England and Wales Court of Appeal (Civil Division) Decisions

IN THE COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM THE HIGH COURT OF JUSTICE
QUEEN'S BENCH DIVISION
COMMERCIAL COURT
MR JUSTICE POPPLEWELL
Claim No. 2011 Folio 58

		Royal Courts of Justice Strand, London, WC2A 2LL
		22nd March 2013

Lord Justice Longmore:

1. This is a dispute between two sets of insurers. The Claimant insurers provided cover to the assured under the terms of an Erection "All Risks" Public Liability and Delay in Start Up Insurance Policy ("the EAR policy"). The Defendant was an insurer providing Marine Cargo Insurance to the same assured pursuant to the terms of a contract of marine insurance ("the marine policy"). In broad terms, the marine policy covered loss in transit; the EAR policy covered loss at the relevant site.
2. The Claimant insurers claimed an indemnity or a contribution from the Defendant, the Claimants having themselves indemnified the assured. The Claimants sought to recover sums from the Defendant on the basis that the damage suffered by the assured whom they had indemnified was in fact covered by the Marine Policy, rather than the EAR policy.
3. The insured, Lakeside Energy from Waste Limited ("Lakeside"), was engaged in 2007 in the construction of an energy from waste facility in Colnbrook, near Slough ("the Lakeside facility"). Certain contractors, Itochu Corporation and Takuma Corporation (together, "the EPC Contractor"), were engaged by Lakeside as the Engineering, Procurement and Construction Contractor for the development of the waste facility.
4. Both Lakeside and the EPC Contractor were named as co-assureds under the EAR Policy. The Marine Policy named Lakeside as the assured and Lakeside's engineering and procurement contractors, which included the EPC Contractor, as co-assureds. Both Policies contained a 50/50 clause providing that in the event that it was not possible to ascertain whether the cause of the damage to the insured's property occurred before or after the arrival of the property at the Lakeside facility, the EAR and Marine Insurers would each contribute 50% of any properly adjusted claim.
5. The waste facility incorporated two boilers, which each housed an economiser, a heat exchanger through which water flows. The water is heated by hot air which is blown through the economiser. Each of the 2 economisers comprises 8 economiser blocks which consist of banks of vertical tubes, 24 tubes long by 40 tubes wide. The outer 12 tubes on each row are "cranked" so to allow them to connect to tubes 2 rows away, thereby allowing the flow of water through the economiser's tubes.
6. The economisers' tubes are welded at the top and bottom of each row of 24 tubes to headers ("the upper headers" and "the lower headers" respectively) and the upper headers are welded together. Each block is about 3.6m wide, 2.8m deep and 6.3m high in dimension and, once installed, suspended from the upper headers so as to be free hanging with space beneath.
7. The economisers were manufactured by a company called Vulcan SA ("Vulcan") at its manufacturing facility in Romania in mid-2007. The 16 economisers were transported variously by a combination of road and sea from Vulcan's facility in Romania to the Lakeside facility site throughout July and August 2007 and installed between August and October 2007. The economisers were installed while the construction of the waste facility was ongoing. 14 of the economisers were transported by a combination of road and sea (via the port of Constanta in Romania to Southampton), whereas 2 of the economisers were transported by road alone.
8. During testing carried out during commissioning of the economisers on 9 February 2008 microscopic cracks were discovered in certain weld joints of the economisers where the cranked tubes met the upper header. Further testing revealed further crack and crack-like indications in a number of the cranked tubes. Ultimately, the weld joints where the cranked tubes met the upper headers were all replaced.
9. The assured notified claims under both the EAR policy and the Marine Policy in February 2009. The Claimants subsequently paid Lakeside and EPC's claims and entered into an agreement on 13 August 2009 with the Defendant which provided that the parties would either agree or litigate between themselves the issue of correct apportionment of the assured's claim as between the two policies.
10. The Claimants assert and the Defendant denies that the loss and damage suffered by Lakeside and/or the EPC Contractor was covered under the Marine Policy. The Defendants assert and the Claimants deny that the loss and damage occurred at the site and was covered by the EAR policy.
11. The main dispute for determination by the judge was, simply, when the damage to the economiser blocks occurred. Was it during road or sea transport, thereby falling within the scope of the Marine Policy, or whilst on-site at Lakeside thereby being covered by the EAR Policy?
12. Both parties' experts agreed that the cause of the damage in the relevant weld joints was fatigue failure due to cyclic stress loading which had been caused by resonant vibration.
13. The Defendant's case at trial was that the necessary vibration could not have occurred during road and sea transport and that it was likely to have resulted from turbulent wind causing the tubes of the economisers to vibrate whilst the economisers were on-site and exposed to the elements at Lakeside. The Claimants' case was that the vibration which caused the fatigue failure occurred during road and/or sea transport by reason of missing packing between the rows of tubes in the economisers and by reason of transport over rough roads in Romania; it did not occur when the economisers were on-site at Lakeside because the wind was not sufficiently strong to excite vibration in the tubes.
14. The Court heard expert evidence from experts from each side in four areas: metallurgy, welding, road transport and wind. Brief factual evidence was also given on both sides.

