Langmaid v Dobsons Vegetable Machinery Pty Ltd - [2013] TASSC 23

[2013] TASSC 23

COURT: SUPREME COURT OF TASMANIA

CITATION: Langmaid v Dobsons Vegetable Machinery Pty Ltd [2013] TASSC 23

PARTIES: LANGMAID, Andrew Kevin

LANGMAID, Kevin George t/as Cherry Hill Coolstores (ABN 87 065 420 794)
LATROBE COOL STORES PTY LTD
v
DOBSONS VEGETABLE MACHINERY PTY LTD t/as Dobmac Agricultural Machinery

FILE NO/S:  495/2009
DELIVERED ON:  31 May 2013
DELIVERED AT:  Hobart
HEARING DATE:  24, 27, 28 and 29 August and 3, 4 and 6 September 2012
JUDGMENT OF:  Tennent J

CATCHWORDS:

Torts – Negligence – Essentials of action for negligence – Damage – Causation – Generally – Onus of proof - Degree of satisfaction required before claim may succeed.

Civil Liability Act2002 (Tas), ss13, 14.
Aust Dig Torts [60]

REPRESENTATION:

Counsel:
             First Plaintiff:  M E O'Farrell SC and T Cox
             Second Plaintiff:  M E O'Farrell SC and T Cox
             Defendant:  K Read SC and T Pilkington

Solicitors:
             First Plaintiff:  Wallace Wilkinson & Webster
             Second Plaintiff:  Wallace Wilkinson & Webster
             Defendant:  Page Seager

Judgment Number: [2013] TASSC 23
Number of paragraphs: 138

Serial No 23/2013

File No 495/2009

ANDREW KEVIN LANGMAID and KEVIN GEORGE LANGMAID t/as CHERRY HILL COOLSTORES (ABN 87 065 420 794), LATROBE COOL STORES PTY LTD v DOBSONS VEGETABLE MACHINERY PTY LTD t/as DOBMAC AGRICULTURAL MACHINERY

REASONS FOR JUDGMENT TENNENT J
31 May 2013

As at March 2006, the first plaintiff was operating businesses at premises at Cherry Hills Road, Latrobe. The first involved the purchase, grading and sale of potatoes to wholesalers, and the second involved the provision of cool store facilities to potato growers. It owned the plant and equipment situated at the premises. The premises were owned by the second plaintiff. The partners in the business conducted by the first plaintiff were Andrew Langmaid and his father, Kevin Langmaid. They were also directors and shareholders of the second plaintiff. As at 28 March 2006, Andrew Langmaid lived in a house on the same site as the business.

Overnight on 28/29 March 2006, a fire destroyed buildings on the site and a large quantity of equipment housed in them. In the period leading up to that night, employees of the defendant were working in the relevant buildings, as were employees of the first plaintiff. At different times on 28 March, various people were doing "hot work" inside the buildings.

The plaintiffs suffered a significant loss as a result of the fire. They claim that the fire was caused by activities of employees of the defendant, and seek to recover their loss from the defendant. In substance, the defendant says that the Court could not be satisfied to the requisite degree that it was activities or failures by employees of the defendant which caused the fire and that, as a consequence, the plaintiffs' claim should fail.

Pleaded bases for plaintiffs' claim

There were three pleaded bases for the plaintiffs' claim. These were:

(a)       breach of statutory duty;
(b)       negligence; and
(c)       breach of contract.

Breach of statutory duty

This was pleaded at par[13] of the plaintiffs' statement of claim in the following terms:

"The fire was caused by breach, by the Defendant its servants or agents, of their duty pursuant to statute.
PARTICULARS OF BREACH OF STATUTORY DUTY
The Defendant, its servants and/or agents committed breaches of statutory duty in that it, he or they, failed to comply with regulation 20 of the General Fire Regulations 2000 by failing to comply with Australian Standard AS1674 in the following respects;
(a)Failed to make any arrangements, before undertaking the works, to prevent any hot products from lodging in places where there was a possibility of starting a fire;
(b)Failed to thoroughly inspect and make safe the area of works before undertaking the works;
(c)Failed to thoroughly inspect the area of works to ensure that it was safe on completion;
(d)Failed to sufficiently contain hot products so that they did not reach or remain in a place where they could start a fire;

(e) Failed to take care to protect combustible material and/or from hot products;
(f)Failed to take care to protect the surface of the drive roller from hot products;
(g)Failed to use protective guards to prevent hot products from falling or rolling beneath the guards or penetrating openings in them, or falling on combustible material;
(h)Failed to undertake adequate final inspection after the works to ensure that no hot products, hot, or smouldering materials or hot debris remained so that combustible material, or the grading line, could catch fire.
(i)Failed to protect combustible material, including timber, adjacent to or below the hot works by wetting or other suitable means."

The defendant denied these allegations.

Relevant provisions of Australian Standard AS 1674 ("AS 1674") were cl 2.1 (Supervision), cl 2.4 (Inspection), cl 2.7 (Off-cuts and electrode stubs), cl 2.8 (Timber), cl 2.10 (Other materials), cl 2.12 (Protective Guards) and cl 5.3 (Final Inspection). These provided:

"2.1   SUPERVISION Hot work shall be carried out under the control of a person who is responsible for the safe execution of all operations and has authority to enforce the requirements of this Standard with respect to other employees, outside contractors and other people in the area. Before hot work is commenced in any location, this person shall ensure that –
(a)     the hazards of the location are identified;
(b)     a means of managing the hazards is in place;
(c)     the equipment complies with the requirements of Section 4;
(d)the equipment is located so that, in the event of malfunction of the equipment, a fire or explosion hazard is not created; and
(e)there is no inherent hazard due to the nature of the item on which the hot work is to be performed.
NOTE:Where hot work is being carried out by a contractor, the person responsible for the hot work should be approved by both the principal and the contractor.

…
2.4      INSPECTION OF SITE Before hot work commences, the site shall be thoroughly inspected and made safe, or alternative methods of carrying out the work shall be adopted.
On completion of hot work, a thorough inspection of the site shall be carried out to ensure that the site is safe.
…
2.7      OFF-CUTS AND ELECTRODE STUBS Before any hot work commences, arrangements shall be made to prevent any work off-cuts, hot metal, slag or electrode stubs from lodging in places where there is any possibility of starting a fire.
2.8      TIMBER Where any hot work is to be carried out adjacent to or above timber, the timber shall be protected, by wetting or other suitable means, from the direct heat of any flame or arc and from sparks, slag and hot metal particles.
…
2.10     OTHER MATERIALS Where hot work is to be carried out in the presence of other materials, care should be taken to protect the surfaces of those materials from damage or from starting a fire.
…
2.12     PROTECTIVE GUARDS Protective guards shall be stable, of ample size and arranged so as to prevent sparks, slag and hot metal particles from rolling beneath them or penetrating openings in them. Protective guards shall be of fire-resistant material.
…
5.3      FINAL INSPECTION After hot work has been completed, inspections shall be carried out to ensure that no smouldering materials remain."

The term "hot work" is defined in AS 1674 to mean:

"grinding, welding, thermal or oxygen cutting or heating, and other related heat-producing or spark-producing operations".

The term also appears in the General Fire Regulations 2000 ("the Fire regulations"). Those regulations provide that the meaning of the term in them is the same as in AS 1674. Regulation 20 of the Fire regulations also provides:

"A person must –
…
(b) in performing the hot work, observe the requirements of AS 1674."

Negligence

This was pleaded at par[14] of the plaintiffs' statement of claim as follows:

"Further, or in the alternative, the fire was caused by the negligence of the Defendant its servants or agents.
PARTICULARS OF NEGLIGENCE
The Defendant, its servants and/or agents were negligent in that it, he or they, failed to;
(a)Comply with regulation 20 of the General Fire Regulations 2000 by failing to comply with Australian Standard AS1674;
(b)The Plaintiffs repeat as particulars of this paragraph the particulars contained in paragraphs 13(a to i) inclusive above;
(c)Failed to refer to and apply the recommendations and observe the requirements contained in AS 1674;
(d)Failed to undertake adequate risk analysis of the fire hazard arising from the works which were required during the course of upgrading the grading line;
(e)Caused hot products to fall or lodge in places where there was a possibility of starting a fire;
(f)Failed to use any or any adequate protective guards to prevent hot products from falling or lodging in places where there was a possibility of starting a fire;
(g)Failed to undertake an adequate final inspection after the works to ensure that no hot products or smouldering materials remained;
(h)Failed to shield or adequately shield material (including combustible material) in the vicinity of the works from hot products;
(i)Failed to shield material (including combustible material) in the vicinity of the works with non-combustible welding drapes;
(j)Failed to wash the rubber belting component of the line down by hand at completion of the hot works."

The defendant denied these allegations.

The plaintiffs, in effect, contended that the defendant owed a duty to them to take all reasonable steps to prevent fire breaking out as a consequence of hot work undertaken by its employees. The scope of that duty required the defendant to comply with AS 1674, undertake an inspection of the work site to identify combustible material, use protective guards to prevent hot products coming into contact with such combustible material, undertake a proper inspection of the site after hot work had been done to ensure no hot products remained, and wet down combustible materials in the area of the hot work. The plaintiffs contended that the defendant had failed to do these things, and hence the duty of care had been breached. The breaches were, they asserted, causative of the fire. As a consequence, the requirements of the Civil Liability Act 2002 ("the Act"), ss13 and 14, had been met. The Court could be satisfied that it was more likely than not that, but for the hot work performed by the defendant, the fire would not otherwise have occurred.

Breach of contract

By par[6] of the plaintiffs' statement of claim, the plaintiffs set out what they asserted was the contract entered into between the parties. Paragraph [6] provided as follows:

"6. In early 2006 the Defendant agreed with the First Plaintiff that it would, for reward;
(a) supply a new mechanical potato sizer and conveyor belt ('the sizer') for Grading Area No 1;
(b) install the sizer and perform additional work at the premises to upgrade the entire grading area, in conjunction with employees of the First Plaintiff.

PARTICULARS OF THE AGREEMENT

(i)The agreement was verbal, comprising of a number of conversations between agents for the First Plaintiff and agents for the Defendant.
(ii)The original quoted contract price for the supply of the sizer was $45,000.00 and was submitted to the First Plaintiff by Mr Phillip Dobson, an agent of the Defendant.
(iii)It was agreed that work performed by the Defendant's agents in installing the sizer and performing additional upgrading work would be on a 'do and charge' basis, that is pursuant to an agreed hourly rate."

