Plaintiff v Defendants

Damages - Goods and services tax (GST) - Whether an award of damages should include GST paid by the plaintiff to third parties as a consequence of rectifying the property damage - Whether damages awarded for loss arising from property damage constitute supply for the purposes of GST - Where the plaintiff is registered for GST purposes and entitled to receive an input tax credit.

Damages - Economic loss in tort and contract from failure to take reasonable care - Concurrent wrongdoers - Proportionate liability - Wrongs Act 1958 (Vic) Pt IVAA, s 24AH.

1 In April 2008, the plaintiff, Fulton Hogan Construction Pty Limited ("Fulton Hogan"), contracted with the Southern and Eastern Integrated Transport Authority ("SEITA") to construct, supply and install a footbridge over the Burwood Highway at Vermont South in Victoria.

2 Fulton Hogan subcontracted the work to the first defendant, then trading as Danfab. One aspect of the work involved applying a protective coating to the trusses and handrails of the footbridge. The protective coating was intended to provide a 25 year period of corrosion protection to the footbridge.

3 Danfab engaged the second defendant, Cox Coating Pty Limited ("Cox"), to carry out coating works on the trusses. During the period December 2008 to January 2009, Cox coated the trusses with a coating system using paint products manufactured by the third defendant, Wattyl Australia Pty Limited. Soon after application, the coating on the trusses exhibited signs of failure.

4 During the period from late 2008 to mid-2009, other contractors[1] applied the same coating system to the handrails of the footbridge. Soon after application the coating on some of the handrails exhibited signs of failure.

5 The coating system used in each case comprised a first coat of Wattyl Galvit EP100 primer (made up of EP100 Parts A and B), a second coat of Wattyl Epinamel EB600 and a third coat of Wattyl Poly U750 (collectively the Wattyl coating system).

6 It was common ground at trial that the Wattyl coating system was a suitable coating system for the purpose of coating the trusses and handrails; and that there was a failure of the Wattyl coating system as applied to the trusses and some of the handrails. The central issue in the proceeding was whether the failure of the coating system was caused by defective products manufactured by Wattyl or by poor application practice by Cox or other contractors.

(1) What were the characteristics of the failure of the coatings on the trusses and the handrails?

(2) Were the coating products used on the Burwood footbridge project defective?

(3) If yes to (2), did the defective coating products cause or contribute to the failure of the coating on the trusses and the handrails?

(4) Was the Wattyl coating system properly applied in accordance with the Wattyl product specifications

(5) Was the Wattyl coating system properly applied in accordance with the specifications under the SEITA agreement, the Danfab agreement and the Cox agreement

(6) If no to (4) or (5), did such application cause or contribute to the defects in the coatings on

(7) If two or more of the defendants caused or contributed to the loss or damage, how is liability to be apportioned between the defendants?

(8) What is the extent of the loss and damage suffered by Fulton Hogan

(9) Did Cox give a warranty to SEITA and Fulton Hogan with respect to the costs and expenses incurred in repairing any defects to the trusses?

8 On issues of liability, the parties relied extensively on evidence given by expert witnesses. Fulton Hogan relied on the evidence given by Dr Vyt Garnys who is the Managing Director and Principal Consultant of CETEC Pty Ltd, a science and technical risk management consultancy.

9 During the trial, Wattyl, through its counsel, made a wide ranging and sustained attack on Dr Garnys and submitted that the Court should not rely on Dr Garnys as an expert on matters to do with the Galvit product specifically and the coating system generally.

10 First, Wattyl challenged his expertise. It was submitted that Dr Garnys' area of expertise was unrelated to three layer coating systems in which the primer is a zinc-rich primer and that he had no expertise or experience in relation to zinc-rich primers and their characteristics.

11 Counsel for Fulton Hogan submitted that the challenge to Dr Garnys' expertise was premised upon a field so narrow as to be meaningless and ignores the fact:

(a) that the process of considering and determining whether the EP100 product is defective is an analytical one which requires a scientific examination into its chemical composition;

(b) that Dr Garnys' doctorate from the University of New South Wales is founded in the discipline of analytical chemistry; and

(c) that Dr Garnys was a consultant analytical chemist with broad experience in conducting chemical analysis of industrial coatings and is a member of the Australian Corrosion Association.

12 In short, it was submitted that Dr Garnys was a highly qualified and experienced scientist.

13 Dr Garnys was academically well credentialed, with a Bachelor of Science degree with honours from the University of Western Australia in the field of chemistry and a PhD from the University of New South Wales. He had over 30 years of technical experience, having been consulted in areas including product and process failure investigations and corrosion prevention and management.

14 Dr Garnys had been involved in about 30 major coating investigations over the last 25 years and minor investigations much more frequently. Over twenty of the major coating investigations had involved the performance of painted coatings on various surfaces. He had conducted investigations over 10 years in relation to multi-layer epoxy coatings for the State Government of Queensland, which involved a close examination of properties of epoxy and polyurethane coatings including issues of adhesion, composition and structure. At the time of the trial, he was involved in a multi-layer coating system investigation. Dr Garnys said during the trial that he was very familiar with epoxy coatings.

15 Dr Garnys gave evidence that whilst he did not have particular qualifications on protective coatings, he had presented academic papers in respect of protective coating issues in multi-layer coating systems. He was a member of the Australian Corrosion Association which examines coatings and the protection and corrosion of the underlying substrates and said that this was of great interest to him. He was also a member of the Australian Chemical Institute and under its auspices regularly discussed coatings and chemicals and their analysis. He was a member of the Australian Institute of Mining and Metallurgy and regularly read its journals and kept up with what coatings do and can do. Finally, Dr Garnys gave evidence that he owns and runs a science and technology investigation organisation and that "all of these things are par for the course for our work".

16 It is clear from the evidence that Dr Garnys had a particular interest in indoor air quality and the properties of materials used in the built environment and that he did not hold himself out as an expert in multi-layer coating systems using a zinc-rich primer. His previous experience relating to the failure of zinc-rich primers involved durability issues and other problems, not specifically a cohesive failure.[2] In my view, this does not disqualify him from being able to engage in an analytical examination of the subject matter adequately for the purpose of this proceeding and to express an expert opinion in relation to the failure of the coating system.[3] I accept that Dr Garnys had adequate expertise for the purpose of expressing an opinion in this case.

17 Second, Wattyl submitted that the Court should not rely on Dr Garnys' evidence on the basis that he had spent only ten minutes on a site examination, did not do much investigative work himself but relied on other people's work and because he did not look at issues to do with application and had no experience with regard to matters of application.

18 It is true that Dr Garnys relied on investigative work carried out by his colleague Dr Robert Schiller. Dr Garnys readily conceded that Dr Schiller had attended the site for longer than he had for the purposes of carrying out the investigation. Dr Garnys gave evidence that he attended the site for "probably less than half an hour, 10 minutes" to see the areas that Dr Schiller had worked on and to gain first-hand knowledge of what the area looked like, not to re-inspect the whole site and trusses.

19 Dr Schiller made observations and collected paint samples at the site visits referred to in the CETEC report of December 2010. These sampling locations were documented photographically. After the samples were brought back, Dr Garnys and Dr Schiller examined them and discussed the mode of failure. Within a week of collection of the samples, Dr Garnys attended the site with Dr Schiller to have a look at the defects. During the site visit, Dr Schiller showed Dr Garnys where the samples came from and where the observations were made.

20 I do not accept that the Court should ignore the evidence of Dr Garnys in relation to what was observed on site and any conclusions drawn by him from site observations. A short visit is better than no visit.[4] Further, the site visit took place after Dr Garnys had been briefed by Dr Schiller and had seen the samples and photographs of the delaminated areas. In that sense Dr Garnys went to the site prepared. Despite the brevity of his visit, Dr Garnys had the opportunity to observe the delamination phenomenon and could see that the Galvit primer was adhering to the steel substrate of both the trusses and the handrails, even though the emphasis of the work of Dr Schiller and Dr Garnys was on scientific analysis at the laboratory.

21 Dr Garnys explained that he and Dr Schiller worked as a team with only one person taking samples in order to avoid duplication of work.

22 Although Dr Garnys took some casual measurements of dry film thicknesses from the trusses, he said he did not report on these measurements because other people had already taken measurements. For example, Mr Athol Stone of Macspec Inspection Pty Ltd had taken 209 readings on the west side of the truss and 315 readings on the east side of the truss. Likewise, Mr Peter Dove of GHD Pty Ltd and Mr Don Bartlett of CTI Consultants Pty Ltd both took extensive dry film thickness measurements. Sensibly, all parties agreed that the measurements contained in these reports could be relied on for all purposes. In Paino v Paino,[5] the New South Wales Court of Appeal noted that factual information in an expert's report may be sourced from a party, from empirical investigations or other data sources and that an expert need not amass all factual data on which an opinion is to be expressed. That is a task which can be delegated. In this context, in ASIC v Rich,[6] Austin J said:

But it is necessary for the expert who is the author of a report to apply his or mind to the analysis and reasoning processes that his or her subordinates have developed, so that when the report is finalised, the whole of the reasoning and conclusions that it contains have been adopted as the expert's own reasoning and conclusions.

23 In this case, the CETEC report of August 2009 was a report signed by Dr Schiller and Dr Garnys. Dr Garnys' attention was drawn to the opening words of the August 2009 report: "I have written this report". It was suggested that Dr Schiller (and not Dr Garnys) was the author of the report, and that as the principal of CETEC, Dr Garnys' signature may simply have been indicative of him taking responsibility for the report rather than authorship. Dr Garnys said that it was incorrect to suggest that Dr Schiller was the sole author. Dr Garnys said:

Well I know it's incorrect. Dr Schiller does not have the analytical insights that I do have and neither does he have the years of experience that I have. On the other hand he has experience that I don't have. So that's why we decided to do it jointly because we saw it was a complex report.

24 Dr Garnys gave evidence that the engagement was with CETEC and that he and Dr Schiller took this case on and that the report was written jointly with his input.

25 I accept Dr Garnys' evidence. Further, to the extent that the CETEC report of August 2009 represents the work of Dr Schiller, I find that Dr Garnys had applied his mind to Dr Schiller's analysis and reasoning processes so that the reasoning and conclusions in the report were adopted as Dr Garnys' own reasoning and conclusions.

26 I am satisfied that he can properly claim to be the joint author of the August 2009 report.

27 Third, Wattyl submitted that Dr Garnys did not carry out the analysis work because the CSIRO operated the scanning electron microscope ("SEM"). I do not accept that it follows that Dr Garnys did not do this investigative work. As Dr Garnys explained, the scanning electron microscope is CSIRO property and both the operator and the person sitting beside the operator can see the screen. Dr Garnys gave evidence that he would direct the operator "where to turn the knobs and what to look at. It's a team effort. We work on a two chair system..." and "we sit alongside the operator and guide the operator as to where to look". Dr Garnys stated that "it's extremely important to do that because the operator doesn't know the background or what to look for, so you're basically telling him where to direct his instrument and field a view".

28 Dr Garnys took the same approach with infrared investigation using the spectrometer, where he operated 'side by side' with the technician.

29 In my view, it would be wrong to say that because Dr Garnys sat alongside a technician operating equipment under his direction, that Dr Garnys has not as a matter of substance carried out the investigative work himself. It was not suggested that Dr Garnys would not be in a position to direct the technician to vary the conditions to optimise the relevant SEM image or infrared spectrum.

30 Fourth, Wattyl submitted that Dr Garnys took a limited approach to the investigation because he undertook no systematic examination of whether there was overcoating and no examination of remedial work to the trusses, that he was unable to comment on verifying observations made in Mr Stone's report about dry spray; that he was unable to comment on the blasting process; that he had not found areas lacking in protective coatings or pockets exhibiting corrosion; that he had seen some orange peeling but could not say how much; that he had not examined the mounting plates on the handrails; that he had not detected inadequate coating thicknesses in places and that he had looked at Cox's Quality Assurance records but did not do a detailed analysis of compliance or non-compliance with what he saw; and that he did not address issues of design or examine the substrate. It was also submitted that he lacked experience and expertise in application matters.

31 In my opinion, Dr Garnys confined himself to areas within his expertise and the limits of his brief. He did not profess expertise that he did not have; he was not an applicator and did not profess to be an applicator. Whilst he gave evidence that the footbridge did not look as though it had a mirror finish or that it was a uniform job, he said that CETEC was not retained to give an opinion in relation to issues of the design of the footbridge or the distribution of coating thicknesses and details of applicator technique. Dr Garnys declined to answer questions beyond his expertise. This is what one would expect of an experienced independent expert. I find that Dr Garnys' evidence should not be rejected on these grounds.

32 Cox relied on the evidence of Mr Gerry Eccleston, Senior Materials Scientist employed by CSIRO in its Manufacturing and Infrastructure Technology Division. Mr Eccleston has a Diploma of Applied Chemistry from Swinburne Institute, a Post-Graduate Diploma of Water Science from Chisholm Institute and a Master of Applied Science from RMIT on the subject of zinc-rich coatings.

(a) Mr Jim Haig, a Materials Scientist employed by Uniquest Pty Ltd. Mr Haig has a Bachelor of Applied Science (Applied Chemistry) and a Post Graduate Diploma in Environmental Studies (QUT). Mr Haig is also a National Association of Corrosion Engineers Coating Inspector Level 1 and has held positions on an Australian Standards Committee for Site Testing of Protective Coatings and the technical working group of the Australian Paint Approvals Scheme.

(b) Mr Mark Dromgool, a Mechanical Engineer specialising in protective coatings, paint, tank linings and the corrosion control of steel and other metal substrates and concrete bodies, paints and coatings. Mr Dromgool is the Managing Director of KTA-Tator Australia Pty Ltd, Protective Coatings Engineers and Consultants. Mr Dromgool is an accredited Protected Coating Specialist[7] and an ACA Certified Coatings Inspector and a NACE Level 3 Coatings Inspector.[8] Mr Dromgool is a member of the Society for Protective Coatings (SSPC), the National Association of Corrosion Engineers (NACE International) and a member of the Australasian Corrosion Association (ACA).

(c) Mr David Cook, an Industrial Analytical Chemist employed by Wattyl. Mr Cook has a Bachelor of Science (Technology) (UniNSW) and a Graduate Certificate in Surface Coating Technology (TAFE).

34 Before the trial, Dr Garnys, Mr Haig, Mr Dromgool and Mr Cook attended a conclave of experts and subsequently compiled a joint report of experts.[9] Mr Eccleston did not attend the conclave and was not a signatory to the Joint Report.

35 During the trial, Dr Garnys, Mr Eccleston, Mr Haig, Mr Dromgool and Mr Cook gave evidence concurrently.

36 By agreement between the parties, several additional expert reports were admitted into evidence without the authors being required to attend court; namely a report of Mr Peter Dove of GHD Pty Ltd dated 23 January 2009, a report of Mr Don Bartlett of CTI Consultants Pty Ltd dated 10 February 2009 and two reports of Mr Athol Stone of Macspec Inspection Pty Ltd dated 27 May 2009 and 9 June 2009 respectively. Accordingly, these individuals were not subjected to cross-examination on their reports.

Question 1 - What were the characteristics of the failure of the coatings on the trusses and the handrails?

37 On this question, it is necessary to determine whether the failure of the coating on the trusses and handrails was a cohesive failure, an adhesive failure or a combination of both.

38 Dr Garnys gave evidence that there was a cohesive failure of the Galvit EP100 primer. He said that the delaminated paint flakes that were received or observed at the site showed similar colour characteristics on the internal faces of delamination suggesting that the failure had occurred within the paint layer rather than between paint layers. When examined microscopically in the laboratory, the paint had the texture of being ripped apart, with residual grey paint material left on the substrate and on the underside of the delaminated paint flakes.

39 Cohesive failure in this context means failure within a single layer of paint whereas adhesive failure could occur at the boundary between two layers, whether they were the same or different coatings. Had there been an adhesive failure between coats of similar materials, Dr Garnys said that there would be subtle differences in appearance which are able to be recognised in the field and without a microscope. In particular, there would have been differences between the two delaminated sides and it would be possible to see the characteristics of the surface of the first coat. Had contaminants caused adhesive failure between coats of similar materials, it would be possible to see residual contaminants on the coating.

40 Mr Eccleston also gave evidence that the failure appeared to be a cohesive failure. When removing samples from the trusses at the site for laboratory analysis he found that the underside of the coating that he had removed was grey and the material left on the steel substrate was also grey indicating that the primer was still there. Mr Eccleston took measurements on the steel surface of the trusses and found that a coating remained on that surface. The variability of the coating on the steel substrate and on the underside of the coating that he had removed indicated a rough surface rather than a smooth surface. According to Mr Eccleston, a smooth surface would have been an indication of two layers losing adhesion. For there to be an adhesion failure between two coats of the primer, Mr Eccleston would have expected a more uniform coating on the steel substrate. Subsequent observation under an optical microscope confirmed that the surfaces on the underside of the paint flakes were fairly rough, consistent with a cohesive failure rather than an adhesive failure.

41 Mr Haig said that for him to be satisfied that in fact there had been a cohesive failure he would need to demonstrate that the residue left on the steel surface was exactly the same material as present on the disbonded flakes. In the Joint Report, Mr Haig stated that the absence of surface preparation impressions or traces of another coating led him to believe that the failure was cohesive rather than adhesive but that this would not preclude a largely adhesive failure between layers of EP100.

42 Mr Haig gave evidence that in three of seven paint samples that he examined, he observed a disbondment between two coats of the primer in that two layers were present on some of the samples but not on other parts of the samples. Mr Haig said he saw evidence of adhesive failure in some places as well as evidence of cohesive failure, but ultimately stood by the opinion he had earlier expressed in the conclave of experts that this was a cohesive rather than an adhesive failure.

43 Wattyl also relied on the conclusion expressed in the first report of Mr Haig where he refers to finding multiple layers of Galvit EP100 during his SEM examination. Mr Haig said, "that adhesive failure was a significant contributor, and by adhesive failure I mean between coats of EP100". Mr Haig gave evidence that "in many cases the samples that I was supplied with had at least two coats of the EP100 (Galvit primer) and that disbondment commonly occurred between those two layers". When asked how many instances of adhesive failure he had observed in the samples, Mr Haig said that at least three of the samples (which included truss and handrail samples) displayed signs of multiple coats. Mr Haig was then asked by Dr Garnys whether a particular SEM image in his report showed multiple layers, to which he replied that that image did not show multiple layers, but then took the Court to an SEM image of a paint sample taken from the truss which he said was indicative of multiple layers of EP100. In his oral testimony, Mr Haig said that there was a subtle discontinuity which was indicative of two layers of EP100 to the left hand side of the photograph. This is to be contrasted with his written testimony. In the commentary relating to the same photograph in Mr Haig's report of 7 July 2011 Mr Haig states, "This sample also showed several plaques of zinc oxidation within the Galvit EP100 layer" and makes no reference to multiple layers.[10]

44 It is clear from the photograph that there is an irregular parting between the zinc oxide plaques. According to Mr Eccleston, this photograph shows the splitting of the zinc with corrosion products forming in the split. Mr Eccleston disagreed that the photograph was indicative of two separate layers. He said that he would be looking for two definite layers and would expect there would be some sort of line between them indicating the two layers. I prefer Mr Eccleston's evidence on this point.

45 In my opinion, Mr Haig's evidence is problematic on this issue in other respects. Whilst he gave evidence that there were multiple coats of primer in three out of the seven paint samples that he examined, Mr Haig did not identify which of the other samples had multiple coats of primer.

46 Dr Garnys was unable to distinguish whether the Galvit primer was applied in one or multiple coats. In addition, the area of discontinuity observed by Mr Haig was approximately 0.2 millimetres across which, as Dr Garnys observed, is an exceptionally small area compared to the areas of peeling paint.

47 In the Joint Report, Mr Dromgool stated that he considered that the characteristics of the failure of the coating system was a disbondment through or within the thickness of the EP100 primer, but he was not able to say whether this was caused by a cohesive failure (within a single applied layer) or between successive applications. Mr Dromgool said it did not really matter as the constituents of the layer itself had failed.

48 Mr Cook was unable to express an opinion on this issue. He said that all he had to look at were paint flakes, being the zinc layer and the other coatings, that what he saw were stuck together and that there was no failure between the two coatings in the samples provided to him.

49 In a National Technical Complaint Form Lab Report completed by Mr Paul Pellegrino, Technical Services manager of Wattyl, Mr Pellegrino stated in relation to the trusses (painted by Cox), that cross-hatch adhesion results were reasonable throughout both trusses and that the failure was clearly occurring within the Galvit EP100 zinc-rich primer layer. In a National Technical Complaint Form Lab Report completed by Mr Pellegrino in relation to handrails (painted by DH Corrosion), Mr Pellegrino stated that inspection clearly confirmed splitting of the zinc-rich primer.

50 Wattyl submitted that Dr Garnys had in fact described an adhesive failure in the CETEC report of December 2010. This submission was based upon a statement of observations made by Dr Schiller, who had made a site inspection and had observed "(t)he coating failure had involved delamination from the Wattyl Galvit EP100 primer coat" and subsequent similar instances of usage of the word "from", including reference to "delamination of the Wattyl Epinamel EB600 from the Wattyl Galvit EP100".[11] It was submitted that the words speak for themselves.

51 I am not persuaded by this submission. In my view, the words relied upon are generally descriptive of the delamination phenomenon but do not precisely define it. The submission focuses too heavily on the use of the word "from" and ignores the fact that all of the samples of delaminated paint collected by Dr Schiller show a greyish coating on the underside. It ignores the results of analysis of a sample selected for analysis using a scanning electron microscope, which confirmed that the underside of the delaminated sample showed different characteristics to the top side and led to the conclusion that the underside was likely to be the primer coating. It also ignores the observation that further work on the other samples was likely to confirm the presence of a zinc-rich layer (Wattyl Galvit EP100) on the underside of the delaminated coating. Further, it ignores the conclusions expressed in the CETEC Report of August 2009 as follows:

The coating failure comprised a delamination phenomenon. The examination of delaminated coating through electron microscopy and infrared analysis has confirmed that the underside of the delaminated coating was Wattyl Galvit EP-100.

For this situation to arise coating failure must had [sic] occurred within the Wattyl Galvit EP-100 coating layer. That is the primer coating layer has split or separated within itself leaving a layer attached to the steel substrate and another layer bound to a subsequently applied coating (eg Wattyl Epinamel EB600).

52 In my view, these passages, read fully and in their context, describe a cohesive failure in the layer of Galvit EP100.

53 Wattyl also relied on the additional expert reports of Mr Dove, Mr Bartlett and Mr Stone in support of the case of adhesive failure.

54 In relation to Mr Dove's report, Wattyl relied on the results of a "pull-off" test which produced an adhesion failure at the interface between the coating and the steel substrate on the south truss. This test involves attaching a dolly to the coating using glue, which is allowed to cure. The dolly is then pulled off the coating using a hydraulic jack. The test is designed to be destructive. In the instance relied on by Wattyl, a pull-off force of 9.6 megapascals was recorded. There is nothing in Mr Dove's report to say whether that pull-off force was a reflection of good adhesion or poor adhesion and Mr Dove was not called to give evidence.

55 Mr Cameron Cooper, a Paint Inspector and Quality Assurance Officer for Cox, gave evidence that he did not believe there was an Australian Standard for the expected pull-off reading but that by reference to American Standards, given the thickness and generic make-up of the coating, suggested that there should be a minimum pull-off force of 5 megapascals and that the relevant figures for the pull-off tests exceeded that minimum pull-off force.

56 In the circumstances, I am not assisted by Mr Dove's test result in determining whether the delamination phenomenon was a cohesive failure or an adhesive failure. However, I note that Mr Dove expresses the opinion that the weakness in the coating was within the epoxy zinc layer and that failures of the top coat over zinc-rich coatings almost always result from cohesive failure or splitting of the zinc film. Mr Dove states that the failure is revealed when appreciable zinc film sections are found still adhering to the underside of a delaminated film and to the steel substrate. In my view, that is a fair description of what occurred in this case.

57 Wattyl also relied on the observations of Mr Bartlett in support of the existence of adhesive failure in the coatings on the trusses. In particular, Wattyl relied on the following passage:

The failure of the coating system is occurring within the zinc epoxy primer layer. The cause of the very poor adhesion of the coating system on the North Truss is not resolved but the grossly excessive zinc epoxy is likely to have been a key factor, as this heavily pigmented coating is unlikely to have the cohesive strength of a coating of similar generic type with less pigment.

58 Mr Dromgool gave evidence that adhesion failure is an overarching term which is used globally and that all the tests to determine whether there is adhesive or cohesive failure are called adhesion tests. Mr Eccleston said he assumed the reference to "very poor adhesion" in Mr Bartlett's report should be to "cohesion", because he did not see any evidence of the coating disbonding from the steel substrate and further, that the reference in the second sentence to the cause of the "very poor adhesion" should be to "cohesion" because it follows from the first sentence. Dr Garnys said he believed that references to adhesion in Mr Dove and Mr Bartlett's reports were described incorrectly. Having regard to the evidence of Mr Dromgool, Mr Eccleston and Dr Garnys on this point, and in the absence of the authors of these reports, I consider it likely that Mr Bartlett and Mr Dove were speaking in general terms about the adhesion issue without focusing on whether the failure of the coating system was technically adhesive or cohesive.

59 Wattyl also relied on the observations of Mr Stone. In his report of 9 June 2009, Mr Stone states that he carried out over 70 adhesion tests over a wide area of the north truss and found that adhesion of coatings to 80% of this truss had failed. He then observes that:

[f]ailure of the coating system is occurring within the zinc epoxy primer (zinc splitting) ... Having stated the above, four adhesion tests failed where the original Pre-weld Primer (Wattyl Sigmaweld) is still evident on some areas of the top chords [sic] and does not have any evidence of a blast profile...

60 It appears to me from the passage reproduced above, that Mr Stone was drawing a distinction between a cohesive failure, i.e. where he observed zinc splitting, and an adhesive failure in the four tests where the Galvit primer failed to adhere to the substrate thus exposing the original pre-weld primer. I note that Dr Garnys and Mr Eccleston each had the advantage of examining the trusses on site and making field observations. In addition, Dr Garnys examined the handrails on site. Although he conceded that he spent a short time on this examination he saw "sheets (of paint) coming off" and adhesion (of the Galvit primer to the substrate). Subsequent measurement of samples from the handrails showed delamination within the layer of (Galvit) primer leaving a residue representing less than 40% of the originally intact layer of primer on the steel substrate and the remaining primer on the back of the delaminated paint samples. This indicated that the primer coating had separated.

61 Mr Cooper supervised the application of the primer on the trusses. He gave evidence that only one coat of primer was applied to the trusses. This evidence correlates with the daily inspection reports completed contemporaneously by Mr Cooper during December 2008 which record that one coat of primer was applied.

62 It is possible that there was some adhesive failure between multiple layers based on Mr Haig's observations on three out of seven sample flakes, but considering the evidence as a whole, particularly the lay evidence of Mr Cooper and the contemporaneous records that show that only one coat of primer was applied to the trusses, in my assessment this is unlikely to have been the mechanism of widespread failure of the coating, at least on the trusses.