The Judge's Reasoning

15. The judge found in favour of the Claimants. His essential reasoning was as follows:

(1) A range of stress cycles suffered by the tubes of between 20,000 – 80,000 stress cycles must have occurred to cause the observed fatigue cracking in the economiser blocks;

(2) Wind excitation could effectively be ruled out as a cause of the fatigue cracking observed in the economiser blocks;

(3) Once one had eliminated wind as a theory for the cause of the damage, it was necessary to assess whether road transport vibration was more likely than not to have caused the damage since, if it was not, then one could not conclude which cause was the proximate cause, both would be equally unlikely and the 50/50 Clause would be applicable;

(4) The Claimants' expert's theory (based upon finite element analysis modelling) that the fatigue cracking could have occurred during road transport if the two factual pre-conditions set out in (5) below had occurred at the relevant time was, on the evidence before him, credible. The Defendant's argument that road transportation could not have resulted in the resonant vibration response of the economiser tubes, (which the parties agreed must have occurred in order to explain the observed damage) was rejected;

(5) The two necessary factual pre-conditions for the required resonant vibration had occurred viz:-

a. there was sufficient packing missing between the tubes in the relevant places of the economiser blocks to have allowed resonant vibration to occur during road transit;

b. each of the economiser blocks was transported over a sufficiently rough road for a sufficient amount of time to have allowed the necessary cycles of stress to have occurred to cause the fatigue cracking;

(6) The judge accordingly found that it was a "realistic and credible possibility" that the observed fatigue cracking occurred during road transport and accordingly found for the Claimants.

16. Permission to appeal against the judge's findings was granted by Stanley Burnton LJ before his recent and much-lamented retirement. The parties agree that the damage to the economisers must have occurred either during the road transportation or while awaiting installation at Colnbrook and that it must have been caused either by a combination of inadequate packing and rough road surfaces on the one hand or by wind excitation at Colnbrook on the other. There was no other potential cause of the damage. In these circumstances there were effectively 3 grounds of appeal:-

(1) there was no or insufficient evidence that the packing between the tubes of the economiser blocks was inadequate;

(2) there was no or insufficient evidence that the road surfaces over which the blocks were transported was rough enough to set up a vibration;

(3) the resonant vibration had to be caused by a narrowband frequency response but the evidence was that any resonant response was a broadband frequency response.

17. Other potential grounds of appeal such as that the judge should have dealt with the points in a different order and that he should not have decided that wind excitation at Colnbrook could be ruled out as a possible cause evaporated during oral argument.

Inadequate Packing

18. The judge relied on 4 matters in particular to establish a credible case for inadequate packing:-

(1) The manufacturing process was conducive to missing or insufficient packing. The packing was inserted while the tubes were being welded to the headers; when the cranked tubes were lifted up to be welded, there could not have been a single strip running the length of the gap between the rows and anything less could easily have slipped out or become displaced.

(2) Packing strips of different lengths were used many of which did not extend to the full length of the gap between the rows of tubes; if a short strip fell out, some of the tubes would be unsupported and could then set up the necessary vibration resonance.

(3) Packing was not applied in even thicknesses across the economiser blocks; so even if only 2 strips out of the 5 packing strips per gap were to be missing or fall out, sufficient movement of tubes would occur to set up the necessary vibration response.

(4) There was what the judge called "cogent" photographic evidence that some elements of packing were missing during transportation.

