FLENDER (AUST) PTY LTD v WARMAN INTERNATIONAL LTD SUPREME COURT OF NEW SOUTH WALES COURT OF APPEAL HANDLEY and BEAZLEY JJA and SHEPPARD AJA 7 and 18 August 1997, 18 June 1998 [1998] NSWCA 81 Contract — Breach of contract — Causation — Non-compliance with requirements of technical codes — Whether non-compliance causative of loss. Referees' report — Review of referees' determination — Procedural fairness — Supreme Court Rules Pt72. In this appeal, the appellant was sued by the respondent for damages for, amongst other things, breach of contract for the supply of certain gear reducers, which included bevel pinion shafts which were to be used by the respondent in a slurry pump installation. The proceedings were referred to a referee for inquiry and report on the whole of the proceedings pursuant to Pt72 of the Supreme Court Rules. The referees found that the gear reducers as supplied by the appellant to the respondent failed in service after a relatively short service life. Eight failures were particularised in the report, two of which involved failure of the shafts. The referees found that the shafts did not comply with the requirements of a technical code known as the FEM rules. The appellant sought a review of the referees' determination. On the review, Giles CJ Comm D held that the referees erred to the extent that they based their determination on the FEM rules, and held that their finding that the shafts had inadequate fatigue strength capacity when assessed according to another technical code, ASME, should be adopted. The consequence of this finding was that the appellant was in breach of its contract with the respondent. This was not challenged. The appellant submitted, however, that the mere finding of breach of contract does not result in a finding that that breach was causative of the damage suffered. It was also submitted that there was no evidence before the referees to establish that the failure to comply with the contractual standard, the ASME code, caused the damage which the respondent suffered, and that this part of the report should therefore have been rejected by the trial judge. On this submission, there being no other evidence before Giles CJ Comm D, the respondent was entitled to nominal damages only. HELD, (per Beazley JA, Sheppard AJA concurring) allowing the appeal: (1) On the review, contrary to his Honour's finding, the appellant had challenged the referee's finding in respect of causation. His Honour's failure to deal with the appellant's challenge on this issue constituted a breach of natural justice. (2) The respondent's contention that even if there was a denial of natural justice, his Honour's judgment should be upheld because there was sufficient evidence before the referees to enable them to find that the breach of contract was causative of the respondent's loss, rejected: The referees had wrongly used a non-contractual specification to determine the question of compliance with the contract. There was no finding by the referees that had there been compliance with the contractual specification, there would have been no loss. (3) In any event, the respondent's contention would have involved the Court in reviewing the referees' report. It is not the function of this Court to do so. This Court's function is to ascertain if appealable error has been demonstrated in the trial judge's review of the referees' report under Pt72 r13 of the Supreme Court Rules. Nine Network Pty Ltd v Kennedy Miller Television Pty Ltd (unreported, New South Wales Court of Appeal, 8 June 1994) applied. HELD, (per Handley JA): The FEM rules were in evidence before the referees, and the appellant did not seek a ruling from the Court under Pt72 r9 of the Supreme Court Rules that the FEM Rules were 2 UNREPORTED JUDGMENTS irrelevant to the issues in the reference, however, reliance on codes other than ASME was not supported at trial by counsel for the respondent. Whether the respondent should now be permitted to withdraw that concession depends on the way the proceedings were conducted, a matter on which the Court does not have complete information. Handley JA In this appeal I have had the benefit of reading the reasons for judgment of Beazley JA. Her Honour sets out the principles which this Court must apply in this appeal, the relevant facts, and the history of the proceedings. I gratefully adopt her statement of these matters. The only issue in the appeal to this Court related to the input bevel pinion shafts (the shafts) in two gearboxes for slurry pumps supplied by the respondent (Warman) to BHP for use at the Escondida copper mine in Chile. The Judge adopted so much of the report of the referees as found that the shafts in the gearboxes supplied by the appellant (Flender) to Warman had inadequate fatigue strength capacity when assessed against a contractual requirement, in the specification, of compliance with ASME B 106.1 (American National Standard "Design of Transmission Shafting" 1985) Industry Code of Practice (ASME). The adoption of the report (paral5.1(a), paral5.2 (a)) in these respects was not challenged on the appeal. Flender was therefore in breach of the relevant term of the contract for the sale of these gearboxes. The referees found that the fatigue failure of these shafts (failures 2 and 3) was a contributory factor in the failure of gearboxes 7 and 8 which occurred on 2 August 1991 after 2,300 hours' service (Table 1 AB 52). The referees assessed the damages recoverable by Warman for breach of contract in respect of these two failures at $533,010. This assessment was adopted by the Judge, and this part of his decision was not challenged on appeal. The only issue on the appeal was whether the Judge erred in adopting that part of the report which found that Flender's breach of contract in supplying gearboxes with shafts that failed to meet the fatigue strength capacity in the specification assessed in accordance with ASME caused or materially contributed to (caused) the failure of the shafts so as to make it responsible for the damages found by the referees. The Judge said that the referees' finding that these breaches had caused the second and third gearbox failures had not been relevantly challenged before him. Mr Walker SC for Flender submitted that this was not correct, that the finding had been challenged and this Court could not find that these breaches had caused those failures. If these submissions were accepted, he submitted that the appeal should be allowed and judgment entered for Warman for nominal damages on this part of its claim. I agree with Beazley JA that the finding of causation by the referees was challenged before the Judge, and with her reasons for that conclusion. It follows that the judgment on this issue cannot be supported on the ground accepted by the Judge but this does not entitle Flender to have the appeal allowed. The appeal is by way of rehearing (Supreme Court Act s75A(5)), this Court has all the powers of the court below (subs(6)), and may make any finding which ought to have been made or which the nature of the case requires (subs(10)). Consistently with this approach Mr Walker SC for Flender attempted to demonstrate that a finding of causation had not been open to the primary Judge and Mr McDougall QC for Warman sought to establish that it was. The Judge noted that the referees had undertaken an extensive examination of the possible reasons for the gearbox failures and had concluded that the exclusion of other postulated causes of failure left "a defect in the gearbox" as a probable URJ FLENDER (AUST) PTY LTD v WARMAN INTERNATIONAL LTD (Handley JA) 3 cause. The referees also undertook an assessment of the capacity of the shafts against the specification