63 On the basis of the lay and expert evidence I am satisfied on the balance of probabilities that the predominant characteristic of the failure of the Galvit EP100 primer on both the trusses and the handrails was a cohesive failure within a single zinc-rich primer layer rather than an adhesive failure from the substrate or between layers.

Question 2 - Were the coating products used on the project defective?

Question 3 - If yes to question 2, did the defective coating products cause or contribute to the failure of the coating on the trusses and handrails?

64 In the CETEC report of December 2010, Dr Garnys refers to the analysis of delaminated coating samples taken from the trusses (samples 68402, 68403 and 68404) and from the handrails (samples 68508 and 68509).

65 Dr Garnys stated that visual observation, electron microscopy and infrared studies indicated that the topside and underside surfaces of the delaminated coating were not the same and that it was likely that the underside surface was Galvit EP100 primer. Further sampling took place on sections of handrails (coated by other contractors) which had delaminated for the purposes of determining coating thicknesses by optical microscopy in the laboratory. The coating thickness (of primer) remaining on the steel substrate was measured at the site of removal of the delaminated coating pieces. Dr Garnys observed that areas of delaminated coating were apparent and were identical to the other coating failures already reported.

66 The coating thickness measurements showed that the Galvit primer had been applied near to but less than the dry film thickness limit of 75 microns recommended in the Wattyl "Specification information" for Galvit primer when it is used under thick, rigid epoxy systems. Where the coating had delaminated, the thickness of Galvit primer remaining on the substrate represented "perhaps less than a third of the original intact layer thickness.[12] Dr Garnys reported that it was not possible to distinguish between the two upper coats being layers of Wattyl Epinamel EB600 and Wattyl Poly U750.

67 Dr Garnys observed that the delamination failure occurred prior to the material entering service. In the case of the protective coating system applied to handrails at DH Corrosion, the delamination occurred soon after application of the intermediate coat (Epinamel EB600) and before application of the top coat (Wattyl Poly U750). Dr Garnys noted that more than one applicator had been involved with the application of the protective coating system on the steel substrate, yet the same delamination phenomenon had arisen.

68 Dr Garnys concluded that the Wattyl coating system was not of merchantable quality for the following reasons:

(1) The product information indicates an unapplied new product life of 12 months ... the used Galvit EP-100 Part A (Batch C60132971) was tested on 5th October 2006 with a water content of 0.030%, and maximum viscosity and drying times indicting early signs of deterioration even in 2006.

(2) The electron microscopy results below show metallic zinc spherules of 0.2 to 5 micron diameter with high oxygen to carbon to ratios of almost 50% indicating likely oxidation or hydration thus weakening the zinc to resin bond and evidence of poor mixing

(3) The infrared analysis shows evidence of hydration and visible carbonation of the zinc and possibly incomplete cure.

(4) The Coating thickness measurements show Galvit EP100 coating thicknesses within the 75 micron specification of 58 to 73 micron and a very uniform delaminated coating of 45 micron on the underside of the epoxy second coat. This indicates a uniform soft delamination irrespective of total coating thickness from 320 to 628 micron even in recent coats. These coats could be easily peeled apart even without delamination indicating incomplete cure well beyond the specified times. This is not characteristic of brittle shear due to varying stresses from topcoats of varying thicknesses.

69 Subsequently, in the Joint Report compiled following the expert witness conclave held on 23 September 2011, Dr Garnys stated:

• that his examination of the Wattyl coating system on site indicated a weakness in the paint flakes that had delaminated or had been recently applied to handrails in continuing to adhere as a system.

• there were islands of zinc corrosion products within the delaminated layer.

• there was no evidence of colour from the underlying primer or metal.

• that since samples of the original were not made available to him, he was unable to compare the material on the delaminated paint flakes to the specified composition.

• that with more detailed examination of the Galvit primer "bulk material" within the flake he observed the presence of zinc oxidation on each zinc particle.

• that following an examination of the original coloured FTIR data[13] there was evidence of no more water in the retained batches of the applied material than there was with new product.

• the quality control sheet indicated the increased age and some performance at the limits of specification.

• the panel of Galvit EP100 applied by Wattyl and left to cure outside by Wattyl showed failure.

70 Dr Garnys concluded "that there may be some deficiencies in the supplied WCS".

71 At trial, Dr Garnys elaborated orally on the site observations which he said tended to indicate that there was an inherent defect. Dr Garnys said that there was a bubbling and breakage of the film; that is, a visual separation of the film in the case of the handrails which indicated a problem which was most unusual; that there was a similar appearance of both coats when the layers were peeled off; that his colleague Dr Schiller had showed him photographs and described exactly how the layers were removed and Dr Garnys was informed there was almost no strength between the layers; that there were upturned corners of delaminated flakes which was not assisted which Dr Garnys described as "spontaneous delamination"; and that there was no evidence of the coating "having come off from the steel" nor any evidence of a pink or reddish primer coat on the steel.

72 As to the indicator of zinc corrosion products within the delaminated layer, Dr Garnys said in relation to the handrails that the samples showed no surface evidence of zinc carbonate, but did show little specks within the coating when looked at by electron microscope. In relation to the trusses, there were areas which Dr Garnys described as "islands" which were relatively large covering several square centimetres. These were samples of white zinc oxide which accumulated in some of the failed areas but were still coated and still covered by the top coats until they were peeled back and revealed. Dr Garnys said:

If you like, there's a blister, and when the blister was cut open and revealed then within that blister there would be some areas which were just straight grey, and not contaminated in this way. Other areas had quite pronounced amounts of the oxidation product. It was subsequently identified as zinc oxidation product and not any other contaminant.

73 Using a scanning electron microscope in "forward and back scatter" modes enabled Dr Garnys to look at the surface of the material and see the spherules of zinc and the resin between them. In back scatter mode, the scanning electron microscope highlighted the heavier elements (e.g. zinc) as opposed to the lighter elements (e.g. carbon and oxygen) and he saw on the surfaces of the spherules a "grading away from brightness into greyness that was suggestive of surface contamination".

74 Dr Garnys then looked at the infrared spectrum which was used primarily to examine the resin component. When Dr Garnys looked at the ratio of oxygen to carbon in the cohesively failed zinc coating, he saw that the oxygen to carbon ratio was higher in the delaminated area than it was on the top surface, namely the Epinamel-EB600 layer. Dr Garnys observed that there was more oxidation product within the delaminated layer than would be considered to be normal, referencing everything to the top layer. Dr Garnys said he was unable to reference the delaminated material against a sample of the same batch material, because a sample of the same batch material was not provided even though it had been requested.

75 Taking account of the rapid delamination, the appearance of the delamination, the visual microscope observation, the electron microscope work and the infrared work together and (at that stage) in the absence of any data from the manufacturer other than the application sheets, Dr Garnys deduced there was a high probability that the Galvit EP100 material was deficient.

76 From time to time whilst giving evidence concurrently the experts had an opportunity to ask each other questions. During such questioning Dr Garnys accepted that a normal zinc-rich product would have some component of zinc oxide; that although the perfect formulation would have no zinc oxide, normal industrial practice would be to have a product which would contain some oxidation. Dr Garnys said he did not know what the specification or ideal amount is for the Galvit EP100. Dr Garnys also accepted that if disbondment had occurred some time before the inspection, the zinc particles would have continued to oxidise during the time before inspection of the surface of the disbondment. Dr Garnys accepted that assuming disbondment had taken place for some time, then given the presence of moisture and air, there would be an accumulation of white carbonate deposits. However, Dr Garnys said in many cases where this disbondment had occurred there was no evidence of any break in quite a heavy strong surface coating, and that he did not observe any entry point for water, so the disbondment under those conditions would be in a dry condition and any water which may have been there in the absence of a break in the coating would have been coming "somehow from within the product".

77 Dr Garnys conceded that he had not examined the interior of any of the Galvit primer layers by removal of the surface layer or by cross-section. Dr Garnys said that the edge was looked at for the purpose of thickness measurement and he used the infrared technique to press against the body of the material, which allowed him to get some insight into the combination of the surface and the body of the material. Dr Garnys gave evidence that he didn't cross-section the material in any systemic manner. When asked whether he would accept that the spherules within the thickness of the Galvit primer layer were not in fact corroded, Dr Garnys said he would have to rely on other people's evidence as CETEC did not have evidence as to what was actually within the layer. Dr Garnys conceded that the comparison of oxygen level between the disbonded Galvit primer surface and the top side of the disbonded flake was a comparison between the Galvit layer and the much higher epoxy resin content material in the epoxy layer of the surface coating.

78 In relation to the infrared work, Dr Garnys accepted that the formulation of the Galvit primer contained amongst other things clay which contain hydroxy-groups and bound water, and epoxy resin which contains hydroxy-groups. In those circumstances, it was suggested that in a weak spectrum, as would be obtained from a highly filtered zinc product, the presence of these hydroxy-groups would make it very difficult to be definitive about the presence of water. Dr Garnys said that he was comparing spectra which gave him good information and accepted that with epoxy there was already a residual signal in the top coat, but that he saw a very much accentuated residual signal in the underlying coat which he considered to be abnormal for the surface that he was expecting to see. In this context, Dr Garnys accepted that he had not compared the new Galvit primer to samples of Galvit primer taken from the failed surfaces.

79 Dr Garnys also accepted that the disbonded surfaces of the trusses could have suffered from oxidation, hydration and carbonation between the time of disbondment and analysis. Dr Garnys said that some of the material was "sort of fresh when it was peeled, others had the white zinc oxide, zinc carbonate, but not in the handrails; on the handrails there was rapid delamination and no evidence of this crusting of abnormal amounts of - or visible amounts of zinc carbonate and zinc oxide on the surface".

80 Wattyl submitted that Dr Garnys' evidence of "spontaneous delamination" of the coating of the trusses and of "bubbling and breakage of the film on the handrails" was not referred to in the August 2009 CETEC report and was unreliable secondhand evidence which should not be accepted.

81 I do not accept the thrust of this submission. It is correct to say that these expressions are not found in the August 2009 CETEC report, but Dr Garnys makes it plain that he is using his own words to describe the phenomenon of apparent spontaneous delamination on the trusses which he observed on site. This is objectively supported by the upturned layers of material in at least one of the photographs in the joint CETEC report.

82 Similarly in the case of the description of "bubbling and breakage" of the film on the handrails. This is objectively supported by a photograph and caption which indicates that the coating on the handrails was removed by cutting using a Stanley knife which "revealed a lifted edge that could be freely peeled away from the primer coating". The observation that the cut "revealed a lifted edge" suggests to me that the coating was bubbling underneath the overcoating and that the primer film had already delaminated underneath such that it could be freely pulled away.

83 It follows that I do not accept that Dr Garnys misinterpreted or embellished Dr Schiller's comments of what he observed on site.

84 Wattyl submitted that the Court should reject Dr Garnys' statement that there were "islands" of zinc corrosion products within the delaminated layer on the trusses on the basis that Mr Haig had established by examination of a cross-section of the interior of the Galvit primer flakes that "there was no evidence of inconsistency through the material and no evidence of contamination through the material that would be indicative of defective product".

85 In relation to the trusses, there may be no real dispute as to the location of the zinc carbonate deposits. Dr Garnys refers to zinc oxide which accumulated in some of the failed areas which were still covered by the top coats. I understand from this evidence that Dr Garnys was referring to an accumulation of zinc corrosion products in the delaminated primer layer. Once the paint flake was removed it appears that the deposits were visible on the surface of the Galvit primer on the flake. I accept that Dr Garnys' description of the corrosion which he saw as "islands" of zinc corrosion products is not inconsistent with Mr Haig's explanation that isolated plaques of deposits of corrosion products which he saw were on the surface of the Galvit primer.

86 This is to be contrasted with Dr Garnys' evidence that he observed no surface evidence of zinc carbonate on the samples from the handrails but did observe "little specks within the coating when looked at by electron microscope".

87 Wattyl submitted that the Court should not accept Dr Garnys' statement in the Joint Report that there was no evidence of colour from the underlying primer or metal where the coating had delaminated.

88 I do not accept this submission. Dr Garnys' statement is expressly referenced to the observations he made on site. I accept the accuracy of these observations. Dr Garnys did not purport to have comprehensively examined the steel substrate, nor did he purport in this respect to extrapolate beyond the areas he examined. That said, and as previously noted above, Dr Garnys acknowledged that evidence of Sigmaweld primer was found in some locations by others.

89 Wattyl referred to Dr Garnys' statement in the Joint Report that he observed the presence of zinc oxidation on each zinc particle and further to Dr Garnys' evidence of observations made using the electron microscope that "[w]e saw on the surfaces of the spherules a grading away from brightness into greyness that was suggestive of surface contamination". Wattyl contended that this evidence effectively confirmed Mr Haig's finding that the contamination of the Galvit primer was a surface contamination and not a contamination throughout the zinc particles which would be suggestive of inherent defects.

90 In my opinion, this overstates the position. Dr Garnys was using an electron microscope in "back scatter" mode which, as he said, enabled the surface of the material to be examined highlighting the heavier elements. On the basis of that examination, Dr Garnys said he saw a grading away from brightness into greyness that was suggestive of surface contamination. Dr Garnys conceded that he did not cross-section the material and did not have evidence as to what was actually within the layer. He said he would have to rely on "other people's evidence" on that issue. However, it does not follow, and Dr Garnys did not concede, that the spherules within the thickness of the Galvit primer were not contaminated by corrosion.

91 Wattyl submitted that Dr Garnys conceded that an examination of the spectra and data provided by Mr Cook of Wattyl at the conclave of experts indicated that there was evidence of no more water in the retained batches of the applied material than there was in the new product. I accept that Dr Garnys made this concession.

92 Wattyl submitted that the four reasons given by Dr Garnys for expressing the opinion that the Wattyl coating system was not of merchantable quality in the CETEC report of December 2010 should not be accepted.

93 Wattyl challenged Dr Garnys' first reason which related to the age, water content, viscosity and drying time of the batch of Galvit primer used on the trusses. With respect to the water content reading in the Galvit Part A batch manufactured on 5 October 2006, Dr Garnys conceded that the water content of 0.03% referred to in the CETEC report was incorrect and that the correct reading for the water content of the batch was 0.01% which was well within the acceptable range. In the circumstances, Dr Garnys accepted that the Dean and Stark water content measurement for the batch could not be relied on to support the statement that there were early signs of deterioration in the Galvit primer in 2006. Nevertheless he maintained that the issue of water content was still a relevant consideration.

94 In relation to the Brookfield viscosity reading of the Galvit EP100 of 24,000 CP (in a range 19,000-24,000 CP) measured in 2006 at the time of manufacture, Dr Garnys noted that the viscosity measurement was right at the limit of the specification. It was suggested to Dr Garnys that the usual procedure in paint manufacture is to manufacture the paint to a higher viscosity than required under the specification so that it can be subsequently "let down" with solvent to the required level. Dr Garnys said he was not aware of the practice of this particular paint manufacturer and what its criteria are. He was not able to comment on whether it was usual procedure in paint manufacture to manufacture paint to a higher viscosity than required under the specification. Mr Eccleston agreed that it was the usual procedure. No evidence was called by Wattyl to say whether it adopted this practice with respect to its paint manufacture generally or with respect to the Galvit primer.

95 As to drying time, Dr Garnys noted that the drying time of one hour was at the limit of specification of 0.00-1.00 an hour. Subsequent information provided during the conclave of experts indicated that, at Wattyl, the drying time is typically reported within 15 minutes. In the circumstances, Dr Garnys said he did not know whether the recorded drying time of one hour means 1.25 or 0.75 of an hour. Dr Garnys maintained his view that, absent the water content figure, there was still a strong indication that the product was at the upper end of its specification at the time of its manufacture. In his opinion, from a quality control point of view, this should have required some attention; that the reading indicated a marginal acceptance of the product and that if he was quality controlling the process he would be concerned about the quality of the product.

96 According to Dr Garnys, another indicator that the product was not of merchantable quality was that the product was old. The CETEC report of December 2010 states that the Wattyl product information (for Galvit EP100) indicated an applied new product life of 12 months. This is not strictly accurate. Under cross-examination, Dr Garnys accepted that the product information for Galvit EP100 states that the product has a shelf life of "at least 12 months". Mr Pellegrino noted, correctly, that the nominated shelf life was not expressed as a maximum and that the product had a "good life expectancy", although he did not say how long that was. Whilst acknowledging that the product life as indicated by Wattyl as being potentially greater than 12 months, Dr Garnys nevertheless gave evidence that for this type of system, extended storage was a likely indicator of problems and from his knowledge of these systems, products aged over 12 months were a concern. Despite challenge, Dr Garnys maintained his position that the age of the product, the high level of viscosity and drying time were indicators that the product was not of merchantable quality.

97 Whilst Dr Garnys' opinion with respect to these indicators is not determinative, it should not, in my view, be rejected, but should be taken into account with the totality of other indicators. Dr Garnys showed a willingness to adjust his opinion as new information came to hand. He was responsive to questioning but firmly maintained his position that the attributes of the Galvit primer (age, viscosity and drying time) were indicative of potential problems and was one of the four reasons listed in the CETEC December 2010 report in support of his opinion that the Wattyl coating system was not of merchantable quality.

98 Wattyl challenged Dr Garnys' second reason, namely his conclusion that the SEM results showed "metallic zinc spherules of 0.2 to 5 micron diameter with high oxygen to carbon... ratios of almost 50% indicating likely oxidation or hydration thus weakening the zinc to resin bond and evidence of poor mixing". Wattyl submits, and I accept, that the conclusion reached in the SEM analysis in the CETEC report is that the underside of the delaminated coating contained Galvit primer and that the delamination occurred within the Galvit primer layer. In addition, the chemical profile of the Galvit primer was analysed by Dr Garnys and was shown to comprise 9% oxygen and 20% carbon giving an oxygen to carbon ratio of almost 50%. The chemical profile of the topside coating (Epinamel EB600) was shown to comprise 10% oxygen and 54% carbon giving an oxygen to carbon ratio of almost 20%. Dr Garnys concluded that the higher oxygen to carbon ratio in the Galvit primer indicated likely oxidation or hydration thus weakening the zinc to resin bond, and was evidence of poor mixing. In my view, this conclusion is problematic in the absence of a reference sample of non-defective Galvit primer, as there is no evidence to suggest that the oxygen to carbon ratio of almost 50% was not a normal attribute of non-defective Galvit primer.

99 Accordingly, I am unable to give this reason any weight.

100 Wattyl submits that Dr Garnys' third reason, namely that the infrared spectroscopy analysis shows evidence of hydration and visible carbonation of the zinc and possibly incomplete cure should not be accepted on two grounds.

101 First, it was submitted that the analytic technique of infrared spectroscopy is unreliable. In the December 2010 CETEC report, Dr Garnys noted that the "peak intensities in the infrared spectra can be unreliable as this parameter is influenced by the physical presentation of the sample in the analysis since the surface reflectance technique has been used." Other experts gave evidence to the same general effect.

102 Mr Eccleston agreed that to do the infrared analysis it is necessary to have a sampler which has a small polished diamond facet of about a millimetre in diameter. The technique involves pressing the sample against the sampler. Mr Eccleston conceded that the small area sampled is liable to provide misleading spectra if there is contaminant on the sample or if the sampler is placed against an unrepresentative area of the sample. Mr Eccleston accepted that the operator would tend to have a look at the spectrum itself to see whether or not it was the expected spectrum and, if not, then the sampler was probably in the wrong spot and re-sampling would occur. Eccleston said, "So you'd have a basic idea of what you're looking for so if you're looking at a epoxy zinc primer and you got a spectrum of zinc oxide you know you've got it in the wrong spot so you'd just move it, essentially." Mr Eccleston agreed that infrared analysis is a bit of a "hit or miss" operation which depends on knowing what one is expecting to see and to match that with whatever sample is being examined. Mr Eccleston agreed that the zinc particles in the paint are opaque to infrared radiation and produce no infrared spectra. Mr Eccleston agreed that the zinc oxide (corrosion products) would provide some spectra. It was suggested to Mr Eccleston that because the zinc was such a high percentage component of the paint in this case, that the spectrum obtained would be weak and noisy. Mr Eccleston said that this would depend on the instrument being used; the sampling time could be extended to account for the noise and there are noise cancelling algorithms built into the instrument. It was suggested that the interpretation of weak or noisy spectra is unreliable and potentially misleading. Mr Eccleston said that the spectrums he saw in this case were not particularly noisy but were fairly good.

103 Mr Haig noted the shortcomings of the infrared analysis technique in his report of 7 July 2011, including the chance of misleading spectra due to contamination or an unrepresentative area in the small field of view of the polished diamond facet, zinc corrosion products forming after sampling, the high percentage of zinc as a component in the paint resulting in a weak and noisy spectra, the possibility that the normal constituents of EP100 may contain hydrates and carbonates as part of the formulation or as normal reaction products and that oxides, hydrates or carbonates might have come from corrosion of the Sigmaweld primer.

104 Having regard to the weight of expert opinion as to the potential unreliability of infrared analysis, the results of the analysis needs to be viewed with a degree of caution. It seems that the reliability of results is likely to be dependent upon the quality of the instrument and the skill and experience of the operator in knowing what to expect and in adjusting the sampling process to ensure that the potential for obtaining misleading spectra is minimised.

105 Second, and more fundamentally, Wattyl submits that Dr Garnys goes beyond the conclusion reached in the CETEC report with regard to hydration and visible carbonation of zinc.

106 In this context, Dr Garnys explained the reasoning underlying his analysis as follows:

the infrared doesn't tell you a lot about the zinc metal at all, it tells you about the zinc oxidation products and tells you about the organic system, if you like, that is there, that was meant to bind the zinc together. Left on their own as a powder obviously the zinc has not [sic] strength, it will come apart and the resin is added to bind all these little spheres together. Obviously the least amount is added to give us the maximum amount of zinc which is there for corrosion protection. When you looked at the ratio of oxygen to carbon in the cohesively failed - in other words the failure within the zinc coating, we saw that the oxygen to carbon ratio was higher in the delaminated area as it was on the top, the top surface, which was the other surface that we had that the - the 600 layer. So we would expect that the resin system which had not dissimilar epoxy resin formulations would have not dissimilar carbon to oxygen ratios when you looked at that component within the layer, as looking at it on the top layer...

...

The bottom line is that we saw that there was more oxidation product within the - within the layer than what would be considered to be normal for a normal western (sic) system which was - the only reference we had at the time of course was the top layer, so we were referencing everything to the top layer. The key reasons for doing this work was (a) to distinguish that the split layer was indeed zinc-rich, which would indicate that it is the EP100 material and not some other material. It was not to generate a comparison for specification, so we were not looking to see whether in fact the material that had been supplied was some way different from the specification, we didn't have that information, we asked for it and we were not supplied with the same batch material, or any batch material which was purported to be the same, so we couldn't do a reference...

107 Both Mr Haig and Mr Cook said, in effect, that Dr Garnys' comparison of high oxygen content in the Galvit primer with a different material was an invalid comparison. Specifically, Mr Cook said that the Galvit primer, a pure epoxy, was a different resinous compound to the Epinamel EB600 which was an epoxy ester. In Mr Cook's opinion, this would not give a valid comparison because it was a comparison between two different systems.

108 Mr Eccleston agreed that it was preferable when trying to draw conclusions about the level of oxidation of the product to compare non-defective Galvit primer on the one hand with "defective" Galvit primer on the other.

109 Obviously it is preferable to compare a sample of like products. Dr Garnys had requested a sample of the same batch of Galvit primer or any batch of Galvit primer purporting to be the same material to make a direct comparison but this was not provided. Of the independent experts only Mr Haig was provided with a sample by Wattyl. In the absence of direct comparison, Dr Garnys made a comparison he was able to make in the circumstances. Dr Garnys' point seemed to be that the systems, though different, were sufficiently similar to make comparison meaningful. However, in my view, there was not sufficient information available to Dr Garnys to make anything other than an educated guess that the difference was significant. Whilst there was agreement that there was a difference, it was not possible on the information available to draw any firm conclusions. Without evidence of the respective properties of the products, it is not possible to know whether a meaningful comparison can be made, despite the differences. In the circumstances, I can give no weight to Dr Garnys' conclusions with regard to the extent or significance of the hydration and visible carbonation of the zinc based on the comparison made, though I do accept Dr Garnys evidence that the infrared spectra shows hydration and possible carbonation. In this regard, Mr Eccleston agreed that the infrared charts show hydration and possible carbonation of the zinc.

110 Wattyl made submissions to the same effect with regard to Dr Garnys' conclusion that the infrared analysis shows "possibly incomplete cure".

111 This conclusion was based on Dr Garnys' assessment that the ratio of the amine to hydrocarbon on the infrared charts of the Galvit primer (EP100) compared with the Epinamel (EB600) were vastly different, suggesting to him that the amine was a "massive component" of the EP100 and above the normal ratio for curing. Dr Garnys said that he did not know whether the amine was a "massive component" of the Galvit primer or whether he would expect the ratio of amine to hydrocarbon to be vastly different in the resin used for the Galvit primer as compared to the top coat. Mr Cook gave evidence that it depends on which hardener was used as some hardeners have a higher hydrocarbon content. Mr Cook said he was not familiar with the formulations and that he believed the formulation of hardeners were different.

112 Mr Eccleston, who routinely carried out infrared analysis, was unable to determine from the infrared charts how they showed the possibility of incomplete cure.

113 In my opinion, Dr Garnys' evidence that the infrared charts showed the possibility of incomplete cure is of no probative value. Without information about the hydrocarbon content of the hardener used on the Galvit primer which might account for the different ratios observed by Dr Garnys, I cannot be satisfied that the infrared charts reveal the possibility of incomplete cure.

114 Wattyl submitted that the disbonded surfaces upon which Dr Garnys carried out his analysis could have suffered from oxidation, hydration and carbonation between the time of disbondment and the time of analysis. It is noted that the failures occurred from about January and February 2009 and the truss samples were collected in July 2009.

115 Dr Garnys conceded these possibilities; in particular, he said this could have occurred in relation to some of the truss samples, namely where the paint had peeled away but not where the material was "fresh when it was peeled" and not in the case of handrails where there was rapid delamination and no evidence of crusting or visible amounts of zinc carbonate and zinc oxide on the surface.

116 Mr Eccleston also accepted that it was possible that material corroded between the time it failed cohesively and the time it was examined. Mr Eccleston accepted the proposition that in order to draw conclusions about the oxidation, the hydration or the carbonation of a sample, it was necessary to note the state of the sample at the time of disbondment and compare it to the state of the sample at the time of analysis.

117 Doubtless some corrosion products may have been produced after delamination where the Galvit primer was exposed to the atmosphere for several months from the time of delamination until the time the sample was taken. However, Dr Garnys said in many cases there was no evidence of a break in the coating.

118 Clearly, there was potential in this case for the Galvit primer to have suffered some corrosion during the period between when it failed and when it was examined by reason of exposure to the atmosphere. However, I accept Dr Garnys' assessment that this would not apply to the "fresh" material from the trusses nor to the material used in the handrails which rapidly delaminated.