19. Mr. Simon Rainey QC (who did not appear below) attacked the conclusion of the judge by submitting that, so far from there being cogent photographic evidence of inadequate or missing packing, the photographic record showed that the packing was entirely satisfactory or, at least, far from unsatisfactory. He produced for the court a schedule of photographs said to have been taken on 28 August 2007 as well as a diagrammatic representation supposedly showing which blocks were being photographed and in what position the photographer must have been standing. He then invited the court to compare that schedule and that diagram with a distribution map (produced for the judge) which showed the rows in the blocks where welds had been found to be cracked. He then invited the court to conclude from the photographic evidence that there had been proper and adequate packing in the gaps either side of the row of tubing in the upper header of which the cracks had been found. He pointed out that most of the damage had occurred in block 3 of boiler Number One (together with some damage in blocks 2, 4 and 5) and that the photographs could be shown to be of precisely those blocks. He submitted further that the court was in just as good a position as the judge to examine the photographs and draw appropriate conclusions from them.
20. There are, as it seems to me, a number of difficulties with Mr Rainey's submission. First, it is based on incomplete material. The distribution map relied on did indeed show (in black) weld joints which were found to be cracked. But it also showed (in red) weld joints where crack like indications were found and it was agreed that these joints may well have included joints which were in fact cracked. Moreover it showed yet further (in yellow) places where joints were not even tested. Some of those joints may also have been cracked.
21. Secondly it is wrong to infer that merely because one knows the position of a crack that it must follow that the place where the packing was missing was the gap either side of the tube at the top of which the crack was found. Mr Anderson's expert report makes clear his view that, if a block has (say) 6 gaps due to inadequate or missing packing, a series of rows of tubes could start vibrating together and set off a vibration that could cause cracking elsewhere in the block. (I have in mind passages in his written evidence at 4/812-3 and in his cross-examination at day 4 pages 11, 20-23 and 27). The judge (para 125) accepted that part of Mr Anderson's evidence and it was open to him to do so, especially in the light of his general preference of Mr Anderson's evidence to that of the defendant's expert, Dr Hunt.
22. Thirdly, the photographic record was not nearly as secure as Mr Rainey claimed. The existence of the photographs on which the Defendant now relies only become apparent on exchange of expert's reports. Although the photographs had been in the Defendant's possession or control since September 2011, they were not disclosed until 20 March 2012 (2 days before the trial started) when the disclosure statement said that the photographs were undated. The trial began on 22 March and Mr Anderson gave evidence on Days 3 and 4 of the trial with a little bit extra on Day 5. Only after his evidence had been given was the computer disk produced which contained certain "metadata" such as the apparent time and date of the photograph. Naturally enough in view of this late disclosure, the judge asked for details of the provenance of the disk. No reliance had up to this time been placed by the Defendant on the metadata and the judge made it clear that he would not draw conclusions from the disk without evidence of its provenance. On the following day (Day 6) Ms Ansell for the Claimants made it clear that it would be unfair if anything from the disk was to be relied on, since whatever might be relied on had not been put to her expert, Mr Anderson. Mr Anderson was in fact recalled for a short time on Day 7 but nothing was put to him about the date and time of the photographs as recorded on the disk. The evidence concluded on day 8 (2^nd April). The judge ordered that written closing submissions should be exchanged by 8.00 AM the next day (3^rd April) preparatory to oral closing submissions being made on 4^th April. The judge and the Claimants were then presented with over 100 pages of written submissions from the Defendant which included submissions about the dates and time of the photographs without evidence of provenance having been given as required by the judge on Day 5.
23. In these circumstances the judge was eminently justified in saying (para 123) in relation to a photograph on which the Defendant placed particular reliance as being of Block 3 in Boiler 1 that the attribution was insecure. He added:-

"The series of photographs, whose source is Takuma, have amongst them photographs of a name plate identifying the block number. The Defendant's argument assumed that the photographs which followed the photograph of the name plate were of that block. But it can be determined that some photos of the block which follows such a name plate must be of a different block. The metadata on the disk of photographs with which I was provided suggests that the photographs are a selection of those taken, not a complete and systematic record".

He also said in response to the submission that the existence of damage in particular blocks was inconsistent with photographs of such blocks showing the packing in place:-

"But all the individual points made in support of the contention started from the false premise that one could identify which block was the subject matter of a particular photograph".