requirement assessed in accordance with ASME and found that "the shaft (sic) was marginally undersized against the requirements of the specification". They also undertook such assessments against Australian standard AS 1403 "Design of Rotating Steel Shafts" (AS 1403) and the Federation Europeenne De La Manutention Rules for the Design of Mobile Equipment for Continuous Handling of Bulk Materials Ch4 "Calculation and choice of mechanism components', 1992 (FEM Rules) which were not contractual standards. They found that assessed according to AS 1403 the shaft "would be marginally undersized relative to the requirements of specification 511-5-001" and that assessed according to the FEM rules, the shaft would be *substantially overstressed according to the specification requirement*. Flender accepts that the shafts did not fully meet (were "marginally undersized") the contractual requirements. They failed after some 2,300 hours' service, and it was common ground that at a steady 1300 Kw the shafts should "last for an infinite life" (Report AB 94) The question for this Court is whether Warman established that the fact that the shafts were marginally undersized against the requirements of the specification assessed in accordance with ASME caused the failure of the gearboxes or whether this was the result of other and unknown causes. Flender had suggested a number of other possible causes during the reference but they were rejected by the referees. The "design life" of the shafts had not been a requirement of specification 511-5-001. Although the referees recognised that the FEM Rules were not a contractual standard they considered that assessment to this and the other codes provided some indication of the adequacy of the design for the specified loadings. Having made their assessments, the referees then stated that although all three codes had been considered and given weight, "the FEM Rules are preferred, since they incorporate recent important developments in fatigue design". However the FEM Rules dated from 1992, whereas the contract between Warman and Flender was negotiated between November 1988 and March 1989 and was entered into in March 1990. The referees said of the FEM Rules that "the incorporation of this high cycle aerating makes (them) much more appropriate for assessing these shafts, and particularly for assessing the question of whether failure at the observed 2,300 hour life could or could not be anticipated". What was not clear to this Court was whether the application of the FEM Rules in this way was merely an appropriate means of determining the significance of the fact that according to ASME the shafts were "marginally undersized" compared with the specification, or whether the use of those rules involved the application of a more onerous standard to the design to the detriment of Flender. There is no finding in terms that the shafts would not have failed even if they had complied with the specification instead of being "marginally undersized" although as will appear there was a finding that corrosion was not significant because the failures would probably have occurred anyway. If fully sized shafts would have failed at about the same time, the fact that they were "marginally undersized" did not cause, or materially contribute to, the loss although it may have slightly accelerated it. On the other hand, the marginal undersizing of the shafts may have made the difference between success and failure. Flender also challenged before the Judge the assessment of the shafts according to ASME made by the referees. It submitted that they should not have applied a safety factor of more than 1.5, it being common ground that if that value had been used in the assessment calculations, the shafts would have been found 4 UNREPORTED JUDGMENTS to comply with ASME instead of being "marginally undersized". The referees rejected Flender's submission that the safety factor was the same as the service factor so that a figure of 1.5 met the requirements of the specification. The assessment of the shafts according to ASME required input assumptions to be made. Assessments were provided to the referees by the experts of each party. The subject matter of the ten relevant assumptions was common ground, but the experts differed as to the appropriate values for six. The referees determined what they described as the preferred values for the six, and arrived at the conclusion that, according to ASME, the shafts were marginally undersized and would be "overstressed". The referees' preferred value for the safety factor was 2. Their adoption of this value was challenged before the Judge but he accepted their conclusion. Nothing further of relevance to the present question can be deduced from the Judge's reasons, and one is driven back to the report of the referees in a search for guidance. The referees' report, including appendices, was 183 pages in length and dealt with a number of highly technical issues, many of which are not relevant to this appeal. It appears from Table 1 (AB 52) that four gearboxes numbered 7-10 inclusive were supplied by Flender, and between June 7 1991 and February 1992 failures of different kinds were experienced in each box. However while the shafts failed in boxes 7 and 8 on 2 August after 2,300 hours of operation, the shafts in boxes 9 and 10 did not fail within this period. The referees reviewed the metallurgical evidence and concluded that material defects (57) and sub standard material (58) played no part in the shaft failures. They found that both failures were the result of fatigue, the cracking initiating from the undercut groove at the junction of the pinion head and shaft. In both cases the base of this groove was extensively marked with corrosion pitting (61). After considering further evidence they concluded that corrosion was a contributing factor, but not a major factor, in the failures because the failures "would probably still have occurred had there been no corrosion present" (62). They then considered the evidence of the fracture mechanics aspects of these failures and concluded that this "would not preclude the possibility that the shafts might have failed as a result of fatigue cracking, propagating from very small cracks around the fillet surface, under normal continuous operation at or about nominal operating power levels" (64). The referees next considered the evidence as to the loadings applied to these gearboxes. Flender argued that the gearboxes had been overloaded and "that the overloading was the primary cause of all the failures". On the other hand Warman argued that the gearboxes had not been overloaded and that the failures were related to other causes. The referees found that power levels were generally in the range 650-850 Kw, compared to the nominal full load motor power of 855 Kw (67), and that, in effect, load fluctuations were not likely to be significant (67). The referees then considered the effect of situations where the pump impellor was caused to stop very rapidly (a crash) which can generate high loads on the gearbox. The possibility of such loading being a major contribution to the failures was advanced by the defendant (68), but the referees concluded that "it would seem that crash stops were unlikely to have contributed significantly to the gearbox loadings, or to have been a major factor leading to the failures" (71). The shaft failures resulted from a very large number of stress cycles which could not be determined precisely by examination of the fracture surface, and for this URJ FLENDER (AUST) PTY LTD v WARMAN INTERNATIONAL LTD (Handley JA) 5 reason also it was "very unlikely that crash stops could have contributed significantly to the fatigue failures" (72). Another