119 Wattyl submits that Dr Garnys' fourth reason, namely that the coating thickness measurements showed the Galvit primer coating thicknesses within the 75 microns specification of 58-83 microns and a very uniform delaminated coating of 45 microns on the underside of the epoxy second coat, should be disregarded. Dr Garnys said this indicates a uniform soft delamination irrespective of total coating thickness from 320 to 628 microns. Dr Garnys found that recent coats could be easily peeled apart even without delamination, indicating incomplete cure well beyond the specified times and that this was not characteristic of brittle shear due to varying stresses from top coats of varying thicknesses. Wattyl submits that Dr Garnys' conclusion should be rejected because it is unknown who carried out the measurements and the measurements were carried out only at four locations on the handrails (and not on the trusses), yet Dr Garnys asserts that an identical delamination phenomenon had occurred on the trusses and handrails. Wattyl submits that this is purely speculative. Wattyl also submits that Dr Garnys has no practical experience with coating applications and cannot comment on the mode of application or whether the application was properly done. In substance, it was submitted that Dr Garnys' investigation was so limited in its scope that he cannot convincingly rule out application as the cause of delamination from the incomplete cure he allegedly saw.

120 Dr Garnys did not carry out the measurements himself and could not positively say whether the measurement was performed by Dr Schiller or by someone at DH Corrosion. Dr Garnys inferred that his colleague, Dr Schiller, made the measurements. I accept that this is likely. The discussion about coating thickness measurements indicates that Dr Schiller returned to the premises of DH Corrosion on 7 September 2009 to inspect the steel handrail panels with the aim to determine the film thickness of the Galvit primer that had been applied to the substrate. Measurements were made using a PosiTector 6000 coating thickness gauge supplied by DH Corrosion.

121 This tends to suggest that Dr Schiller was supplied with a PosiTector 6000 coating thickness gauge to carry out coating thickness measurements at the time of his inspection. On being questioned whether the delamination of the coating on the trusses was identical to the delamination of the coating on the handrails, Dr Garnys agreed that it could not be said by measurement that they were the same but said that from field observations they were "visually identical".

122 Dr Garnys was the only expert in a position to say that the delamination phenomenon on the trusses appeared to be identical to the delamination phenomenon on the handrails. Mr Eccleston and Mr Dromgool only examined the trusses. Mr Haig and Mr Cook did not attend the site and only examined paint samples.

123 I accept Dr Garnys' evidence that the delamination phenomenon on the trusses and the handrails appeared to be identical. As previously noted, Dr Garnys did not profess to have practical experience with coating applications so was not able to comment in depth on matters concerning application. It is correct to say his investigation was therefore somewhat limited. However, the criticism which was levelled at Dr Garnys to the effect that he "cannot convincingly rule out application as the cause of the delamination" is, in my opinion, misdirected. Dr Garnys did not purport to rule out application as the cause of delamination. His conclusion was that "there seems to be a low probability that coating failure has arisen through an issue of application". Whether the coating failure has arisen through an issue with application will be dealt with later in these reasons.

124 It is convenient here to deal with a discussion which occurred between the experts at trial in relation to calcium oxide. Wattyl submits that the discussion regarding calcium oxide is a "red herring" which illustrates that Dr Garnys lacked expertise and understanding of the zinc-rich primer product. The discussion about calcium oxide took place in the context of a consideration of the constituent parts of the Galvit Part A resin. Dr Garnys was asked by counsel about the properties of calcium oxide. Dr Garnys gave evidence that calcium oxide is a powerful water scavenger which has a dual function; to take away free water (either water globules or water vapour) and combined water (water which is chemically attached in some way to one or more of the chemicals in the mixture). Dr Garnys was then asked about the effect on combined water of calcium oxide. Dr Garnys said that "when calcium oxide is hydrated, it forms calcium hydroxide" which "is very alkaline". According to Dr Garnys, although hard to measure, calcium oxide creates an alkaline environment and is corrosive to metals, for example, zinc. The more calcium oxide in the mixture and the more finely it is dispersed throughout the mixture, then the more residual alkalinity there is in the formulation. Dr Garnys gave evidence that if there is less calcium oxide, then, assuming a certain amount of water, either combined or free, there would be more residual water in the formulation. Dr Garnys said that if there is too much calcium oxide in a formulation, the calcium oxide will "abstract as much water as it can from the system, the free water and some of the combined water, and will also be there to have capacity for further water absorption in the future." Dr Garnys gave evidence that zinc is an amphoteric metal, which dissolves in acid and alkali. In this case, Dr Garnys said that "playing around with the alkalinity of the system renders the zinc vulnerable to further reaction, further oxidation".

125 Dr Garnys gave evidence as to the possible presence of combined water in the formulation of Galvit Primer Part A; that one component was the calcium oxide which was added to the mixture to combine with free water and bentonite clay (another component in the formulation) where water can sit between plates of clay, "which you could almost consider that to be combined because it is not readily available". According to Dr Garnys, the "combined" water from bentonite clay can be relatively easily released under moderate temperatures, but under normal laboratory conditions it would not be released in the test conducted at the time of manufacture for the presence of free water. Dr Garnys was asked what possible effect combined water attaching to each of those compounds could have on the quality of the batch under discussion. Dr Garnys said:

Well if the water was so combined if you like that it could not be further released for any other reaction then that would, if you like, take it out of circulation and not be available for other reactions. However, in the combination with water things, compounds like calcium becoming... calcium hydroxide, would become present. That starts to change the characteristics; it swells, we heard that it was extremely finely distributed throughout the product, so it would be in contact with the zinc, so that if you like alkaline product, and that's still alkaline when it's hydrated, it's not as alkaline as lime but still alkaline, could attack zinc... It would affect the integrity of the zinc keeping in mind that zinc is very small spirials [spherules] and large surface area that reaction would be relatively rapid... the zinc itself is zinc metal so if we can imagine a sphere of zinc metal, if it's attacked by oxygen or moisture or alkaline chemicals or even acids there are deposits that are built up on the surface. So whatever was... bonding to that metal would potentially be disbonded, or whatever is trying to bond to that metal sphere would not - would have to go through the corrosion products to bond and it's a bit like putting a coating over a rusted iron. The integrity of that bond strength would be compromised.

126 Dr Garnys was asked what the consequences would be of combined water being released from the bentonite clay or alternatively if calcium oxide caused an alkaline effect on the zinc. Dr Garnys gave evidence that the ultimate [effect] would be a weakening of the layer and it would also potentially affect the Part B by weakening it as well. Dr Garnys said, "in layman's terms you would get low cohesive strength".

127 Wattyl submitted that Dr Garnys' evidence about the effect of combined water and calcium oxide and bentonite clay and their effect on the formulation of Galvit Primer Part A is pure speculation and that there was no evidence to support a suggestion that this in fact occurred to the Galvit Primer Part A in this case. Further, it was submitted that Dr Garnys' speculation as to the effect of calcium oxide on the formulation is wrong.

128 Mr Haig denied that if too much calcium oxide was added to the mixture, the effect would be that the formulation could become alkaline and an alkaline coating would then form around the zinc particles and have a degrading effect on the zinc particles. Mr Haig gave evidence that without liquid water there cannot be alkalinity, that the presence of combined water in the calcium could not become alkaline and could not degrade the zinc.

129 Mr Cook also denied that if you put too much calcium oxide into the formulation, it will become alkaline when it absorbs water. According to Mr Cook, alkalinity will only exist if there is water present. Mr Cook disagreed with Dr Garnys' evidence that bentonite clay would release water and with Dr Garnys' conclusion that there would be a weakening of the layer of Galvit primer and that there might also be a weakening of the Part B used in the Galvit primer.

130 Mr Eccleston agreed that calcium oxide in the mixture would attract only as much water as is present in the mixture and that there would not necessarily be a damaging chemical imbalance in the mixture. When asked whether if there were more calcium oxide than free water in the mixture, he would accept that there would not necessarily be an alkaline coating on the zinc particles in the mixture, he replied, "There might just be calcium oxide surrounding the surface ... there might be a layer of calcium oxide around the surface of the zinc itself". Mr Eccleston also accepted that unless calcium oxide is dissolved in water, it would not have an alkaline effect on the mixture and would not act to corrode the zinc particles.

131 It was also submitted that Dr Garnys contradicted his own evidence on the corrosive effect of calcium oxide by acknowledging, "If there's no free water then it [the calcium oxide] will just rest if you like without reaction". It was submitted that by this acknowledgment, Dr Garnys' evidence is in fact consistent with the evidence given by the other experts about excess calcium oxide and serves to highlight the unreliability of his evidence.

132 At times there was a frank exchange of views between experts about the effect of combined water in calcium oxide and the combined water component of bentonite clay and the capacity of the combined water to bring about oxidation of the zinc particles. Insofar as calcium oxide is concerned, the weight of expert opinion favoured the view that adding too little calcium oxide to the mixture would have the consequence that there would be a risk of degradation of the zinc particles due to a higher amount of free water available in the mixture. Adding an excess of calcium oxide would scavenge all the available free water, which would become combined water in the calcium oxide. The other experts disagreed with Dr Garnys' proposition that adding too much calcium oxide would have the consequence that the mixture could become alkaline and that an alkaline coating would form around the zinc particles, having a degrading effect on them.

133 As for the combined water in the bentonite clay, only one expert, Mr Cook, expressly disagreed with Dr Garnys' evidence about bentonite clay releasing combined water and ultimately having the effect of the weakening of the Galvit primer.

134 Unfortunately, there is no evidence before the Court as to how much calcium oxide was put into the mixture, or as to the level of humidity at the time the product was manufactured, which could form the basis for accurately testing or at least modelling how the primer mixture might be expected to behave in the field.

135 Wattyl withheld documents relevant to this issue. It failed to discover the "batch tickets" for the relevant product batches which would, in all likelihood, have recorded this information. Accordingly, the experts were left to discuss the issue on a hypothetical level only. In the circumstances, the discussion is of no assistance in determining the likely cause of the failure in this case. In any event I am not persuaded that the expression of a different opinion by Dr Garnys in the discussion about calcium oxide illustrates a lack of expertise or understanding.

136 Mr Eccleston, Senior Materials Scientist from CSIRO, carried out an investigation of the coating failure on the trusses in July 2009. He concluded that the cohesive failure of the Galvit primer occurred because the primer coat had not sufficiently cured to achieve the tensile strength necessary to withstand the curing stress of the intermediate and top coats.

137 Mr Eccleston's investigation commenced with an on-site inspection of the trusses prior to their erection. Trusses A and B had been bolted together and the concrete flooring was in place. Mr Eccleston states that the inspection showed that the paint was delaminating, revealing a layer of zinc on the steel work and on the underside of the removed coating. The coating had been previously inspected revealing the adhesion was quite poor. In some areas the paint had peeled away in large areas.

138 At three locations Mr Eccleston checked the coating thickness using a film thickness gauge and using a method prescribed for the determination of dry film thickness under Australian Standard AS3894.3. He checked the coating for adhesion in those locations using Australian Standard AS1580 Method 408.2 which involves scribing a cross with a knife and gentle prising of the paint. The adhesion was rated according to the amount of paint removed ranging from 1 for no paint removal through to 5 for removal of the paint past the cuts. The inspection revealed that the total dry film thickness ranged from 292 microns to 516 microns. The adhesion in the areas inspected was quite poor with a minimum rating of 4. Samples of the delaminated coating were taken for laboratory examination.

139 At the laboratory the samples of paintwork removed from the site were examined by Mr Eccleston for pencil hardness and degree of cure.

140 Mr Eccleston found that delamination of the coating was occurring irrespective of the coating thickness. The coating had a pencil hardness of less than 8H and zinc was deposited on a cloth soaked in methylethylketone (MEK) when the test area was subjected to 10 double rubs when testing for degree of cure. These tests indicated that the epoxy zinc primer had not fully cured.

141 Mr Eccleston states that the fact that the zinc was adhering to the steel substrate is an indication that the surface preparation was satisfactory as inadequate surface preparation would have resulted in adhesion loss between the steel substrate and the Galvit primer. Adhesion failure was not evident to Mr Eccleston at the site inspection or on any samples removed for laboratory examination. Mr Eccleston noted that adhesion failure can occur if the coating is applied at less than 3°C above the dew point, since moisture can condense on the surface of the metal and can lead to delamination of the coating. Mr Eccleston states that there was no sign of adhesion failure of the epoxy zinc primer to the steel substrate and that the inspection records supplied to Mr Eccleston indicated that the coating conditions during application were within the recommended application range.

142 Mr Eccleston reported that in some of the areas inspected, the coating thickness was well in excess of the 375 microns thickness specified, whilst in other areas the coating thickness was well within the specified range. With one of the samples, the dry film thickness of the primer exceeded the 75 microns specified. Mr Eccleston states that the coating inspection records indicate that the coating was initially applied at the correct thickness. Mr Eccleston found that the delamination was occurring in both the low and high thickness areas and was not dependent on the total coating thickness.

143 Referring to Mr Dove's report, Mr Eccleston gave evidence that Mr Dove had found that adhesion was "quite poor in both low total thickness and high total thickness which is what I found". Further, Mr Eccleston noted that Mr Dove found that where the zinc primer was within specification (between 74-80 microns) the adhesion was also quite poor.

144 Mr Eccleston agreed with Mr Dove's conclusion that the weakness in the coating was within the epoxy zinc layer. As to Mr Dove's opinion that "this could be possibly due to the excess film thickness causing solvent to be trapped in the intermediate coat as it cures on the surface and affecting the epoxy zinc", Mr Eccleston said he had looked at a cross-section of the coating that he took from the site and saw no evidence of solvent entrapment. He also said there would be a smell associated with the solvent but there was no evidence of that either.

145 Wattyl does not challenge the validity of Mr Eccleston's evidence or his testing. Rather, it is submitted that his evidence is of limited effect and that elimination of application as the cause of failure could only apply at the test locations on the trusses and not to the trusses overall. Wattyl contrasts Mr Eccleston's testing at three locations in June 2009 with Mr Stone's comprehensive survey of dry film thicknesses of the coating on the trusses set out in his reports.

146 In its submissions, Wattyl emphasises the fact that Mr Stone took hundreds of dry film thickness readings and there were conspicuously high dry film thickness readings found on each of the various parts of the trusses.

147 I accept that Mr Stone carried out a comprehensive survey of dry film thicknesses on the trusses. As a coating inspector this was his task. It was not Mr Eccleston's role as a materials scientist to do that kind of work. However, the fact that Mr Eccleston did not carry out such a survey does not negate Mr Eccleston's evidence. He did not set out to survey the film thickness. He set out to ascertain the probable cause of failure. Mr Eccleston was alive to the possibility that excessive dry film thickness might play a part in the failure of the zinc-rich primer and accordingly checked the dry film thicknesses of his samples to see whether the failure in the Galvit primer correlated with the dry film thickness of the coating. He found that the coating failed irrespective of dry film thickness.

148 This result is not inconsistent with the results of Mr Stone who, for example, found that average dry film thickness

• on the north truss (west side internal) of 486 microns exceeded the specified thickness[14] with the lowest reading of 267 microns and the highest reading of 1345 microns;

• on the north truss (east side internal) of 321 microns was below the specified thickness with the lowest reading of 109 microns and the highest reading of 865 microns;

• on the south truss (Mr Stone combined readings on the east with readings on the west side) of 456 microns exceeded the specification with a lowest reading of 155 microns and highest reading of 910 microns.

Mr Stone found the dry film thickness to be inconsistent, with the majority of the dry film thickness readings exceeding 350 microns. On the underside of both trusses, dry film thickness readings showed film thicknesses to vary considerably with some areas showing low film build. In his first report, Mr Stone reports that the coating had disbonded at various locations on both trusses.

149 Mr Stone's second report contains an even more comprehensive survey of dry film thicknesses on the trusses. After a two-day inspection, Mr Stone found that the coating on the bridge structure exceeded the film thickness beyond the manufacturer's recommendations to 50% of the bridge and had underfilm thickness to approximately 40% of the bridge. At no point in these reports does Mr Stone correlate the areas of failure of the coating on the trusses to areas which had excessive coating thickness.

150 In my view, it does not follow that Mr Eccleston's elimination of application as a cause of failure could only apply to the three locations he inspected on the trusses. More general inferences may be able to be drawn depending on the totality of the evidence and the opinions of other experts.

151 It is true that Mr Eccleston acknowledged that cure problems could be caused by the over-building of coatings or as a result of poor mixing during application.

152 In relation to over-building of coatings, it is clear that Mr Eccleston was aware of this possibility at the time of sampling and at the time he conducted the tests. Some credence must be given to Mr Eccleston's considerable expertise as a materials scientist with the CSIRO. Part of that expertise is to determine the number of samples necessary to draw conclusions. Mr Eccleston's expertise was not challenged, nor were the results of the pencil hardness test and the degree of cure test.

153 Wattyl submits that the Court cannot be satisfied that Mr Eccleston considered any of the following factors adequately in order to properly discredit them or credit them as causes of or contributors to the failure. The factors were solvent retention, contamination of the substrate, applying coatings in cold weather or in damp weather, not stirring the pot properly, exceeding pot life and dry spray overcoating.

154 Wattyl submitted that the position when the trusses initially came to site in January 2009 might have been different to when he did his report in May 2009. Mr Eccleston accepted this was a possibility and admitted that he was not familiar with the detail of the work that was undertaken on the trusses between January 2009 and the end of May 2009.

155 In my view, it is unlikely that anything turns on this timing difference. Samples of delaminated coating were taken from areas showing distress in May 2009. Whether samples were taken from areas which had been previously repaired and had again failed or whether samples were taken from areas on the trusses which had not previously failed probably doesn't matter. In each case, there was a failure. In each case, the failure was in the Galvit primer layer. The failure seemed to be a failure of the same character. As previously mentioned, Mr Eccleston had looked at a cross-section of the coating taken from the site and saw no evidence of solvent entrapment either visually or by reason of the smell which he said would be associated with solvent entrapment.[15] Mr Eccleston also addressed Mr Bartlett's comments in Mr Bartlett's report with respect to solvent retention. I accept Mr Eccleston's evidence that Mr Bartlett had speculated on solvent retention as a possible mechanism for the failure but there was no evidence to suggest that Mr Bartlett had observed on site that this was the problem.

156 In relation to contamination of the substrate, Mr Eccleston gave evidence that he saw no evidence of the presence of any pre-weld primer on the substrate. Wattyl submits that the evidence clearly points to the presence of Sigmaweld and submits that the contamination of the substrate by the presence of Sigmaweld is a likely and probable cause of the delamination which was not adequately considered by Mr Eccleston. In summary while Mr Eccleston did investigate contamination of the substrate, I accept that the most comprehensive analysis of adhesion issues on the trusses relating to Sigmaweld was carried out by Mr Stone. This issue is addressed later in these reasons in response to question 4. In the circumstances I give greater weight in relation to the contamination of the substrate to Mr Stone's survey results and confine Mr Eccleston's evidence on this issue to the areas on the trusses that he examined.

157 In relation to applying coatings in cold or damp weather, Mr Eccleston reported that the Cox "inspection records indicate that the coating condition [sic] during application were within the recommended application range".

158 Mr Eccleston agreed that he assumed that the inspection records were correct and had accepted them at face value. Wattyl submits that these quality assurance records show an "inexplicable error" in the recorded average micron depth of the second coat.[16] When Mr Eccleston was asked whether he could explain the error in the inspection records, he suggested that it was likely to be a transposition error with the average figure having been incorrectly placed where the minimum figure should have been. I accept this is the likely explanation. When this error was brought to Mr Cooper's attention in cross-examination, he immediately acknowledged it was a mistake. It was then suggested to Mr Cooper that in order to get the job done quickly he simply wrote down figures which were within an acceptable range in order to justify the work continuing. Mr Cooper emphatically denied this.

159 Wattyl relied on the Bureau of Meteorology records for Laverton RAAF for December 2008 which show relative humidity of 99% at 9.00am and 95% at 3.00pm on 13 December 2008, whereas the Cox quality assurance records record relative humidity at those times at 61.5% and 60.8% respectively. Mr Eccleston was directed to these records, and in particular to the fact that the relative humidity readings on 13 December 2008 at 9.00am and 3.00pm are substantially higher than the relative humidity readings recorded for Cox's premises at those times. Mr Eccleston agreed that he did not undertake any auditing process to compare the records with temperature records of the weather bureau.

160 In my opinion, nothing useful would have been achieved had he done so. The Bureau of Meteorology data includes relative humidity readings taken (apparently) outside (apparently) at Laverton RAAF base. The trusses were coated inside Cox's premises. The relative humidity readings taken by Mr Cooper were of the internal atmospheric conditions within those premises at specific times and on the relevant days show that the relative humidity within Cox's premises remained in a narrow range throughout the day of between 60.8% and 62% in the case of one truss section and between 50% and 53.6% in the case of the other.

161 During the trial, it was suggested to Mr Cooper that the relative humidity readings he had recorded in the Cox inspection records did not accurately reflect the humidity on that day and were incorrect. Mr Cooper disagreed. Mr Cooper gave evidence that the instrument that he used to ascertain the relative humidity and the steel temperature was a compound instrument with a probe to put on the steel to register the steel temperature. In relation to the relative humidity, Mr Cooper said:

The instrument gives me the relative humidity at that particular place when I do my tests. ... I can walk 20 metres with the instrument and get a totally completely different reading in 20 metres.

162 I accept Mr Cooper's evidence. There was no evidence that Mr Cooper falsified the Cox inspection reports. Mr Cooper was not a principal or employee of Cox. He was a contractor engaged as Cox's consultant coating inspector and there is no reason to suppose he was not doing his job diligently.

163 In the circumstances, I accept the accuracy of Mr Cooper's relative humidity readings taken on 13 December 2008. In my opinion, no meaningful comparison can be made of the relative humidity readings taken inside Cox's premises with the relative humidity readings taken outside at a different location.

164 Mr Eccleston did not specifically verify whether the paint had been adequately mixed. He did not see the coating being applied. However, he was aware that Cox were experienced coating specialists and presumably knew how to mix paint. Mr Dromgool saw no evidence of poor mixing.

165 Mr Cooper was dismissive of any suggestion of poor mixing, noting there was no likelihood of error in mixing the proportions in this case because the correct proportions were produced after one can of Part A was added to one can of Part B.

166 In relation to excessive pot life, Mr Eccleston thought it would be very difficult to apply the coating in such circumstances. He thought that Cox would not have got to the stage where they applied the coating if the pot life had been exceeded. None of the other experts challenged or contradicted this evidence.

167 In relation to the phenomenon of dry spray, Mr Eccleston conceded that he had not investigated how the coating of the trusses had been physically done.

168 Finally, in relation to Mr Eccleston's evidence, Wattyl submits that Mr Eccleston's conclusion that the Galvit primer may have been past its shelf life when it was applied and had resulted in the incomplete curing of the primer was not based on tests to establish whether the primer had developed defects. When asked at trial, "Did you do any tests yourself to establish whether this primer had developed defects?", Mr Eccleston replied, "No, the wet material was no longer available so that's as far as I could proceed with my investigation." At trial it became apparent that the wet material was available because Wattyl retained a sample of each batch manufactured but had refused to supply the plaintiff with a sample for testing.

169 In the circumstances, Mr Eccleston's conclusion that the Galvit primer may have been past its shelf life is based on his experience as a scientist rather than the results of comparative testing. Nevertheless, it is clear from tests which were conducted by Mr Eccleston that the Galvit primer layer had failed to cure well outside the full 5-day cure period stated in the manufacturer's product information sheets for Galvit EP100.

170 Wattyl's expert Mr Haig gave evidence that he examined the Galvit primer on sample flakes taken from the trusses and handrails and on test panels sprayed both from retained samples and freshly manufactured Galvit primer.

171 In the Joint Report, Mr Haig states that his examination produced identical laboratory results from which he concludes that the Galvit primer used on the footbridge project was free from inherent defects. According to Mr Haig, using infrared spectroscopy, optical microscopy and scanning electron microscopy produced results which showed consistent values for oxygen content throughout the samples, commensurate with the levels of oxygen that he would have expected. In addition, Mr Haig said the spectroscopy results indicated no abnormal material in the nature of contaminants which would indicate an inherent defect in the product.

172 Further, Mr Haig said that by examining the interior of the flakes of the Galvit primer he was able to establish that there was no evidence of inconsistency through the material and no evidence of contamination through the material.

173 However, Mr Haig did observe corrosion products in some of the paint film samples that had been taken from the trusses and the handrails and supplied to him for examination.

174 In his report of 7 July 2011, Mr Haig examines these corrosion products on a truss sample flake and states:

The presence of zinc hydroxycarbonate is clearly seen. Zinc corrosion proceeds through oxidation in wet environments to yield zinc oxide and zinc hydroxide. These slowly react with carbon dioxide in the air to produce zinc hydroxycarbonate (ZHC). The importance of this is that ZHC takes months to form. It would not have formed after disbondment of the flakes and would not form whilst in a paint vehicle (wet component). It must have formed prior to application of the Galvit EP100, i.e. formed by corrosion of the zinc-rich Sigmaweld MC and therefore represents contamination painted over by the Galvit EP100.[17]

175 In the Joint Report, Mr Haig expresses the opinion that the corrosion products developed after application and therefore do not constitute an inherent defect in the Galvit primer. In relation to zinc corrosion products that Mr Haig found in some of the paint film samples, Mr Haig said:

The nature of these corrosion products is that they were present as isolated plaques or deposits, which in my opinion can only have formed upon a surface of cured or partially cured EP100, and are not at all indicative of contamination of the product whereby the zinc corrosion products would be intimately mixed through the whole of the material. [18]

176 It was put to Mr Haig that his evidence that the corrosion products "must have formed prior to the application of the Galvit EP100" contradicted his evidence that the corrosion products "can only have formed upon a surface of cured or partially cured" Galvit primer. Mr Haig denied this. He said there were circumstances where both could be satisfied logically which he elaborated upon in the following terms:

the particular circumstance I refer to and which is supported by the observations that I've made would be where a layer of the Galvit EP100 has been applied perhaps to one side of an element of the fence, or to one part of a truss element, and then left overnight. Subsequently there has been contamination by water splash, or dew, or such like, and that the corrosion develops on that layer. In that case you're talking about 'after application' of the first layer of Galvit. Subsequently the painter comes back the next day and paints over that and the inference there is that it has been 'before application'. In effect, the corrosion products have been trapped between two layers of Galvit EP100 which is what my laboratory examination has shown me.

177 Whilst I accept that in theory corrosion products could form between layers of Galvit primer where there are multiple applications, I find that this is an unlikely explanation for the presence of corrosion products in the circumstances of this case.

