 1998 Meeting and the issue PINs 100 and 102 were appropriate and reasonable. He submitted that, such as it was, the evidence about the maintenance regime in the BHS boiler house was relevant in two respects. First, if Nu-Way had given the advice it is alleged should have been given, it is likely that it would have been ignored and the explosion would not have been avoided. Secondly, the incompetence of BHS’s maintenance regime was such as to have broken the chain of causation.

The law on negligent failure to warn is conveniently set out in Clerk & Lindsell on Torts (19th Ed) §11-30 to 32. At §11-30, under the heading ‘The Duty To Warn’, the Editors write:

Whether a duty to warn arises will depend on the circumstances, including the level of danger and its obviousness to the reasonable user, though the necessity of avoiding undue alarm will also be taken into account. Where the duty applies the warning must be such as to make the user as safe as reasonably possible.

In §11-31 the editors deal with the duty to warn of dangers appearing after manufacture.

... if a clear indication of a serious danger is received, the manufacturer may be under a duty to take reasonable steps to bring it to the attention of those likely to be affected. Thus in E Hobbs (Farms) Ltd v. Baxenden (Chemical Co) Ltd manufacturers selling supposedly flameproof wall-coverings discovered subsequently that they were more flammable than they seemed. They were held to owe a duty to bring this to the attention of previous customers ...

Finally, at §11-32, under the heading ‘Effect of Warning’

In either case of failure to warn, the claimant still has to prove causation: ie that had there been a warning it would have been heeded and hence his loss would not have been suffered. Since warnings are often disregarded in practice, this is likely to be a heavy burden which will make it difficult for many claimants to succeed on a failure to warn basis.

It should be noted that the observations about warnings being disregarded are made in relation to the particular facts of smoking cases. I do not assume in the present case that the CPP face ‘a heavy burden’, only the usual burden of proving a case.

At the core of CPP’s argument is the contemporaneous response of Messrs Yarker and Cadwell to the Nynex incident; and the recognition that there were old and unmodified models of the GB3500 which were still in use, of which BHS’s equipment were two. It is these contemporary views that enable CPP to say that the need for a warning was implicitly recognised at the time. The reason for the changes in Mr Cadwell’s draft remain obscure; but it is clear that both Mr Yarker and Mr Cadwell recognised the danger of damage to the casing caused by the impeller coming loose.

The issue is whether the omission to say anything beyond what was conveyed in PINs 100 and 102, taken together with the handbook, was negligent. Mr Edwards-Stuart submitted that something similar to what had been set out in PIN 108 should have been issued in August 1998. However, that had been sent out after the BHS incident and in the light of the additional information that the BHS incident had provided. In the course of his closing submissions Mr Edwards-Stuart advanced a wording in the following form:

There is a risk that boosters manufactured before February 1998 can suffer catastrophic failure of the casing in the event of damage to the spindle or the impeller or serious vibration from excessively worn bearings.

It seems to me that there was a duty to give a warning in August 1998 which specifically addressed the risk to the fan casing: by this stage there had been 3 incidents (Marks & Spencer, Brunswick Road and Nynex) in which the fan casing had broken. The risk had been addressed by modifications both to the fan casing and the spindle; but there were still boosters in operation which had never been modified. It was the risk in relation to these which was not specifically addressed in July 1998 (PINs 100 and 102). On the other hand Nu-Way was entitled to take the view that it would not be practicable to call-in all unmodified boosters; and that a warning of the need for proper and effective maintenance would make the continued use of such boosters as safe as was reasonably possible.

The problem with Nu-Way’s PINs 100 and 102, taken with the handbook was that they did not in my judgment focus sufficiently on the risks if inspection and maintenance were not carried out regularly and effectively. PIN 100 stressed the importance of regular inspection to ensure the fan unit bearings and drive belt were in good condition; but in the context of the installation of replacement fan casings. PIN 102 was expressed in more general terms but did not provide any advice, save by reference to the new version of the Handbook. The 8/98 version of the Handbook referred to the importance of the inspection and maintenance of the booster in accordance with good engineering practice, but this was not addressed specifically to the unmodified bosters.

I have concluded that Nu-Way should have distributed a warning in the following terms, most conveniently in PIN 102,

There is a risk that boosters manufactured before February 1998 may suffer failure of the casing if not properly maintained. It is particularly important to carry out proper inspection and maintenance of these boosters.