possible cause of the failure suggested by the defendant was that the discharge location could be significant but this seems to have been rejected by implication (73). The referees also considered that it was unlikely that bogged starts, in the absence of torsional oscillation at least, would cause any abnormal loading on the gearbox and in view of the very low frequency of occurrence they were also unlikely to have contributed to the pinion shaft failures unless the associated oscillation considerations were significant (75). Without a significant contribution from torsional oscillation, none of the other load cases were likely to produce loads either high enough or frequent enough to be responsible for the kind of fatigue failures that occurred (75). In the end even Flender's experts concluded that torsional resonances due to any mechanical source should not be significant (78). At 85-7 the referees summarised the evidence on the torsional oscillation issue which included four possible causes of electrical excitation of such oscillation and concluded that none of them was likely to be a significant cause of the fractures. The referees then noted that both parties had addressed the issue of overheating of the gearboxes since there was reasonably convincing evidence that this had occurred at least occasionally. They concluded however that the basic capacity of the pump coolers was adequate (89). They then addressed the adequacy of the gearbox component design and asked whether the components had adequate static and fatigue strength for the required duty (91). They approached this task by enquiring whether the design met the requirements of the specification in terms of capacity, and whether it provided a capacity consistent with that represented in Flender's catalogue. The referees asked themselves whether the observed failures could be anticipated based on the design capacity, in the absence of other possible contributing factors (91). The specification stated that continuous 24 hour operation was requested (92), but it did not nominate any specific design standards for shafts and was generally silent on other requirements. The design life of the shafts was not specified (93). Capacity of the shafts as represented by Flender's catalogue, when operated at a steady 1300 Kw, was that they should last for "an infinite life" (94). The referees assessed the design requirements of the specification and Flender's catalogue in accordance with AS 1403, ASME and the FEM Rules. As to AS 1403 they concluded, at 107, that the shaft would be marginally undersized relative to the requirements of the specification but there was no doubt that it was undersized according to AS 1403 in terms of satisfying the capacity represented by Flender's catalogue. At 111 they referred to "the agreed reasonable expectation that shafts should survive forever at the nominated power levels". Having assessed the shafts against the requirements of the specification assessed in accordance with ASME, they concluded that it was "marginally undersized". The equivalent overstress factor is about 1.13 indicating that stress levels were about 13% higher than required for complete conformance. In the case of the more stringent test where the shaft is assessed against the capacity represented in the catalogue "the shaft is overstressed, according to this analysis, by about 26%". The referees then assessed the shafts in accordance with the FEM Rules and concluded, at 119, that the pinion shaft would be overstressed by a substantial amount on both the specification requirement and the catalogue representation. See Table 12 (120). They then addressed the third question, referred to at 91, 6 UNREPORTED JUDGMENTS whether the observed failures of these shafts could be anticipated based on their design capacity and concluded at 121 that "according to these calculation procedures, there would be a significant chance of these shafts failing at lives as short as 2,300 hours, without the influence of corrosion, requiring only that the equivalent loading imposed on the gearboxes was approximately 1.2 times the nominal motor power because of torsional vibration, or the influence of occasional overloads, or for whatever reason". It is not at all clear that the process of reasoning which culminated in these findings was rendered invalid by the fact that the FEM Rules were not a contractual standard. On one view the referees were merely stating that the more sophisticated calculations provided for by the FEM Rules demonstrated that what was shown to be a marginal undersizing of the shaft according to the specification, and the proportionally greater undersizing according to the catalogue assessed in each case in accordance with ASME, were highly likely to produce the failures that occurred. On the other hand I cannot, on the report alone, find that the referees' reasoning in this passage was not affected by the legal error identified by the Judge, and we were not referred to any expert evidence before the referees, or before the Judge, which would enable this Court to decide those questions. In s16 of their report, commencing at 145, the referees analysed the various factors in order to assess Flender's responsibility for the various breakdowns. The shaft failures were dealt with at 146-9, where the referees found that both were high cycle fatigue failures "requiring a very large number (hundreds of thousands to millions) of loading cycles to propagate the fatigue cracks from initiation at the surface to final fracture". They concluded that "the most probable cause of failure is the high level of rotating bending stress generated at the groove at the base of the pinion head during normal continuous running of the drive". The referees next referred to the results of the assessment of the shaft according to the FEM Rules, and again noted the prediction derived from this assessment that the shafts could fail after 2,300 hours at gearbox loadings equivalent to about 1.2 times the load corresponding to the drive motor nominal power with a higher predicted failure probability at a higher service factor (147). They concluded therefore that the primary factor contributing to the failure was likely to be "the limited design capacity of the pinion shaft" (147). They again stated that corrosion was not significant. This finding may have been influenced by the prediction of failure at 2,300 hours, derived from the assessment according to the FEM Rules, and it may have been legally flawed for that reason. They then stated that "additional comments in support of the above conclusions" were that equivalent gearbox loadings were consistently 20% - 30% above those corresponding to the nominal motor power of 855 Kw (147); that it was improbable that load starts resulting from crash stops, bog starts or normal starts would have been a primary cause of the failure (148); that microcracks from manufacture below the threshold of detectability by inspection techniques could have initiated and propagated at normal loading (148); that there was no evidence that shaft material properties were inadequate or that the failures initiated from material defects; that problems with the lubricating oil were unlikely to have had any impact on the shaft failures, and it was improbable that the condition of the bearings could have had any significant influence (148-9). While the referees concluded that, according to the specification and ASME, the shafts were marginally undersized so that stress levels were 13% higher than they should have been, and according to the catalogue and ASMF, were 26% URJ FLENDER (AUST) PTY LTD v WARMAN INTERNATIONAL LTD (Handley JA) 7 higher, a reading of their report