By par[6] of its defence, the defendant pleaded:

"The Defendant admits the allegations contained in paragraph 6 of the Amended Statement of Claim, save and except that those acting on behalf of the Defendant were its employees. It was an implied term of the agreement that:
a)        Andrew Langmaid would be, for the purposes of Australian Standard AS1674.1, the person who was responsible for the safe execution of all works and operations to be performed at the premises and had the authority to enforce the requirements of AS1674.1 with respect to any persons working at the premises.
This term is to be implied by reason of:
(i)        The work to be performed by the Defendant was a part only of work performed on grading line no 1;
(ii)       The First Plaintiff engaged others to work on grading line no 1 including its own employee Michael Cooper and employees of contractors Shane Hill Electrical, Central Coast Electrical and Dennis McCormack;
(iii)      The First Plaintiff had their own welding equipment on site which was used by the First Plaintiff's employees and the Defendant's employees;
(iv)      Andrew Langmaid in fact controlled all operations on grading line no 1 as throughout the course of completing those works he was present at the premises co-ordinating and supervising the works in relation to setup, providing deadlines and giving directions as to the completion of the works which determined the First Defendant's start and finish times while completing the works;
(v)       Andrew Langmaid was overseeing the work of the First Defendant's employees and had numerous conversations with them each day in relation to heights, drop points and the layout of the grading line;
(vi)      Andrew Langmaid gave directions to the employees of the First Defendant in relation to fire safety by reminding them of the fire that had occurred at the premises earlier in March 2006 and ensuring that they incorporated the practice of ceasing hot work, cleaning up and checking for signs of fire;
(vii)     On the day of the fire Andrew Langmaid instructed all those working on the line that it had to be finished that day and in that regard he also directed Michael Cooper to assist in completing the works at the northern end of the grading line by installing a waste conveyor at approximately 5:00pm on 28 March 2006 and Andrew Langmaid gave directions as to the standard of the works being completed;
(viii)     Australian Standard AS1674.1 required the appointment of a responsible person;
(ix) Andrew Langmaid was the person responsible for the direction and management of the business of the First Plaintiff at the workplace within the meaning of s. 10 of the Workplace Health and Safety Act 1995; and
(x)       Andrew Langmaid was responsible for ultimately deciding how the grading line would be completed and how the grading line layout would look so that it met the First Plaintiff's production requirements."

By par[7] of the statement of claim, the plaintiffs also asserted the following:

"It was an implied term of the agreement and/or it was the duty of the Defendant that all work performed by the Defendant, its servants and agents at the premises would be undertaken;
(a)       in a safe and competent manner,
(b)       take reasonable care for the safety of the Plaintiffs' property;
(c)       provide a safe system of work;
(d)       in compliance with;
(i)        the provisions of the Fire Service Act;
(ii)       The General Fire Regulations 2000
(iii)      Australian Standard AS1674."

By par[7] of its defence, the defendant pleaded:

"In relation to paragraph 7 it says that what terms were implied into the agreement or what elements were part of its duty is a matter of law and thus the Defendant does not plead further to paragraph 7 of the Amended Statement of Claim."

The defendant further pleaded at pars[20] to [22]:

"20. At all material times Andrew Langmaid was the person in control of all operations at the premises within the meaning of AS1674.1 and was the person responsible for the direction and management of the business of the First Plaintiff at the workplace within the meaning of s. 10 of the Workplace Health and Safety Act 1995 including the direction and management of the employees of the Defendant.
PARTICULARS
a)Andrew Langmaid was the person provided by the First Plaintiffs with authority to direct and control works on the premises;
b)The First Plaintiff engaged various contractors including Shane Hill Electrical and Central Coast Electrical and its own employees including Michael Cooper and Dennis McCormack to work on the premises and ensured that their works were co-ordinated by being present at the premises as the owner and occupier throughout 28 March 2006 giving verbal directions and instructions to all of those people undertaking on the completion of the works;
c)Andrew Langmaid gave directions to those working at the premises including the First Plaintiff's employees and in that regard he directed Michael Cooper to assist in completing the works at the northern end of the grading line by installing a waste conveyor at approximately 5:00pm on 28 March 2006 as well as giving directions to the employees of the Defendant including that the works were to be completed by 28 March 2006 as well has [sic] directing them in relation to heights, drop points and the layout of the grading line;
d)Andrew Langmaid gave directions to the employees of the First Defendant in relation to fire safety by reminding them of the fire that had occurred at the premises earlier in March 2006 and ensuring that they incorporated the practice of ceasing hot work, cleaning up and checking for signs of fire;
e)Andrew Langmaid supervised work carried out by employees of the Defendant and other persons working at the premises as he gave directions as to the standard of the works and the deadline for the completion of the works which determined the employees of the First Defendant's start and finish times; and
f)Andrew Langmaid was responsible for ultimately deciding how the grading line would be completed and how the grading line layout would look so that it met the First Plaintiff's production requirements."

21The work performed by the First Plaintiff's employee Michael Cooper on 28 March 2006 was hot work and as likely a source of ignition as any work carried out by the Defendant's employees.
22The Defendant says if it were in breach of its agreement and/or in breach of statutory duty and/or negligent in that it is determined by the Court that its actions or omissions caused the fire at the premises (all of which is denied), then it says the First Plaintiff was in breach of its statutory duty and contributorily negligent."

In response to those pars[6], [20], [21] and [22], the plaintiffs asserted at par[2] of their reply:

"2Further, the Plaintiffs deny paragraphs 6, 20 and 22 of the Defendant's Amended Defence and each and every particular therein and say:
a)   At no material time did the Defendant act as the servant or agent of the First or Second Plaintiff.
b)   The Defendant agreed to and did, at all material times, execute the work it undertook in accordance with the Agreement outlined in paragraph 6 of the First and Second Plaintiffs' Statement of Claim ('the Agreement'), as an independent function.
c)   The First and Second Plaintiffs relied on the expertise, knowledge and skill of the Defendant to perform all aspects of the Agreement.
d)   Any obligation pursuant to Australian Standard AS 1674.1 to appoint a responsible person for the safe execution of hot works was an obligation of the Defendant.
e) Any obligation to observe the requirements of Australian Standard AS 1674.1 and appoint a responsible person for the safe execution of all works and operations was the obligation of the Defendant, pursuant to reg.20 of the General Fire Regulations 2000, as the person performing the hot works.
f)   At the time of the works, the property in the sizer had not passed to the First Plaintiff and remained at the Defendant's risk."

The plaintiffs' plea in respect of this cause of action appeared in par[15] of their statement of claim in the following terms:

"Further, or in the alternative the fire was caused by the Defendant's breach of the implied term of the agreement referred to in paragraph 7 hereof
PARTICULARS OF BREACH OF AGREEMENT
The Plaintiffs repeat as particulars of this paragraph the particulars ~~contained in paragraphs repeat the particulars set out in~~ paragraphs 13 and 14 above."

The defendant denied this assertion.

Particulars of loss and damage were supplied. There was almost complete agreement between the parties as to what that loss was and the quantum of it in the event the plaintiffs' claim succeeded.

Issues

Notwithstanding that there were three bases for the plaintiffs' action against the defendant, what underpins them all is that the plaintiffs need to prove that:

·    The defendant carried out certain activities, and/or
·    The defendant was responsible for certain failures, and
·    there was a causal connection between any activities or failures of the defendant found to have occurred and the fire.

The issues and who bears the onus of proof

There was no dispute between the parties that the central issue on this trial was causation.

At the start of his closing address, counsel for the defendant addressed issues of proof. He did so at that point to address what he perceived to be a particular approach to be adopted by counsel for the plaintiffs. In his written closing, under a heading "Causation", counsel for the plaintiffs submitted at par[97] that:

"If it is found that the breach increased the risk of damage the court is entitled to find that the conduct caused the damage unless the defendants establish that the conduct had no effect at all or the damage would have occurred in any event."

Foot noted to that paragraph was the decision of Crockett v Roberts & Anor [2002] TASSC 73 and a number of cases referred to by Underwood J (as he then was) at par[63] of that decision, which included Betts v Whittingslowe (1945) 71 CLR 637. Counsel for the defendant read that paragraph as an attempt by the plaintiffs to shift the burden to the defendant to, in effect, prove its employees did not cause the fire.

Counsel for the defendant submitted that Crockett's case was decided prior to the introduction of the Act, s14, and that the introduction of that section put all of the material referred to by Underwood J to rest. The Act, s14, provides as follows:

"In deciding liability for breach of a duty, the plaintiff always bears the onus of proving, on the balance of probabilities, any fact on which the plaintiff wishes to rely relevant to the issue of causation".

The High Court recently dealt with the New South Wales equivalent of s14 in Strong v Woolworths Ltd [2012] HCA 5. While the wording of s14 and s5E of the relevant legislation in New South Wales is not identical, it is sufficiently similar for the case to be relied on in this State. The case involved a customer who slipped and fell in a Woolworths supermarket. In the High Court, the customer's appeal was dismissed. Both counsel before me referred to the dissenting judgment of Heydon J, although to different paragraphs. His Honour identified that the question in that appeal was causation. He considered arguments raised by the appellant about the "evidential burden". He considered, in particular, statements relied on of Lawton J and Megaw J in Ward v Tesco Stores Ltd [1976] 1 WLR 810 at 814 and 816; [1976] 1 All ER 219 at 222 and 224, and the arguments made by reference to those statements. He said at par[60]:

"The better view is that the 'evidential burden' to which Lawton LJ referred was the 'provisional' or 'tactical' burden of meeting the plaintiff's evidence or facing the possible peril that the trier of fact would draw inferences from it sufficient to satisfy the legal (ie persuasive) burden resting on the plaintiff. This is what Jacobs J was referring to when he said that in some circumstances 'the plaintiff need only produce slight evidence of negligence before a factual onus may shift to a defendant.' That is an 'evidential burden' in the second sense discussed above."

His Honour went on to say at par[63] about that evidential burden:
"It follows that if 'evidential burden' is used in the second sense to mean that the strength of a plaintiff's causation case may imperil the defendant who fails to answer it, then it was possible that the strength in the appellant's causation case here would imperil the first respondent, if it went unanswered. But that begs the question whether there was strength in the appellant's case on causation."