That was also a remark which is eminently justified.

24. As Mr Andrew Bartlett QC for the Claimants pointed out, it was also a curious fact that whereas at trial and in the Defendant's Skeleton Argument for this court it was said that the block shown in the photographs was the block after the photograph recording the number of the block, Mr Rainey in his oral submissions asserted, at least in some cases (e.g. photograph 713) that it was of a block before the number of the block (number 2) which was recorded on the photograph.
25. In the circumstances the Defendant has not, in my judgment, even now securely proved which photograph is of which block, let alone that the gaps between the relevant rows were adequately packed.
26. Furthermore, while this is of significance in itself what is even more significant is that the Defendant's case, based (as it now is) in tying particular photographs to particular gaps between tubes in particular blocks, was never put to the Claimants' expert witness, Mr Anderson. It is just not right to say that unprovenanced photographs can speak for themselves and that this court can examine them in just the same way as the judge can. If a case was to be made that particular photographs showed particular blocks, that is a case which should have been placed before the Claimants' expert with time given to him so that he could assess that case in the context of the litigation as a whole. It is not right for such a case to be mounted in written closing submissions for the first time, let alone for it to be mounted even better, after time for reflection, in this court.
27. That is sufficient to dispose of the positive arguments in favour of the first ground of appeal but I would add that the first 3 of the matters relied on by the judge are important. His findings in relation to manufacturing process, the length of the strips and this uneven thickness were open to him on the evidence. Moreover the fact that the Defendant cannot call on the photographs to justify its positive assertions that the packing was sufficiently tight and workmanlike to avoid the risk of vibration does not mean that the photographs produced cannot be relied on to justify the various points made by the judge in sub- paragraphs 122 (3) - (6).

Roughness of road surface and journey times

28. Mr Rainey relied on the judge's finding that the range of cycles of resonant vibration required to cause the observed fracturing was a minimum of 20,000 cycles and a maximum of 80,000 cycles. A frequency (which was likely) of between 4.4 Hz and 5Hz would then give the following times necessary to accumulate the relevant number of cycles:-

(1) for 20,000 cycles, between 67 and 75 minutes;

(2) for 50,000 cycles, between 2 hours 47 minutes and 3 hours and 10 minutes;

(3) for 80,000 cycles, between 4 hours 17 minutes and 5 hours 3 minutes.

29. The judge further said that the journey from the factory in Bucharest to the port of Constanta was about 215 kilometres of which 65 kilometres was on urban roads of poor quality (where the average speed would have been about 25 kilometres/hour and the journey time would have been over "2 ½ hours") and 150 kilometres on a motorway. The judge accepted hearsay evidence that two years after the transport the motorway was of tolerable quality but only after improvements had been made in the intervening period so that the surface of the motorway might have been worse when the economisers were transported.
30. Mr Rainey submitted that, if one ignored that hearsay evidence as the judge ought to have done, the journey on the urban roads was too short for the necessary vibrations to have occurred.
31. That is not necessarily right because with a minimum of 20,000 cycles there would have been plenty of time for the vibrations to have occurred and, even if 50,000 cycles was a more approximate figure, 2 hours and 47 minutes equates, respectably enough, to the time the judge found would have been taken of over 2 ½ hours.
32. There was, moreover, evidence that the motorway had a rumble strip at the border between the road itself and the hard shoulder. Since the economiser blocks were extremely large and the lorry transporting them needed to be correspondingly wide, it is quite likely that a great part of the motorway journey would have been over that rumble strip which would itself set up vibrations if there had not been adequate packing.

Narrowband response

33. Here the submission was that Dr Hunt had given evidence, with the assistance of a model made for the purpose, that for the damage to have occurred the frequency would have had to have been a narrowband resonant response and that there was no way in which it had been shown that such a narrowband response could have been achieved.
34. The judge understood this submission but did not accept it for the reasons given in paragraph 103 of his judgment; he also made clear in paragraphs 88-91 that he could not accept Dr Hunt's model. Mr Anderson dealt with the point in his evidence saying that vibration could be set up if only one or a few tubes vibrated and such vibration could have been at a narrowband frequency (Vol 4, tab 38, pages 830-1). Moreover Dr Hunt never went as far in his evidence as to say that damage occurring during road transport was impossible; he could only say that in his view it was very unlikely. In the circumstances the judge's rejection of the submission seems to me to be amply justified.