without the assistance of expert evidence directed to the point suggests that they were not prepared, without more, to find that this alone could explain the shaft failures after 2,300 hours of operation. Nor it would seem were the referees able to conclude that this breach, coupled with the rejection of other suggested causes, especially overloading of the gearboxes which the defendants had asserted was "the primary cause of all failures" (66) enabled the inference of causation to be drawn. Instead the referees thought it appropriate, and possibly necessary, to assess the shafts according to AS 1403, the FEM Rules, and the Niemann Winter Procedure to determine whether the breach of the contractual standard in the specification was a probable cause of the failures. Their approach suggests that they thought that a straightforward inference of causation from the found breach and their rejection of all other suggested causes, was not reasonably open. On the other hand it is possible that they were merely giving additional reasons to confirm a straightforward finding of causation that they were prepared to make in any event. If this were the case they have not given that explanation in their reasons, and I am in no position to make that finding. The appeal book does not include the transcript of the proceedings before the referees or the exhibits. However we have been furnished with the written submissions of counsel in the proceedings before the Judge, and these include extensive references to the transcript before the referees. The High Court has repeatedly emphasised the importance of the conduct of the trial. In Coulton v Holcombe (1986) 162 CLR | at 8, the joint judgment of four Judges approved an earlier statement by the Court that: "Tt is elementary that a party is bound by the conduct of his case. Except in the most exceptional circumstances, it would be contrary to all principle to allow a party, after a case has been decided against him, to raise a new argument which, whether deliberately or by inadvertence, he failed to put during the hearing when he had an opportunity to do so". In McAllister u Richmond Brewing Co (NSW) Pty Ltd (1942) 42 SR (NSW) 187 at 191, Jordan CJ quoted with approval this statement by Lord Carson: "T do not think that... any appellate tribunal can be too careful in ascertaining by an examination of the proceedings what was the course of the trial, what was the real issue fought out by the assent of the parties on both sides". The pleadings in the reference were included in the appeal book. Flender's statement under SCR Pt72, r8(5) included assertions that its catalogue was part of the contract (para7, para9, paral2, p180) and Warman made the same claim in its statement (174). This formal admission certainly appears to support the referees' finding to that effect. Paral7-paral9 of Flender's statement related to "other technical standards". These stated: "17. For the purposes of comparative assessment of the KEN 400 reducers, the relevant duty is that referred to in para5, paral3 and paral4 hereof. 18. By utilising a correct engineering analysis and the relevant duty, the KEN 400 reducers complied with: (i) the relevant AGMA standard; (ii) ASME; (iii) AS 1403; (iv) The relevant ISO practice; (v) General mechanical engineering principles and calculations. 8 UNREPORTED JUDGMENTS 19. On the basis of a reasonable engineering assessment of the shafts, there was adequate safety factors against fatigue for: (i) The relevant duty; (ii) The actual "normal" duty to which the system was subjected (excluding dynamic or torsional vibration)". The next section (para20-para27) dealt with the actual operational use and maintenance of the pumps prior to August 1991. Flender claimed that "the whole system was subject to operational abuse" (para20), with this general allegation being particularised in the paragraphs that followed. The following section (para28-para39) dealt with the reasons for the failure of the reducers (gearboxes). Para28-para33 were as follows: "28. The first failure in June 1991 was: (i) Probably due to lubrication failure associated with either inadequate or overheated lubricant; (ii) Due to poor maintenance practices by the mine operators; (iii) Unrelated to any deficiency in the design and manufacture of the reducers; and was (iv) Probably associated with overloading due to the destruction of the pump liners and the jamming of the pump. 29. The failures at the pinion shafts were due to a multiple application of dynamic overloads to the pinion shafts at or near the fatigue limit. 30. The dynamic overloads to the pinion shafts occurred during start up. 31. The dynamic overloads by torsional vibration were probably due to the inherent characteristics of the motor and the dynamics of the whole system. 32. Warman did not check or investigate the dynamics of the system prior to the installation of the pumps, reducers and couplings. 33. Warman has never carried out an investigation or analysis of the dynamics of the whole system, including the assessment of the torsional vibration imposed by reason of the motor characteristics during start up". It is evident that Flender's case before the referees as pleaded raised the issue of compliance with standards other than ASME, including general mechanical engineering principles and calculations" relevant to the making of "a reasonable engineering assessment of the shafts". It is also evident that Flender sought to establish a case that the system had been subjected to operational abuse, and that the shaft failures were due to causes other than inadequate design. The proceedings for the adoption of the report before the Judge were conducted on pleadings in the form of points of claim and points of defence. Warman again alleged that Flender's catalogue formed part of the contract (para5.1, p213) and this was admitted by Flender (para5.1, para6, p224-p225) with what appear to be irrelevant qualifications. Although the Judge did not consider it necessary to decide whether the catalogue was part of the contract and refrained from adopting the report on this issue, the proceedings appear to have been conducted on that basis before the referees and the Judge, and in my opinion there is no reason why this Court should not now adopt this part of the report and make this finding. (See also Flender's outline submissions paral, p232 and Warman's reply paral.2, p252.) Warman's submissions in reply (p252) included paral.6 as follows: "The reference proceeded on the basis that the Referees clearly regarded the use of the abovementioned codes and standards for the abovementioned purpose as relevant to determining the adequacy of the shafts as designed. At no stage did the Defendants submit to the Referees that those codes were not relevant matters URJ FLENDER (AUST) PTY LTD v WARMAN INTERNATIONAL LTD (Handley JA) 9 to the issues arising in the reference. Indeed they introduced evidence from fresh technical experts to address the issues and the need to do so was one element in their seeking and obtaining an adjournment of the reference. See Transcript p504 to p506". Para4.1 of those submissions provided (p253) as follows: "This criticism of the Referees' procedure ignores the Referees' own stated purpose in having recourse to the standards (which other than ASME are not identified in para2 of the defendant's submission). See p68 to p73 of the Referees' Report. The standards were merely used by the Referees as a tool for measuring objectively the actual capacity of various gearing components particularly shafts in order to determine whether such capacity meets the capacity specified in the Specification and the catalogue". These submissions are supported