His Honour then made clear his view about s5E at par[64] where he said:
"There is one submission of the first respondent which must be rejected. That submission was that s 5E, when read in light of the report that led to its enactment, was a complete answer to the appellant's submissions about the evidential burden. In fact s 5E addresses a different problem. It deals with a proposition stated thus in Bennett v Minister for Community Welfare:
'[G]enerally speaking, if an injury occurs within an area of foreseeable risk, then, in the absence of evidence that the breach had no effect, or that the injury would have occurred even if the duty had been performed, it will be taken that the breach of the common law duty caused or materially contributed to the injury.'

As the appellant correctly pointed out, that passage concerned a doctrine permitting the shifting of the legal (ie persuasive) burden of proof of causation, and that doctrine was what s 5E abolished. The submission of the appellant about the 'evidential burden' does not seek to shift a legal (ie persuasive) burden of proof. Its goals are less ambitious."

There was also discussion about a further case, that of Amaca Pty Ltd v Booth [2011] HCA 53. That was an asbestos case. Under the heading "Risk of harm and factual causation", French CJ said at pars[41] – [43]:

"41Causation in tort is not established merely because the allegedly tortious act or omission increased a risk of injury. The risk of an occurrence and the cause of the occurrence are quite different things. That proposition is obvious enough and not determinative of these appeals.
42It is necessary, nevertheless, to reflect upon the relationship between risk and causation. In ordinary usage 'risk' refers to a hazard or danger or the chance or hazard of loss. Assessment of the risk of an occurrence is prospective in character. It can be expressed as an ex ante probability that the occurrence will occur. If quantifiable, that probability may be expressed numerically as a figure greater than 'zero' up to 'one' which denotes certainty. The range of probabilities may be traversed by terms such as 'mere possibility', 'real chance', 'more likely than not', 'highly likely' and, ultimately, 'certainty'.
43The existence of an association or a positive statistical correlation between the occurrence of one event and the subsequent occurrence of another may be expressed as a possibility, which may be no greater than a 'real chance' that, if the first event occurs, the second event will also occur. The mere existence of such an association or correlation does not justify a statement, relevant to factual causation in law, that the first event 'creates' or 'gives rise to' or 'increases' the probability that the second event will occur. … An after-the-event inference of causal connection may be reached on the civil standard of proof, namely, balance of probabilities, notwithstanding that the statistical correlation between the first event and the second event indicated, prospectively, no more than a 'mere possibility' or 'real chance' that the second event would occur given the first event. There may of course be cases in which the strength of the association, as measured by relative risk ratios, itself supports an inference of a causal connection."

His Honour went on to say at [49]:
"49In summary, a finding that a defendant's conduct has increased the risk of injury to the plaintiff must rest upon more than a mere statistical correlation between that kind of conduct and that kind of injury. It requires the existence of a causal connection between the conduct and the injury, albeit other causative factors may be in play. As demonstrated by medical evidence in this case and in particular by Professor Henderson's evidence, a causal connection may be inferred by somebody expert in the relevant field considering the nature and incidents of the correlation. The Bradford Hill criteria provide a guide to the kind of considerations that lead to an inference of causal connection. As noted above, they may include reference to relative risk ratio as an indicator of the strength of the association. Where the existence of a causal connection is accepted it can support an inference, in the particular case, when injury has eventuated, that the defendant's conduct was a cause of the injury. Professor Henderson offered that inference of specific causation by reference to Mr Booth's exposure to the products of both Amaca and Amaba. Where such an inference is drawn, the probability that it is correct is not to be determined only by reference to epidemiologically based ex ante probabilities. In Betts v Whittingslowe, Dixon J employed apposite logic when he said:
'the breach of duty coupled with an accident of the kind that might thereby be caused is enough to justify an inference, in the absence of any sufficient reason to the contrary, that in fact the accident did occur owing to the act or omission amounting to the breach of statutory duty.' (emphasis added)

That logic encompasses the case of an ex ante probability, of accident given breach, supported by a causal explanation linking breach and accident. In this case an explanatory causal mechanism was proposed in the medical evidence."

Counsel for the defendant also referred to remarks by his Honour in Amaca's case which appeared at par[47] where his Honour said:

"47 … Factual causation which can be established by the application of the 'but for' test is 'the threshold test for determining whether a particular act or omission qualifies as a cause of the damage sustained.' That threshold must also be surmounted in the case of concurrent or successive tortious acts:
'it is for the plaintiff to establish that his or her injuries are "caused or materially contributed to" by the defendant's wrongful conduct ... Generally speaking, that causal connexion is established if it appears that the plaintiff would not have sustained his or her injuries had the defendant not been negligent'."

Counsel for the defendant submitted by reference to the remarks at par[47] that in the present case, if there was one tortious act, there might be two, that is the performance of hot work by Mr Cooper and Mr Maine. That is, there could have been concurrent acts or at least successive acts very close together. It is for the plaintiff to establish that the fire was caused or materially contributed to by the defendant's wrongful conduct (the acts of Mr Mayne). The question must be, he argued, would the plaintiffs have sustained the damage they did had the defendant not been negligent? Counsel submitted that I could not possibly draw that conclusion because there were workers from both the first plaintiff and the defendant doing the same type of work in the lead up to the fire.

In essence, counsel for both parties were not really at odds as to the law to be applied. Neither cavilled with the impact of the Act, s14. Counsel for the plaintiffs identified the manner in which he sought to use Crockett's case. He said that what was to be considered was a question of inference. At what point in a case is a trier of fact entitled to draw an inference that the plaintiff has established the causal link? There was no attempt to shift the ultimate burden to the defendant.

The trial

The trial commenced with a view of the site. That was a useful exercise in that, despite the fact that the buildings and equipment on site at the time were different from those in 2006, it enabled a clearer understanding of the way in which the area was set out in 2006, and assisted with orientation of buildings and equipment on site when witnesses gave their evidence. The principals of both parties gave evidence, as did various employees and contractors of the first plaintiff and the defendant who had been working on site on 28 March. There was also evidence from a Tasmania Fire Service ("TFS") investigation officer, fire investigation experts for each side and other experts.

The focus of the trial was upon how and where the fire started.

A plan was tendered which showed the layout of the building containing the cool stores and grading line at the time of the fire. It is annexed to this judgment. It showed that the grading line was almost hard up against the eastern wall of the building. On the western side of the machinery which made up that line, there was a space. On the other side of that space, about 10 to 12 metres from the line, were cool store facilities. The eastern wall of the building did not go right to the ground and there was effectively no wall on the southern end of the shed. The electronic sizer, or E-sizer, about which evidence was given, was a component of the grading line. There was also what was described as a computer room, external to the building where the grading line was, which was on the north-east corner of the building. That room was entered via a door from outside the main building.

Also tendered at the trial were three sets of photographs. One set was taken by TFS representatives. Each photograph was numbered, and, whenever it is necessary in these reasons to refer to one of those photographs, it will be referred to as TFS and whatever the appropriate number is. The second set was taken by Colin Thomas, a fire scene examiner, called as a witness for the plaintiffs. Each photograph in that set was also numbered and will be referred to, where necessary, as CT and whatever the relevant number is. The third set was one produced by the defendant. Those photographs were also numbered and are referred to, where necessary, as DM and the relevant number.

General factual background

The first plaintiff's business involved it in receiving potatoes from various producers, grading them and storing them. The fire occurred in a building which housed cool store areas and grading equipment. The grading equipment was fed from outside the eastern side of the building by an elevator which allowed trucks to unload potatoes and for them to be directly carried into the building. In the building, running along the eastern side oriented north/south, was a grading line. Very generally, that consisted of a conveyor belt which carried potatoes through sizers. Those sizers sorted potatoes into various size potatoes, with each size exiting the line on another different conveyor belt. One of those sizers was the E-sizer, which was operated by computer which was in the computer room. The grading line was about 40 metres in length.

The defendant was a business which specialised in the manufacture and installation of equipment used by the first plaintiff. The first plaintiff and the defendant have had a long and mutually acceptable commercial history and, I infer, still do so. The defendant from time to time supplied and installed equipment for the first plaintiff, and its employees were on site perhaps two to three times a year.

The E-sizer, on the grading line at the time of the fire, had been purchased by the first plaintiff some years before through the agency of the defendant. It was, as I have already indicated, a computer operated sizer and very accurate. It assisted the first plaintiff in developing an overseas market for small potatoes. The development and maintenance of that market relied on the first plaintiff being able to accurately size small potatoes. When the machine was purchased, the first plaintiff had been given assurances that the E-sizer could grade potatoes at the rate of about 15,000 tonnes per hour. That claim ultimately proved to be overstated. The accuracy of the sizer was adversely affected by the volume of potatoes put through it, in that the greater the volume the less accurate the machine became. Further, the machine was made up of a number of parts. They were expensive to replace, and it was generally expensive to maintain the sizer.

The first plaintiff and the defendant over time had discussions about how this problem might be remedied. It seems the conclusion reached was to somehow take away from the E-sizer the volume which was affecting its accuracy. The plan eventually devised was that another sizing machine would be manufactured and installed with associated conveyor belts to, in effect, catch, before they reached the E-sizer, a significant quantity of larger odd shaped potatoes. The defendant supplied the first plaintiff with specifications for new equipment and the modifications needed to the existing grading line so the new equipment would fit in. The defendant was then engaged to manufacture this new equipment while the first plaintiff undertook the necessary modifications. The first plaintiff was to do the actual installation of the new equipment.

The preparatory work to be done by the first plaintiff needed to be done in the grading area before the new equipment could be installed. This was begun by employees of the first plaintiff and a contracted electrician in the weeks before 28 March. The existing grading line had to be cut to enable it to be lengthened to fit the new sizer in. That involved a physical cutting of the framework of the line as well as the unbolting of bolts and the pulling of the separated pieces apart to allow a space. It also involved the disconnection of wiring to motors which operated the grading line generally, and the installation of new wiring to accommodate the changes to the line and the new sizer. The changes resulted in the northern end of the grading line being closer to the northern wall of the building than it used to be. It ended about a metre from the wall.

The employees of the first plaintiff principally involved in this work were Michael Cooper and Dennis McCormick. The work the men were doing involved, at times, "hot work", that is the men were involved in grinding, oxyacetylene cutting and welding. As a consequence, there had been an oxyacetylene torch, a Mig welder and other tools in the grading area.