178 By way of illustration, Mr Haig gave evidence as to the presence of zinc corrosion products in a sample flake taken from the truss sections which he said shows corrosion products "between layers" of Galvit primer. He suggested that a time difference of at least one hour would be needed to ensure that the second layer would not readily amalgamate into the first layer. As previously noted, the observation about multiple layers is problematic. Mr Eccleston disagreed that there were two layers of Galvit primer shown in the SEM images of the sample paint flake taken from the trusses. The lay evidence of Mr Cooper, which was corroborated by contemporaneous records, established that only one coat of Galvit primer was applied to the trusses. There was no evidence to suggest that a second layer of Galvit primer was applied to a first layer at least one hour after the first layer had been applied. Mr Eccleston agreed that there are corrosion products, but that they are within a split layer of Galvit EP100. Mr Eccleston examined Mr Haig's SEM images (Figure 8 and Figure 9) and described what he saw as "a splitting of the zinc itself. It's an irregular parting of the EP100". I do not accept Mr Haig's opinion that these SEM images show multiple layers of Galvit primer with corrosion products in between the delaminated layers. I prefer Mr Eccleston's opinion. In my view, on the balance of probabilities, the corrosion products formed within a single layer of Galvit primer.

179 Further, Mr Haig's suggestion that a layer of Galvit EP100 may have been applied to one part of the truss and then left overnight, allowing for contamination by water splash or dew before the painter returned to apply a second layer of Galvit EP100 has no correspondence with the facts in this case. The inspection records show that each truss section was coated with a single coat of Galvit EP100 during the course of a single day where the ambient substrate temperatures were well above the dew point. In the circumstances, I find Mr Haig's explanation that corrosion products could have formed between layers of Galvit EP100 primer applied at different times, unconvincing.

180 Further doubt was cast on Mr Haig's analysis when in cross-examination it became apparent that Mr Haig was not aware whether the sample of "old primer" and the sample of "new primer" had been made to the same formulation. Mr Haig was supplied with a sample batch of Galvit primer applied to a metal test panel. It emerged during the trial that Wattyl had changed the formula for the manufacture of the Galvit primer on two occasions following application to the footbridge components, once on 30 July 2009 and again on 29 September 2009. It follows that it is not possible to know whether Mr Haig was comparing like with like. Mr Haig said in effect that he didn't know whether he had the new or the old formulation, that this was not the purpose of getting a sample; that he is interested in what contaminants and what inherent defects he could observe in the samples he was supplied with and didn't need to know the ingredients. However, in Mr Haig's second report, it is clear that Mr Haig has used a sample of "new" and purportedly non-defective EP100 as a reference to indicate what the Galvit EP100 product should look like.

181 In his report, Mr Haig set out in a table the list of elements present and their approximate amounts. The analysis is questionable as the reference sample appears to be Galvit primer made to the new formulation.[19] What remains is a comparison between retained samples of the product used on the trusses by Cox and on the handrails by DH Corrosion and paint from samples taken from the trusses. I accept the submission of Fulton Hogan that in light of the changes made to the formulation of the primer revealed in the later discovered documents from Wattyl, it is evident that any comparison of paint flakes from the trusses and handrails to a new batch of Galvit primer would be a comparison of product made to different formulations. Further, it was submitted by Fulton Hogan that the utility of any comparison between retained paint samples and paint flakes used on the trusses and the handrails would be doubtful, because if the control sample is defective paint, any comparison between retained paint would naturally be consistent with paint used on the trusses and handrails.

182 Wattyl submitted that tests done using a new non-defective sample of Galvit primer as a benchmark show that all the samples of paint flakes and paint stock retained by Cox and DH Corrosion contained no extra oxygen to cause concern about the integrity of the product used on the project.

183 I accept Fulton Hogan's submission that tests performed using "new" Galvit primer as a benchmark are of questionable utility as the formulation of the Galvit had been changed. Wattyl failed to adduce any evidence to say what the effect of the changes to the formulation were, why the changes were made and whether despite the changes, a comparison with the old formulation may have validity. In the circumstances, it cannot be assumed that the "new" Galvit primer can provide a proper benchmark for the purpose. There was nothing in Mr Haig's second report which disclosed that the "new" Galvit was made to a different formulation. It seems that Wattyl failed to disclose this to Mr Haig, as Mr Haig gave evidence that he did not know whether he had the later formulation or the earlier formulation. Whilst this is not the fault of Mr Haig, I regard this as a significant omission which reflects adversely on the reliability of Mr Haig's second report. It should have been disclosed. Further, Wattyl's suggestion that Dr Garnys should have purchased a can of Galvit primer to use as a reference sample was disingenuous in circumstances where Wattyl knew that the formulation had been changed twice.

184 Mr Haig also made a comparison between Galvit retained stock and the samples of the paint used on the project. I accept Fulton Hogan's submission that if the retained stock of Galvit primer is defective this will amount to a comparison of wet defective primer product with dry defective primer product which would be expected to provide similar results. In the circumstances, it does not help determine the ultimate question.

185 Mr Haig's evidence is problematic for another reason. He was engaged by Wattyl to give evidence as an independent expert, but throughout the course of the ten day period during the trial when the experts were giving evidence concurrently, it became clear that Mr Haig had firmly aligned himself to Wattyl's case. At times he seemed to be an advocate for Wattyl. This attitude was, to some extent, signalled in his first report in which he exposed a tendency to use unmeasured language in his critique of Dr Garnys' work,[20] but it became obvious by his demeanour in Court. Mr Haig appeared to be aggressively opposed to Dr Garnys. This was in contrast with the demeanour of Mr Dromgool, the other independent witness retained by Wattyl, who maintained his composure and demeanour as an independent expert. In my opinion, Mr Haig had lost sight of the fact that he had an overriding duty to assist the Court impartially on matters relevant to his expertise and that he was not retained to resist the primary case of Fulton Hogan (of product defect) nor to advance the case of Wattyl (of application failure).

186 Mr Haig resisted any suggestion which might support the case for material defect. When shown documents that indicated that Wattyl had withdrawn a batch of Galvit Primer Part B, the following exchange occurred:

COUNSEL: "Well doesn't that suggest that there might be something wrong with the product?"

MR HAIG: "It suggests a responsible attitude on the part of the manufacturer".

187 Mr Haig's first impulse was to defend Wattyl's position. At other times, Mr Haig was non-responsive and evasive. He was frequently dogmatic, tending to express himself in absolute, rather than moderate, terms. In my assessment, he was insufficiently disengaged from the interests of his client. Accordingly, I have given less weight to his evidence than would otherwise be the case.

188 In my view, Mr Haig's conclusion that because he observed no inconsistencies in the material there was no defect in the product cannot be determinative of the question whether there was a defect in the Wattyl product. Ultimately, Mr Haig's evidence left unreconciled the presence of corrosion products within a single layer of Galvit primer which had failed.[21] If these corrosion products were not sourced from contamination on the substrate at the time of application (as Mr Haig originally suggested) or from contamination on the first of two layers before a second layer was applied (as Mr Haig subsequently suggested) it is likely they were sourced from within the Galvit primer layer itself. And if that is so, it suggests that some attribute of the product caused or enabled the corrosion products to form within the layer.

189 Mr Cook was employed by Wattyl as the supervisor of Wattyl's analytical laboratory. In the Joint Report, Mr Cook states that his analysis showed there was no detectable variations between the Galvit primer from the flakes, the retained samples and the more recent product. He concluded that there appeared to be no chemical variation between the cured films that he examined for the purposes of his analysis nor any foreign contamination present.

190 For the purpose of conducting his tests, Mr Cook was provided with paint flakes from the trusses, paint flakes from the handrails, a steel panel coated with Galvit primer mixed from the parts A and B used by Cox on the truss and a steel panel with the Galvit primer mixed by the parts A and B used by DH Corrosion on the handrails and four liquid samples (comprising the part A used by Cox, part B used by DH Corrosion, a retained sample of Part A designated as "New A" and a newly formulated part B).[22]

191 Spectrograms were obtained from scrapings from the Galvit primer used by Cox on the trusses and from the Galvit primer used by DH Corrosion on the handrails and laboratory cured film on metal plates. Mr Cook concluded that all spectrograms appeared similar to each other. Spectrograms were also obtained with respect to Part A used by Cox on the trusses and new Part A and were observed to be similar to each other. The spectrograms obtained of the Part B used by DH Corrosion and a new Part B were compared and found to be similar to each other.

192 Spectrograms obtained of cured films of the liquid samples were observed by Mr Cook to be similar to each other. Spectrograms were also obtained from scrapings from the Galvit primer on the flakes from Cox (trusses) and DH Corrosion (handrails) which were compared to each other and to laboratory preparations and were found to be similar to each other.

193 Mr Cook concluded that the analysis indicated that the current and older liquid samples appear to be consistent in chemical composition and that the analysis of the dried films prepared in the laboratory and those taken from the supplied flakes also appear similar in composition. This suggested to Mr Cook that there are no chemical variations in the cured films or any foreign contamination present.

194 Fulton Hogan submitted that Mr Cook lacked the independence and candour expected of a reliable expert evidence.

195 Mr Cook certainly lacked independence. He was employed by Wattyl and the tests he undertook for the purpose of this proceeding were done in Wattyl's laboratory in the course of his employment. It was suggested to Mr Cook that there was nothing in his curriculum vitae to indicate that he worked for Wattyl and that his expert witness statement referred to instructions provided to Mr Cook "from Mr Bo Padkjaer of Wattyl Group" conveys a misleading impression that he was independent and not employed by the Wattyl Group. Mr Cook agreed that his curriculum vitae did not disclose that he worked for Wattyl. Mr Cook's explanation was that his curriculum vitae was "three years old". Mr Cook also conceded that the statement in his witness statement referring to instructions provided from Mr Padkjaer of the Wattyl Group could convey a misleading impression that he was independent and was not employed by the Wattyl Group. However, Mr Cook gave evidence that he had never not said that he worked for Wattyl and that all his work had been done on Wattyl letterhead. In any event, it was submitted that at the commencement of the trial, counsel for Fulton Hogan referred to the fact that Mr Cook was an employee of Wattyl.

196 I reject the suggestion that Mr Cook intended to create a misleading impression that he was not employed by Wattyl. It is true that his curriculum vitae suggested that he is not employed at Wattyl, because it states that he was employed at Nuplex Industries (Aust) Pty Ltd from "October '97 to Present". I accept Mr Cook's explanation that this is because his employment details had not been brought up to date. On the other hand, Mr Cook's report was written on Wattyl Group letterhead and though it does not state that Mr Cook is an employee of Wattyl, it is entitled "Group Analytical Report: CONFIDENTIAL: For Internal Laboratory Use Only" which tends to suggest that the Cook report is not an independent report.

197 It is unfortunate that Mr Cook's employment details did not reveal that he was an employee of Wattyl, but I attribute this to oversight rather than lack of candour on his part. Of more concern is the fact that Mr Cook attended the joint conference of experts notwithstanding he was an employee of Wattyl at the time. A court order was made with respect to expert witnesses which provided, among other things, that the experts shall confer and provide a joint report to the Court and to the parties and that "(t)he conference of experts and all further communications between them in relation to the preparation of the joint report shall be conducted in the absence of the parties, their employees or agents, or any practitioner associated with any party".[23] Had the Court been informed at the time those orders were made that Mr Cook was an employee of Wattyl, the orders made with respect to the joint conference of experts would have expressly excluded Mr Cook as an attendee to ensure that the process of a free exchange of views by independent experts could occur in the absence of an employee of any party. It is regrettable that the integrity of process contemplated by the order was compromised by his attendance.

198 Next is the question of the effect of instructions given by Mr Padkjaer to Mr Cook.

199 Fulton Hogan submitted that Mr Cook was requested to achieve a "pre-ordained outcome which should invite suspicion on all conclusions favouring Wattyl".

200 Counsel for Fulton Hogan focused on instructions given to Mr Cook by Mr Padkjaer when he requested the analysis. Mr Padkjaer, who was Mr Cook's superior at Wattyl, wrote:

Aim is to confirm that the product or cure does not deteriorate with age (some of the batches were several years old at the time of use and 'expert' consultants have indicated that the product was not fit for purpose and cure is incomplete. That would be a significant claim to refute.

Analysis should focus on the EP100 layer only. There are no issues relating to other coatings used in the system.

Please discuss results with me prior to issuing report (report required by mid-Friday afternoon at the latest).

201 Mr Cook gave evidence that there was a very high likelihood that he did discuss the results with Mr Padkjaer prior to issuing the report but he did not recall the conversation. Mr Cook also in substance accepted that if he was acting independently he would not have discussed the report with Mr Padkjaer before issuing it.

202 However, Mr Cook denied having tailored his results to suit Mr Padkjaer's objectives. He insisted that he read the instructions from Mr Padkjaer as a request to either confirm or deny that there was a problem, that he believed in producing the facts and he would not say something that is not right and that if he had found an issue with the product, the right thing for him to do would be to advise that he saw something wrong with it.

203 I accept Wattyl's submission that there is no evidence that Mr Cook manipulated his tests or test results to suit Wattyl. I also accept Mr Cook's denial of having deliberately tailored the results to best suit Mr Padkjaer's objectives, though in my view he may have done so unconsciously, having been put on notice that the aim of the analysis "is to confirm that the product or cure does not deteriorate with age".

204 Further, there are some issues of concern with the analysis of the results.

205 In substance, Mr Cook compared the laboratory-made and retained samples of the product with samples of the product taken from the trusses and the handrails looking for differences between the product in the samples and the product from the flakes.

206 It became evident that the two control samples made up at the Wattyl laboratory used to compare the Cox sample (from the trusses) and the DH Corrosion sample (from the handrails) were differently constituted samples. The first control sample was a blend comprising Part A of the Galvit primer used by Cox on the trusses mixed with Part B used by DH Corrosion on the handrails (rather than the Part B used by Cox on the trusses). Mr Cook accepted that in consequence he had not made a precise comparison.

207 The second control sample was a blend comprising Part A which had been manufactured in 2011 to a different formulation from the Part A used by Cox and DH Corrosion and a Part B newly manufactured to the old formulation. Mr Cook accepted that there was no ageing of this product.

208 It follows that to the extent that comparisons were able to be made they were not precise comparisons. According to Mr Cook, using the old Part B hardener used by DH Corrosion (instead of the old Part B used by Cox) with the Part A used by Cox was close enough to being the same.

209 Neither Dr Garnys nor Mr Eccleston agreed that if the results set out in Mr Cook's report were accepted, that those results suggested there was no problem with the Galvit primer. Mr Eccleston gave evidence that the method used by Mr Cook to determine the differences (using KBr pellets) was:

not a very specific method of determining other differences within the product itself. There may be differences in the amount of cross-linking the amount of Part A to Part B and that really hasn't - wouldn't really show up specifically with a KBr disc of samples taken from the site because you've got a lot of zinc in there, so you're masking a lot of the peaks. So you haven't got very distinct peaks in the KBr pellets, so you wouldn't be able to pick up subtle differences within the resins.

210 Mr Eccleston also disagreed with Mr Cook's conclusion that all spectrograms appeared similar to each other. He referred to the spectrogram which compared old and new blends of the Galvit primer with the Cox and DH Corrosion scraping samples. Mr Eccleston drew attention to the difference between the spectrogram from the Cox sample and to the other three samples. He gave evidence that the difference would warrant some further investigation. When asked about whether the broader peak under discussion was related to the presence of moisture in whatever form, Mr Cook said, "It could be moisture and it could be the nitrogen, and the nitrogen was part of the Part B system, which has actually cured it". According to Mr Eccleston, there is a significant difference in the peaks.

211 The asserted similarity of all spectrograms is a key path to Mr Cook's conclusion that there is no defect in the Galvit product. I do not accept Mr Cook's conclusion that all spectrograms appear similar to each other. There is an obvious visual difference. I accept the evidence of Mr Eccleston that there is a significant difference in the peaks shown in the spectrograms and the evidence of both Dr Garnys and Mr Eccleston that one cannot conclude that there is no defect in the Galvit material from the spectrogram analysis conducted by Mr Cook.

212 I also formed a strong impression that Mr Cook was steadfastly resistant to any suggestion that the Wattyl product might have been defective. During the trial, any concession against Wattyl's interests was made begrudgingly. When responding to questions Mr Cook gave the bare minimum of information. I formed the view that he was seeking to avoid any risk of adverse inferences being drawn against his employer. Mr Cook's obvious lack of independence has meant that I must view his evidence as a whole with caution.

213 Wattyl also relied on the evidence of Mr Pellegrino, the Technical Services Manager at Wattyl. Mr Pellegrino has a Bachelor of Applied Science and has worked in the paint industry for over 30 years, 26 of which were in various departments at Wattyl. Mr Pellegrino's duties included handling customer complaints.

214 After complaints were received from Cox in relation to the trusses and from DH Corrosion in relation to the handrails, Mr Pellegrino became involved in the investigation.

215 In relation to the trusses, Mr Pellegrino gave evidence that he carried out an inspection and an on-site test over a period of two hours; that he had conducted MEK double rub tests, each with 50 rubs on the Galvit primer which showed a minimal transfer of primer and no softening of the primer; that in his opinion the Galvit primer had fully cured; that there was evidence of Sigmaweld on the substrate; that in the case of two areas on the trusses, one around the corner from the other (W# 4 and W# 5), one failed the cross-hatch adhesion tests whilst the other did not; and that there were considerable film build variations across the trusses and total dry film thicknesses in excess of the 375 micron dry film thickness which was the recommended specification.

216 Mr Pellegrino was criticised for his comment made at the end of his lab report:

NOTE: CSIRO report stated/alluded the cause of the problem was related to excessive film builds and also the GalvitEP being under cured - note, they do not imply that this is faulty.

217 It was suggested to Mr Pellegrino that he was not objective and that he had misrepresented the CSIRO report (of Mr Eccleston) in order to put the best gloss on the complaint in favour of Wattyl rather than objectively considering the merits of the complaint. Mr Pellegrino denied this was the case.

218 Wattyl submitted that this criticism of Mr Pellegrino was unwarranted, essentially because the CSIRO report was not couched in definitive terms, stating that there could be a material problem; that this may be due to the use of primer that is past its shelf life or has not been stored appropriately; and that the batch numbers indicate that the primer may have been past its shelf life when it was applied.[24] In that context, it was submitted that the statement "they do not imply that it is faulty" was not inappropriate.

219 In my view, nothing determinative turns on this evidence which reflects more upon the perception of protagonists in dispute rather than on matters of substance. I do not draw any conclusion on this issue adverse to Mr Pellegrino about the manner of his expression at the end of his report. I accept that this was his view at the time.

220 Of more significance, in my view, is the reliability of the results of tests conducted by Mr Pellegrino on the trusses.

221 Mr Pellegrino carried out an MEK double-rub test using 50 double rubs, not the 10 double rubs performed by Mr Eccleston when he conducted the same test. Mr Eccleston gave evidence that testing using 10 double rubs was in accordance with the method prescribed by Australian Standard test methods. Mr Pellegrino did not disagree that the standard prescribed 10 double rubs rather than 50 double rubs. He said that he did 50 double rubs because that was the way he had been taught. Mr Eccleston gave evidence that he followed the Australian Standard test method by wetting a surface of a white cloth with a solvent, methylethylketone, rubbing the wet cloth 10 times and then observing the cloth. Mr Eccleston said that if the coating has cured and it's a full cure then nothing will be left on the white cloth. When he conducted the tests on the primer, he found that the Galvit primer was depositing on the cloth after 10 rubs which indicated that Galvit primer had not achieved full cure.

222 Mr Pellegrino did not refer to the Australian Standard test method and did not use it. On areas where cross-hatch adhesions removed the coating exposing the Galvit primer, he performed "heavy MEK double rubs" to determine whether the Galvit primer had fully cured. Mr Pellegrino gave evidence that, "if it's not properly cured it turns - it goes really soft, you can actually scratch it with your finger nail, and also you get a lot of paint transfer on a rag if it's not fully cured". Mr Pellegrino gave evidence that in this case there was "minimal paint transfer and it was hard as a rock when I tried to scratch it, so in my opinion it had been fully cured".

223 Mr Pellegrino's evidence as to the degree of cure involves highly subjective elements; as to what is meant by a "heavy" double rub, his evaluation of the extent of the paint transfer to the MEK soaked cloth and whether, and if so, the extent to which the material being rubbed had softened.

224 There was no evidence from which it could be ascertained whether the paint which had deposited onto the MEK soaked cloth was deposited after Mr Pellegrino had rubbed the paint film ten times which would have been evident had Mr Pellegrino conducted the test in accordance with Australian Standard test methods. No evidence was given by Mr Pellegrino as to the process by which he tried to scratch the paint surface, although I infer he may have scratched it with his fingernail.

225 In the circumstances, I prefer the evidence of Mr Eccleston, who I accept conducted a degree of cure test and a hardness test in accordance with Australian Standard test methods.

226 It is also necessary to note that Mr Pellegrino is a longstanding Wattyl employee, having dealt with customer complaints over many years. In my view, Mr Pellegrino tended to understate the extent of removal of the Galvit primer from the truss structure when the MEK double rub procedure was performed. When questioned as to the extent of removal of the Galvit primer, I found him to be quite reactive and unwilling to concede that much Galvit primer had been removed in the rubbing process. Further, at the same time as he completed his lab report dated 30 July 2009, Mr Pellegrino recorded in a Complaint Follow Up Form dated 30 July 2009, that Wattyl expected a claim in relation to the complaint in the order of $500,000. The inevitable conclusion to be drawn is that the lab report was written at a time when its author was conscious of an impending claim against his employer for a substantial monetary amount.

227 For all of these reasons, in my view Mr Pellegrino's conclusions must be given less weight.

228 I prefer the evidence of Mr Eccleston and accept his conclusion that the Galvit primer had not cured sufficiently to achieve the tensile strength necessary to withstand the curing stress of the intermediate and top coat.

229 In my view, Mr Pellegrino exaggerated the extent to which he observed Sigmaweld on the substrate. It is noteworthy that although Mr Pellegrino observed the presence of Sigmaweld, he does not expressly correlate the failure of the Galvit primer with the presence of Sigmaweld, nor does he conclude in his lab report that the presence of Sigmaweld was a cause of the cohesive failure of the Galvit primer. I also note that Mr Pellegrino did not refer to the Wattyl product specifications for Sigmaweld, which state that Galvit EP100 is recommended for use with Sigmaweld.

230 Mr Pellegrino expressed the opinion that if there is a product fault, the product should fail everywhere. This opinion was expressed without any attempt to establish a scientific basis for it. Mr Pellegrino cited as an instance where failure had occurred on one face of the truss but not on another, even though the locations were in close proximity, one around the corner from the other on the same pillar.

231 Dr Garnys rejected the proposition that if the Galvit primer was defective it would fail everywhere. He gave evidence to the effect that the Galvit primer had failed in areas, but not universally throughout the structure and thought that it would depend upon the stressors applied at various locations on the trusses with a split in the coating occurring where those stressors were too large for the integrity of the coating. In Dr Garnys' opinion, the majority of the coating would fail with the passage of time.

232 I accept Dr Garnys' independent opinion on this issue in preference to that of Mr Pellegrino.

233 Mr Pellegrino's observations in his lab report that there were considerable film build variations across the job and total film thicknesses in excess of the recommended specification of 375 microns were consistent with observations made by the other experts. Likewise, Mr Pellegrino found areas where the dry film thickness of the coating was well in excess of the specification, but where no zinc splitting was detected.[25] This is consistent with Mr Eccleston's findings that the delamination was not dependent on the total coating thickness.

234 Wattyl submitted that it would make no sense for Mr Pellegrino to ignore or avoid the possibility of a defective product, relying on a passage in Mr Pellegrino's evidence where he said in effect that he would not say a complaint was not justified to save costs, "because a lot of the batch sizes are quite big, and the ramifications for other complaints is horrendous, so your argument of trying to cut costs is incorrect. I really need to justify whether it's [the complaint is] justified or not and remove it from the market place if it is faulty".

235 In my opinion, this logic cuts both ways. At the time of completing the lab report, Mr Pellegrino was conscious of the fact that there was a likely claim to be made against Wattyl in the order of $500,000. Mr Pellegrino had extensive experience in the area of product complaints. He must have known that what he wrote in his lab report would have been subjected to critical examination later and that anything he might say could expose his employer to a substantial claim for damages. In my view, this is very likely to have affected the impartiality of his analysis.

236 In relation to the handrails, Wattyl relied on Mr Pellegrino's lab report dated 20 June 2009 in which Mr Pellegrino recorded having investigated the cause of a paint failure at DH Corrosion in order to determine whether the failure occurred due to a product fault or an application issue.

237 Mr Pellegrino's on-site inspection confirmed zinc splitting on the handrails he inspected, but Mr Pellegrino could not confirm why the zinc splitting had occurred as there was an acceptable blast profile on the substrate, the quality assurance records did not indicate anything untoward, and dry film thicknesses of the Galvit primer and the top coat were not excessive. Mr Pellegrino conducted a double rub test with a MEK soaked rag using 50 rubs. Mr Pellegrino gave evidence that there was no softening or transfer onto the rag, indicating that the primer had fully cured. This evidence was inconsistent with his lab report which recorded that there was a slight transfer onto the rag.

238 Testing was carried out in the Wattyl laboratory at various temperatures and film builds using different batches of Part A and Part B of the two-pack mixture which makes up the Galvit primer. Twelve panels were sprayed with various combinations of Parts A and B used at Cox and Parts A and B used at DH Corrosion. Two samples of each combination were allowed to cure in three locations; the Wattyl laboratory, in a refrigerator at 5°C and outside. The Galvit primer was deliberately sprayed at low film builds at one end of the panel working up to high film builds. On a portion of the panels there was overspray left on bare metal. The overspray was then sprayed again the following day. Each panel had a high build and oversprayed section.

239 Zinc splitting was found to have occurred on the high build section of two panels that were cured outside. Apart from both having been cured outside, the common factor with these two panels was the use of the batch of Part B Galvit primer that had been used by DH Corrosion.

240 There was no evidence of any film thickness measurements having been taken of the "high build section" of the two panels which failed. There is nothing to suggest that the Galvit primer coating on the "high build sections" of these panels was so excessive as to exceed specification for the product. Presumably Wattyl did not exceed its own product specification in carrying out the test. There was also no evidence to suggest that any top coating had been applied over the Galvit primer. Accordingly, it would appear that the Galvit primer spontaneously failed on two panels left to cure outside. There was no evidence to suggest that any of the test panels had failed on the section of the panels where the Galvit primer had been deliberately oversprayed and later over coated with primer.

241 Mr Pellegrino gave evidence that he sent paint flake samples taken from DH Corrosion for analysis, together with samples of Parts A and B of the Galvit primer used by DH Corrosion. The paint flakes were sent to measure the film builds and the Part A and B were sent for testing to make sure they were consistent with their formulation. Mr Pellegrino gave evidence that everything looked in order, consistent with the formulation on Parts A and B; and the examination of the film builds indicated dried film thicknesses which raised no alarm bells and did not indicate that DH Corrosion had carried out anything incorrectly. Mr Pellegrino also sent the Part B used by DH Corrosion (the common factor with the two failed panels) for analysis, together with the Part B used by Cox. The laboratory was asked to compare both batches for consistency. The batch used by DH Corrosion was found to be very close to the standard formula mixture. The batch used by Cox appeared to be lower in Ancamine K54. Mr Pellegrino gave evidence that Ancamine K54 is an accelerator to speed up the curing process of the product especially under colder conditions. Mr Pellegrino said:

Whether it's in the formula or not it's irrelevant really, but it's in there just to speed up especially under colder conditions. So if it was all missing out there altogether it wouldn't - it wouldn't affect the coating.