by the referees' statement in their report (95-6): "While conformance of the shaft design to any of the standards or procedures is clearly not a contractual requirement, assessment to these codes does provide some indication of the adequacy of the design for the specified loadings". The standards, particularly the FEM Rules, now said to be completely irrelevant, were in evidence before the referees (para4.2, p254) and it is claimed that they were referred to by Flender's experts, Moore and Nicholson. The FEM Rules were dealt with in paral4.1 as follows (256): "The position in relation to the FEM rules is that they were raised by the Referees as a standard which the Defendant being effectively a German entity ultimately might be more comfortable with using. This somewhat flexible approach of the Referees towards an appropriate standard to use only serves to highlight the relatively limited use to which the standards were being put namely to assist the Referees to determine on the balance of probability whether the shaft and the gearboxes were underdesigned for the duty contracted for". Moreover it appears that Flender itself tendered the FEM Rules which became Exh 27 (paral4.3). The relevance of the rules was also dealt with in paral4.4-paral4.7 of Warman's submissions which indicate that this topic was referred to in the transcript before the referees at 576 and foil (25/11/93), 742, 896 and foll, 899-90, 915-6 and 920. It appears to me to be highly significant that Flender, having been put on notice of the possible use of the FEM Rules the referees were considering, did not seek a ruling from the Court under SCR Pt72 r9 that they were irrelevant to the issues in the reference. It allowed the trial before the referees to continue on the basis that those rules could be relevant. Senior counsel for Flender, in further written submissions to the Judge, submitted that the referees should not have had reference to FEM for the purpose of determining causation (paral, p259) but then submitted that if they were entitled to do so, they should have given Flender a further opportunity to lead evidence in relation to these rules (para4, p259). The Judge rejected Flender's other claims that the referees had denied it procedural fairness. Although he held that the FEM Rules were irrelevant, his findings on other claims of denial of procedural fairness and the many references to the transcript in Warman's submissions to the Judge demonstrate that this particular claim should also fail. It was not renewed in argument before this Court and may be disregarded. However a later written submission by senior counsel for Flender to The Judge contained the following (267): "12. His Honour asked Mr Gee whether he wanted to support the reliance by the Referees on the codes other than ASME. 10 UNREPORTED JUDGMENTS Mr Gee did not thereafter seek to support that reliance". But for this submission I would have concluded with some confidence that the parties conducted the trial before the referees on the basis that other technical standards were relevant to the making of "a reasonable engineering assessment of the shafts" (181), that the FEM Rules were suggested by the referee, Dr Morrison, as appropriate for this purpose, that both parties accepted the relevance of this issue, fought the trial before the referees on that basis, and that the referees' use of the FEM Rules did not, in any sense, depend on their having contractual status. On that basis I would have upheld the relevant section of the report, ordered that it be adopted, and dismissed the appeal. However the apparent concession by Mr Gee QC for Warman, referred to in paral2, makes that course inappropriate at this stage of the proceedings. If my analysis is correct, one wonders why the concession was ever made. There may, however, have been material before the Judge which has not been reproduced which made that course appropriate, if not inevitable. It is also possible that the concession was more limited than would appear from the terms of paral2, read without knowledge of the way the proceedings before the Judge were conducted. It is also possible that the concession was made in error. In that event the Court would have to consider whether Warman should now be permitted to withdraw it, and this again would depend upon the way the proceedings were conducted before the Judge, a matter on which the Court does not have complete information. In these circumstances the appropriate course in my view is for the Court to publish reasons and afford the parties a further opportunity to make submissions and perhaps refer to material that was before the Judge that has not hitherto been placed before this Court. It may or may not then become clear that this Court should not attempt to determine the outstanding issues, but should set aside the orders and judgment entered below, and remit those issues for determination by the primary Judge. Beazley JA This appeal raises a narrow question as to whether the trial judge afforded the appellant procedural fairness in respect of one aspect of a review of a referee's report under Pt72 of the Supreme Court Rules. The appellant was sued by the respondents for damages for, amongst other things, breach of contract for the supply of certain gear reducers, which included bevel pinion shafts (the shafts) which were to be used by the respondent in a slurry pump installation in Chile. Relevantly, for the purposes of the appeal, the matters in issue on the reference were the terms and conditions of the contract, in particular, what documents constituted the contractual documents and whether the gear reducers supplied conformed with the terms and conditions of the contract. The proceedings were referred to a referee for inquiry and report on the whole of the proceedings pursuant to Pt72 of the Supreme Court Rules. The referees determined that the contract included an extract from specification 511-5-001 and a particular catalogue. The referees reported on 21 March 1994. The referees found that the gear reducers as supplied by the appellant to the respondent failed in service after a relatively short service life. Eight failures were particularised in the report, two of which involved failure of the shafts. The referees found that the shafts did not comply with the requirements of a technical code known as the FEM tules (Federation Europeene De La Manutention Rules for the Design of Mobile Equipment for Continuous Handling of Bulk Materials). They stated in paral5.I(a) of their report: URJ FLENDER (AUST) PTY LTD v WARMAN INTERNATIONAL LTD (Beazley JA) 114 "15.1 The defendant was obliged to but failed to provide gear boxes which met the capacity requirements of specification 511-5-001 which specification we conclude to be part of the contract. In particular - (a) input bevel pinion shafts as supplied by the defendant had inadequate fatigue strength capacity when assessed against accepted industry codes of practice viz:- AS 1403, ASME B106.1 and the FEM rules." The appellants sought a review of the referees' determination pursuant to Pt72 of the Supreme Court Rules. On the review, Giles CJ Comm D held that the referees erred to the extent that they based their determination on the FEM rules. His Honour stated at 16: "\.. [I]t was common ground that the specification required that the shafts conform to ASME, and Warman really did not attempt to explain how the referees might have thought that it required that the shafts conform to AS 1403 or the FEM rules. The report should not be adopted so far as the referees found breaches of contract in the terms of paral5.1(a) of the report." ASME (American Society of Mechanical Engineers) is a reference to another technical