The project began to run behind time. Discussions were held with a view to the defendant assisting with the installation. On about 23 March 2006, two employees of the defendant, Rodney Maine and Scott Nalder, arrived with the new equipment. They had both worked at the Cherry Hills site before. As a consequence, in the few days leading up to 28 March, employees of both the first plaintiff and the defendant, and an electrician contracted to the first plaintiff, were working in the grading area on the installation of the new equipment. Employees of both parties were at times engaged in what was called "hot work". According to Andrew Langmaid, the installation was being done by Maine and Nalder with Cooper assisting.

Work done in the grading area on 28 March 2006

The evidence focused primarily on the activities of Michael Cooper, Dennis McCormick, Rodney Maine and Scott Nalder on the afternoon and early evening of 28 March. There was some evidence about the activities of a Peter Dutton, an electrician contracted to the first plaintiff.

Michael Cooper and Dennis McCormick

Michael Cooper was employed by the first plaintiff to run the grading shed. He was also responsible for general maintenance. He had no trade qualifications. Prior to the work commencing on the changes to the grading line in 2006, he discussed with Andrew Langmaid what was going to be done. He knew that the southern end of the existing grading line had to be raised, that the line had to be lengthened to accommodate some new machinery, and that one section of the line actually had to be removed. He also knew that, to achieve that removal and lengthening, dyna bolts had to be cut off the floor and the machinery unbolted and that electricians had to come in and change the wiring to motors. Some of the work involved cutting and grinding because the bottoms of legs on some machinery needed to be cut and sections added to make them longer. This preparation work was predominantly done by him and Dennis McCormick.

When the defendant's employees arrived on site a few days before the fire to install the new sizer, Cooper said that he and Maine and Nalder discussed what needed to be done. There was no agreement that any one person would be in charge. Cooper said everyone knew what needed to be done, and when it had to be done by, and they just got on with it. He did not give any instructions to anyone and he did not see Andrew Langmaid do that either.

On 28 March, Cooper, McCormick, Maine and Nalder were all working in the area of the grading line. By about 4pm, McCormick had finished his work and gone.

McCormick had primarily been involved in the lengthening of the grading line and the raising of the southern end. On that day, he was working near the box tipper (the green machine apparent in TFS7) for the purpose of raising the elevator. He recalled having to cut legs on the machinery for that purpose. He believed he was working no more than about five metres from the southern end of the line. He believed he may have performed cutting there, that the upturned wooden box shown in TFS8 had been used to put tools on, but that he did not do any cutting there. He recalled finishing work about 4pm.

TFS24 showed the northern end of the grading line after the fire. In the centre of the photograph was the main conveyor frame. It was supported by a metal framework. At the northern end of it were the remains of a drive roller and, to the western side of the frame beside the roller, a chain guard can be seen. Leading down at an angle from the conveyor frame is a conveyor with one end either on or almost on the ground. At the point where the conveyor hits the ground, there is what was identified as a piece of rolled hollow steel.

By reference to the conveyor seen in TFS24, Cooper said that, probably some time that morning, he had used a forklift to put a large wooden box under the conveyor to prop it up. On top of the box, he put a piece of rolled hollow steel. The purpose of putting the box and the piece of steel there was to raise the end of the conveyor to the height and angle it would ultimately need to be positioned in. Cooper estimated that the start of the wooden box would have been about a metre from the chain guard. Cooper was asked whether he used any PVC buckets in the area that day. He said it was very possible he had, to get a bit more height on the conveyor. If so, they would have been about 500 to 600 millimetres from the chain guard.

Cooper said that, at around 4pm, Nalder was working in about the middle of the grading line on a divider conveyor. At or around the same time, Maine wanted to do some work on the northern end of the divider conveyor to fit the chain guard on the side to protect the chain seen in TFS24. Cooper was however already working at the northern end of the line trying to fit the conveyor to the line. Cooper went ahead with his work. Once he finished it, a bit after 4pm, he moved to the southern end of the line where he worked to fit a handrail on a feeder bin. He did that so Maine could work on the chain guard. Cooper said he finished his work at the southern end of the line around 7pm. There was virtually no evidence about precisely what work Cooper did at the southern end of the line or what it involved.

In his work to fit the conveyor at the northern end, Cooper said he made a cut in the conveyor which he identified in CT13. He made that cut from the eastern side of the conveyor using a five inch angle grinder. The cuts would have taken a matter of seconds. His work was about 500 to 600 millimetres off the ground. He cut straight down and the sparks generated fell straight to the ground. While he was doing that work, the edge of the wooden box was positioned about 700 to 800 millimetres to his right. The PVC bucket he thought was about the same distance although he was not entirely sure there was even one there.

Cooper identified Maine as using a grinder in his work to fit the chain guard. He did not make any particular observations about what Maine was doing, but simply said goodbye to him as he left.

Cooper told the Court he had made use of wooden boxes before to prop things up when he did hot work. He had not seen any problem with fire being generated by hot work near the wooden boxes, PVC buckets or belts on conveyors. Cooper was asked about TFS43. He agreed it showed an oxyacetaline bottle standing beside a wooden box. He agreed tools and the like were stored there but not that hot work was done there. That was consistent with what McCormick had said.

Rodney Maine and Scott Nalder

Maine was a qualified fitter and machinist and, as at the time of the fire, experienced in welding, grinding and cutting metal although he did not have any formal qualifications in welding. He had been employed by the defendant for some years. That company had procedures in relation to safety for hot work which employees were required to sign. That signed by Maine was tendered in evidence. Maine described the type of safety precautions he was used to taking around hot work, such as checking the immediate area for flammable or combustible materials, checking for people in the immediate area, making sure the area was clean and isolating the area to ensure it was safe to work.

His recollection was that he was on-site at Cherry Hills for about five days before the fire. While working there, he had done welding and grinding work near conveyor belts but had not noticed any sign of fire being generated by that work. He was unsure whether he had ever done welding or grinding work near the wooden bins used at Cherry Hills. He had had experience before of doing hot work near the PVC bins, and had never seen one burn. He was asked about the cleanliness of the grading area, and described it as generally clean with no obviously flammable things around.

The work Maine did in relation to the chain guard included cutting a piece of metal on the ground and then welding it, and then welding a block to that. He used a five inch grinder to do the cutting which took him about 30 seconds. He did not notice or smell any sign of fire. He said he finished the cutting at about 6pm. That cutting was done on a pallet about two to three metres to the west of the grading line. The sparks mainly went straight onto the floor, although he agreed some may have gone to the east towards the conveyor and the wooden box. For the welding which followed, he used a 175 amp gasless MIG welder. The welds took a matter of seconds. To protect the conveyor belt while he welded, he put a piece of plate in between it and where the welding was. After doing the welding, Maine also used a cordless drill to screw some screws in. He believed he finished the work on the guard at about 7pm.

Maine was asked about the presence of chargers for cordless drills, and he said there were a couple on a bench on the wall against the cool stores. They were switched on when he left.

Maine said he tidied up before he left, packing tools away. He put the MIG welder he had been using in the wooden box which Cooper had propped the conveyor on, simply to get it out of sight. When he left, he neither saw nor smelled any indication of a fire. Maine agreed that, while he looked for combustible materials, such as fluids and the like, he did not perceive the conveyor belt and rollers to be combustible. He also did not wet down the wooden box before he did the hot work he did, and he did not specifically inspect for any signs of fire before he left the grading area that night. He was however aware that there was a fire extinguisher just inside the door to the shed and they had a fire extinguisher and fire blankets in their mobile workshop. He and Nalder then left at the same time. Before they did so, they had to walk up and down the building to find out where the light switch was to turn the lights off. They left the shed about 45 minutes after Maine finished the work he was doing, which would have made it about 7.45pm.

Maine also told the Court that he watched Cooper do his cut on the conveyor and neither saw nor smelt any burning after that. Maine acknowledged that, as at March 2006, he was not aware of the Australian Standard relating to welding activity, and had not received any training in relation to it. A number of the provisions of the standard were read to him and he agreed they all represented sound practice.

Nalder was also using welding equipment and working about half way along the grading line (it was described as being near the letter "A" at the beginning of the word "Area" in the term "Grading Area" on exhibit P10). Prior to starting the work he had checked to see if the area was clean – he said it was a cement floor and there was no debris such as wood or paper. He finished the hot work he was doing at around 7pm, and did not then see or smell any sign of fire. After cleaning up and putting tools away, he left with Maine at about 7.45pm and still neither saw nor smelt any sign of fire.

Peter Dutton

Dutton was an electrician employed by Shane Hills Electrical. He had been working on site in the days before 28 March. He said they turned the power off to the grading line and cut the electrical wiring so that the grading line could be cut and pulled apart. After that was done, he had to replace all the cabling and reconnect all the wiring between the parts of the existing machinery, and make allowances for interconnection to the new machinery.

By reference to exhibit P10, a plan of the buildings pre-fire, Dutton was shown where a switchboard appeared on the wall of cool store 8 opposite the grading line. He said he had not worked on that switchboard on 28 March. He was instead working on a switchboard on the original grading line. He identified that area as being where the letter "A" at the beginning of the words "Grading Area" appeared on the plan. That was about the middle of the old grading line. It was, he said, the interconnect point between the two machines. He finished doing the wiring up and did some testing of a few motors ready for the commissioning the next day.

When he had finished that work, he packed up all his tools and equipment and isolated the electrical work he did (or effectively turned the power off to it).

During the course of his cross-examination, the term "control box" was introduced. Dutton confirmed that that was the same as a switchboard although, for the purpose of answering the questions being put to him, he designated the box referred to in the first sentence of par[61] as a switchboard and the boxes on the actual grading line as control boxes. It was suggested that Dutton had put an isolation tag on the switchboard. He said he could not remember if he had put it there, or on one or more of the control boxes.

Andrew Langmaid

Andrew Langmaid said he went into the grading area where the works were being done briefly on maybe two or three occasions on 28 March. The last time he saw the area before the fire was at about 7.45pm. There was no-one there. He turned the lights on and walked from one end of the shed to the other just to get an idea of where the work was at and what had to happen the following morning. He could have been in the shed for maybe ten minutes just going through in his mind the anticipated commissioning of the new equipment the next day.

Mr Langmaid next saw the area in the early hours of the following morning. When he saw it, the cool stores had very active flames leaping from them. The shed over the grading line had collapsed down onto the line and that area was smouldering and smokey. The computer shed, he said, was destroyed. The roof over the cool store area was still standing and was not collapsed as was shown in the TFS photographs.