242 In July 2009, Mr Pellegrino arranged for a trial to be conducted at the premises of DH Corrosion on two sections of the handrails. These sections were given a 2½ class blast and sprayed on one side and then on the other with the Galvit primer. Mr Pellegrino did not observe the application of the coating but attended the site to inspect the coated handrails within 24 hours of application. He conducted a cross-hatch adhesion test, checked the profile of the metal and did some MEK double rubs to check that the coating had fully cured. Mr Pellegrino gave evidence that he found nothing out of the ordinary and nothing that concerned him.

243 A few days later on his second visit he found that the coating was delaminating.

244 Mr Pellegrino conducted further cross-hatch adhesion tests which showed zinc splitting. He found that, "a lot of the flat sections had not failed and some of the louver sides or the wings had failed, so it was quite puzzling again what was going on." Mr Pellegrino expressed his opinion as to why the adhesion was good on the flat surfaces in contrast to the surface of the louvres or wings in the following terms:

By that stage all the laboratory work that we'd carried out and so forth, yes, I did have a view that it was due to overspray. Those flat sections did not have any overspray, so the coating performed excellent in all the flat spots. All our testing that we did and carried out both in the lab and on the field finally come to the conclusion that the louvres or those wings that fail in areas due to overspray, so the zinc splitting was the cause for that reason.

245 This opinion was premised on Mr Pellegrino's view of how the primer was sprayed onto flat elements of the handrails in comparison to the louvred elements. According to Mr Pellegrino:

flat sections would have one coat, so would the louvres. The only difference is when you spray the louvres they're on about 45 degree or so... so when you spray one side, the whole panel and you will get some bounce back on the back of the panel's overspray [sic]. The next day after 24 hours of cure when they turn the whole thing around, and do the first coat - or they only do one coat on the other side you end up spraying on the overspray. If that overspray has not been cleaned, or properly removed it's just basically loose powder there, so you'll get adhesion failure.

246 In essence, Mr Pellegrino concluded that the primer coating on the trial hand rail sections failed to due overspray caused by "bounce back of spray" and not due to any fault in the product.

247 In my opinion, this overspray theory was speculative and was not established on the evidence.

248 As Mr Pellegrino did not see the spray being applied he cannot reliably comment on the likelihood of overspray being caused by a "bounce back" effect. Before any conclusion could be drawn, it would be necessary to have some evidence of how the spray was applied and to know whether any, and if so what, measures were taken by the operator to avoid a bounce back effect. Mr Pellegrino did not say he observed overspray on the handrails; he was in no position to say and did not say that overspray had not been cleaned or properly removed. Further, he conceded that on his analysis, there would be overspray causing adhesion failure only on one side of the louvres. However, there was no evidence to suggest that the delamination occurred only on one side of the handrails.

249 In my view, if overspray was an operative cause of failure one would have expected that the primer coated over deliberately oversprayed panels in the 12 test panel experiment conducted at Wattyl would have delaminated. Yet the results of the 12 test panel experiment did not refer to delamination occurring on the oversprayed areas.

250 Wattyl submitted that Mr Eccleston's acknowledgment that the fence mounting points and broader surfaces were less likely to have an issue with overspray gave Mr Pellegrino's opinion on overspraying some support. However, Mr Eccleston's acknowledgement of the risk of overspray was dependent upon Mr Eccleston's own assumptions about how the panel sections were sprayed. Mr Eccleston did not observe the coating being applied. No evidence was given by anyone from DH Corrosion to say how the coating was applied to the test panels.

251 To the extent that Mr Eccleston's remarks could be regarded as support for the overspray theory, in my opinion it is support of a very general kind, falling a long way short of any endorsement of Mr Pellegrino's theory. Further, when the potential of overspraying was raised with Mr Eccleston, it was not suggested to him that there would or could be a "bounce back" effect. In my opinion, Mr Pellegrino's theory lacked a cogent evidentiary foundation.

252 Wattyl submitted that test panels showed failures which were inconsistent with product defect because only two test panels out of twelve failed, that in one instance only 50% of a cross-hatched area failed in an adhesion test on two test sections of handrail, and because flat areas on these handrails showed good adhesion. I am unpersuaded by this submission. In my view, the failure of two panels out of the twelve test panels prepared in Wattyl's laboratory was a very significant failure. It is reasonable to infer that in the preparation of the test panels, the Wattyl technicians eliminated application errors by careful preparation and application of the coating in laboratory conditions. There is no evidence which would suggest any shortcomings in this regard.

253 There were two features in common with the two test panels on which zinc splitting occurred. Only panels left to cure outside failed. Secondly, both panels which failed used the same Part B batch that had been used by DH Corrosion.

254 This tends to suggest at least that this batch of the Galvit primer product was unable to perform adequately in normal drying conditions, notwithstanding that the preparation and application are likely to have been carried out properly. [26]

255 In my view, this evidence is consistent with a likely product defect. The fact that the zinc splitting has not occurred in all instances at the same time is not decisive as previously noted.

256 Mr Peter Lowe, the State Sales Manager (Industrial) for Victoria and Tasmania for Wattyl, gave evidence that there were two complaints made in relation to the Galvit primer supplied on the Burwood footbridge project. The tenor of his evidence was that Galvit primer was a reliable product which had been in the market place for approximately 30 years.

257 The first complaint was an internal complaint from the Wattyl Rocklea warehouse dated 9 May 2008. The complaint relates to the Part A Galvit primer batch C60132971-10 supplied from Wattyl to its distributor from about August 2008 and used by Cox on the trusses later that year.

258 The complaint was characterised as "low visc/settling". The lab report was as follows:

The viscosity of the returned complaint sample was found to be within the product specification. The product re-incorporated easily and no particulate matter was found on straining. No other complaints have been received on this batch.

259 Wattyl submits that the lab report conclusion that the sample was within product specification means that it was not likely that this batch was defective and that the cause of the delamination must lie elsewhere.

260 The second complaint was an internal complaint from Wattyl in Blacktown, New South Wales dated 15 May 2009. It related to the same Part A Galvit primer batch C80151394 used by DH Corrosion on some of the handrails later that year.

Please check warehouse stock (earliest batch made of resource possible) for condition and saleability.

Product was mixed with a drill - was settled to about the same degree as B/N C60130597. Product was able to be mixed and incorporated - when sampling for mixing by QC, it was noted it was slightly lumpy. QC lab mixed product with Part B and sprayed it out - no problems reported.

Product okay to sell.

263 Wattyl submitted that the conclusion that the product was of saleable quality meant that it was not likely that this batch was defective, and that the cause of the delamination must lie elsewhere. In my view, this does not necessarily follow. In relation to the report on Part A later used by Cox, no precise details of testing are given, save that one can infer that the sample was tested for viscosity, re-mixed and strained for particulate matter. No spraying out test appears to have been performed to determine whether the product when mixed with Part B cured properly and without delamination. The lab report on its face appears to be very superficial.

264 In relation to the lab report on the Part A later used by DH Corrosion, no precise details of testing are given. The product was noted to have settled but was able to be "mixed and incorporated". In this instance, the product was mixed with Part B and sprayed out. There is no evidence to suggest that any adhesion tests were performed; there is no evidence of the dry film thicknesses involved; there is no evidence of the conditions under which the coating was dried and whether drying occurred in the lab or outside; and there is no evidence that any hardness or degree of cure tests had been carried out. In my view, the lab report on its face also appears to be superficial.

265 What does seem remarkable is that a product with an apparently enviable track record of reliability over a 30 year period should have internal complaints made about the very batches later used on a project where serious delamination issues occurred. In my view, it is trite to submit that the results of the Rocklea and Blacktown internal complaints suggest that the cause of the delamination does not lie with the product. On the contrary, in my opinion, the pointed complaints history tends to raise rather than quell concern. Furthermore, there was no evidence given by Mr Lowe or any other witness for Wattyl as to whether any other part of the batches in question had been sold and distributed to customers and, if so, when they were sold and distributed. Mr Lowe gave evidence that from time to time an older batch of product remained on the shelf and slipped through the process of "first in, first out". He acknowledged that it was possible that this happened with the batch of product manufactured in 2006 which was supplied to Cox in late 2008. In the absence of evidence as to whether or when other product from these batches reached the marketplace, the complaints history evidence (of only two complaints) has no probative value.

266 Fulton Hogan drew attention to the fact that Wattyl reformulated Part A of the Galvit primer on two occasions in 2009 after the zinc splitting had occurred, and before Part B of the product was withdrawn. The reformulations with respect to Part A occurred on 30 July 2009 and 29 September 2009 respectively.

267 Fulton Hogan submits that the reformulations would not have occurred had the product not been faulty.

268 Wattyl submits that the evidence suggests only that this was a possibility. I accept Wattyl's submission on this point. Mr Pellegrino gave evidence that it was normal for Wattyl to upgrade formulas for raw material rationalisation or to improve the product, but that it does not mean that there was anything wrong with the product in the first place.

269 It is significant, in my view, that there was no evidence before the Court to indicate the precise effect of the changes made to the formulation;[27] how the chemicals in the formulae functioned; and whether the changes to the formulae would have improved the performance of the product or simply reflected a change of supplier or the availability of cheaper ingredients. Despite the fact that the dates of reformulation closely aligned with the investigation of the complaint in this matter and therefore raises concern, there is no sound evidentiary basis to infer that changes to the formulation on each occasion were made because the product was known to be faulty.

270 Once manufactured, Wattyl products are receipted into Wattyl's warehouse ready for distribution. Customers place orders through a centralised customer service department and the product is then distributed as required. Mr Lowe accepted that the Galvit primer batch C60132971 had been manufactured in October 2006 and was supplied to Wattyl's distributor for on-supply to Danfab (the first defendant) and Cox progressively from August 2008 to late 2008.

271 Mr Lowe conceded that it was possible that this batch may have deteriorated during the time it was stored in Wattyl's warehouse. The Wattyl product information sheet states that Galvit EP100 has a "shelf life (cool and dry place) ... at least twelve months". Wattyl did not produce any records, nor give any evidence as to the conditions of storage for the period of almost two years that it was stored at Wattyl's warehouse.

272 In the absence of any explanation (even as to the usual storage conditions in Wattyl's warehouses), I infer that Wattyl's evidence as to the conditions of storage of the product would not have helped Wattyl's case. Further, I note that the Galvit product did not have a "use by" date on the cans.

Wattyl's failure to disclose relevant documents

273 It was common ground that Wattyl is a ISO9001 company. As such, it subscribes to the maintenance of quality assurance standards. This includes the requirement that it retain and maintain a document trail with respect to the manufacture of batches of product.

274 This documentation in all likelihood includes laboratory certificates verifying the quality of the constituent parts of each ingredient in each product and production records which record how each batch has been manufactured. Wattyl admitted that it kept production records known as "batch tickets".

275 Wattyl failed to discover the relevant laboratory certificates and the relevant batch tickets. The failure to produce these documents was the subject of comment during the trial. Wattyl can have been under no misapprehension as to their relevance. Nevertheless, no documents of this kind were forthcoming. I infer from their non-discovery without explanation that neither category of document could have assisted Wattyl in the defence of its case.

276 On 20 July 2009, Mr Pellegrino completed his lab report with respect to the failures at DH Corrosion and recorded that the two test panels prepared by Wattyl which were cured outside had suffered from zinc splitting. On the same day, the batch of Part B of the Galvit primer used by DH Corrosion was withdrawn from the market. This was stated to be as a precaution.

277 Approximately two months later, on 25 September 2009, all of Wattyl's remaining stock of Galvit EP100 Part B was withdrawn from the market. This was stated to be for reasons of "product rationalisation". Mr Lowe gave evidence that the withdrawal was not really for product rationalisation but was for a precautionary reason; that product rationalisation to achieve economies of scale was an ongoing process within Wattyl, but that the move towards economies of scale with the manufacturing operation had no impact on the products in issue in this proceeding.

278 Wattyl submitted that Mr Lowe's understanding of the reason for withdrawal of the Part B product was only his personal understanding, it should not be treated by the Court as evidence of anything more. I disagree. As State Sales Manager, he gave evidence on behalf of Wattyl and his understanding can be reasonably attributed to Wattyl. In this regard I take into account, as I have said, that Mr Lowe gave unqualified evidence to the effect that product rationalisation had no impact on the products in issue in this proceeding. On balance, I take the view that the term "product rationalisation" was a euphemism used to obscure the true reason. I accept Mr Lowe's evidence that the reason for the product withdrawal was as a precaution. I am satisfied that Wattyl took that step because it was concerned that the primer was defective.

279 In my opinion, on the balance of probabilities, the batch of Galvit primer used by Cox on the trusses (batch C60132971-10) and that used by DH Corrosion on some of the handrails (batch F80156328-05) which spontaneously failed in tests conducted by Wattyl were defective.

280 In reaching my conclusion, I have substantially accepted Dr Garnys' field observations and have given weight to his evidence that the age, high level of viscosity and drying time of the Galvit primer batch used on the trusses were indicative of potential problems with the primer; that freshly disbonded material from the trusses and the delaminated material from the handrails would not have suffered corrosion between the time of failure and the time of examination; and that the delamination phenomenon on the trusses appeared to be the same as the delamination phenomenon which appeared on the handrails subjected to testing at the premises of DH Corrosion.

281 I have also substantially accepted Mr Eccleston's evidence. I found him to be a most impressive witness. He was methodical, measured and scientific in his approach. At the three sites he tested on the trusses, he found that the Galvit primer had failed to cure and that the failure to cure had occurred irrespective of coating thickness. This evidence was unchallenged and in my opinion could not be satisfactorily explained away by applicator error. Mr Eccleston's finding that the failure had occurred irrespective of coating thickness was corroborated by the report of Mr Dove and was not inconsistent with the reports of Mr Stone.

282 In this context, I have accepted the evidence that there are some Sigmaweld remnants on the substrate of the trusses, but I have rejected Mr Dromgool's opinion evidence relating to the permeability and porosity of the remnants of the Sigmaweld and migration of the Galvit primer into remnants of the Sigmaweld which is discussed later in these reasons.

283 In relation to the evidence given on this issue by witnesses called by Wattyl, suffice to say that, save where I have otherwise indicated, I prefer Dr Garnys and Mr Eccleston's evidence over the evidence of Mr Haig, Mr Dromgool, Mr Cook and Mr Pellegrino.

284 It is not possible to say precisely why the batch of Galvit primer used on the trusses was defective. It may have deteriorated due to effluxion of time since it was manufactured in October 2006 and the date of its supply to Wattyl's distributor in the latter months of 2008; because the warehousing conditions at Wattyl during storage prior to supply to Wattyl's distributor were unsatisfactory; because there was a mistake made in the manufacturing process; because inferior quality ingredients were used; or a combination of the above. Wattyl withheld relevant documents which might have shed light on these issues. Wattyl also declined to call witnesses whose evidence might have shed light on those issues.

285 In reaching this conclusion, I have taken into account the evidence relating to application of the primer to the trusses discussed in response to question 4.

286 For the reasons which are set out below in response to question 6, I am satisfied on the balance of probabilities that applicator error did not cause or contribute to the cohesive failure of the Galvit primer on the trusses, although it did cause some isolated instances of adhesive failure.

287 For reasons set out below, in response to question 6, I am unable to determine on the balance of probabilities whether or not applicator error caused or contributed to the cohesive failure of the Galvit primer on the handrails, even though I have found that batch C80156328-05 was defective.

Question 4 - Was the Wattyl coating system properly applied in accordance with the Wattyl product specifications (a) by Cox to the trusses; and (b) by other contractors to the handrails?

288 Fulton Hogan submitted that the Wattyl coating system was properly applied in accordance with the Wattyl product specifications.

289 In relation to surface preparation, the Wattyl product specifications for Galvit EP100 recommend that the substrate should be blast cleaned in accordance with AS1627.4 to Sa 2½ minimum (AS1627.9), with a surface profile between 40 and 70 microns.[28]

290 Fulton Hogan relied on the Cox quality assurance records to show that the trusses were blast cleaned with a surface profile of approximately 80 microns based on contemporaneous readings taken of the substrate profile using Press-O-Film tape and the evidence of Mr Cooper, a Paint Inspector and Quality Assurance Officer for Cox, who completed the records and supervised the job. The plaintiff further submitted that in his report, Mr Eccleston observed that since the zinc was adhering to the steel substrate, the surface preparation on the substrate was satisfactory.

291 In relation to the thickness of the Galvit EP100 primer, the Wattyl product specifications for Galvit EP100 state:

typical film thickness (per coat) 15 microns (dry), 75 microns max, 90-135 microns (wet) depending on the system. Dft's [Dry film thicknesses] of more than 75 microns are not recommended underneath thick, rigid epoxy systems

292 Fulton Hogan submitted that Cox's quality assurance records show that only one coat of EP100 was applied in the original application, that intermediate and top coats were applied in accordance with Wattyl's recommended overcoating intervals and that all other atmospheric criteria were in order. Fulton Hogan further submitted that the wet film thickness measurements taken by Mr Cooper at the time of application are the most informative measurement and indicate that the application of the Galvit primer by Cox when applied was within specification, namely within the range 70-75 microns for dry film thickness and in the range 120-135 microns for wet film thickness. Mr Cooper gave evidence that Cox applied the Galvit EP100 in accordance with the Wattyl product specifications for Galvit EP100. It was submitted that the subsequent dry film thickness readings which would indicate a higher dry film thickness of the Galvit primer layer could be explained by the fact that the trusses had had spot repairs during the period from January 2009 to May 2009 at numerous locations where the primer had cohesively failed. This point was made by Mr Tanti,[29] the managing director of Cox.

293 Prior to carrying out the work on the trusses, a specification was given to Mr Tanti which specified a first coat of Galvit primer at 75 microns, a second coat of Wattyl EB600 at 200 microns and a third coat of Wattyl Poly U750 at 100 microns, giving a total system build of 375 microns.

294 The dry film thickness readings in the Cox inspection reports suggest that the paint was applied generally in accordance with these specifications.[30] Cox also relied on evidence from Mr Cooper to the effect that he used a paint comb to measure the wet film thickness of the Galvit primer when he tested it on or about 13 December 2008 and Mr Cooper recalled that the wet film thickness was between 120 and 135 microns and that he would expect 120 microns to dry to about 75 microns.

295 Wattyl submitted that Cox failed to apply the Wattyl coating system to the trusses in accordance with the Wattyl product specifications.

296 In particular, Wattyl submitted that the product specifications for Galvit EP100 specified a maximum dry film thickness of 75 microns. Wattyl relied on measurements taken by Mr Dove in January 2009 using a Positector 200 Ultrasonic coating thickness gauge in five locations with three to five measurements taken at each location which showed readings for the zinc epoxy layer ranging from 72 to 248 microns; the measurements taken by Mr Pellegrino on 23 June 2009 which showed readings between 88 and 121 microns; and three zinc layer thickness measurements taken by Mr Eccleston which showed readings of 23, 64 and 104 microns.

297 Mr Eccleston gave evidence that this Positector 200 instrument is not recommended for measuring coating thicknesses on steel. Mr Eccleston doubted the accuracy of this instrument in determining the coating thickness of the Galvit primer layer.[31] This accorded with the anecdotal evidence of Mr Tanti that when he tried to take dry film thickness readings with a Positector coating thickness gauge alongside Positector readings taken by Mr Dove found differences of 500 or 600 microns. I am not satisfied that the readings taken by Mr Dove of the thickness of the Galvit primer layer using a Positector 200 Ultrasonic coating thickness gauge are reliable and in the circumstances, I place little weight on those readings. However, I am satisfied that the measurements taken by Mr Pellegrino from the paint flakes are reliable and produced readings which exceed the maximum recommended dry film thickness of 75 microns, being 88, 90 and 121 microns. I am also satisfied that one of Mr Eccleston's readings showed that the recommended maximum thickness of Galvit primer was exceeded.

298 Further, Wattyl submitted that the product specifications for Galvit EP100 state that dry film thicknesses of more than 75 microns are not recommended underneath thick, rigid epoxy systems and that Mr Dove measured dry film thickness in five locations with three to five readings at each location ranging from a total micron depth of 390 to 798 microns. Mr Don Bartlett on 10 February 2009 noted that the total dry film thickness of the coating on the trusses was consistent with the results obtained by Mr Dove which showed considerable variation around an average somewhat higher than the minimum dry film thickness specified; Mr Dromgool on 27 May 2009 measured readings on truss B (the north truss) ranging from 62.7 to 1233 microns and on truss A (the south truss) ranging from 73.5 to 1073 microns; Mr Stone took a large number of dry film thickness readings on 26 May 2009 and recorded readings ranging from 267 to 1345 microns on the north truss on the west side, from 109 to 865 microns on the north truss on the east side internal and from 155 to 910 microns on the south truss east and west sides. In Mr Stone's second report, based on readings taken on 1 and 2 June 2009, the highest reading on the north truss was 1383 microns on the external east side, 1587 microns on the south truss on the external west side and 1133 microns on the underside of both trusses. Wattyl relied on the fact that Mr Eccleston's readings at three locations ranged from 292 to 651 microns and that the readings in Dr Garnys' report taken at four locations ranged from 376 to 628 microns.

299 Wattyl also relied on the evidence of Mr Gilchrist that it was difficult to remove the Wattyl coating system in the blasting of some areas because of thickness which was getting up over 2 mm in some areas and on Mr Haig's evidence that he found on one disbonded sample of the truss coating at least 11 layers of coating.

300 Wattyl submitted that the recommended substrate conditions for mild steel was to blast clean in accordance with AS1627.4 to Sa 2½ minimum (AS1627.9) with a surface profile between 40 to 70 microns. Wattyl relied on evidence from Mr Tanti and Mr Cooper to the effect that Cox was instructed by Mr Spiteri of Danfab to do a whip blast, which is a class one blast. Wattyl also submitted that the fact that pre-primer coats were visible beneath the top coats shows that the blast was ineffective to clean the substrate.

301 It is highly probable that the trusses had Sigmaweld primer on them when they arrived at the premises of Cox as there was a photograph taken of (one of) the trusses which shows a distinctive reddish colour on the steel which the experts and the parties accepted indicated the presence of Sigmaweld. This photograph was taken prior to blasting by Cox in preparation for applying the Wattyl coating system.

302 Mr Stone noted the presence of Sigmaweld on the trusses in some instances post-blasting. These instances have been referred to above.

303 Mr Pellegrino took three photographs which appear to show the presence of Sigmaweld on the trusses. Other photographs taken by Mr Pellegrino of the steel substrate appear to show bare steel with no Sigmaweld present.

304 Mr Pellegrino's evidence as to the extent of the Sigmaweld remaining on the substrate is problematic. Initially he agreed with counsel for the plaintiff that the only Sigmaweld he saw on the trusses was "contained" in his photographs. He then said that "everywhere he took sampling and testing in most cases revealed the Sigmaweld."

305 It was suggested that this was not evident in the photographs. When it became apparent in the course of cross-examination that very few photographs showed the presence of Sigmaweld and it was in substance suggested to Mr Pellegrino that he was exaggerating about the presence of Sigmaweld, Mr Pellegrino said he did not take photographs of everywhere he did testing.

306 I found Mr Pellegrino to be most unconvincing during this part of his evidence. In my view, he gave an opportunistic answer which exaggerated the extent of the observed Sigmaweld; certainly only a few photographs indicate its presence on the substrate. I accept that Sigmaweld was present but not to the extent suggested by Mr Pellegrino.

307 I prefer the independent evidence of Mr Stone which in my opinion is more likely to be representative of the true position. Where Sigmaweld was observed by Mr Stone, the characteristic of the failure was an adhesive failure rather than a cohesive failure. Mr Eccleston gave evidence that if the steel substrate was contaminated by Sigmaweld he would expect to see a loss of adhesion of the Galvit primer to the steel or the Sigmaweld, which would reveal a pink layer (being the colour of Sigmaweld) on the substrate and a grey layer (being the colour of Galvit primer) on the delaminated coating. This is consistent with Mr Stone's adhesion test results which revealed a loss of adhesion to the substrate and the presence of Sigmaweld in four locations on the trusses. In my opinion, these test results serve as an indicator of the extent to which remnant Sigmaweld caused a failure of the coating and as an indicator of the extent to which poor preparation may have played a part in the delamination.

308 The Court was directed to some general observations made by Mr Haig about contamination of holding primers. Mr Haig said that holding primers were notorious for becoming contaminated during transport, storage and handling, quite apart from contamination by welding fumes, grinder spatter, oils and greases and the like. Mr Haig notes that extended periods between the application of holding primers and application of further layers of paint can result in degradation of the holding primer by UV attack or by corrosion or reaction of holding primer constituents with environmental agents. These broad observations are of general interest only.

309 In relation to Sigmaweld, the Court was also directed to the evidence of Mr Dromgool in the following passages:

If the Sigmaweld primer was retained and was not removed during the blasting process, but was sweep blasted to what is known as a "Sa1" there is a difference in permeability and porosity of the remnants of the Sigmaweld MC as opposed to a blasted surface. So in the case where the Sigmaweld MC is overcoated with a product like EP100 there would be some of the resin component from the EP100 layer that would migrate its way into the remnants of the Sigmaweld MC layer potentially making the resin to pigment ratio within the film of the EP100 slightly deficient.

...

I raise it as a possibility primarily because that is the nature of pre-weld zinc silicate primers that are used very widely in marine construction and are overcoated with zinc-rich epoxy coatings thereafter.

...

No I still believe that the Sigmaweld primer or retention of the Sigmaweld primer could well be a possible cause, and I would stress that the way this would manifest itself would not be a splitting necessarily of the EP100 off the Sigmaweld MC, thereby showing the red colour of the Sigmaweld primer, but the point of failure or the locum of the failure would be more logically through the centre of the film of EP100 as in zinc splitting.

310 Wattyl submitted that the Court should accept Mr Dromgool's evidence as to the effect of Sigmaweld and the manner in which the zinc primer would fail. It was further submitted that of the experts, Mr Dromgool had the most experience in real-life coating failures and his observations are based on years of experience and knowledge of zinc-rich primers and their characteristics. In his curriculum vitae, Mr Dromgool states that he is:

A competent and knowledgeable corrosion engineer and consultant ...over 33 years of continuous involvement with protective coatings and paint of all types for the mitigation of corrosion on a variety of substrates and materials in many parts of the world...the last sixteen years have involved extensive work in coating and lining failure analysis, forensic investigations, specification preparation, coating system evaluations, performing coatings inspection, acting as an expert witness, project management, inspector training and lead paint management as an independent protective coatings consultant.

311 Although Mr Dromgool's curriculum vitae does not expressly state that he has particular experience with zinc-rich primers, I infer from the reference to "paint of all types for the mitigation of corrosion" that his experience would include experience with zinc-rich primers.

312 Despite having extensive practical experience, Mr Dromgool is not an analytical chemist or a materials scientist, nor has he academic qualifications in chemistry. His assertions on the "difference in permeability and porosity of the remnants of the Sigmaweld as opposed to a blasted surface" and to the migration of the Galvit primer into remnants of the Sigmaweld presumably reflect genuinely held opinions. But there is no evidence that Mr Dromgool has adequate qualifications to express such opinions, nor had he undertaken any tests to verify his assertions.