code, which on his Honour's finding, was the contractual standard to which the shafts were required to conform. His Honour held, however, that the referees' finding that the shafts had inadequate fatigue strength capacity when assessed according to ASME should be adopted. The consequence of this finding was that the appellant was in breach of its contract with the respondent. That finding is not challenged. The appellant however, contends that the mere finding of breach of contract does not result in a finding that that breach was causative of the damage suffered. It was submitted that there was no evidence before the referees to establish that the failure to comply with the contractual standard, that is, failure to comply with the ASME code, caused the damage which the respondent suffered. It followed on this submission, there being no other evidence before Giles CJ Comm D, that the respondent was entitled to nominal damages only. Before dealing with the specific challenge raised by the appellant, it is convenient to deal with the principles which govern an appeal to this Court of a judge's review of a referee's report. S124(2) of the Supreme Court Act 1970 (NSW) provides: "(2) The rules may make provision for or with respect to - (a) the cases in which the whole of any proceedings or any question or issue arising in any proceedings may be referred by the Court to an arbitrator or referee for determination or for inquiry and report; (e) the consequences of a determination or report by an arbitrator or referee; (f) the manner in which such a determination or report may be called in question; (g) whether or not, or to what extent, a determination or report may be called in question on a matter of fact or law; Pt72 12 of the Supreme Court Rules provides that the Court may, in any proceedings, on application by a party, or by its own motion, make orders for a reference to a referee appointed by the Court for inquiry and report by the referee, on the whole of the proceedings, or any question or questions arising in the proceedings. 12 UNREPORTED JUDGMENTS Pt72 r8 provides for the manner in which proceedings may be conducted before a referee. Pt72 r13 provides: "13(1) Where a report is made, the Court may, of its own motion, after 5 notice to the parties, or on application by any party, on a matter of fact or law or both - (a) adopt, vary or reject the report in whole or in part; (b) require an explanation by way of report from the referee; (c) on any ground, remit for further consideration by the referee the 10 whole or any part of the matter referred for a further report; (d) decide any matter on the evidence taken before the referee, with or without additional evidence; and shall give such judgment or make such order as the Court thinks fit (2) Evidence additional to the evidence taken before the referee may not be adduced 15 before the Court except with the leave of the Court. " S124 and Pt72 provide for a special mode of trial of issues: see Buckley v Bennell Design Constructions Pty Ltd (1978) 140 CLR 1 at 15; Nine Network Pty Ltd v Kennedy Miller Television Pty Ltd (unreported, New South Wales Court of Appeal, 8 June 1994) per Gleeson CJ at 9. Pursuant to Pt72 r13, the 20 Court is required "to consider and determine afresh any material question of law arising out of the referee's report which either party desired to agitate. Subject to that obligation, [the Court has] a discretion as to the extent to which [it will] re-consider, and determine... matters upon which the referee had reached a conclusion and reported": Nine Network Pty Ltd per Gleeson CJ at 9-10. 25 An appeal lies to the Court of Appeal in respect of error by a judge exercising the powers of review under Pt72 r13. See s101 of the Supreme Court Act. The nature of that appeal was also considered by Gleeson CJ in Nine Network Pty Ltd. His Honour stated at 11: "Tt is not for this Court to review the referee 's report... In the exercise of the 30 power of review given by the rules, the judge at first instance may fall into appealable error. If that can be demonstrated to the Court of Appeal, then ordinarily the judgment at first instance will be set aside and consequential relief granted. However, what the Court of Appeal is concerned with is error on the part of the judge. If the judge's decision to adopt (or vary or reject) the referee's report 35 in whole or in part cannot be shown to be based upon a material error on the part of the judge, then there will be no ground for attacking the judgment based on that decision. If on the other hand, that decision can be shown to be based upon a material error on the part of the judge then a different result will follow. If the point at issue is one of law, it may not be difficult to demonstrate such error. If 40 the judge can be shown to have made an error in the approach taken to the exercise of the discretion conferred by the rules (as was contended unsuccessfully in Super Pty Ltd v SJP Formwork [(1992) 29 NSWLR 5490 then that also may constitute a ground for setting aside the judgment. It may even, in a given case, be possible to demonstrate that the judge's decision to adopt, or vary 45 or reject, the report was based upon an appealable error of fact made by the judge. An example might be a case where the judge embarked upon a consideration of new evidence, or a fresh consideration of evidence that was before the referee, and could be shown to have reached a wrong conclusion. The important point is that it is the judge at first instance who reviews what the referee did; the Court 50 of Appeal, within the limits of the ordinary rules governing appeals, reviews what the judge did. URJ FLENDER (AUST) PTY LTD v WARMAN INTERNATIONAL LTD (Beazley JA) 13 ... However, if the judge, in the proper exercise of the discretion given by the rules, declines to consider afresh questions of fact that have been decided by the referee, then it is not open to the party aggrieved to invite this Court to re-visit those questions on the basis that, by virtue of the adoption of the referee's report, the judge's decision is vitiated by any errors in it. " Senior counsel for the appellant submitted that as his Honour determined that the contract required compliance with the ASME code, he was required to consider the appellant's case which it made before the trial judge, that it was not open to him (or to the referees) to use the FEM rules to determine causation. It was submitted that his Honour's failure to consider this aspect of its case was an appealable error within the principles in Nine Network Pty Ltd v Kennedy Miller Television Pty Ltd. It was further submitted that there was no evidence before the referees establishing that non-compliance of the shafts with ASME was the cause of the shafts' failures. Accordingly, it was submitted that this part of the report should have been rejected by the trial judge. These submissions, if made out, are correct. A trial judge's failure to consider a material argument constitutes a breach of natural justice. Giles CJ Comm D, after having found breach, stated at 30: "In the manner earlier described, the referees found that Warman suffered loss as a result of the breaches of contract. The findings that the inadequacies in the shafts, gearing and the bearings were causally connected with the eight failures, and the assessments of the relative contributions of the various factors... were not challenged before me." The appellant submitted that it had not left that matter unchallenged. The Court's attention was directed to the appellant's written argument before Giles CJ Comm D. In a document entitled "Errors" and noted to have been handed up by Mr Emmett QC on 12 December 1994, the