Tasmania Fire Service involvement and investigation

The fire was reported by a security guard driving towards the site in the early hours of 29 March. He called the fire in at 2.59am. He then drove to the site and alerted Andrew Langmaid whose house was nearby. I have described in the previous paragraph what Mr Langmaid saw. TFS employees were on site fairly quickly. It took some few hours to contain the fire. A fire investigator, Claudio Muench, arrived on scene at about 6am on 29 March. He prepared a report and gave evidence. No other attending fire officer gave evidence although the log maintained by the service was tendered which showed the time the fire was reported, the time that it was contained and the names of various people who attended at different times.

At 3.09am, the fire was seen to be burning well in the centre of the building which housed the grading line and cool stores. The front section of the complex had already collapsed. The fire was observed to be moving from the eastern to western side of the complex, that is from the grading line area towards the cool store area.

Mr Muench in his evidence identified certain signs which he used to identify the area of origin of the fire. The first was what was described as a "V" pattern on the outside of the wall on the eastern side of the building, that is the wall adjacent to the grading line. A "V" pattern can be formed on a vertical surface in a situation where a fire burns at a low level and then burns up and out in the shape of a "V". The second indication was the extent of overhead damage in a particular area. In an area above the northern end of the grading line, there had been complete destruction by fire of the timber batons to which the metal supports of the roof were attached, which resulted in the collapse of the roof down onto the grading equipment. This indicated that there had been prolonged burning in that particular area, although he could not indicate a time-frame beyond saying it could have been hours. This was reinforced by the obvious heat damage to the metal part of the roof structure, which was discoloured and bent.

Mr Muench determined that these various fire indicators suggested that the northern end of the grading line was the area of origin of the fire. He then conducted what he described in his report as "A process of elimination" to identify other possible ignition sources. He considered an electrical switchboard. In his report he said as to that:

"A switch board was positioned toward the northern end of the wall dividing Grading Area 1 and Cool Store 8. The potato grading machine had recently been connected to the switchboard by Peter Dutton (Shane Hill Electrical). The switch board cabinet was located laying flat on the floor face up with the door lying partially across it. The switch board cabinet had been affected by the fire, however the original coating was still evident and in reasonably good condition on the bottom of the cabinet. This indicated the fire had progressed from the top of the wall down to the bottom of the wall. The cabinet was lying on its back, on top of charred timber remains, indicating it had fallen from the wall after a considerable amount of fire destruction had taken place. The door to the cabinet was removed exposing the internal circuit breakers and wiring. The circuit breakers were still identifiable and in reasonably good condition indicating the fire had not originated from this area."

In his evidence, Mr Muench identified, by reference to a floor plan of the building prior to the fire, where it was the switchboard had been found. It was some distance from the northern end of the grading line to the west of it.

Further, in his report, Mr Muench said he could not identify what he described as the "lowest point of charring" due to the extensive damage. When he gave his evidence, he described the lowest point of charring as being the lowest point that had been affected by fire. Mr Muench adhered to that statement in his oral evidence but agreed that the remains of a wooden box shown in TFS147 was a low point. The remains of that box were just near the end of the conveyor, previously referred to as that on which Cooper was working on 28 March.

Mr Muench considered the MIG welder which had been placed in the wooden box propping up the conveyor by Maine when he went home. It was eliminated as a possible ignition source. He considered the computer room. He described the room as being of mostly timber and metal construction and observed that the internal framework had been totally consumed by fire. He said in relation to the room:

"Examination of the computer, monitor and the area where these items were located failed to provide any evidence of the fire originating from this area. The office was determined to have been an exposure, damaged as a result of heat transfer from the main fire."

Mr Muench, at page 10 of his report, set out his conclusion. It was as follows:

"It was evident that the type of hot works being carried in close proximity to potato grading machine left combustible components susceptible to igniting. The most probable cause of the fire was either weld slag or grinding sparks coming into contact with the combustible components of the machine. These in turn smouldered undetected for a period of time before generating sufficient flame and heat to enable the fires spread.
With consideration to all evidence gathered at the incident including witness statements, physical evidence and the elimination of any natural, mechanical, chemical cause the fire is determined to have been caused by work practices and deemed to be accidental in nature."

Mr Muench was unable to determine a point of origin for the fire. Nor was he able to be precise about just what combustible material may have ignited first. He suggested however that it may have been the rubber on the conveyor belt. Mr Muench initially adhered to that opinion in Court notwithstanding the deletion of certain material in his report.

It was subsequently put to Mr Muench that, if the combustible components on the grading machine were in fact not combustible, would his opinion remain as stated. He said it would not, and that he would need to look for another ignition source.

Extracts from the seventh edition of a text Kirk's Fire Investigation, which Mr Muench accepted was an authoritative text for fire investigators, were tendered to the Court. Mr Muench agreed with statements made in that which were read to him. These were as follows:

"For a fire to occur the following conditions must exist;
·Combustible fuel must be present.
·An oxidiser (such as oxygen in air) must be available in sufficient quantity.
·Energy as some means of ignition (e.g., heat) must be applied.
·The fuel and oxidiser must interact in a self-sustaining chain reaction.

The first three elements listed have long been described as the fire triangle, but the fourth must also be present if the fire created is to be continuous (self-sustaining), thus creating what is called the fire tetrahedron….
…
It must be remembered that removal of any one of the four elements results in extinguishment, and a careful analysis of all the factors is required to establish what happened, since mechanisms of ignition, combustion, and even sources of fuel, may not be obvious at first glance. Some are transient and require careful analysis of possible mechanisms before their presence and effects can be reasoned out.
…
Proper interpretation of a fire situation requires knowledge of the properties of the fuel, its physical state, and the character and dynamics of the fire itself. Thus, the knowledge of availability and suitability of fuels in specific instances may require more than every day experience.
…
No matter what the nature of the fuel, its susceptibility to ignition, or the duration of the suspected ignition source, it is important to remember that an ignition source must be of sufficient energy and in contact with (or at least capable of transferring heat to) appropriate fuel long enough to raise the fuel to its ignition temperature to trigger the combustion of that fuel under the ambient conditions present. That combustion may be in the form of self-sustaining smolder or open flames, and each requires particular conditions. As we have seen, flames, arcs, sparks, self-heating, and heated objects can all initiate fires under the right conditions, and all must be considered whenever a fire scene is being examined. It is not enough simply to have an ignition source with a given temperature. The cause of a fire is the source of energy, the first fuel ignited, and the circumstances under which those are brought together that resulted in ignition.

…
The analytical fire investigator must ensure that a postulated sequence of events in a fire not only fits observed conditions but also follows these laws. Reliance on published data to confirm some ignition hypotheses may be enough if the conditions of the published tests are a reasonable match to the fire conditions. If they are not, properly designed and executed tests may be necessary to prove or disprove a particular hypothesis."

The purpose no doubt of having Mr Muench acknowledge those statements was the defendant's subsequent argument based on the evidence it led about tests done on various materials taken by Mr Muench to be combustible, which suggested their level of combustibility was either nil or extremely low. The defendant had arranged for some tests to be conducted with a view to determining just how combustible the rubber belt of a conveyor, the wooden boxes used in the grading area and plastic bins used to collect potatoes, were. A video recording of the various tests was played in Court and in particular so that Mr Muench could see it.

Mr Muench accepted that the level of intensity and duration of the heat source from sparks and welding slag he observed in the videos was far higher than he had assumed for the purpose of the preparation of his report in relation to the fire.

Mr Muench was not specifically asked at any time, over what sort of period combustible materials may have smouldered before ignition and the presence of obvious flames.

A number of other matters were raised with Mr Muench in cross-examination and re-examination. As a consequence Mr Muench told the Court:

·     He did not consider there had been a flashover in the fire.
·     He did not agree that certain vertical spikes in a V pattern identified in TFS23 indicated an area of origin of the fire.
·     While he did not search that area for combustibles or look for severe low level burning, he did look all along that area for overhead damage, and the only area of significant overhead damage he identified was in an area of one to two metres around the left-hand side of the roller identified in TFS26.
·     He did not agree that this was a multi area of origin fire.
·     The most substantial overhead damage he saw was in the place already identified.
·     The computer room was not in his view an area of origin.
·     Because the first people at the scene had identified that the collapse of the roof over the grading area had already occurred, and that the roof over the cool-store area was still standing, it was unlikely the fire started in the wall between the grading area and the cool store area, notwithstanding that wall may have had battery chargers adjacent to it and the wall itself was constructed of highly flammable material.

Plaintiffs' fire expert

The plaintiffs, through insurers, initially instructed Mr Colin Thomas, a fire scene examiner, to investigate the fire. He attended the scene on the morning of Friday 31 March 2006. His evidence was contained in his scene notes, a report dated 18 April 2006, a proof of evidence dated 16 May 2011, a supplemental proof dated 18 August 2011 and his oral evidence. It was Mr Thomas' view that the area of origin of the fire was the north-east corner of the grading shed "in the immediate vicinity of the front of the grading machine". More particularly, he said that "fire damage and burn indicators suggested to him that the fire originated "very near the right side of the roller fitted to the new grading equipment within the area of fire origin". The reason for this view was the presence of severe low level fire damage in the area, V patterns which extended to the rear and western side of the complex and the presence of isolated severe overhead damage directly above that part of the grading line. He had no doubt about that view.

No evidence indicative of the introduction of flammable liquids was found. He also concluded that examination of electrical circuitry throughout the complex offered no evidence of electrical overheating prior to the fire.

Mr Thomas said at par[12.4] of his report:

"Taking into consideration the nature of the combustible materials in the determined radius of fire origin the most likely heat sources responsible for this fire would be molten metal or by products of cutting or welding in the immediate vicinity of the front roller of the grading equipment. The main reason I detail this as being the most likely heat source is that such hot materials may have inadvertently lodged in a location where they have retained heat due to the confined areas concerned and eventually raised available combustibles to their ignition temperature. The burn indicators are totally consistent with a 'slow heat build-up fire'. The severe overhead damage directly above the roller was restricted to an area of approximately 450 millimetres in width. A 'baking effect' has occurred prior to significant flaming combustion eventuating."

As to low level burning, Mr Thomas identified the remains of the wooden box which had been propping up the conveyor. He said the box had been totally consumed, or at least reduced to charcoal, which indicated either exposure to fire for a long period or exposure to extreme heat for a longish period of time. As to overhead damage, he said there was severe overhead damage almost directly above (Mr Thomas accepted that it was not directly above) where the low level damage was. He described the main features of that damage as complete consumption of the timber involved in the construction of the roof immediately above the end of the grading line, with heavily charcoaled pieces of timber either side of the area of complete consumption which became less severe the further away from that area you moved. The steel parts of the roof were also most heavily distorted and discoloured in the same area.