This would have alerted recipients to the risks associated with the design of the casing, and would also have discharged their duty of care. It was not necessary specifically to draw attention to the risk of ‘catastrophic failure’ since that might have induced undue alarm. Nor was it necessary to draw attention to particular parts of the bearing housing, since at that stage it was uncertain where the particular vulnerability might lie. A warning in the terms I have described would also have given proper prominence to Nu-Way’s justifiably held view that poor maintenance may have been the substantialcause of the Nynex and earlier failures.

CPP point out that decisions about replacing equipment were for BHS rather than their contractors; and that Mr Kerswill’s first witness statement at §30-31 makes clear that he would have taken appropriate action if he had received warnings about possible explosions. Mr Edwards-Stuart also makes the point that it is inherently unlikely that BHS, as the owner of several retail stores fitted with boosters, would be prepared to take the risk with public safety that ignoring any warning about the possible risk of catastrophic failure of the unmodified boosters would involve.

Mr Kerswill’s statement is emphatic that, if there had been warnings about the risk of the bearings becoming life expired and the possibility of this leading to a gas leak and possible explosion, he would have heard about it and would have taken appropriate action. He was not specifically cross-examined about this; and it is difficult to see how Mr Moody could have done so. In my judgment the important issue is whether BHS would have reacted in the way Mr Kerswill describes if the warning had been given in the terms I find that it should have been. This involves looking at the wider picture; and, at least to some extent, at BHS’s maintenance regime.

Mr Kerwsill’s evidence was generally unimpressive. Although he understood that there was an annual servicing of the boilers in the Oxford Street branch he had never been there and the maintenance contractors did not report to him. I was left with the impression that his evidence about annual maintenance was evidence about what should have happened, rather than evidence of his personal knowledge as to what in fact happened.

He gave evidence that he would expect inspections to be carried out at least once a year, at a time chosen by the maintenance engineers. However he acknowledged that the frequency should have increased in the light of the system designer’s recommendation that regular inspections should increase in frequency as the system aged. He gave evidence that a Planned Preventative Maintenance Programme should and would have involved work-sheets in triplicate. BHS would have records of work done by maintenance engineers (both in the maintenance log at the Store and at Head Office); and so would the engineers. He agreed that he would expect competent engineers to fill in the labels attached to the control panels of the two boosters. He also accepted that the Integral Agreement required an Asset Register to be produced by January 2001, and that it was important to have an Asset Register which summarised the condition and life expectancy of each item of equipment. However he acknowledged that he had not checked whether this had been done; and there is no evidence that it was.

The labels recorded inspection and maintenance in 1993, 1995, 1997 and 1999; but nothing in 1994, 1996, 1998 or 2000. CCP invites me to find that inspection and maintenance was carried out in the even numbered years which would have been recorded in the maintenance records.

As already noted, any maintenance records kept by either Integral or BHS are no longer available. I accept that this is no fault of CPP; however, CPP relies on the evidence of BHS and there is an inexplicable lack of documentation of a type which one would expect to support the picture presented by Mr Kerswill.

There is also the evidence about the highly unsatisfactory, inexplicable and unrecorded modification of booster No.2; and the unexplained running of booster No.1 at midnight in August.

In the light of this evidence I have concluded:

In all these circumstances I am not satisfied that, if the warning had been given in the terms which I find that it should have been given, it would have been heeded and the loss would been avoided. Nor, for similar reasons, am I persuaded that the loss would have been avoided if a warning had been given in the broader terms advanced by Mr Edwards-Stuart.

This issue does not arise in the light of my prior findings. However, as part of its argument, Nu-Way submitted that it was not in any event liable for the loss since it was caused by the intervening act or omission of the maintenance contractor which broke the chain of causation. Three specific matters were relied on: (1) the running of booster No.1 without maintaining it with reasonable care, (2) causing or permitting the bearings to fail, and (3) failing to appreciate that the bearings were failing.

The law on the intervening conduct of a third party is set out in Clerk & Lindsell (19th Ed) §2-82

The question of the effect of a novus actus ‘can only be answered on a consideration of all the circumstances and, in particular, the quality of the later act or event’. Four issues need to be addressed. Was the intervening conduct of the third party such as to render the original wrongdoing merely part of the history of events? Was the third party’s conduct either deliberate or wholly unreasonable? Was the intervention foreseeable? Is the conduct of the third party wholly independent of the defendant i.e. does the defendant owe the claimant any responsibility for the conduct of that intervening third party?