313 In my opinion, in this respect Mr Dromgool advanced a speculative theory premised on the assumption that the trusses were subjected to a sweep blast (class 1). In this case, there is evidence that prior to applying the Galvit primer Cox blasted the trusses to at least a class 2, and possibly a class 2½ level blast which produced a substrate profile of between 76 and 80 microns. This exceeded the surface profile recommended in the Galvit EP100 product specification sheets of between 40 and 70 microns. This evidence, which I accept, overwhelmingly supports the conclusion that the Galvit primer was applied to a substrate profile which ensured its adhesion to the substrate except in isolated instances. In addition, I note that Mr Dromgool seemed to regard Sigmaweld as a contaminant yet failed to refer to the fact that the Wattyl product information material recommended Galvit EP100 for use with Sigmaweld.

314 For all of these reasons, I am unable to give any weight to Mr Dromgool's opinion as to the effect of remnant Sigmaweld.

315 In relation to the trusses, in my opinion, considering the evidence as a whole, the Wattyl coating system was applied in accordance with Wattyl product specifications in some places but not in others.[32]

316 The reliability of the Cox quality assurance records depends not only on the accuracy of the readings taken at the time the trusses were being coated, but also upon whether the readings taken with respect to each coat (between 50 and 80 readings) are representative samples of the thickness of the coat overall.

317 Some readings of Mr Eccleston and Mr Pellegrino show that the recommended maximum thickness of Galvit was exceeded. However, it is not possible to say with any confidence to what extent the Wattyl coating system was not properly applied in accordance with the Wattyl product specifications.

318 There was no reliable systematic survey of the dry film thickness of the Galvit primer layer and the dry film thickness of the upper coating layers.[33]

319 The issue is complicated by reason of the fact that many repairs were carried out after the coating had initially been applied and had failed. Further, the repair methodology agreed between Cox and Wattyl's representative, Mr Ash Ramlakan, involved spot blasting in affected areas and reapplying the Galvit primer onto bare steel, tying the two coats together with another Wattyl product, EP250 (which was a binding coat), and then reinstating the EP600 and Poly U750 coatings. That is to say, the remedial work involved four coatings of product instead of three coatings. The spot repairs and the variation from the original Wattyl coating system to the remedial system suggested by Wattyl's industrial sales representative may have accounted for some of the thicker readings.

320 Wattyl emphasised the highest and lowest readings which tended to give a skewed picture overall. A more representative overall picture can be gleaned from looking at average dry film thickness readings derived from reports of the experts. Mr Dove, Mr Dromgool and Mr Stone recorded average dry film thicknesses of 441, 439.5 and 456 respectively for the south truss and 409, 404 and 386 respectively for the north truss.[34]

321 Considered in this light, the average dry film thicknesses exceeded the specified dry film thickness of 375 microns, but not to the extent overall that might be suggested by reliance on the highest readings.

322 Nevertheless, in relation to the trusses, having regard to the number of readings taken by the different experts, I am satisfied on the balance of probabilities that there were occasions where the dry film thickness of the Galvit primer exceeded 75 microns under layers of coatings subsequently applied which could be described as "thick, rigid epoxy systems", contrary to the recommendation contained in the Wattyl product specifications for Galvit EP100.

323 I also accept the evidence of Mr Cooper's contemporaneous measurements taken of the surface profile of the steel to the effect that the substrate was sufficiently blast cleaned to meet the recommended substrate conditions, save in some minor respects.

324 In summary, and subject to qualifications previously expressed, in my opinion the Wattyl coating system was for the most part applied on a properly prepared substrate. I accept there is evidence that the Galvit primer was on occasions applied too thickly, that is at a thickness exceeding the recommended maximum of 75 microns and that on occasions the coatings subsequently applied were applied at thicknesses which could be fairly characterised as a "thick, rigid epoxy system". The Wattyl product specification for EP100 states that dry film thicknesses of more than 75 microns are not recommended underneath thick, rigid epoxy systems. In my opinion, it is likely that in some areas, Galvit primer had been applied too thickly in places where the top coatings had also been applied too thickly, thereby exceeding the recommended dry film thickness for the product.

325 In the absence of a comprehensive survey of coating thicknesses, it is not possible to quantify the extent of the departure from specification. Nevertheless, I am satisfied that the Wattyl coating system was not properly applied in accordance with the Wattyl product specifications on some areas of the trusses.

326 In relation to the handrails, Fulton Hogan submitted that the cohesive failure of the Galvit primer occurring in multiple locations on the handrails applied by applicators other than Cox suggests that application played no part in the coating failure.

327 Wattyl submitted that there is no evidence to show who applied the Wattyl coating system to the handrails which failed. I accept Wattyl's submissions in this regard. Eleven handrail panels failed out of 149 handrail panels. There were numerous other contractors involved in the application of the Wattyl coating system to the handrails. These included Bottles Galore of Thomastown Sand Blasting, Deer Park Engineering and Belkblast. No evidence was given by any of the contractors who applied the Wattyl coating system to the handrails. There was no evidence to connect any batch of product with any of the 11 handrails which failed. There was no evidence to connect any particular contractor or contractors with any of the handrails which failed. In the circumstances, it is not possible to make an informed assessment as to whether the Wattyl coating system was properly applied in accordance with the Wattyl product specifications.

Question 5 - Was the Wattyl coating system properly applied in accordance with the specifications under the SEITA agreement, the Danfab agreement and the Cox agreement (a) by Cox to the trusses; and (b) by other contractors to the handrails?

328 Fulton Hogan submitted that the Wattyl coating system was applied properly in accordance with the SEITA specification which, it submitted "ultimately resembled the Wattyl specification", and repeated and relied on its submissions in relation to question 4 in response to this question.

329 Cox submitted that it was never given any of the specifications relating to the SEITA agreement and that the only specification sheet that Cox had was the Wattyl product specification and a letter from Goldfields Surface Coatings which requested the work.

330 Wattyl submitted that SEITA approved of the application of Wattyl coating system to the footbridge instead of the coating system of another manufacturer (IPC) which had originally been specified under the SEITA agreement with Fulton Hogan on the basis that the Wattyl coating system was applied in accordance with the manufacturer's recommendations. Wattyl submitted that this approval forms part of the SEITA agreement and the agreement with the first defendant, Danfab. In the agreement between Fulton Hogan and Danfab, Danfab agreed that in the course of carrying out the works on the footbridge all steel would be coated as "per revised specification (provided in tender)". In short, these specifications were the same as those agreed to by Fulton Hogan with SEITA. Wattyl submitted that the Wattyl coating system was not applied to the trusses in accordance with the Cox agreement because the Cox agreement required a "whip blast to AS1627.4 Sa 2.5" (which meant that the standard of blast specified was to Sa 2½) and a system build of 375 microns and that Cox was clearly in breach of the Cox agreement in that regard. In substance, Wattyl's position was that to the extent that the substrate of the trusses was blasted to less than a Sa 2½ class blast and to the extent that the system build exceeded 375 microns the Wattyl coating system was not properly applied in accordance with the Cox agreement.

331 There was evidence given by Mr Tanti to the effect that the specification document supplied to Cox specified a whip blast to Sa 2½. Mr Tanti said there was no such thing; that a whip blast was a class 1 blast, not a class 2½ and this was clarified by Mr Spiteri who said it was only necessary to carry out a whip blast. However, in my view, it is clear that due to the high performance of the blasting equipment used by Cox, the level of blast greatly exceeded a class 1 blast. As Mr Tanti put it, Cox did a very good blast, much more than Cox was asked to do. Mr Tanti's evidence in this regard was corroborated by Mr Cooper's evidence.

332 Accordingly, to the extent that Cox performed the blast of the substrate exceeding Mr Spiteri's instructions to do only a class 1 blast, it was strictly not in accordance with Danfab's specification because it exceeded what was asked for.

333 In summary, I broadly accept Wattyl's submissions to the effect that the Wattyl coating system was not properly applied to the trusses in accordance with the specifications under each of the agreements to the extent that the system build exceeded 375 microns as originally coated. Further, in my opinion, Cox did not comply with Mr Spiteri's instruction to carry out a class 1 blast on the trusses.

334 In my view, it is unproductive to further address this question. The more important issue is whether any non-compliance with specifications caused or contributed to the defects in the coatings on trusses.

335 In relation to the handrails, I accept Wattyl's submission that there was no evidence to allow any party to state whether or not the 11 handrail panels which failed (out of a total of 149 handrail panels) were coated in accordance with any particular agreement or specification, because it is not possible to correlate the handrail panels which failed with any one of the five sub-contractors. Except in respect of Cox, there was no evidence of the agreements made with, or specifications provided to, these sub-contractors.

Question 6 - If no to question 4 or question 5, did such application cause or contribute to the defects in (a) the trusses; and (b) the handrails?

336 In relation to the trusses, the views of the experts were polarised. Dr Garnys said that the weakness of the Galvit primer was the cause of the premature failure of the Wattyl coating system irrespective of the quality of the application.

337 Mr Eccleston gave evidence that although he saw "a case of overbuild" this was not the direct cause of the cohesive failure, though movement and increased stresses in the coating would, over time, further accelerate the failure.

338 Mr Haig and Mr Dromgool state that the incorrect application of the Wattyl coating system caused the truss defects and the handrail defects. On the basis that he could not find any apparent fault with the Galvit primer, Mr Cook concluded that "one of the more probable causes is due to issues related to application".

339 On application issues, Wattyl relied primarily on the evidence of Mr Dromgool including his reports of 27 May 2009 and 10 December 2010.

340 Essentially, Mr Dromgool concluded that the application of the coatings by Cox to the trusses was very poorly executed, resulting in both grossly excessive and inadequate film builds; that there was coating over abrasive and grit and that there were uncoated areas. Mr Dromgool gave evidence that the variation of the film build and other non-conformances were such that it was almost certain that the cause of the problems up to the point he saw it was application related. As previously noted, Mr Dromgool also expressed an opinion that the presence of Sigmaweld on the substrate "could well be a possible cause" of failure which would also manifest itself as zinc splitting through the centre of the film.

341 In his first report dated 27 May 2009, Mr Dromgool states that the application of the coating system was extremely and unpredictably variable with wide inconsistencies in film build and coverage; that on many steel sections there was embedded surface grit that had not been removed after abrasive blasting though it was not possible to establish exactly where in the coating system this material was located. Further, he states that the film thickness readings of 800 microns were exceedingly common on both trusses and that a principal concern for overbuild is a progressive tendency for the coating system to fail due to a gradual increase in the cumulative stress in the coating system above the level of the zinc primer layer as the thick film builds of epoxy and polyurethane slowly become brittle as they lose the plasticiser in each coating layer. Mr Dromgool states that all forms of zinc primer are susceptible to shear stress along the plane of film when overcoated with thick film, rigid paint systems; that this is due to the high zinc particle loading and minimum level of resin binder in the coating and often this ultimately manifests itself as cohesive film fracture or film splitting. Mr Dromgool states that the site damage and its result indicated very strongly that there was a fundamental weakness in the applied coating system on the trusses and that a major cause of the then recent and probable future breakdown would be directly related to overbuild in the existing system.

342 In his second report dated 10 December 2010, Mr Dromgool states in relation to film thickness that specific instructions on a product data sheet are there for good reason; that the failure to carry out the instructions may result in failure of the coating system or may prevent the coating system from performing properly; that the reason for a cautionary note in the Wattyl product data sheet for Galvit EP100 about dry film thicknesses above 75 microns was due to the risk of cohesive failure of the zinc primer; that the consequences of overbuilding the zinc film may be apparent quite soon after application, for example, within a few weeks, but it could easily take quite some time - even years - depending on how long the overcoat system takes to lose flexibility and increase in its internal stress levels; and that it can be hard to predict how long before the coating system will crack and then disbond within the thickness of the zinc primer.

343 In Mr Dromgool's opinion, Cox failed to observe and abide by the instructions contained on the Galvit EP100 product data sheet and failed to exercise good workmanship by failing to control the zinc film thickness and by having a large percentage of the total surface area of the trusses having dry film thicknesses that ranged to over 1200 microns. According to Mr Dromgool, the failure of the coating system applied by Cox on the trusses was a "totally predictable consequence" in the circumstances.

344 In relation to preparation of the truss substrate, Mr Dromgool concludes from the Cox quality assurance records that all Cox appeared to have done was perform a class 2½ blast to the welds and a class 1 blast to the balance of the steel work. Mr Dromgool concludes that performing the full blasts only on the welds and joints and the class 1 blast on the balance of the steel work was a major non-conformance with the project specification. According to Mr Dromgool, the poor integrity of the Cox applied coating system on the trusses and the mode of disbondment that occurred was mostly caused by the failure to perform a full class 2½ blast to the steel work combined with the gross overbuild that occurred of the Galvit primer and subsequent high build topcoat layers.

345 In a nutshell, Mr Dromgool attributed the likely cause of the failure of the Wattyl coating system to poor preparation of the substrate and excessive film build.

346 I have earlier referred to the issue of preparation of the substrate. I have accepted the evidence of Mr Cooper to the effect that even though the specification called for a class 1 blast, in fact the substrate was blasted in excess of that level due to the power of the high pressure blasting equipment used by Cox.

347 According to Mr Cooper, the specification was a bit of a contradiction because Mr Spiteri had specified a "whip blast to AS11607.4 Sa 2½"; that a whip blast is a class 1 blast and that a 2½ is a 2½ class blast. Mr Cooper said that the finish on the truss shows that it was "in fact a 2½ blast". Later he said "much closer to a 2½" blast. According to Mr Cooper, to control the blast with the equipment they used would have been "very, very hard" because of the "volume there that we're using and the pressure that we're using"; that the blast actually performed was at a higher standard than had been requested and that it would have taken longer to blast to a lower level. According to Mr Tanti, Cox did much more than they were asked to do in relation to blasting; that it was "quite a good blast" more of a class 2 and that most of the pre-primer came off.

348 I note that the product specifications for Galvit EP100 state that all surfaces to be coated must be clean, dry and free from chalking and contamination and that oil and grease should be removed from all surfaces. The product specifications do not say that pre-primer must be removed; rather it specifies a recommended level of blast clean (Sa 2½ minimum and a recommended surface profile of between 40-70 microns).

349 The objective evidence in this case based on Press-O-Film tape subsequently measured by a micrometer, suggests that the surface profile of the trusses not only achieved the surface profile recommended by the manufacturer but exceeded it. I have no reason to doubt the accuracy of these measurements. The quality assurance reports prepared by Mr Cooper indicate that the surface profile of the trusses exceeded 75 microns and that at the time of application the trusses were free of dust, spent abrasive, oil, contaminants, and that the surface of the trusses had been degreased with a solvent wash and were "okay to paint". I do not accept that Mr Cooper signed off knowingly on inaccurate quality assurance records. Mr Cooper did acknowledge one error, but he adamantly denied a suggestion that in order to get the job done quickly, he simply wrote figures which were within an acceptable range according to the manufacturer's data sheet in order to justify the work continuing.

350 Mr Cooper was an experienced paint inspector and quality assurance officer. He had worked in the protective coating and abrasive blasting industry for approximately 36 years during which he had been hands-on in blasting and painting and had graduated from there to being a quality assurance inspector and supervisor. He had worked on the ANZAC project on eight frigates at Williamstown and on a desalination plant at Wonthaggi in regional Victoria. He had provided consulting services to Cox for 20 years on and off. He was not an employee. In my opinion, it is extremely unlikely that Mr Cooper would have put his reputation on the line by knowingly signing off on inspection reports which were inaccurate or falsified.

351 I also formed a favourable impression of his demeanour when he gave evidence. He was responsive to questions. He made numerous concessions under cross-examination. In my view, if Mr Cooper had any doubt about the adequacy of the level of blasting to the trusses he would have made appropriate concessions.

352 Finally, on the issue of preparation of the substrate, I note that the Wattyl product data for Galvit EP100 does not specify or recommend that pre-primer must be removed - it merely specifies that the substrate should be clean and indicates a recommended level of blast.

353 In this context, I note that in the areas he examined, Mr Eccleston found no evidence of adhesion failure from the substrate. According to Mr Eccleston, the fact that the zinc was adhering to the steel substrate is an indication that the surface preparation was satisfactory; and that inadequate surface preparation would have resulted in adhesion loss between the steel substrate and the Galvit primer. I accept this evidence which tends to suggest that the surface preparation was satisfactory. In my opinion, the surface preparation of the substrate of the trusses was generally satisfactory and on the balance of probabilities did not cause the widespread failure of the Wattyl coating system.

354 On the issue of overbuild, it is necessary to moderate the extent of the criticism levelled at Cox for departing from the specification. Mr Dove was one of the first experts who examined the trusses in early January 2009. Mr Dove appears to have been supplied by SEITA with a VicRoads specification which specifies a total dry film thickness of the coating of 325 microns with a micaceous iron oxide intermediate coat. This was not the correct specification. Cox had been provided with and had worked to a specification which required a dry film thickness of 375 microns and an intermediate coating which did not contain micaceous iron oxide.

355 Mr Dove's observations based on the results of dry film thickness readings when compared with the specified dry film thickness are, as a consequence, based on an incorrect assumption which significantly distorts the extent of non-conformance. For example, Mr Dove contends that most of the structures had high (more than 50% greater than specified) dry film thickness as indicated by the average thickness and the standard deviation. Mr Dove was not called to give evidence nor were his calculations revisited or his conclusions reconsidered in light of the fact that Cox had been instructed to work to a dry film thickness of 375 not 325 microns. In the circumstances, it would be unsafe to draw conclusions from such observations.

356 On the other hand, taken at face value the average dry film thickness of the readings taken by Mr Dove of truss A (441 microns) and of truss B (409 microns) shows that the average dry film thickness was less than 20% over the dry film thickness of 375 microns specified.

357 Nevertheless, Mr Dove's readings do suggest there was excessive dry film thickness on parts of the trusses.

358 In his report to SEITA dated 23 January 2009, Mr Dove expresses the opinion that the weakness in the epoxy was within the epoxy zinc layer and that as the intermediate coat continues to harden it will develop internal stress with delamination occurring after several years; that the weakness may be seriously detrimental to the long term performance of the coating system; and that excessive thickness and under thickness may be detrimental to the long term integrity of the coating system.

359 In Mr Stone's reports of 27 May 2009 and 9 June 2009, it is acknowledged that extensive remedial works had been carried out on the trusses after the initial application of the protective coatings. Mr Stone states that excessive thickness on various members will also be subjected to breakdown "well before the 25 year life expectancy" if left unattended, without specifying how long that breakdown would take to occur.

360 In Mr Dromgool's first report dated 27 May 2009, Mr Dromgool states that failure of the coating system might not manifest itself for some years because failure will occur as a result of a gradual increase in the cumulative stress in the coating system above the zinc primer layer as the coatings slowly become brittle as plasticiser is lost in each coating layer. In an example given of a previous investigation Mr Dromgool had carried out on a coating system using a zinc-rich epoxy and polyurethane coating system, the problem took about five or six years to appear (notably the failure on the bridge in that instance occurred only in the zones where the dry film thickness was above 600 to 650 microns and that in all other areas that had dry film thicknesses below this level were in excellent condition).

361 In his second report dated 10 December 2010, Mr Dromgool states that the consequences of overbuilding the zinc film may be apparent quite soon after application, for example within a few weeks, but it could easily take quite some time, or even years depending on how long the overcoat system takes to lose flexibility and increase in its internal stress levels. In the example he cited in his first report from his own experience, the problem took five or six years to appear. There was no evidence that he had had any experience of a failure of a coating system within a matter of weeks as a consequence of overbuilding of coatings. Accordingly, I give less weight to his opinion that failure from overbuilding a coating can appear within a few weeks.

362 Mr Eccleston gave evidence that cohesive failure would occur over a longer period than within a few weeks. He stated that it would require a period of time for stressors such as expansion, contraction and seasonal contractions to start to build up and for cohesive failure to become apparent. Mr Eccleston thought it was unlikely in the short term for the Galvit primer to fail cohesively within a few weeks as a consequence of overbuilding, stating "I think it's unlikely in the short term with a well formulated product. Over a period of time it would start to show up". Mr Eccleston conceded that the consequence may be apparent soon after application, even though it was unlikely.

363 Dr Garnys gave evidence that even after taking into account overbuilding of the coating, the coating should have performed for a decade or more. Dr Garnys drew attention to the statement in Mr Dromgool's first report where he had said that the coating failure had taken about five or six years to appear. Dr Garnys gave evidence that failures of this kind are measured in years rather than spontaneously, as occurred on the Wattyl test panels, what he saw with the handrails at DH Corrosion and what he believed was the case with the trusses where the failure occurred before they were installed. Dr Garnys conceded that spontaneous self detachment of coatings could occur, but that this would have been an adhesive failure which was not observed. Mr Dromgool said it had been a concern in this case that the manifestations of failure were so rapid.

364 In my opinion, the evidence strongly indicates that cohesive failure of coating systems of the kind considered in this case, if caused by overbuilding of coatings, is likely to occur over a time span measured in years rather than in days or weeks.

365 In the present case, Cox completed the coating of the trusses on 23 December 2008. Failure of the coating on the trusses was observed before completion of the application of the Wattyl coating system. According to Mr Tanti, Danfab had to carry out additional works on the trusses by installing kick plates and that whilst Danfab personnel were at Cox's premises putting the kick plates on, the coating was found to be delaminating; and that the paint failure started while both trusses were at Cox premises.

366 It was suggested to Mr Tanti that no issue was raised with the truss coating until after the trusses had gone to site at Burwood and that no repair work had been done on the trusses until they had been at the Burwood site. Mr Tanti denied these suggestions and gave evidence that problems arose with the coating prior to the trusses leaving Cox's premises.

367 Mr Cooper gave evidence that the first indication of a problem with the coating on the trusses was possibly two weeks after the coating had been applied when it started delaminating. Mr Cooper also disagreed with the suggestion that no repair work was done until after the trusses had gone to the Burwood site for the first time.

368 I accept that the cohesive failure of the Wattyl coating system became apparent within two weeks of its application, resulting in the need to carry out extensive spot repairs at Cox's premises prior to the transportation of the trusses to the Burwood site. Repair works took place in consultation with Wattyl's representative, Mr Ash Ramlakan. After the trusses were transported to the site at Burwood the problem re-emerged with more delamination occurring. At some point Mr Ash Ramlakan gave suggestions as to a four-coat repair system to repair the delaminating paint. According to Mr Cooper, it became obvious that the system had had a major breakdown.

369 I accept that the paint was delaminating even in areas which had been repaired on truss B when it was returned to Cox premises for rectification following repairs using the Wattyl four coating repair system. In my opinion, there was rapid delamination of the coating system, including rapid delamination of the repaired coating system.

370 In this case, not only has the Wattyl coating system failed very rapidly, but it appears that the failure has been widespread and to have occurred indiscriminately in areas which were thickly coated and areas which were thinly coated.

371 In Mr Eccleston's tests, the adhesion of the coating was "quite poor" in areas where the total dry film thickness range from 292 microns to 516 microns in the examples tested. In his laboratory examinations, Mr Eccleston found that in respect of the three samples of paintwork taken for laboratory examination, delamination of the coating occurred irrespective of the coating thickness.

372 Considering the evidence overall, I am satisfied on the balance of probabilities that overbuilding of the coatings did not cause the failure of the Wattyl coating system.

373 For completeness, I note that other criticisms levelled at the standard of workmanship of the coating applied to the trusses were put to Mr Cooper. In particular, Mr Cooper was asked about Mr Dove's comments:

• That dry spray (overspray) was observed on 30% to 40% of the "visual faces" of the trusses. Mr Cooper said that dry spray was aesthetically not pleasing and that because it sits above the top coat would account for some of the high dry film thickness readings. He disagreed with the 30% to 40% estimate, but said that with the structure of the size and complexity of the trusses it would be expected there would be some overspray.

• That excessive orange peel was observed on 3% to 5% of the faces of the trusses. Mr Cooper agreed there were some examples of orange peel on the surface which would be expected and would also cause high dry film thickness readings. He disagreed with the 3% to 5% estimate.

• That runs and sags were observed on 2% to 3% of the members. Mr Cooper agreed there were some runs and sags, but estimated the percentage at 1%. He said they would also cause extremely high dry film thickness readings in the order of 1000 or 900 microns if a probe was put on the area of a drip or sag. He accepted that runs and sags would be evidence of poor application but only to a minor degree.

• That there were inclusions, possibly dirt, blast media and weld spatter in the coating. Mr Cooper disputed there were any inclusions in the paint.

374 In my opinion, whilst there were clearly some justifiable criticisms in the standard of workmanship on the trusses, it is neither necessary nor appropriate to make findings about the extent of these defects. Neither party saw fit to call oral evidence from Mr Dove and to the extent that the issues were contested, it is not productive to comment further. In my view, this was a sensible, forensic decision because it is evident that none of these defects caused the widespread failure of the coating on the trusses.

375 Mr Dromgool's criticisms of the application of the Wattyl coating system from his inspection of the trusses on site on 22 May 2009 were also put to Mr Cooper. In particular, Mr Cooper was asked about Mr Dromgool's comments:

• That the application of the coating system were extremely variable and characterised by wide inconsistencies in the film build and the coverage. Mr Cooper disagreed with that extreme but conceded that a structure of the size and complexity of the trusses will have different film builds and that the orientation of the steel section to the spray operator would make a difference in the dry film thickness. He disagreed with Mr Dromgool's conclusion that "the coating application operator was not particularly experienced in applying high performance coating products to within a narrow film thickness range, over all surfaces of a complex fabricated item and achieving good coverage".

• That a number of pockets where scale, weld slag and mill scale still appeared to be present. Mr Dromgool concluded firstly that the surface preparation in these areas had probably not been fully performed and that the film build in the hard to access areas was also grossly inadequate. Mr Cooper disagreed that the surface preparations in these areas had not been fully performed or that the film build in the hard to access areas was grossly inadequate. Mr Cooper gave evidence that Mr Dromgool was mistaken; that what he saw was in fact grindings from welding, grinding and drilling which had all taken place on the bridge after it had come to site at Burwood; that when the trusses came to the site, panels were installed on the trusses which required 12 to 20 holes to be drilled per panel so that the panels could be attached; that this caused drill swarf to fall on the structure which settled in the wells and looked a mess.

• That bare steel in some of these areas was corroding. Mr Cooper denied that what Mr Dromgool saw was corrosion of the steel; that in fact what Mr Dromgool saw was rust stain which had come from drill swarf or the grinding dust that had settled on the paint and that moisture in the air had caused it to corrode and stain the top coat of the paint. Mr Cooper said he observed this himself. Mr Cooper said that these were only rust stains which had occurred as a result of engineering works on site. He rejected the criticism.

• That there were pockets of accumulated abrasive (in quantities equalling about half a teaspoon or more in some locations). Mr Cooper gave evidence that considering there had been some spot blasting on the truss, he would expect that some stray abrasive might have got into those areas but that it was not in an area where a coating was being applied. He agreed that if there were pockets of accumulated abrasive material, that would suggest poor application practice.