Defendant's fire expert

The defendant instructed Mr Neil Barnes, a private forensic consultant from Victoria, to provide advice as to this fire. His curriculum vitae was impressive. I was urged by counsel for the defendant to accept his evidence in preference to that of Mr Thomas in particular. Mr Barnes' evidence consisted of a proof of evidence dated 23 July 2010, a second proof of evidence completed in September 2011 and oral evidence.

In his evidence-in-chief, Mr Barnes was taken to some portions of his first proof for the purpose of making corrections to it. The first corrections were at pages 7 and 8. Mr Barnes told the Court that in par(iv)(a) and par(c), where the word "sparks" appeared, it should more correctly be "spatter". Further in the fifth dot point on page 8, he said that the reference to "1000 degrees" should instead be "between 1500 and 1600 degrees maximum".

Mr Barnes was then taken to par[33] of his first proof of evidence where he was dealing with the likelihood of work undertaken by Nalder being the cause of the fire. He said:

"In any event I note that Mr Nalder was not working at the northern end of the machine but further south. I note that in this particular area he was working in there was little fire damage and is very unlikely that the fire originated there."

Mr Barnes gave as his reason for this view an observation of TFS43. He said that clearly fire had not spread to a piece of equipment he pointed out which still had green paint on the steel frame. He also said that you could see a gas bottle in the centre of the photo which still appeared to have valves and did not appear to have been exposed to a great deal of heat. He also pointed to the fact that there were two intact wooden crates which, had the fire started there, would have been destroyed.

As to TFS43, the green painted steel frame referred to was at the southern-most end of the grading line. That is apparent from TFS41. That was not where Nalder was working according to his evidence. He was working about the middle of the grading line. As to the gas bottle and wooden crates to which Mr Barnes referred, while, according to Mr Langmaid, these were at about the middle of the grading line, all were to the west of the grading line and not where Nalder said he was working. The wooden crates referred to were also not that which was propping up the conveyor in the vicinity of where both Maine and Cooper had worked. As to where Maine was working, Mr Barnes said at page 7 of his report in point (iv)(b), that Maine was working near the centre of the grading line. That was not Maine's evidence. The chain guard he was fitting was near the northern end.

Mr Barnes also expressed the view in his report that it was likely the fire started between 11pm and 2am, and that it was very unlikely the fire would have smouldered from about 7.30pm until then. In general terms he said:

·     The area of origin of the fire was likely to be in the north end of the building.
·     The point of origin of the fire could not be identified. In this regard, Mr Barnes gave as one of his reasons that it could be "the fire started to the north of the machine in the area where it was reported the batteries were being charged on a wooden bench". In fact the bench on which the battery chargers were was to the west of the grading line against the wall of cool store 8.
·     It is unlikely that Maine's activities caused or contributed to the fire.
·     It is unlikely the activities of Cooper caused or contributed to the fire.
·     Had the fire started as Mr Thomas believed, Maine, Nalder and Mr Langmaid would have smelt the gases produced which would have been a particularly acrid smell.

As to his second proof, that dealt with a product testing report commissioned by the defendant with AWTA testing and responses to proofs of Mr Thomas. Mr Barnes referred to the testing identified in that report as having been conducted in accordance with Australian Standards. The standard referred to was document E9 in exhibit P2.

Mr Barnes said at page 2 of his second proof when dealing with Mr Thomas's proof of 18 August 2011:

"As to the wooden bin due to the size of such a bin it is to be expected that a large pile of charcoal would remain in the area where it burnt. Examination of the photos provided by TFS shows that there was no identifiable material left. Hence in my view the plastic bucket was not the source of combustion."

Mr Barnes was referred to TFS147 (and CT9) and then the statement above was read to him. When it was read, counsel read the words "the plastic bucket" in the last line as being "the wooden bin". While it was not made clear, I suspect that is what Mr Barnes had intended to put. CT9 is a much clearer representation of the area, and I am unsure why counsel examined Mr Barnes primarily by reference to TFS147. In any event, Mr Barnes indicated he had seen neither photo, and that, having now looked at them, he accepted there had been a wooden crate near the end of the conveyor. However, his view was that the crate was too far away from the machine and that a MIG welder and a grinder would not ignite material such as that. The spatter and sparks from that equipment would have insufficient heat or energy output to ignite such material. Mr Barnes seemed to be unaware of the evidence to the effect that the wooden crate identified had actually, prior to the fire, been propping up the end of the conveyor, putting it well within range of the hot work done by both Cooper and Maine.

Mr Barnes was also asked to express some opinions about the room in which the computer which operated the E-sizer was situated. He agreed it was a possibility that the computer room was "an area of the fire". He spoke of a V pattern extending along the northern wall away from the computer room towards the entry door, I assumed, of the grading shed.

He also said:

"And in the actual room itself on the outside the external part of the roof is partially collapsed, which would indicate that there's been a considerable fire or fire load burning against the outside northern wall of that area, and also when you look inside the – it's almost total destruction of what would appear to have been timber wall linings and any furniture that was combustible is virtually burnt away. So you've had one considerable fire which if it had spread to that room may not have been as severe."

He then said by reference to copy photos which were exhibit D2 that:
"… the top bit of the computer appeared to have been on a stand or a small table which would have had a timber top and that's disappeared. You can also see around the sides of the building there were obviously studs to support the walls and there would've been a lining in that room, which is – knowing these portable buildings they are usually particle board or hard board."

As to his experience with fires started by computers he said:
"In eighty five percent of cases that we've looked at they are the monitors that go up rather than the computers … the damage to the computer is more consistent with it being a monitor because there's no evidence of any monitor in any of the photographs, so we can't make a comment on the monitors or the monitor."

Evidence about the layout and construction of the computer room came from Mr Muench, Mr Andrew Langmaid and photographs post-fire. The room was a stand-alone construction which abutted the northern wall of the grading line shed. It seemed to have been made of the same external material as that shed. The room was accessed by a door on its western wall. Mr Langmaid said the room had a glass window on the eastern and northern walls. Under the window on the eastern wall was a desk and beside that was a metal fabricated stand with a computer on it which had a clear Perspex cover over it. There were shelves on the northern and southern walls where consumable items were stored. There was power connected to the room and the computer was left running. Mr Langmaid was not asked anything about the configuration of the computer by reference to the photos which were D2, for example whether there was a monitor.

Mr Muench was however asked about the copy photos which became D2. He said that the top photo on the second page showed "the computer and monitor" which had been on the table but had ended up on the floor after the fire. He also referred to the computer and monitor in the TFS report. It is unclear why Mr Barnes seemed to think there was no monitor present.

Following this evidence, Mr Barnes was cross-examined, and I will deal with material that emerged in more detail shortly. Firstly however I should indicate that I found Mr Barnes to be a most unimpressive witness. His summary in his first proof of what was in the statement of claim was inaccurate. His enquiries about some matters appeared superficial. He disputed observations made by Mr Muench in circumstances where the observations were made first hand, and Mr Barnes did not attend the scene at all, using what he saw in photographs as his basis for disputing matters. He did not consider the impact of there being a wooden box or crate propping up the conveyor (a significant fact in the trial), saying he must have missed that. He was dismissive of the fact that he had made errors. He had not seen the video recordings of the tests done by AWTA Product Testing. As a consequence, I have approached Mr Barnes' evidence with significant caution.

As to the computer being a possible cause of the fire, Mr Barnes was referred to his evidence-in-chief to the effect that fires related to computers were more likely to arise in the monitor. His response was that without electrical expertise in examining both the computer and the monitor, the computer could not be eliminated as a source of the fire. He then accepted that he had had no evidence there was even a monitor present. As to the V pattern referred to which Mr Barnes initially referred as being evidence of a fire coming out of the computer room, he accepted it could relate to a fire coming out of the computer room or coming out of the main door of the grading shed.

Mr Barnes also conceded that, at no stage, did he ever inspect the scene of the fire and he did not speak to witnesses Cooper, Dutton or Andrew Langmaid. He had seen written statements of Maine and Nalder, none of which were in evidence. As to the existence of battery chargers, he had been told by the solicitor for the defendant that there were at least two, he thought Hitachi ones, somewhere in the north end of the building, but he did not speak to any witness about them to establish they were there and where. By reference to the plan which was P10, he understood that the battery chargers were somewhere to the north of where the plan was marked with a position of a switchboard. He could be no more precise than that.

It was put to Mr Barnes in cross-examination that the fire started in grading area number 1 (leaving aside his possibility about the computer room) in about the north east corner of the building and then spread across the complex (to the cool store areas). Mr Barnes responded:

"No, I can't say where the fire started. That is my proposition: you can't tell where the fire started."

He then however agreed that the area of origin was at the north end of the shed of grading area number 1.

Mr Barnes was then referred to a number of TFS photographs, in particular TFS5 (which showed a V pattern on the outside of the eastern side of the building adjacent to where the grading line was), TFS15 which showed badly burned timber rafters above the end of the grading line with one part of one being completely burnt through, and TFS16 which showed damage to a steel purlin. Mr Barnes agreed that they showed an area of severe overhead damage but then said that, "Yes, it is severe overhead damage all along the north end of the building". He was asked to indicate what photographs showed the "severe overhead damage all along the north end of the building". He referred to TFS12 and TFS83 (which were in fact the same photo) and TFS97 and TFS113.

TFS12 and TFS83 were photographs which appear to have been taken from a crane overlooking the building which housed the grading line and the cool stores. The photographs show the extensively collapsed roof above the cool store areas to the right. I was not able to identify, nor did Mr Barnes explain, what evidence those photographs show of internal severe overhead damage apart from the collapsed roof. As to TFS97 and TFS113, Mr Barnes was not actually asked any questions about these. He interrupted while counsel for the plaintiff was checking a query I had raised about other photographs and said:

"97 shows severe internal damage, and 113"

TFS113 shows the inside view of the collapsed roof, and TFS97 shows the same area from another angle. Mr Barnes did not identify in either the severe overhead damage which had been the subject of discussion by reference to TFS12 and TFS83. The only obvious thing was again the collapsed roof.

An observation of Mr Muench was also put to Mr Barnes about severe overhead damage which resulted in the following exchange:

"… Mr Muench said from his examination of the fire scene that the area that I showed you in photographs 13 and 15, I think it was, was the area of the most severe overhead damage of any part of the complex?     If that's what he says then that's his opinion.
Yep. No, it's his observation, Mr Barnes.           Yes
Do you accept that observation?           No. There's severe roof level damage at the north of the building."