following argument was put: "1. The Referees should not have had reference to FEM for the purpose of considering whether there was any causal connection between the failures and the design capacity without first giving Flender the opportunity to lead evidence as to the inappropriateness of FEM for that purpose: see Report 68-9 and paral4.3.5.4 2. The Referees should not have held that Flender failed to meet Specification 511 - 5 - 001 because the shafts Jailed to meet the standards of AS 1403 or FEM since neither of them was a contractual document: see Report 109. " Likewise, in a document entitled "Notes for Reply " it was argued that: "(Counsel for the respondent] refers to p43-p67 and the material where the Referees deal with a number of postulated causes of the failure other than design shortcoming These were matters advanced by Flender as possible causes of failure. However, there is no suggestion and certainly no finding that they are the only possible cause of the failure other than inadequacy of design and it was Warman's onus to establish what the cause of the failure was. Warman must establish that the shafts failed by reason of some characteristic or inadequacy the presence of which was a breach of the contract. The only findings were that the shafts did not satisfy criteria which were much more stringent. 14 UNREPORTED JUDGMENTS [Counsel for the respondent] suggested that causation proved itself once there was a negative answer to questions | and 2 on p68 of the Report and that question 3 was unnecessary. That proposition, however, ignores completely the necessity for Warman to establish not only that there was a breach of contract but that the breach of contract caused the damage or resulted in the failure. Each question must be answered. [Counsel for the respondent], in answering a question from his Honour concerning causation, gave two responses: (a) When every reasonable alternative postulated by Flender was rejected there should be a conclusion that the failure was caused by breach of contract. But that was not the reasoning adopted by the Referees. The only reasoning exposed in the Report is based on the answer to question 3 by reference to FEM. (b) If you establish that the design was inadequate, you can conclude that there is a causal link But that begs the question as to the standard by reference to which you measure inadequacy." It was not contended by the respondent that these arguments were not in fact put. In my opinion, they clearly put in question the issue of causation. It follows that his Honour erred in stating that the matter was not in issue and thus failed to consider a fundamental part of the appellant's case. Appealable error can be demonstrated. However, the respondent submitted that the trial judge's decision should be upheld because, leaving aside any question of the use of the FEM rules on the causation issue, the appellant's case must fail, for the simple reason that there was ample material before the referees to enable them to find that the appellant's breach of contract was causative of the respondent's loss. This argument was based on the following propositions: 1. The referees found that the contract comprised an extract from Specification 5511 - 5 - 001 and Flender's catalogue K24DENS.88. 2. The terms of the contract included: "[T]hat the defendant in assessing the mechanical components of the reducers should use a motor power of 855KW; that there was to be a service factor of at least 1.5 applied to the maximum operating power; that the reducers should have complied with the capacity as represented in the Catalogue; that the catalogue nominated a power rating of 1300KWfor the reducers. " 3. By reason of the incorporation of the catalogue into the contract, the shafts were required to have a capacity of 1300KW. 4. The referees found that notwithstanding that the shafts failed in service at 2,300 hours, they had not been loaded beyond the required capacity. 5. Accordingly, the failure of the shafts in service either was or proved a breach of the contractual requirement as to capacity. 6. If the appellant sought to challenge propositions (4) and (5), it bore an evidentiary onus to establish that the failure occurred by reason of some matter for which it could not be held responsible. The appellant accepted as common ground that Flender's catalogue was a contractual document. However, it submitted that the contract must be read as a whole, and that, as Giles CJ Comm D found, the contract as a whole included a specification which required compliance with the ASME code, not the FEM rules. It further submitted that the respondent's argument on this point involved an assertion that the question whether the shafts could meet the catalogue capacity of 1300KW should be measured by the FEM rules. The competing URJ FLENDER (AUST) PTY LTD v WARMAN INTERNATIONAL LTD (Beazley JA) 15 positions of the appellant and respondent on this point are best considered by referring directly to the respondent's submission. It asserted: "The Catalogue did not specify, or by implication dictate the use of, any particular standard or procedure to assess the design capacity of a gearbox component against the capacity which, according to the Catalogue, that component was required to have. Accordingly, it was open to the Referees to select whatever standard they regarded as the most appropriate in determining the answer to that question." The submission continued that the referees had selected the FEM rules as being "much more appropriate for assessing these shafts, and particularly for assessing the question of whether failure at the observed 2,300 hour life could or could not be anticipated". In my opinion, the appellant's resistance to the validity of this approach is correct. Giles CJ Comm D held that the design specification of the shafts required compliance with ASME. It is not open therefore, for the respondent to assert that the design specification could be assessed as against some other standard. The respondent's alternative argument was that the referees used the FEM rules to assist them to determine "whether the observed failure of these shafts could be anticipated based on the design capacity". The referees approached this question by using the FEM calculation procedures. This is apparent in s14 of the report. The referees stated: "A central issue in these proceedings has been the question of whether the various components which have been the subject of failures have been designed adequately, in the sense of having adequate static and fatigue strength for the required duty. In the present context, adequacy can be assessed in several ways to different benchmarks. Specifically, two separate tests have been applied as follows:- 1. Does the design of these components meet the requirements of the specification in terms of capacity. 2. Is the design of these components likely to provide a capacity consistent with that represented in the manufacturer's (Flender's) catalogue. A separate related question, which arises in the event that design capacity does not satisfy either of the above tests has also been considered specifically; 3. Could the observed failures be anticipated, based on the design capacity, in the absence of other possible contributing factors (overloading, corrosion, overheating etc). Answers to this third question are relevant to consideration of whether there is any causal connection between the failures which occurred and the design capacity of components of the gearbox. (Emphases added) " The respondent submitted this approach was open to the referees. It is not easy to discern why this is so. Points | and 2 raised in this section were the contractual issues raised in the case. Point 3 was the causation issue. In order to be entitled to more than nominal damages, it was necessary for Warman, who was asserting breach, to prove that the breach was causative of the loss. The approach to this issue by the referees gave rise to a question on the Pt72 review whether their use of the FEM rules was the mere use of a more precise tool to measure compliance or whether it was the use of a technical standard sufficiently different from the contractual requirement of compliance with the ASME code. However, his Honour did not deal with that question, having, mistakenly as it appears, assumed that there was no challenge on that point. 