Standing Back

35. It is not disputed that the judge correctly directed himself in accordance with the Popi M [1985] 1 WLR 948 and Ide v ATB Sales Ltd [2008] EWCA Civ 424. In the latter case at para 4 Thomas LJ said:-

"As a matter of common sense it will usually be safe for a judge to conclude, where there are two competing theories before him, either of which is improbable, that having rejected one it is logical to accept the other as being the cause on the balance of probabilities."

Of course the judge has to be satisfied that the second theory is, on the balance of probabilities, correct. That is what Popplewell J decided and that was a conclusion which was open to him on the evidence. In any event I agree with him that it was indeed more likely than not that the damage occurred during the transportation to Colnbrook. In the circumstances I would dismiss the appeal.

Lord Justice Moses:

36. I agree that the appeal should be dismissed for the reasons given by Longmore LJ. I wish to underline the audacity, if not effrontery, with which the appellants have advanced this appeal.
37. The foundation of their argument lay in the appellants' assertions as to what should be derived from the photographs referred to in Longmore LJ's judgment at [19]. In my view the appellants ought never to have been permitted to rely on those photographs either at trial or on appeal.
38. As Longmore LJ has pointed out, the appellants had the photographs in their possession in September 2011. They were not disclosed, as they should have been, at that time, six months before the trial. That would have given all parties and their experts ample time to determine their provenance, the times and dates when they were taken and obtain the views of the experts as to their significance. On the contrary, not only were the photographs hidden from the respondents' scrutiny, the appellants did not even obtain for themselves such evidence as would have been necessary before any reliance could be placed upon them. Without accurate evidence of precisely which loads were illustrated in the photographs, they seem to me of little use and, possibly, misleading. That evidence depended on accurate evidence of when, where and from what position they were taken. They were digital photographs; there can have been no difficulty in obtaining such information. No explanation was given either to Popplewell J or to this court as to how or why the appellants allowed this state of affairs to arise.
39. The photographs were disclosed shortly before the trial. They should not have been disclosed without at the very least the details I have identified. Some of those details, some of "the metadata", but by no means all, were disclosed five days into the trial by which time it was far too late for witnesses or the parties and still less the judge to assess their significance. The judge did his utmost to be of assistance on the sixth day. He warned that he would be unable to draw conclusions without proper evidence as to provenance and what they illustrated. But he was not invited, at that stage, to rule that they should not be admitted. He asked for further information.
40. The following day, the seventh day of the trial, counsel for the respondents, emulating the courtesy of the judge, put down what she described as "a marker". She pointed out that if reliance was to be placed on the dates of the photographs, they had not been put to the witnesses. The judge noted the comment and the trial continued on a different point.
41. Despite the manner of their disclosure, the appellants' counsel, who appeared below, did rely on the photographs in his final submissions as a major method of attacking the respondents' arguments as to the cause of damage. The judge was focussing on the dense and technical expert evidence as the clarity and cogency of his exposition in his judgment demonstrates. But he was prepared at least to consider the photographs before dismissing the arguments based on what they were said to illustrate.
42. The respondents' counsel was placed in a position which ought never to have been allowed to occur. It is difficult for counsel in a civil case to object to evidence on the basis it has been adduced far too late and without adequate background evidence as to whether it is reliable and significant. Too strident an objection may prompt a Gertrude-like judicial response to the protestations of the Player Queen. But the judge had every right to invoke the overriding objective and exclude the evidence of the photographs pursuant to CPR 32.1(2). He should have been invited to take that course. That would, at least, have flushed out the full extent of the defects in their reliability. If the appellants had wished to make good those defects, they would have been forced to ask for an adjournment. That might well have been refused, putting a quietus on their deployment as a tool for challenging the respondents' theory.
43. But far from being consigned to the dustbin of last-minute evidence, the photographs were resurrected on appeal, with the assistance of Mr Rainey QC, who cloaked what was in reality impermissible fresh evidence with beguiling assertion and with a bi-coloured sketch plan which he had drawn. Had Stanley Burnton LJ appreciated the circumstances in which those photographs emerged and the reliance which would be placed upon them, he would surely never have given permission to appeal, however imminent his retirement.

Sir Alan Ward:

44. I agree.