This passage was cited with approval by Buxton LJ (with whom Laws LJ agreed) in Roberts v. Bettany [2001] EWCA Civ 109 at §18-19.

Two questions arise on this part of the case. First, did the booster run on worn bearings, and if so for how long? Secondly, was it negligent of BHS to allow the booster to run on worn bearings and to prevent the bearings failing?

The answer to the first question is relatively straightforward. It is clear that the booster must have been running on worn bearings for a considerable period. I accept the evidence of Dr Crofton (the engineering expert called on behalf of Nu-Way) that the bearings would probably have been showing detectable signs of wear for ‘many hundreds of hours’ before their failure.

The second question is not susceptible to a ready answer. It is plainly a foreseeable consequence of running a machine with worn bearings that it will eventually break down and have to be replaced. There is however no basis for concluding that a competent person in authority at BHS or a reasonable engineer employed to inspect and maintain the plant would have foreseen that if a bearing failed it could lead to an explosion. In my view, it was a reasonable approach to maintenance not immediately to replace bearings at the first sign of wear, providing there was a system in place for keeping such wear under review. However at the point at which the bearings were showing demonstrable signs of wear with a high degree of noise, vibration and increased heat, it would have been negligent to allow the booster to continue in operation, although the consequences of failing to do so may not have been precisely foreseen.

One of the unsolved mysteries of this case is how it was that booster No.1 was running in August at night. I accept that during the final stages of the breakdown of the bearings there would have been a screeching sound as the races were damaged; and I also accept Dr Crofton’s evidence that this was likely to have taken place in the final ‘several tens of hours’ before the explosion. It is highly surprising that the noise was not detected; but I note that there is no allegation that it actually was detected running in such a condition. Although it is likely that some BHS staff were in the boiler room while this noise should have been noticed, it may (if it had occurred in March when the booster was running for 8 hours a day) have been only running for a few days in this condition. In all these circumstances I decline to find that some unspecified employee, who was not trained as a maintenance engineer was negligent in failing to detect the worn bearings in their final stages of breakdown. I therefore would have found that there was no break in the chain of causation such as to disentitle CPP from recovering if they were otherwise entitled to do so.

Again this point only arises if I am wrong on the issue of causation.

This is a trial on liability, but I am asked to decide one issue of principle on quantum. CPP was the freeholder of the building, but its claims include the cost of repairs to the boiler house, which was in part of the building leased to BHS.

The issue here is whether CPP’s measure of damages should be calculated by reference to the cost of repairs (as CPP contends) or by reference to the diminution in the value of the reversion of its freehold (as Nu-Way contends).

The general and long-established rule, set out in McGregor on Damages (17th Ed) at §34-034, is that parties with a reversionary interest may only recover damages to the extent of the injury to the reversion, see also Rust v. Victoria Graving Dock Co (1887) 36 Ch.D 113 CA. On this basis it is by reference to the value of the reversion that CPP are entitled to recover damages.

Mr Edwards-Stuart recognised the existence of the general rule but submitted that was not a rule of universal application. He relied on the observations of Nolan LJ in Ehmler v Hall [1993] 1 EGLR 137, at 138 to the effect that the general rule is intended to ensure that a defendant’s liability is not increased by the sub-division of interests in a building (see for example at 138B). This was not an issue in the present case since there would only be one claim for repairs: namely the subrogated claim of the building insurers with whom CPP had been obliged to insure the building under the terms of the lease with BHS against fire and explosion. In addition he submitted that in the present case CPP had a continuing present interest in the property in the form of its liability to pay the cost of repairing damage caused by insured perils.

In my judgment, and for the reasons advanced by Mr Moody, this claim falls within the general rule. CPP are confined to a claim for damages to the extent of the diminution of the value of the freehold reversion. It cannot recover the cost of repairs which it was obliged to pay out pursuant to the lease, because that would be a claim for economic loss, which would not be recoverable. Ehmler v Hall concerned a claim by a landlord for loss arising from damage to its own property. In the present case, CPP’s claim is for the cost of repairing BHS’s property.

For these reasons I have concluded that the claim fails and that there must be judgment for Nu-Way.