16 UNREPORTED JUDGMENTS Notwithstanding this omission by the trial judge, the respondent submitted that the referees had dealt with the causative issue correctly when they considered the "critical question" of overloading and concluded that "the primary factor contributing to the failure is therefore likely to be the limited design capacity of the... shaft". The appellant submitted that that finding did not assist the respondent as the referees assessed the design capacity by reference to the FEM tules. As the appellant submitted: "The referees made no finding... about whether compliance by Flender with the stipulated capacity according to the ASME code would have led to a different outcome from what actually happened, Flender having delivered shafts which were 'marginally undersized" according to the ASME code." In the absence of this differential analysis, it was submitted that causation could not be proved. The appellant further submitted: "Given the referees' finding that 'power levels were generally in the range of 650 to 850kW", there is obviously a very important gap between the conditions under which these shafts failed and their falling short of the stipulated capacity (which was marginal). It cannot logically be asserted that the fact of failure under those conditions necessarily means that the shafts failed because they were marginally undersized ". This is because their contractual shortcoming (ie the breach which must exist for loss to be compensable) will not have caused the loss unless compliance with the contract or the contractual stipulation would have prevented the loss." These submissions are, in my opinion, correct. Furthermore, this survey of the respondent's submissions reveals its misconception as to the nature of the task before this Court. As stated earlier in my judgment, it is not this Court's function to review the referees' report: see Nine Network Pty Ltd. Acceptance of the respondent's submissions would involve the Court in doing precisely that. In my opinion, the appellant having successfully established error in Giles CJ Comm D's decision, the appeal should be allowed. The appellant submitted that it was open to the Court to enter judgment for the respondent in the sum of $1, by way of nominal damages for breach of contract or to dismiss the respondent's claim altogether. However, as the appellant's case was based on breach of natural justice, and given the nature of the review by a judge under Pt72 and the limited scope of an appeal to this Court from that review, I am of the opinion that the appropriate course is for the Court to set aside his Honour's judgment in so far as he found Flender liable for the shaft failures and to remit to the Construction List in the Common Law Division the question of whether the findings in the report with respect to the shaft failures should be adopted or rejected. The respondent should pay the appellant's costs, but have a certificate under the Suitors Fund Act 1951, if so entitled. The costs of the original hearing before Giles CJ Comm D should be in the discretion of the trial judge on the rehearing. Sheppard AJA In this matter I have had the advantage of reading the judgment to be delivered by Beazley JA. I agree in the orders which she proposes and, generally, with her reasons. The problem in this case arises because the referees used a standard which was not part of the contract, namely the FEM rules. As the learned primary Judge said, there is confusion in the referees' report so far as they found breach in failure of the shafts to meet the capacity requirements of the specification. When they began their consideration of capacity requirements according to the specification, the referees seem to have considered that none of the three codes URJFLENDER (AUST) PTY LTD v WARMAN INTERNATIONAL LTD (Sheppard AJA) 17 had been given contractual force. They also said that assessment, according to the codes, provided "some indication of the adequacy of the design for the specified loadings". His Honour overcame the problem with which he was confronted by adopting the report to the extent to which the referees found that the shafts had inadequate fatigue strength capacity according to the applicable contractual standard, the ASME code. He said that, although they had erred in the contractual setting in which they placed their assessments of the shafts, the assessment of capacity according to ASME was not affected by that error and might be adopted and given its proper contractual effect. It was then that his Honour concluded that a safety factor of 2 was justifiable in the circumstances and was not limited to the 1.5 relied upon by the appellant. Although I follow what his Honour has done, my problem is that the referees having made some use, really a substantial use, of the FEM tules, notwithstanding their recognition that they did not form part vi the contrast. This reflected confusion in their thinking. Instead of putting the FEM rules aside, as his Honour endeavoured to do, they continued to have regard to them. Thus, in a passage from their report quoted by his Honour, they said that, while it was recognised that the FEM rules were not "preferred" in any contractual sense, the incorporation of this "high cycle aerating" made them much more appropriate for assessing the shafts, and particularly for assessing the question whether failure at the observed 2,300 hour life could or could not be anticipated. They added that a further attraction of the FEM rules was the removal of subjectivity associated with establishing the required factor of safety, since this was expressly related to "the applicable Wohler curve for the component being considered". It seems to follow from this that they did use the FEM rules in order to arrive at their safety factor of 2. That really is the essence of my problem with the approach taken by his Honour. These matters were plainly raised in argument before his Honour and were the subject of submissions to him as well as to us. Accordingly, I agree with Beazley JA that the appeal should be allowed and the matter remitted to the Construction List for the purpose of the determination of the question whether the findings in the referees' report with respect to the shaft failures should be adopted or rejected. 1. Appeal allowed. 2. Judgment of Giles CJ Comm D in so far as his Honour found Flender liable for the shaft failures set aside. 3. Remit to the Construction List in the Common Law Division the question of whether the findings in the report with respect to the shaft failures should be adopted or rejected. 4. The respondent to pay the appellant's costs, but to have a certificate under the Suitors Fund Act 1951 (NSW), if so entitled. 5. The costs of the original hearing before Giles CJ Comm D should be in the discretion of the trial judge on the rehearing. Counsel for the appellant: B Walker SC/I Bailey Solicitors for the appellant: Gilbert & Tobin Counsel for the respondent: R C McDougall QC/S C Kalfas 18 UNREPORTED JUDGMENTS Solicitors for the respondent: Phillips Fox