• That embedded foreign materials can provide a major compromise to the integrity and durability of an applied coating system, because they provide a weak zone where moisture and oxygen can more easily permeate through the reduced effective film build. Mr Cooper agreed with this proposition in principle, but said that there was a lot of welding, grinding and drilling going on which could cause an inclusion of weld slag into paint. He said there were no inclusions in the paint when Cox coated the trusses. When shown a photograph and asked whether he would accept that it showed some embedded abrasive that had been painted over, Mr Cooper denied this and said the photograph showed melted weld slag that had adhered to the coating which no amount of blasting would remove. According to Mr Cooper, other photographs which showed abrasive not completely blasted or coated were work in progress and not defects, but reflected work done on site in the course of repairing the coating.

376 In broad terms, I accept Mr Cooper's evidence that the trusses looked a lot worse than they actually were, because works had been carried out on site and because repair works were in progress at the time Mr Dromgool did his inspection. In any event, I am not satisfied that the application issues were the cause of the failure of the Wattyl coating system on the trusses.

377 Mr Stone's criticisms were also put to Mr Cooper. In particular, Mr Cooper was asked about comments made in his first report dated 27 May 2009.

• That if there were a large number of low dry film thicknesses on various members of both trusses that would result in premature coating breakdown well before the 25 year life expectancy of the coating. Mr Cooper accepted that this would probably be the case.

• That excessive thickness on various members will also be subjected to premature breakdown if left unattended. Mr Cooper disagreed with this comment.

• That there are documented cases where excessive film builds of zinc under epoxies puts undue stress on the underlying zinc primers causing them to fail. Mr Cooper agreed with the statement but said there was no evidence that the zinc coating on either of the trusses was in excess of what was in the specification, other than by measurement with an instrument that was not designed for that purpose.

• That under thickness on a large number of members on both trusses would be detrimental to the life expectancy of the coating system. Mr Cooper disagreed with the reference to a large number of members, but otherwise agreed with the proposition.

• That the contractors' QA documentation had been viewed and the dry film thickness results conflict with what was obtained during the inspection and by other inspection authorities. Mr Cooper denied the inspection reports were inaccurate. Mr Cooper said:

"we don't know what was happening with the - with the zinc layer. It obviously was delaminating, or splitting within itself. And once again you would not expect that the readings would be same as when I took my readings if that zinc layer was in fact splitting".

• That the contractor has failed to apply the coatings in accordance with the manufacturer's recommendations. Mr Cooper disagreed with this proposition for the same reasons.

• That overall the protective coatings applied to the two bridge trusses by Cox does not meet the specification requirements, manufacturer's instructions or the relevant standards. Mr Cooper said he agreed with that at the time Mr Stone inspected the trusses and took readings for the reasons previously mentioned (i.e. that the zinc layer was splitting within itself giving different readings to those taken when Mr Cooper took his readings).

378 Mr Cooper was also asked about comments made in the second report by Mr Stone dated 9 June 2009. In particular, that the coating contractor had failed to adhere to the contract specifications in all aspects of the coating procedures. Mr Cooper denied this.

• That the coating contractor had failed to achieve the specified class of blast prior to the epoxy zinc application. Mr Cooper said that was not correct and that he did not know how Mr Stone would have arrived at that statement. I interpolate that it is apparent that Mr Stone wrongly assumed that a class 3 blast was required and relied (understandably) on Cox's inspection reports which showed that the main structure had a light class 1 blast, presumably not realising that the Cox blasting equipment was so powerful that a much higher class of blast was actually achieved on most of the structure.

• That the coating exceeded dry film thickness far beyond the manufacturer's recommendations to 50% of the bridge. Mr Cooper disagreed and said a much smaller percentage had film thicknesses beyond the specification. Mr Cooper referred to his earlier reason and noted that the readings were taken in June almost six months after the trusses had been painted. Mr Cooper said "Who knows what the zinc coating was doing underneath the other two layers during this period of time".

• That there was an under film thickness to approximately 40% of the truss. Mr Cooper disagreed with that. He repeated his earlier statement that with a structure of that size and complexity it would be expected that there would be a few readings under specification, but not to the extent of 40%. Mr Cooper disagreed that the quality of the application work was poor regardless of paint quality, although he did agree that it was not good coating practice to have a very broad range of dry film thicknesses.

379 In my opinion, considering the evidence as a whole, the only aspect of application which is shown to have caused or contributed to the defects in the coating on the trusses relates to the inadequacy of blasting in areas detected by Mr Stone as a consequence of carrying out the adhesion tests. Mr Stone found four instances out of more than 70 adhesion tests where Sigmaweld was still evident on areas of the top chords of the trusses where there was no evidence of a blast profile.[35] Dr Garnys was shown images from Mr Stone's report that revealed pre-weld primer on the substrate. Dr Garnys acknowledged that in those instances it appeared that the coating was "adhesively coming off the Sigmaweld".

380 Mr Eccleston also acknowledged that there might be a problem if Sigmaweld had not been removed and that in those circumstances he would expect to see a loss of adhesion of the primer to the steel. Mr Eccleston agreed that this would reveal a pink layer on the substrate (being the colour of Sigmaweld) and a grey layer on the peeled away part (being the colour of the Galvit EP100).

381 There was minimal lay evidence regarding the extent of Sigmaweld remaining on the substrate. Mr Michael Gilchrist, the supervisor at McElligott's Partners who re-blasted and re-coated the trusses, said that in his blasting work he saw a grey primer, a red oxide primer and a black filmy coating on the smaller section of the footbridge that he worked on.[36] Mr Gilchrist did not endeavour to quantify the amount of or give the location of the coatings that he had observed. In the circumstances I have found his evidence to be of little assistance.

382 No analysis was done by any expert to quantify the amount of Sigmaweld primer remaining on the steel substrate of the trusses and the percentage remaining is unknown. In my view, the fact that only four out of over 70 adhesion tests carried out on the north truss and that (apparently) no Sigmaweld primer was revealed as a result of adhesion tests on the south truss suggests that adhesive failure from the Sigmaweld, to the extent that it remained on the trusses occurred only to a minor extent.

383 On those areas, I find that surface preparation was inadequate and resulted in the failure of the Galvit primer to adhere to the substrate (as distinct from cohesive failure within the layer of primer).

384 In relation to the handrails, this issue does not arise as there was no evidence of Sigmaweld or any other pre-weld primer remaining on the substrate of the handrails. Indeed, there being no evidence to show who applied the Wattyl coating system to the handrails which failed, or how it was applied, I am unable to say whether issues with application caused or contributed to the defects in the handrails.

Question 7 - If two or more of the defendants caused or contributed to the loss or damage, how is liability to be apportioned between the defendants?

385 Fulton Hogan submitted that the damages sustained by Fulton Hogan arising from the failure of the Wattyl coating system was caused solely by Wattyl and that the issue of apportionment therefore does not arise. Cox submitted that apportionment can only occur if the alleged defective application caused or contributed to the coating becoming defective. Cox submitted that an apportionment would only be appropriate if the evidence suggested that the application by Cox of the paint to the trusses in some way contributed or aggravated the defect. Consistently with the plaintiff, Cox submitted that if the paint was found to be defective, no apportionment should be made as between defendants.

386 Cox further submitted that (in substance in the alternative) in favour of apportionment

• against Danfab for supplying paint which was defective or past its shelf life and for Danfab's failure to provide it with all the technical specifications in the head contract between Danfab and Fulton Hogan; and
• against Wattyl allowing or permitting the manufacture, sale or supply of paint which was either defective or past its shelf life.

387 Wattyl submitted that Cox or Danfab should be responsible for any loss established by Fulton Hogan. In particular, it was submitted that Danfab had a contractual responsibility to supply, fabricate, coat and deliver the main steel truss bridge and handrails and to engage subcontractors to perform aspects of this work. It was submitted that Danfab did not supervise these subcontractors or take any, or any effective, measures to ensure that the subcontractors did the work in accordance with the agreement between Danfab and Fulton Hogan. Further, Danfab instructed Cox to do a level 1 blast rather than a class 2½ blast required by the specification and the Wattyl product data sheet. In relation to Cox, Wattyl submitted that the most likely problem with the trusses was the application work carried out by Cox.

388 Wattyl submitted that if it is to be liable for any proportion of the loss and damage it should be minimal.

389 In my opinion, the proportionate liability provisions in Part IVAA of the Wrongs Act 1958 (Vic) apply in the circumstances of this case.[37] None of the parties contended otherwise. Fulton Hogan's claim is for economic loss in an action for damages (brought in tort and contract) arising from a failure to take reasonable care as contemplated in s 24AF of the Wrongs Act. [38]

390 In my view, Cox and Wattyl are concurrent wrongdoers because their actions or omissions caused the loss or damage that is the subject of the claim.[39] It is not at all clear that that there was any operative omission by Danfab. Danfab had engaged Cox to coat the trusses. Mr Tanti gave evidence that Cox had an accredited in-house quality control system and was recognised as highly skilled in the heavy coating industry. I accept that Cox were competent coating specialists and would not require supervision by Danfab in carrying out their task. Mr Eccleston gave evidence that assuming there had been a class 2½ blast to the substrate of the trusses after construction and before the application of Sigmaweld, then a class 1 (whip) blast would have been sufficient to achieve the same blast profile as a class 2 ½ blast. Mr Cooper gave evidence that Mr Spiteri of Danfab had asked for a whip blast as the truss structure had already been blasted to class 3 and coated with 15 microns of Sigmaweld. Whilst I do not propose to make findings on the basis of hearsay, it is not possible without further evidence to be categoric that Mr Spiteri was wrong to specify a class 1 blast instead of a class 2 ½ blast. It appears this may depend on whether the trusses had already been subjected to blasting at an appropriate level.

391 Putting that issue aside, in my opinion, as events transpired, any acts or omissions of Danfab did not cause or contribute to the loss or damage that is the subject of the claim and therefore Danfab is not a concurrent wrongdoer within the meaning of s 24AH(1) of the Wrongs Act.

392 Fulton Hogan's primary submission was that the evidence overwhelmingly points to the Galvit primer being the sole cause of the defects giving rise to the damage and that no contribution should be required from Cox. I disagree. In my opinion, poor preparation of the substrate caused or contributed to the loss, though only to a minor extent. Given the extent of adhesive failure of the coating, I consider that it is just in all the circumstances to limit the liability of Cox to an amount reflecting 5% of the loss suffered by Fulton Hogan with respect to the adhesive failure of the coating on the trusses and to limit the liability of Wattyl to an amount reflecting 95% of the loss suffered by Fulton Hogan with respect to cohesive failure of the coating on the trusses.

393 As Fulton Hogan has failed to establish liability with respect to the handrails, the question of apportionment between defendants with respect to the handrails does not arise.

Question 8 - What is the extent of the loss and damage suffered by Fulton Hogan (a) in relation to the trusses; and (b) in relation to the handrails.

394 The claim for loss and damage totals $1,896,887.92. Fulton Hogan divided its claim into the categories set out below:

A. Site Management & Supervision, Facilities $39,332.14 $726,998.14

B. Damages from 16 January 2009 $27,500.00 $302,500.00

works, redundant works due to coating $34,639.22 $381,031.38

D. Blasting and Painting of trusses $31,409.54 $345,504.88

of Chain Boxes to allow for recoating $5,932.95 $65,262.45

395 Wattyl submitted that the damages claim by Fulton Hogan is excessive, unjustified, and was formulated with a casual attitude and contained errors.

396 To establish this aspect of its claim Fulton Hogan relied principally on the evidence of Mr Dean Riha, Civil Department Manager, who was involved in management of the project during the period from about December 2008 until its completion at the end of October 2009.

397 Mr Riha gave evidence that he compiled the material relied on by Fulton Hogan to substantiate its loss and damage; that in doing so he reviewed all of the invoices relevant to the coating issue to satisfy himself that they were part of the claim, spoke to relevant personnel, in particular Mr Adem Cengiz, Project Manager and Mr Andrew Hume, Site Manager with respect to their involvement on site and the time they expended on the project, and had regard to the terms of the contract project timelines, minutes of meetings and correspondence from the client. In my opinion, Mr Riha's approach to the collation task was methodical and, for the most part, accurate. The general criticism that he took a casual attitude to the task is, in my opinion, unfounded. The fact that there were a number of errors (which were readily conceded by Mr Riha when challenged) is not surprising given the scope and complexity of the task. These errors did not detract from my impression of Mr Riha as honest and professional in every respect.

398 Wattyl relied on expert evidence given by Mr Rodney Alsop, Senior Quantity Surveyor of WT Partnership. Cox adopted the costings advanced by Wattyl and the evidence given by Mr Alsop as part of its case. Wattyl disputed the quantum claimed under categories A, B and C. Mr Alsop did not dispute the amount claimed under categories D, E and F, though some items under these categories were challenged by Wattyl in final submissions. It is convenient to deal with categories A, B and C separately.

Category A: Site management and supervision, facilities

399 This claim for $726,998.14 inclusive of GST is largely made up of labour costs for Project Managers, a Site Engineer and a Site Manager and Foreman for a period of 37 weeks commencing from 6 June 2009 and concluding on 22 September 2009 and off-site overheads during that period.

400 The claimed delay period of 37 weeks was calculated by determining when the trusses first arrived on site on 6 January 2009, coated and ready to be installed to the date they were again on site re-coated and ready to be installed on 22 September 2009. Mr Riha conceded he had initially made an error and claimed 38 weeks instead of 37 weeks.

401 Mr Alsop gave evidence that the delay period should be 10 weeks instead of 37 weeks, because Fulton Hogan had negotiated an extension of time to the Date for Practical Completion to 21 August 2009. In my opinion, this analysis fails to take into account the fact that Fulton Hogan incurred costs on site and off site during the period to the end of the extension of time (21 August 2009) and was not recompensed by SEITA for those costs.

402 Mr Alsop was not in a position to dispute Fulton Hogan's contention that it was unable to recover costs for the delay from SEITA despite being granted an extension of time for a revised Date for Practical Completion of 21 August 2009. I accept the evidence from Mr Riha that the extension of time was given without a corresponding entitlement to claim costs from SEITA. This evidence is consistent with a letter dated 29 April 2010 from Linking Melbourne Authority (formerly SEITA) to Fulton Hogan which objectively reflects that position.

403 Although there were significant construction works being carried out on the site during this period which involved re-design of guardrail footings that interfered with a drainage system, those works were not on the critical path of the project; that is to say they did not cause the delay. Mr Riha took into account the fact that minor concurrent works were being carried out during the relevant period. When Mr Riha was asked whether he was surprised that the initial coating could be done in about two weeks and the re-work took about three months, Mr Riha said:

I guess it doesn't surprise me. It is what it is. McElligotts were working seven days a week at some stages and I personally witnessed them working long hours and certainly productive hours, so I can't say that McElligotts are- were inefficient or working to extended [sic] the duration if you like...we were certainly pushing McElligotts. McElligotts were certainly aware that we had to get the truss up as quickly as possible, so they were working the best they could. There was certainly more work in the re-coating because...the actual re-blasting phase was something that was a bit unknown to McElligotts because although there was delamination in some areas, there was still adhesion in other areas. So they were unsure as to how long it would take to remove the paint, and that was a considerable effort. They also had to erect the scaffolding and the plastic encapsulation around the whole of the trusses in comparison to what Cox Coating would have done - just simply placing the trusses and painting them. So there was quite a lot of preparation work involved in the re-coating exercise as opposed to the first coat.

404 Overall, I am satisfied that for the purposes of calculating loss and damage, the delay period of 37 weeks is the appropriate period. In reaching this conclusion, I am satisfied that it was reasonable for Fulton Hogan, in the first instance, to endeavour to repair the coating. Had those repairs solved the problem, substantial costs and delay would have been avoided. As it transpired, the repairs were unsuccessful and it was necessary to re-blast and re-coat the trusses. Mr Alsop did not take issue with the time taken by McElligotts to re-blast and re-coat the trusses. In the circumstances, Fulton Hogan did what it could to mitigate its loss and damage. The plaintiff is only required to act reasonably to mitigate loss.[40]

405 Mr Adem Cengiz gave evidence that he worked full time managing the project from October 2008 until June 2009, then from June 2009 spent 50% of his time managing the project to completion.

406 His duties included attending all meetings with the client SEITA, monitoring the progress of the works, reviewing site diaries, day to day contact with the site by telephone three to five times a day (maybe more), keeping abreast of all long term issues, visiting contractors' premises including Cox's premises and attending meetings in relation to the coating issues.

407 After 29 June 2009, Mr Cengiz spent 50% of his time on the project based on a 10 hour day with Saturday as a normal work day (58 hour week). This translated to about 29 hours per week. Fulton Hogan's claim is for 20 hours per week during this period rather than 29 hours to make allowance for time spent on concurrent activities unrelated to the coatings issue. In my opinion, this is a conservative and reasonable approach.

408 In respect of the period from June 2009 onwards, Mr Cengiz gave evidence that as project manager he was responsible for whatever was happening with the project; that he would review site diaries on a daily basis and would communicate with regard to the progress of the painting or blasting of the trusses; that he would be involved in procurement assisted by the project engineer including "signing out", final approvals and commercial decisions; that he had direct communications with the client SEITA; that he needed to keep the client informed and needed to be engaged in daily programming. Mr Cengiz said with the construction project it is necessary to know what is happening on a day to day basis.

409 Mr Andrew Hume gave evidence that he worked on the footbridge project in the capacity as a site engineer from January 2009 until the Date of Practical Completion of the project in November or December 2009. He reported to the project manager Mr Adem Cengiz. Mr Hume went to the site most mornings and would spend a couple of hours there and would then go into the office and continue working on such things as programming, liaising with clients or traffic management issues such as obtaining memoranda of consent from VicRoads to allow speed reduction or the closure of traffic lanes as required. Mr Hume estimated that during the delay period he worked on the Burwood project for at least eight hours a day, that he was working on some other small projects mainly on long weekends or doing odds and ends, but that he spent eight hours of the day working on the footbridge project and that he normally worked ten to twelve hours each day, mostly six days but up to seven days a week. For most of the time during the delay period there was a site manager/foreman on the site who reported to Mr Hume and Mr Cengiz. At times when there was no one to look after the site on a day to day basis, Mr Hume attended the site and carried out the role of site manager. In that role he supervised contractors and ensured that all safety and quality systems were in place and operating. When he worked as site manager, Mr Hume arrived on site at between 6.00am and 6.30am and left at about 6.00pm or 6.30pm.

410 Fulton Hogan's claim with respect to Mr Hume is for a 37 week period on the basis of a minimum 40 hour week, assuming an eight hour day, not claiming the ten hours actually worked each day to allow for any minor concurrent unrelated works that were performed during the delay period. I accept Mr Hume's evidence that he normally worked a ten hour day six days a week. In my opinion, Fulton Hogan's claim for eight hours, making allowance for some time spent on concurrent activities unrelated to the coatings issue and scaling back from a potential six day per week to a five day per week claim, is a conservative and reasonable approach.

411 Fulton Hogan also claims for the cost of a site manager/foreman. Mr Hume gave evidence that Mr Neil Robbins fulfilled this role during the period from January 2008 until around May 2009. Mr Robbins' job was to look after the site; supervising works on site making sure everything was being done safely, checking environmental controls and traffic controls on the main highway. Mr Robbins reported to Mr Hume and the project manager Mr Cengiz. Mr Hume said it was necessary to have a foreman on site. In order to mitigate supervision costs, Mr Robbins was moved to a different project in June 2009 and Mr Hume took over his role.

412 In his revised report of 5 August 2011, Mr Alsop challenged the reasonableness of Fulton Hogan's claim for these labour costs, on the basis that in his opinion there was not sufficient work to occupy Fulton Hogan personnel for the time claimed. There is a major difficulty with Mr Alsop's analysis. He assumed incorrectly that both trusses were returned to Cox's premises in February 2009 and were not returned until 8 April 2009, that there was no work being carried out on site. In fact only one truss was removed from the site. The other truss remained on site while rectification works were carried out there.

413 In Mr Alsop's original report, he stated, among other things, that without formal declarations that the project manager, site engineer and the site manager/foreman were working full time on the project, he was unable to agree to the amounts claimed for labour.

414 When Mr Alsop was informed that Mr Hume had given evidence that he was engaged on the project full time from 6 January to 22 September 2009; that Mr Cengiz had given evidence that he was engaged on the project full time from 6 January 2009 until mid-June 2009 and then 50% of his time from mid-June through to completion on 22 September 2009; and that Mr Riha had given evidence that Mr Robbins, the site foreman, worked full time on the project from 6 January 2009 until some time in May 2009, Mr Alsop said he could not disagree with that evidence.

415 Mr Alsop accepted that it was a requirement under the contract for the site to be supervised daily by an experienced and qualified engineer.

416 All labour costs were charged at the day work rates provided for in the contract for the relevant roles. Mr Alsop accepted that the rates were reasonable.

417 Mr Alsop also accepted that the 8.9% claim for off-site overheads was reasonable. Indeed he adopted the rate for the purposes of his own calculation. He accepted that the issue with the off-site overheads was not so much the rate but rather the period of time over which the calculation was made. In its written submissions, Wattyl asserted that administrative costs would have substantially diminished because of the reduced level of activity on site. The assertion was not supported by any analysis. I am not satisfied that off-site overheads would necessarily have diminished. The project still had to be administered through to its conclusion, notwithstanding a reduced level of activity on site whilst the parties tried to arrive at a solution to the coating failure. In light of Mr Alsop's concession that the rate claimed for off-site overheads was reasonable, I propose to allow the claim in full.

418 Mr Alsop conceded the reasonableness of all of the remaining items claimed under category A with the exception of item 5 which related to container hire for two containers for the 37 week delay period. Mr Alsop accepted that it was standard practice to allow contractors to leave their tools for the duration of a job on site and to provide a facility for that purpose. Mr Alsop accepted that there would have been tools and equipment needed to repair and re-coat the truss which remained on site during the period from February through until June 2009 and that the contractors engaged to remove all of the coating and re-coat the trusses would have required somewhere to store plant and equipment and tools overnight. Mr Alsop accepted that Fulton Hogan would have items to store on site, but said that Fulton Hogan would not have had "major equipment like expensive levels" to store on site. In my opinion, it was not unreasonable for Fulton Hogan to have a container on site as a storage facility for its own equipment for use as and when required and a separate container to serve as a storage facility available for the use of contractors who might come and go during the course of the works. Accordingly, I will allow this item at the rate claimed for the 37 week period.

419 In his report of 5 August 2011 Mr Alsop notes that Fulton Hogan's claim for hire pool stores for a 19 litre water container was duplicated under the claim for water and under the claim for sundry items resulting in an overstatement of the claim by $326.37. Mr Riha readily conceded there was duplication of these items in his spreadsheet which had resulted in double counting.

420 This claim has two components, a claim for the recovery of damages equal to the amount of liquidated damages retained by SEITA under the contract for delay ($195,000) and a claim for an amount retained by SEITA ($80,000) in lieu of Fulton Hogan providing SEITA with a 25 year warranty with respect to the coating on the handrails to the bridge.

421 Under cross-examination, Mr Alsop accepted that it would be reasonable for Fulton Hogan to claim compensation for the liquidated damages imposed on Fulton Hogan.

422 In a payment certificate dated 29 April 2010, SEITA deducted $195,000 from payments to Fulton Hogan claiming liquidated damages under the terms of their contract for delay from 30 January 2009 to 30 October 2009. This amount was calculated on the basis of a $5,000 per week rate for liquidated damages specified under the terms of the contract for a period of 39 weeks.

423 Fulton Hogan submitted that if a delay period of 37 weeks is accepted, then Fulton Hogan should be entitled to recover the liquidated damages retained by SEITA.

424 Wattyl submitted that Fulton Hogan had reached a commercial settlement of various disputes with SEITA and if Fulton Hogan had made a commercial agreement which did not meet its expectation on claims which were due, that disappointment could not be attributed to Wattyl. I am not persuaded by this submission.

425 But for the coating failure which delayed the works for 37 weeks, SEITA could not have deducted liquidated damages for delay under the contract. The fact that later negotiations about other matters caused some amounts to be set off against the amount retained as liquidated damages does not detract from the fact that the loss and damage was suffered by Fulton Hogan. As Mr Riha said:

The delay period from the coating is what gave rise to the liquidated damages, so... the reason that we had liquidated damages in the first place was because the delay in completing the project, which was a direct result of the coating.

In my opinion, the fact that in negotiations between Fulton Hogan and the client SEITA Fulton Hogan was able to set off other claims against the amount retained by SEITA is not to the point. Fulton Hogan is not making a complaint about the quality of the bargain made with respect to other claims which had settled. The complaint is that the amount which SEITA has retained was only able to be retained because of the delay caused by the coating failure. The only necessary adjustment to the claim in my view is to reduce it from 39 weeks to 37 weeks, which reflects the period attributable to delay caused by the coating failure. Accordingly, I propose to allow $185,000 for the proven delay period of 37 weeks.

Retention of $80,000 by SEITA in lieu of Fulton Hogan warranty

426 By letter dated 16 November 2009, Linking Melbourne Authority (formerly SEITA) wrote to Fulton Hogan regarding the provision of a 25 year warranty protective coatings on the railings on the footbridge.

There is approximately 15 linear metres of railing that has been coated with the International Protective Coatings system. The majority of the railing remains coated with Fulton Hogan's original system which was found to contain a weakness in the epoxy zinc layer.

I have been advised that the original coating system will continue to harden and will develop internal stress with delamination resulting after several years. I am also advised that the estimated replacement cost for the protective coatings on the railing is of the order of $80,000 (based on present day figures).

Please advise whether Fulton Hogan will be able to provide the specified 25 year warranty on the railing protective coatings. Alternatively LMA would be prepared to accept a deduction of an appropriate amount from the contract sum to cover the potential future coating replacement costs, should Fulton Hogan choose to pursue this option.

428 Subsequently in April 2010, Linking Melbourne Authority (formerly SEITA) withheld $80,000 "until such time as this matter is resolved".

429 Under the terms of the contract with SEITA, Fulton Hogan was required to provide "from the Painting contractor and the Coating supplier a twenty-five (25) year corrosion protection warranty for the system".

430 Mr Riha gave evidence that Fulton Hogan has not been able to obtain a warranty and has therefore been unable to comply with its contractual obligation and that SEITA continues to retain $80,000 on account of this non-compliance.

431 I am not satisfied that Fulton Hogan has made out this claim. Whilst it is clear that Fulton Hogan is currently incurring a cost arising from the fact that SEITA has withheld $80,000, it is not clear to me that Fulton Hogan was ever in a position to give a 25 year corrosion protection warranty for the system from the coating supplier.

432 The early correspondence from Goldfield Surface Coatings, being Wattyl's authorised distributor, refers to the coating system being suitable for a project which required a 25 year life span, but does not go so far as to provide a warranty to that effect. I am not satisfied that Wattyl is responsible for Fulton Hogan's failure to comply with its contractual obligation to provide a warranty from the "Coating Supplier". Nor, if it matters, am I satisfied as to the quantum of this element of the claim. On any view, Fulton Hogan would not be entitled to damages for the entire $80,000, because the retention amount may be returned by SEITA at any time. In a letter from Linking Melbourne Authority (formerly SEITA), Fulton Hogan is advised that the $80,000 is withheld "until such time as the matter (of the warranty) is resolved". There is no indication as to when or if that will occur. Realistically the claim is a claim for the loss of use of the $80,000 and with an allowance to cater for the contingency that the capital sum may not be returned. No evidence was called by Fulton Hogan which would enable an evaluation of the loss of use of the amount involved or which would enable a sensible and appropriate allowance to be made for contingencies. Accordingly, in my view, this head of loss and damage is not capable of informed assessment. For both reasons I disallow this claim.

Category C: Transport of truss for repairs, rectification works and redundant works

433 The first component of this claim is for costs associated with transporting truss B from the site to Cox's premises for repairs and the cost of those repairs, payment of $114,731.38. Fulton Hogan paid the amount of $114,731.38 to Cox in order to secure the release of truss B after it was returned to Cox's premises for repairs and after those repairs were effected.

434 Fulton Hogan submits that this payment was made in order to promptly complete the works and mitigate its position. Fulton Hogan submits that it was left with no other alternative but to pay Cox. Wattyl submits that Fulton Hogan had a contract with Danfab, but no agreement with Cox; that Cox claimed a tradesman's lien over truss B in order to obtain payment for work performed on the truss; that Fulton Hogan denies that it has any liability to Cox to make any payment with respect to these works and that the payment was made ex gratia to obtain the immediate release of the truss. Wattyl submits that it should not be held liable for monies which Fulton Hogan was not obliged to pay to Cox.

435 On this issue Mr Carecos, then state manager of the construction division of Victoria for Fulton Hogan, gave evidence of his dealings with Mr Tanti of Cox regarding the re-delivery of truss B back to the site. Mr Carecos gave evidence that Mr Tanti asked for payment of the full monies which he was owed by Danfab ($104,301.25) for coating works originally performed on the trusses and ($40,437.72) for repairs effected to them both on site and at Cox's premises. Mr Carecos gave evidence that Fulton Hogan:

paid the money essentially to get the truss back as Mr Tanti was refusing to release the truss to us. We understood that, you know, it was a difficult situation and we were under pressure from SEITA to get the job completed, so we took a commercial decision to take that - to pay the money and get the truss.

Upon payment of $104,301.25 plus GST or a total of $114,731.38 the truss was released.

436 Under cross-examination, Mr Carecos agreed that Fulton Hogan only paid the money in order to keep going with the project. Mr Alsop agreed that Fulton Hogan was left with no choice but to pay Mr Tanti for the release of the truss. Mr Alsop also agreed that in the circumstances it was reasonable for Fulton Hogan to seek to recover the amount of $114,000 it had paid to Cox; that Fulton Hogan could not leave the truss sitting at Cox's premises indefinitely while liquidated damages were being incurred under the SEITA contract; and that it was necessary for Fulton Hogan to move promptly if it was to get the project completed in a timely manner.

437 In my opinion, it was reasonable in the circumstances for Fulton Hogan to pay Cox $114,000 even if Cox was opportunistically purporting to exercise a tradesman's lien over truss B for the works originally carried out on the trusses. Fulton Hogan is obliged to mitigate its loss. In my opinion, the steps Fulton Hogan took were practical and reasonable in the circumstances. Had it not done so and left truss B sitting at Cox's premises while the dispute was being resolved, it is likely that Fulton Hogan would have been criticised for failing to mitigate its loss. Decisive action was warranted. Decisive action was taken. In my opinion, Fulton Hogan is entitled to recovery of this amount.

438 The second component of this claim is for cost of the repairs effected to the trusses on site and on truss B after it had been returned to Cox's premises.

439 Wattyl contends that Fulton Hogan cannot recover monies which it says it has no obligation to pay, has not in fact paid and presumably will not pay in the future.

440 Fulton Hogan arranged for and requested the repairs to be effected to the trusses. On the face of it, Fulton Hogan is liable to pay the reasonable cost of the work requested unless the repairs were required as a consequence of defective workmanship by Cox. In light of my findings that the delamination was caused (predominantly) by defective primer, in my opinion Fulton Hogan is liable to pay an amount representing 95% of these repair works. Whilst Fulton Hogan has not yet paid for these repairs, in my opinion it does have an obligation to do so and it should be indemnified against that liability. No issue was taken by Wattyl as to the reasonableness of the charges for these repairs. There is no dispute that the work was performed. I infer in the circumstances that the rates were reasonable.

441 Fulton Hogan claims for the cost of transporting truss B from the site back to Cox's premises on 17 February 2009 for repairs and for the return of the repaired trusses to the site on 16 April 2009 after repairs were effected.

442 There was no dispute about these costs having been incurred. The costs associated with haulage of the truss to and from the site were a necessary incident of effecting repairs to truss B. I propose to allow this item.

443 Fulton Hogan claims for the cost of proofing of stool supports at Cox's premises, the cost of changing over beams due to non-compliance of Cox's supports and the cost of the use of replacement supports to be used during repairs. There is no dispute that these expenses were incurred.

444 Mr Riha gave evidence that SEITA required that supporting stools at Cox's premises be proof-engineered before their inspector would go under the truss to inspect the coatings. SEITA only required this when it inspected the coating repairs. Mr Riha said that SEITA were happy to inspect the trusses in December 2008 without "a fine tooth comb" check; that they could carry out an inspection without getting underneath the truss. But once the coating failures occurred, the inspection regime intensified such that the inspectors wanted to inspect underneath the trusses. Previously they were happy to inspect without going underneath the trusses. The beam and stools were subsequently design checked and supplied and used. The beam and stools were returned after use. There was a charge of $20,000 for these items. Mr Riha said the products themselves cost more than that, but that was the rate agreed and that was the amount which Fulton Hogan paid. The reasonableness of the $20,000 fee was not challenged. In the circumstances, I infer the amount charged was reasonable. Wattyl submits that the stools and the design work for them and the beam were needed for coating whether or not the coating was done well and irrespective of paint quality. I reject this submission. The unchallenged evidence of Mr Riha, which I accept, is to the contrary. I propose to allow this item and the other related items for truss supports.

445 Fulton Hogan claims the costs which relate to rotating the truss on site for installation in late April 2009, splicing the trusses together in early May 2009. This work involved traffic management expenses, scaffold hire, lifting and rotating the truss to provide access for the splicing process, working platforms, splicing materials and labour. In late May 2009, jacking equipment, crane and transport equipment for jacking and skating the structure was required for the intended bridge lift.

446 Wattyl submitted in substance that, given the history of the coating problems on the trusses, there was no reasonable basis for Fulton Hogan to decide that the trusses were ready to lift into position on 27 May 2009. Wattyl submits that Fulton Hogan should not be able to recover the cost of making abortive plans concerning the trusses in light of the history of difficulties with the coatings and submitted there was no evidence that the coatings had materially improved to the point where Fulton Hogan and the client SEITA would have been satisfied to erect the trusses in their then current state; that Fulton Hogan was not justified in making contractual arrangements to organise the erection of the trusses; and that such arrangements should not have been made at all or only on a conditional basis so that they could be cancelled without financial penalty.

447 On this issue, Mr Carecos gave evidence that this part of the claim related to work which was done in anticipation of putting the truss into position before that work was cancelled. Mr Carecos conceded there were obvious problems with the coating on the trusses from the time they came to site in January 2009; that attempts were made to solve the problem with various investigations by experts and that concerns were being expressed by the client SEITA in early February 2009. He accepted that a decision was not taken until late May 2009 to completely re-coat the trusses. In response to the suggestion that this decision should have been taken sooner, Mr Carecos said:

That's clearly not the way we looked at it at the time. We wanted to give our painter every opportunity to remedy the product. We were getting advice that was different to that from our painter... he [Mr Tanti] was indicating that the painting of the truss was acceptable and he could remedy it... it was certainly a difficult process, there's no doubt, and we had a number of people telling us different things. But at the end of the day we needed to provide the opportunity to the installer to remedy the product, and simply removing it - you know, making the decision to remove it straight away I believed was the incorrect decision to start with. Now, as it turned out, the product did fail and we had to do it. But I don't necessarily agree with the - with your assertion that we should have made that decision earlier.

448 On the specific issue of whether there was a proper basis in May 2009 to go ahead and arrange for the lifting and installation of the trusses over the highway, Mr Carecos said:

We certainly had to make a call that - we had to make a call on when we wanted to erect - erect these trusses and we certainly made that call and made provisions in case, you know, we could get it to a situation where they were acceptable...so I guess what I'm saying is that we had to make preparations a fair way in advance to lift these trusses. They could not happen a week beforehand, we had to get approval from road authorities, from SEITA itself and therefore we had to make those preparations quite a way in advance for that lift to occur.

449 As to the suggestion that at the time these arrangements were made Fulton Hogan did not have any reasonable basis to think that the trusses would be a in state satisfactory to erect over the highway, Mr Carecos said:

No, that's not correct. My - my view of the world was that we were in a position to be able to correct the truss at that stage. That was my state of mind... right up until we made the decision to re-coat the truss...SEITA were also not directing us.

In summary, Mr Carecos said that all the costs claimed were incurred, and denied they were incurred prematurely in the circumstances. He said that arrangements for the lifting of the trusses would generally be made at least four weeks or up to two months in advance.

450 Mr Riha gave evidence that Fulton Hogan decided on 27 May 2009 or thereabouts that it was not going to go ahead with the lifting of the truss and it was going to re-coat both trusses.

451 Mr Riha gave evidence as to the lead up to late May when the trusses were scheduled to be lifted as follows:

So the preparation is quite extensive, it's quite a big logistical exercise. As far as procedures in the lift itself, safe work method statements, and the procedural side of things needs to be ironed out. All the guys need to go through risk assessment workshops so everyone understand the risks involved, there is also a fair amount of logistics in just confirming craneage and transport, because they're items that are not regular, or not regularly in the industry, you need to book those in advance to make sure you've got access to that equipment... I would suggest a minimum of three or four weeks given that notification as to public and community as far as the closure goes and it might be up and - the reality is they will probably notify and talking to OD group well in advance, probably a couple of months in advance and would have locked in dates I would say a minimum of three weeks in advance, maybe four.

452 Under cross-examination, Mr Alsop agreed that between 6 January and 27 February 2009 there were numerous meetings between representatives of Fulton Hogan, Danfab, Wattyl and Cox regarding the failure of the coating; that there were a number of inspections and reports were compiled; that spot repairs were being undertaken during the period; that as at 27 February 2009 Wattyl had prepared a repair specification and that repairs were being undertaken to the truss which remained on site and the truss which had been taken back to Cox's premises; and that as at mid-April 2009, it was anticipated that the spot repairs might have worked with a view to the truss being installed some time in early to mid May 2009. Mr Alsop also accepted that installation of the truss over a public highway was not an easy task requiring a considerable degree of time and planning to get all of the appropriate approvals and organising cranes and labour and occupational health and safety and VicRoads approval.

453 On this issue, I accept the evidence of Mr Carecos, Mr Riha and Mr Alsop. In my opinion, Fulton Hogan, in conjunction with its client and the painting contractor, made genuine attempts to resolve the issues with delamination of the coating which included attempting repairs following a repair specification prepared by Wattyl. Presumably Wattyl thought the defects were repairable when it provided the repair specification to Cox.

454 With the benefit of hindsight, it may have been better to immediately strip the trusses of all coatings and re-coat the trusses with coating products supplied by another manufacturer as ultimately occurred. But at the time the decision was made, Fulton Hogan obviously did not have the benefit of hindsight. With one exception, in my opinion all the expenses were reasonably incurred notwithstanding they proved to be redundant. The exception relates to the fee for cancellation of a photographer. Whilst I understand that Fulton Hogan would want to photograph the lifting event, in my opinion it was an elective expense which Fulton Hogan is not entitled to recover.

Invoice 0122 from North Fitzroy Constructions Pty Ltd

455 In final submissions, Wattyl stated that it should not be liable for form work removal said to have been carried out by North Fitzroy Constructions Pty Ltd, on the basis that it was not clear from the relevant invoice precisely what work had been carried out or when.

456 The relevant invoice is dated 25 January 2009 and refers to works having been completed at the site and refers to the supply of labour and tools to install the deck to the truss. It includes a hiring charge for "Superslims" for the period 12 January 2009 to 25 January 2009. The details of the invoice were entered into Fulton Hogan's journal entry under "FHC Victoria East Civil Burwood Highway Pedestrian".

457 There is some lack of precision. I note that the claim is only with respect to parts of invoices rendered by North Fitzroy Constructions Pty Ltd for large amounts (Part 122 day works $19,053.34 and Part 127 day works and materials $34,595.20). Mr Riha has extracted from the total of $53,648.54, a component for removal of the deck formwork before transportation of truss B to Cox for repairs, allowing $8,000 for that component. Mr Riha allowed the same amount for re-installation of the formwork after the truss was returned to the site, again a component from much larger invoices (Part 139 day works and materials $15,589.40 and Part 141 day works and materials $27,357.70 or a total of $42,946.00). Whilst this claim does involve some matters of judgment and degree, I accept Mr Riha's evidence that he had reviewed all of the invoices relevant to the coating issue and satisfied himself that they were part of the claim after speaking to relevant personnel. It was not suggested by Wattyl that it was not necessary to remove the truss decking formwork before truss B was transported back to Cox's premises or that the decking formwork would not have to be re-installed after the truss was returned to the site. In addition, I note that Wattyl's expert Mr Alsop did not suggest that this was an unnecessary or unreasonable expense. In my opinion, Wattyl is liable for the decking formwork removal and re-installation costs. I propose to allow this item.

458 Wattyl also submitted that Fulton Hogan should not be able to recover the cost of removal and reinstatement of cladding panels. This relates to the claim for removal of architectural panels by Access Platforms Systems Pty Ltd and their reattachment in September 2009. Wattyl submitted that the condition of the trusses was such that it was premature and unreasonable to put the external panels on before the coating was satisfactorily completed and that if the panels had not been put on until the truss coatings were completed and ready for cladding then there would have been no need to reattach the panels.

459 I do not accept that it was premature and unreasonable to attach the cladding panels when they were attached. Again, Wattyl's submissions are made with the benefit of hindsight. Genuine efforts were being made to avoid delays on the project and it was not unreasonable to carry out these works. Fulton Hogan were not to know that delamination, even in areas which had been repaired using a repair specification recommended by Wattyl, would continue. Again, I note that Wattyl's expert Mr Alsop did not say that these works were premature, unnecessary or unreasonable. I propose to allow this item.

Do damages awarded for loss arising from property damage constitute supply for the purposes of GST?

460 Fulton Hogan seeks an order for damages of $1,896,887.72 inclusive of $145,685.74 to allow for GST.

461 Wattyl contends that any damages awarded to Fulton Hogan should be exclusive of any allowance for GST for two reasons: first, that neither an award of damages or payment by a judgment debtor of a judgment sum constitutes a "supply" and therefore does not attract a GST under the provisions of the A New Tax System (Goods and Services Tax) Act 1999 (Cth) ("GST Act");[41] secondly, the plaintiff had failed to prove an entitlement to have GST included.

Taxable supplies

You make a taxable supply if:

(a) you make the supply for consideration; and

(b) the supply is made in the course or furtherance of an enterprise that you carry on; and

(c) the supply is connected with Australia; and

(d) you are registered, or required to be registered.

However, the supply is not a taxable supply to the extent that it is GST-free or input taxed.[42]

463 Senior counsel for Wattyl referred to and relied upon Ruling GSTR 2001/4, which states that 'supply for consideration' is determined by three criteria. There must be a supply, there must be a payment and there must be a sufficient nexus between the supply and the payment.[43] In Ruling GSTR 2001/4, the Commissioner of Taxation concludes that neither a court ordering a party to pay damages pursuant to a judgment,[44] nor the extinguishment of a judgment debt by its payment by a judgment creditor constitutes a supply for GST purposes.[45]

464 Counsel for Fulton Hogan submitted that compensation was being sought for what was ultimately a supply of services where GST was paid for those services, so GST should be included in the award of damages. Counsel for Fulton Hogan referred to Peet v Richmond (No 2), where Hollingworth J stated with reference to Ruling GSTR 2001/4, that:

payment made under a court order can constitute consideration for an "earlier supply". Whether there is a sufficient nexus for a court order to constitute consideration for the earlier supply is a question of fact in each case. The fact that payment is made by compulsion under a court order rather than in the ordinary course of commercial dealings does not affect the liability to GST.[46]

465 Although Peet v Richmond (No 2) involved a quantum meruit award for services rendered at an earlier point in time, it was submitted that in this case, orders made by the Court would constitute consideration for an 'earlier supply' and so may give rise to a GST liability.

466 Wattyl submitted that Peet v Richmond (No 2) is distinguishable on the facts in that it involved a quantum meruit claim where the plaintiff had provided goods or services to the defendant and no payment had been made. In that case, the plaintiff had sought an allowance for GST to be included in the order because otherwise the award of damages would have been diminished by 1/11th which was less than the court intended due to GST needing to be remitted.[47]

467 Wattyl submitted that Fulton Hogan has already paid to third parties the amounts which it now seeks to recover as damages, including GST. It was submitted that, due to the usual practice of businesses routinely setting off GST collected against GST paid, unless there was evidence to the contrary, the Court should infer that Fulton Hogan has already claimed an input tax credit in respect of the GST amounts it paid. Counsel submitted that the only evidence that Fulton Hogan had not claimed GST related to an amount of $40,437.72 (exclusive of GST) owed to Cox, since it denied liability for these payments. Wattyl submitted that an amount which included GST paid would result in double recovery of that component of the claim.

468 I agree that Peet v Richmond (No 2) is materially distinguishable. As a consequence of the defective coating Fulton Hogan had to pay third parties for goods and services plus GST with respect to those goods and services. I accept that in the ordinary course, Fulton Hogan would have received an input tax credit to the extent of the GST it had paid against the GST it collected on services it had supplied to third parties. Without the delamination of the coating, Fulton Hogan would not have needed to pay the GST on those goods and services and would not have received an input tax credit equal to that payment. In my opinion, it is reasonable to infer, and I do infer, that Fulton Hogan has already been compensated by the input tax credit it would in the ordinary course have received equivalent to the GST paid. Damages are compensatory; they are to recompense the plaintiff for loss suffered in paying third parties for goods and services. I accept that if damages were awarded for an amount equivalent to the GST component already paid, then Fulton Hogan would be doubly compensated.[48]

469 I further accept Wattyl's submission that the only evidence that Fulton Hogan had not claimed GST is with respect to the amount of $40,437.72 (exclusive of GST) currently owed to Cox but unpaid. Fulton Hogan has denied liability for these payments. I have already allowed 95% of the costs of these repairs ($40,437.72 by .95 or $38,415.83) as part of the claim under category C above. In my opinion, an allowance for GST must be included in the award of damages with respect to this amount as Cox has an outstanding entitlement to the payment of GST of $3,841.58 for these repair costs. It follows that I propose to allow $3,841.58 for this item and that no allowance should be made for GST already paid.

470 In final submissions, counsel for Fulton Hogan contended that there was no notice that this legal issue would be raised and that, unlike in Peet v Richmond (No 2), the plaintiff had not had 'ample opportunity' to address the matter. In the circumstances, counsel for the plaintiff submitted that a declaration could be made that any GST awarded in damages and remitted within a certain period of time by the Commissioner of Taxation be refunded to the defendants.[49] Having specifically claimed an entitlement to an allowance for GST as a component of its loss and damage, Fulton Hogan always had the burden of making out this aspect of its case at law. Save with respect to the unpaid repair costs, it has failed to make out its case on this issue. No request for an adjournment was requested by Fulton Hogan to enable it to further consider the matter or to make additional legal submissions. As a matter of discretion I decline to make the declaration sought.

Question 9 - Did Cox give a warranty to SEITA and Fulton Hogan with respect to the costs and expenses incurred in repairing any defects to the trusses?

471 Pursuant to terms of settlement entered into between Fulton Hogan and Cox, Cox agreed to provide Fulton Hogan with a warranty "correctly addressed to Fulton Hogan in accordance with the format already communicated to Cox by Fulton Hogan and to Fulton Hogan's satisfaction".

472 Subsequently, Fulton Hogan provided a warranty and indemnity to SEITA and SEITA accepted the warranty and indemnity on the terms and conditions set out in a Deed of Warranty and Indemnity. Clause 2 of the Warranty and Indemnity is in the following terms:

(a) Cox Coatings represents and warrants that the application of the paint coating system applied to the Bridge pursuant to the terms of the Contract will not breakdown, corrode or fail (other than due to fair wear and tear), due to application, for a period of twenty five years commencing form the date of this deed (Warranty Period).

(b) If, during the Warranty Period, as a result of the application of the coating system it corrodes, breaks down or fails, Cox Coatings will promptly repair any relevant areas identified by SEITA as affected by the corrosion, breakdown or failure (as the case may be), at no cost to SEITA, or fully indemnify SEITA for its costs and expenses incurred in repairing the affected areas as a result of any application deficiencies.

473 Mr Tanti of Cox gave evidence that he signed the Cox warranty and said that the warranty was for workmanship. I agree. By its terms, the warranty covers break down, corrosion or failure of the paint coating system as a result of any application deficiencies. It follows that the warranty does not cover costs and expenses incurred in repairing defects due to failure of the materials used to coat the trusses.

474 Having regard to my earlier findings as to proportionate liability, Cox is required to indemnify Fulton Hogan under the Cox warranty for the extent of 5% of the loss and damage suffered by Fulton Hogan with respect to the trusses.

475 In this case, I have found that the batch of Wattyl Galvit EP100 primer used by Cox to coat the trusses on the footbridge was, on the balance of probabilities, defective and caused widespread delamination of the coating which ultimately necessitated re-blasting and re-coating of the trusses. Some delamination of the coating applied by Cox was attributable to inadequate preparation of the substrate on the trusses and by making due allowance, I have apportioned liability 5% to Cox and 95% to Wattyl.

476 In relation to the handrails, there is insufficient evidence to determine whether the defects to the coatings on 11 out of 149 handrails was caused by defective product or poor application practice by other contractors.

477 Accordingly, Fulton Hogan has proven liability with respect to the trusses but not with respect to the handrails. The proven loss and damage with respect to the trusses may be summarised as follows:

1. Transport of truss for repairs, rectification works, redundant works

E. Reinstallation of cladding and decking truss form work etc

478 Having regard to the apportionment of liability, Cox shall pay $79,690.29 (or 5%) of the proven loss and damage and Wattyl shall pay $1,514,1115.54 (or 95%) of the proven loss and damage.

[1] Such as DH Corrosion Pty Ltd. These other contractors are not parties to this proceeding.

[2] This is perhaps not surprising. According to Dr Garnys, cohesive failure within a coat is very rare unless there is a formulation or a mixing issue.

[3] The only independent expert who gave evidence of having particular expertise in zinc-rich coatings was Mr Eccleston.

[4] Cf Wattyl's expert Mr Haig did not attend the site.

[5] [2008] NSWCA 276 [66], [67] (Hodgson and McColl JJA (Young CJ in dissent but agreeing on this point)).

[6] [2005] NSWSC 149; (2005) 53 ACSR 110, 184 [329].

[7] With SSPC (the Society for Protective Coatings USA) and NACE (National Association of Corrosion Engineers USA).

[8] ACA (Australasian Corrosion Association) and NACE (National Association of Corrosion Engineers USA).

[9] Joint Report of Dr Garnys, Mr Haig, Mr Dromgool and Mr Cook dated 20 October 2011 ("Joint Report").

[12] In three out of four measurements the thickness of the primer remaining on the substrate represented less than one third. In one instance the thickness was slightly more than one third (38%).

[13] Provided to Dr Garnys on 23 September 2011 by Mr Cook of Wattyl.

[14] Which Mr Stone incorrectly assumed was 350 microns when it was in fact 375 microns.

[16] An average of 205 micron when the maximum micron thickness was 300 micron and the minimum micron thickness was 250 micron.

[19] No submission was made to the contrary by Wattyl.

[20] E.g. "fatally flawed", "a more honest sketch of Dr Garnys' scheme", "totally refute" and "Dr Garnys' has totally ignored".

[21] Cf Mr Haig's opinion that the corrosion products represent deposits of corrosion between successive layers of Galvit primer product.

[22] It is not clear from the email from Mr Bo Padkjaer to Mr David Cook dated 23 March 2011 when the retained sample designated as "New A" was manufactured.

[23] Orders [3] and [5] made by the Honourable Justice Almond on 12 September 2011.

[25] E.g. Truss A test area W# 5 with a dry film thickness of 782 microns which did not fail; cf Truss A test area W# 4 with a dry film thickness of 518 microns which did fail.

[26] Cf the panels drying in the Wattyl laboratory or in a refrigerator.

[27] Cf some non-specific remarks made by Mr Cook, Mr Lowe and Mr Pellegrino.

[28] Exhibit D2-3: Letter from Goldfields Surface Coatings dated 1 July 2008.

[29] Mr Tanti also advanced a theory that when the Galvit primer cohesively failed the coating would expand and get thicker. This theory was not put to any of the experts and I give it no weight.

[30] Though there is an obvious error in the average dry film thickness recorded in the Cox inspection reports with respect to the second coat on the north section of the trusses which Mr Eccleston thought was an obvious transposition error. Mr Cooper gave evidence that it was "obviously a typo... obviously a mistake".

[31] Cf the use of the an alcometer 456 for determining total coating thickness which was the appropriate instrument for that purpose.

[32] Found in Wattyl product information sheets for the respective coating products.

[33] Mr Dove took approximately 20 measurements of the primer and intermediate and topcoat layers but see the discussion about reliability of these readings above. Mr Bartlett states that his measurements were consistent with Mr Dove's measurements without stating how many measurements he took. Mr Stone took extensive measurements of total dry film thickness.

[34] The reading of 386 is an average of the average readings on the east (321 microns) and west internal (486 microns) sides of the north truss in Mr Stone's first report.

[35] See the detailed discussion of adhesive failure and Sigmaweld above.

[36] The reference to the smaller section of the bridge was a reference to truss B.

[37] Wrongs Act 1958 (Vic) Part IVAA ("Wrongs Act").

[38] Wrongs Act 1958 (Vic) s 24 AF.

[39] Wrongs Act 1958 (Vic) s 24AH.

[40] McGregor on Damages (Sweet & Maxwell, 18th ed, 2009) 273-275; British Westinghouse Electric and Manufacturing Co Ltd v Underground Electric Railways Co of London Ltd [1912] AC 673, 689.

[41] A New Tax System (Goods and Services Tax) Act 1999 (Cth).

[42] A New Tax System (Goods and Services Tax) Act 1999 (Cth) s 9-5.

[46] Peet v Richmond (No 2) [2009] VSC 585, 16 [76].

[48] See also Gagner Pty Ltd Trading As Indochine Cafe v Canturi Corporation Pty Ltd [2009] NSWCA 413, 25 [151] and Merringtons Pty Limited v Luxottica Retail Australia Pty Ltd [2006] VSC 525, 12-13 [40]-[41].

[49] Relying on the formulation of Peet v Richmond (No 2) [2009] VSC 585, 17 [82].