With respect, Mr Barnes' responses in that exchange demonstrated a level of surprising arrogance. What had been put to him was an observation by an experienced fire fighter who had attended the scene of the fire. Yet he effectively dismissed it, having not personally examined the scene of the fire, and been confined to photographs not all of which he had actually seen.

The next portion of Mr Barnes' cross-examination elicited that he had omitted to deal in his proofs with the fact that the conveyor near the end of the grading line, where both Maine and Cooper had worked, had been propped up on a wooden bin. He eventually conceded that, when he wrote his report, he had in particular Mr Thomas's first proof which disclosed the bin was there. He said that he must have missed it. He in effect conceded also that he was wrong to say that the first time it had been raised was in Mr Thomas' second proof. The consequence of this was that, when he wrote his first proof, he did not consider the question of whether a wooden bin had played any part in this fire.

Mr Barnes also accepted that the bins were constructed of combustible material, they were a possible fuel source, they were made of soft wood timber which could have knots and splits, and that they were supported by skids to allow them to be moved by forklifts and that therefore things could roll under them.

Mr Barnes' second proof commented upon the AWTA Product Testing report. It transpired however that he had never seen the videos of the tests undertaken .

Mr Barnes was also taken through a number of statements in texts and agreed with most. He however disputed that it was likely that the fire could have begun in a crack or split in one of the wooden boxes. At the end of his cross-examination, he agreed that spatter and sparks from a welding machine could travel up to 10 metres.

Report and evidence of Mr Glen Allan (for the plaintiffs)

Mr Allan was employed by the Welding Institute of Australia as the manager of OzWeld Technology Support Centres Network. He was commissioned by the plaintiffs to answer a number of specific questions relating to welding and grinding, its by-products and their effects. Mr Allan has certificates in Metallurgy, Pressure Welding, Welding Supervision and Welding Technology. He also has a Masters of Engineering Practice in Materials Welding and Joining from the University of Wollongong.

His report dated 14 June 2012 was tendered to the Court. He expressed a number of views in that report. He said at page 7:

"The hot metal from cutting with a 5 inch grinder could travel up to ten metres. The hot metal particles and sparks would rapidly lose temperature and velocity as they travelled through the air. Maximum travel depends on angle of trajectory and tip velocity of the grinding disc. While grinding particles could travel more than 3 metres from the point of cutting, it is not expected that the particles would retain sufficient heat to ignite solid flammable material beyond that distance."

At page 7, he also said:
"The spatter and sparks from welding could travel up to ten metres. The hot metal as spatter from welding would lose temperature and velocity as the particles travelled through the air. The spatter particles would be generally significantly larger and hotter than the particles produced by grinding. The initial velocity of spatter particles would be significantly lower than the particles from grinding. The size of spatter particles produced is dependent on a number of welding variables including; welding consumable type, welding consumable size, welding current, welding voltage, contact tip to work distance, arc travel speed, welding position and direction of travel. Heat retention of the larger spatter particles possibly produced by the FCAW(N) process would be considerably longer than for smaller spatter particles or the consistently finer particles produced by grinding."

Further on the same page, he said:
"The grinding particles and spatter would lose their heat very quickly.
Spatter is initially molten, which in the case of steel requires temperatures > 1500˚ C, and loses heat more slowly due to the larger volume of each particle. The size and volume of spatter is not known, however the larger and hence more heat retentive globules would tend to accumulate close to and below the welding point."

And on page 8, he said:
"The proximity of the wooden bin and PVC bucket to the welding point are not clear from the information to hand, however if it were possible for larger globules of spatter (if they were produced at all under the welding conditions used on the day) then it is considered possible that combustion could occur."

Mr Allan was also asked about gases and vapours produced and their smell and whether any such smell could mask the smell of a small smouldering fire. He responded:

"The grinding operation on clean steel would not produce any significant gases, vapours or fumes, however there is a characteristic smell produced by hot steel and bonding agents in the cutting/grinding disc.
The 'mig 175 AMP welder' used as described with a flux-cored welding consumable with no shielding gas would produce various gases due to decomposition of fluxing agents and their reactons with the molten weld pool. FCAW(N) welding consumables typically produce more fumes than welding consumable for FCAW with gas shielding or GMAW. There would be a characteristic smell associated with the process but no strong residual smell would be expected."

It was possible he said that that characteristic smell might be sufficient to mask the smell of a small smouldering fire.

Testing commissioned by defendant as to the consequences of hot work

The defendant commissioned a firm, AWTA Product Testing, to conduct some tests during which grinding, welding and oxyacetylene equipment was used near wood, conveyor rubber and PVC buckets. About 40% of that firm's business involved testing products for combustibility.

A Mr Sean Bassett wrote a report following the tests. His instructions were to attend at the premises of the defendant on 11 April 2011 to witness and evaluate certain tests. He was told the tests were to be conducted:

·     on a conveyor belt assembly unit which was a smaller replica of that destroyed in the fire,
·     by an operator present and working on the actual machinery at the time of the fire. In fact this was Mr Maine, and
·     using equipment similar to that used on the day of the fire.

Maine was involved in the construction of the replica used for the testing and indeed took part in those tests in that he did the welding, grinding in them and used an oxyacetylene torch to replicate things he had done in and around the grading line on 28 March 2006.

The report completed by Mr Bassett was in the following terms:

"The re-enactment tests involved welding on the outside of the chassis of the test unit, followed by grinding off the excess weld. Welding / grinding were carried out approximately 7 cm from the belt, but with the chassis between the welding area and the conveyer belt. Welding was carried out for approximately 10 seconds, and grinding for approximately 5 seconds. Sparks from the grinding operation hit the conveyer belt but self extinguished immediately upon impact. The area where the sparks made contact with the conveyer belt was cool to the touch immediately after the completion of the welding.
I was informed that in an actual work situation a scrap piece of metal would be clamped into place to act as a barrier to sparks hitting the belt. The welding / grinding were thus repeated with the barrier in place. Less sparks hit the belt than the previous test and once again there was no ignition.
The next test carried out was designed to check the effect of a large number of sparks falling onto the belt from close range. This was carried out by welding a piece of metal onto the frame of the test unit directly over the conveyer belt. I was informed that this was not standard practice – welding and grinding would not be carried out with a conveyer belt in place due to not only the danger of fire but also the potential for damage to the belt. Welding was carried out for approximately 40 seconds with a large number of sparks and molten slag landing on the belt. There was a significant amount of smoking, pitting, and charring to the conveyer belt but there was no ignition of the belt. The belt was cool to touch within 5 minutes. The welded metal was then ground off, with the grinder in operation for approximately 60 seconds, once again causing a large number of sparks to hit the belt. There was no ignition and the belt was cool to touch within 2 minutes.
The third test involved welding a piece of steel plate onto the outside chassis of the test unit in such an orientation that sparks from both the welding and grinding would be directed into the small gap between the conveyer belt and the rubber roller liners. This served a dual purpose – firstly to see if either the edge of the belt or the roller cover would ignite, and secondly to see if any spark that lodged between the roller cover and the conveyer belt would smoulder and ignite after a period of time. This test was carried out in duplicate (on either side of the conveyer belt). During the first test welding was carried out for approximately 20 seconds with the sparks causing ignition of the belt / roller. Welding was ceased immediately following the occurrence of ignition. Flame was sustained for 2 seconds after which it self extinguished. The area was cool to touch within 5 minutes with no signs of any further smouldering or combustion that would have led to any further ignition. The flame did not spread from the area where ignition occurred. The weld was then ground away for approximately 40 seconds. No ignition occurred during this process. The second test again caused an ignition of the roller cover / conveyor belt combination following 20 seconds of welding. After flame time in this instance was 25 seconds after which the flame self extinguished was cool to touch within 5 minutes and showed no further signs of smouldering or combustion that may have led to a second ignition to occur. In both instances the flame was very small and did not spread from the area of ignition.
Following on from this point testing was designed to subject the conveyer belt and roller to aggressive conditions to check their likelihood to ignite and propagate flame. All tests from this point were not direct simulations of what could be reasonably expected to occur during actual works on the unit.
The fourth test involved applying the flame from an oxy-acetylene torch directly onto the surface of the conveyer belt above the roller. The flame was applied for increasing periods up to 5 seconds and then removed. Ignition occurred in all instances as soon as the flame contacted the belt but self extinguished within 2 seconds of removal of the flame. The oxy-acetylene torch was then applied to the edge of the conveyer belt for increasing periods up to 10 seconds in duration. The belt ignited as soon as the flame touched it but self extinguished within 5 seconds of the flame being removed. The oxy-acetylene torch was then applied to the combined edge of the belt / roller for increasing periods up to 10 seconds. The combination ignited and had an afterflame time of 58 seconds after which time it self extinguished. The belt was then removed from the roller and the flame applied directly onto the face of the roller cover for periods up to 20 seconds. Afterflame time from removal of the flame was 58 seconds before self extinguishing. The flame was then applied directly to the edge of the roller cover for periods up to 30 seconds with an afterflame time of 22 seconds. In all instances there were no signs of smouldering or other combustion that may have caused a re-ignition. There was however a particularly strong odour given off by the combustion of the roller linings that lingered long after the cessation of flaming.
The fifth series of tests involved heating a hexagonal machine nut with an oxy-acetylene torch to a red hot state and then placing it onto various positions on the mock up. This was done to check the performance of the product with an ignition source that retained heat for an extended period of time. Previous tests had shown that sparks from the welder were certainly hot enough to cause an ignition if applied for a long enough time to the edge of the belt, but cooled off relatively quickly after contacting the surface of the belt. The oxy-acetylene flame was hot enough to cause a rapid ignition, but flaming self extinguished relatively quickly following it's [sic] removal. The first hot nut test was carried in the centre of the conveyer belt. The conveyer belt ignited upon the hot nut touching it and burnt for 140 seconds before self extinguishing. The flame did not penetrate the conveyer belt. There were no signs further signs [sic] of after glow or smouldering following flame out. A second hot nut test was carried [out] on the conveyer belt directly over the roller. Again the belt ignited upon the nut being placed on it and burnt for 135 seconds. There were no further signs of combustion, and no penetration through the belt. The conveyer belt was then removed exposing the roller cover and a hot nut placed directly onto the centre of the roller cover. The roller cover ignited upon the nut touching it and burnt for 50 seconds before self extinguishing. There were no further signs of combustion, no penetration through the roller cover. In all instances damage was only slightly larger than the nut.
The last on-site tests carried out again used a machine nut heated red hot with an oxy-acetylene flame. In these tests the belt was removed, the hot nut placed onto the roller cover, and the belt placed back on top of the hot nut effectively sandwiching it in place. This test was designed to check to see if ignition of the assembly could possibly have occurred by an ignition source becoming lodged between the roller cover and the conveyer belt. The test was carried out in triplicate, with the nut being located initially in the centre of the roller, then the edge of the roller, and finally up against the metal chassis of the equipment. In all instances ignition occurred as the nut was placed onto the roller, and the placement of the conveyer belt over the roller caused a hole to be burnt through the conveyer belt. Any flame that was present was smothered by the placement of the conveyer belt. The assembly did continue to smoulder and smoke for a period of up to a maximum of 450 seconds after covering with the belt, but at not [sic] time did any flaming occur. After removal of the conveyer belt there were no signs of any other glowing or smouldering that may have led to a subsequent ignition.
The results of the on-site tests are tabulated below:

Test

Test Location

Ignition Time (s)

Sample Flame Time (s)

Observations

Weld

Outside of chassis

5

No ignition

Grind

Outside of chassis

5

No ignition

Weld

Outside of chassis – scrap metal barrier in place

5

No ignition

Grind

Outside of chassis –scrap metal barrier in place

5

No ignition

Weld

Above conveyer belt

40

No ignition

Pitting and scorching of belt

Grind

Above conveyer belt

60

No ignition

Weld

Adjacent to edge of roller / conveyer belt – Test 1

20

2

Self extinguished

Grind

Adjacent to edge of roller / conveyer belt – Test 1

40

No ignition

Weld

Adjacent to edge of roller / conveyer belt – Test 2

20

25

Self extinguished

Grind

Adjacent to edge of roller / conveyer belt – Test 2

No ignition

Oxy-acetylene Torch

Conveyer belt – above roller

1

1

Self extinguished

8

2

Self extinguished

Oxy-acetylene Torch

Conveyer belt – edge

2

1

Self extinguished

10

2

Self extinguished

Oxy-acetylene Torch

Edge of roller / conveyer belt

3

3

Self extinguished

10

58

Self extinguished

Oxy-acetylene Torch

Face of roller cover

5

5

Self extinguished

10

10

Self extinguished

20

20

Self extinguished

Oxy-acetylene Torch

Edge of roller cover

2

5

Self extinguished

12

10

Self extinguished

30

22

Self extinguished

Hot nut

Centre of conveyer belt

140

Self extinguished

Conveyer belt above roller

135

Self extinguished

Roller cover centre

50

Self extinguished

Centre of roller cover, with conveyer belt over the top

No ignition – smouldering/smoking for 300 seconds. Hole in conveyer belt after 10 seconds of placing belt over nut

Edge of roller cover, with conveyer belt over the top

No ignition – smouldering/smoking for 420 seconds. Hole in conveyer belt after 20 seconds of placing belt over nut

Edge of roller cover adjacent to metal chassis, with conveyer belt over the top

No ignition – smouldering/smoking for 450 seconds. Hole in conveyer belt after 10 seconds of placing belt over nut

In order to finalise the testing of the conveyer belt a sample was delivered to AWTA Product Testing to carry out some controlled laboratory tests. Two tests were carried out: firstly a hot nut test in accordance with BS 4970-1987, and secondly, an ease of ignition test in accordance with AS 2755.1-1985. Unfortunately there was no more sample of the roller cover available for laboratory testing. BS 4970-1987 is a test that involves heating a 30g metal nut to 900 °C and then placing it onto the face of the sample for 30 seconds before removing it. The burn time after application of the nut and burn radius is then measured. The maximum time for extinction of the flame from application was 38.7 seconds, and the maximum radius of burn was 20mm. These results confirm those determined during the on-site testing with the hot nut. AS 2755.1-1987 is a test for determining how easily a sample will ignite when a small gas flame is applied for a period of time from 1 to 20 seconds. For the purpose of this test an ignition is defined as a flame time following removal of the test flame of longer than 5 seconds. There was no ignition recorded for any time period up to 20 seconds.
In summary, there were no instances during the re-enactment testing where any ignitions occurred on any part of the test unit, nor were then [sic] any indications of smouldering or other signs of combustion that led to a later ignition. There was also no ignition when sparks and slag were intentionally directed onto the surface of the conveyer belt from above, and the area was cool to touch within 5 minutes of the completion of welding and grinding. Although there were two ignitions following the intentional direction of welding into the area between the conveyer belt and the roller cover, the flame self extinguished within 25 seconds, did not spread from the area where ignition occurred, and was cool to the touch within 5 minutes.
The intentionally aggressive tests using the oxy-acetylene torch and the hot nut all led to ignition of the conveyer belt and / or roller cover, but in all instances the sample self extinguished, flame did not spread from the ignition area, there was no sign of smouldering, and the area was cool to touch within 5 minutes. It should be noted that in all instances of an ignition occurring there was [sic] significant amounts of smoke produced and a very strong burning odour.
Based on the on-site and laboratory testing carried out, it is my opinion that it is highly unlikely that a conveyer belt unit that was built in the same manner as the test unit could have been ignited by sparks from either a welder or angle grinder of the same type as was used for this on-site testing. Both the on-site and laboratory testing showed that it was relatively difficult to get either the conveyer belt or the roller cover to ignite, and even when they did they self extinguished and the flame did not spread from the damaged area. In no instance did either component exhibit any smouldering or glowing that led to a later ignition, with the areas being cool to the touch within a maximum of 5 minutes. It is also unlikely that the smoke and odour that was given off when either the conveyer belt or the roller cover did ignite could have not been noticed by any operator that was nearby. Even when the conveyer belt / roller cover were smouldering during the last hot nut tests large amounts of smoke and odour were produced.
It should be borne in mind that the opinions expressed in this letter are based on a limited number of observations made on a single sample and may be subject to alteration if more detailed testing was to be carried out. We recommend that you have further testing conducted if the information above is critical to your decisions on this product.
Yours faithfully
Sean Bassett
Laboratory Controller"

When he was cross-examined, Mr Bassett agreed that there were a number of variables which might affect the outcome of the tests he observed. However what he also said was that:

"... the one thing to note on that though is even if smouldering had increased or had gone for longer or the ignition had increased or gone for longer our testing showed that there was still smoke and odour …"

This has some relevance given that none of the witnesses who were in the grading area around 7.30pm to 8pm on 28 March 2006 saw or smelled any sign of fire.

Video recordings of the tests conducted were shown in Court. I have also had an opportunity to look at them again before writing this judgment. It is clear that the use of hot work in the tests was far more prolonged than any witness described as happening near the grading line in 2006. Efforts were obviously being made to show more spatter, sparks and heat than were likely to have been generated by the work on 28 March 2006 prior to the fire.

Discussion

As I have already outlined, three separate causes of action have been pleaded by the plaintiff. These are breach of statutory duty, breach of contract and negligence (involving a breach of duty). Even if the plaintiffs satisfy the Court that there was some form of breach, they cannot succeed in this action unless the Court is satisfied on the balance of probabilities that any breach resulted in the fire. It is that last issue which is central to this case.

To determine whether any particular breach, if such breach be proved, caused the fire, it is first necessary to determine, if possible, how and where the fire started. If it cannot be determined how the fire started, it would not be possible to establish a causal connection between failures or activities of the defendant and the fire.

It is the plaintiffs' case that the area of origin of the fire has been established on the evidence to be that area surrounding the drive roller near which both Cooper and Maine worked at different times on 28 March 2006. It does not automatically follow that, if that finding is made, there should be a finding that hot work by Maine caused the fire, although the plaintiffs would say there is no alternative.

It is the defendant's position that, while it may be found that the area of origin is in the northern end of the building, there are other possible causes of the fire simply not properly explored in the evidence. The result is that the Court could not be affirmatively satisfied that hot work by Maine was the cause of the fire.

None of the experts was prepared to state categorically what the point of origin of the fire was, although Mr Thomas came closest when he confined the area of origin to the right-hand side of the roller near which both Cooper and Maine worked on the day.

The possible causes of the fire raised on the evidence were:

·     Hot work;
·     A malfunction in the computer and/or monitor in the computer room;
·     A malfunction in the battery chargers;
·     A malfunction in the switchboard appearing in TFS17 and 18; or
·     Possibly, an unidentified burnt object in the extreme north-eastern corner of the building.
All these possible causes could fit within an area of origin limited generally to the northern end of the shed housing the grading line.

What therefore is the evidence about the area of origin? Mr Muench and Mr Thomas both identified the V pattern on the outside of the eastern wall of the shed, where the base of the V was near the northern end of the grading line, as an indication of the area of origin. While Mr Barnes did not dispute the existence of that particular V pattern, he said there was another on the outside of the northern wall of the shed between the entrance to the computer room and the entrance to the shed. He said that could indicate the computer room as the source of the fire, although did concede it could also be evidence of a fire moving out of the grading shed to the computer room.

Mr Muench and Mr Thomas also identified severe low level burning in the form of the remains of the wooden crate used by Cooper to prop up the conveyor. Mr Barnes had not considered that. He agreed it could be evidence of an area of origin. Counsel for the defendant also pointed to what appeared to be charred remains of something in the extreme north eastern corner of the grading area, although nobody was able to identify what that might have been.

The third indication pointed to by both Mr Muench and Mr Thomas was the severe overhead damage in the area almost directly above where the roller was. There was clearly significant damage there in the form of complete destruction of timber rafters, and bending and discolouration of metal supports in the roof. Mr Barnes' evidence as to other possible areas of severe overhead damage was unconvincing.

There was also the evidence of the observations of fire fighters and Mr Langmaid on arrival at the scene. That evidence was clearly to the effect that fire had already been through the grading area and left the roof there collapsed. That area was not actively burning by this stage. The computer room was however still burning and there were flames leaping from the cool store area of the building. The clear inference from these observations is that the fire began in the grading area and moved north and west.

I am satisfied, having regard to this evidence, that the area of origin of this fire was the northern end of the grading line. The observations of fire fighters and Mr Langmaid, in my view, mean that the possibility of this fire starting in the computer room should be discounted.

As to the battery chargers being a possible cause of the fire, there seemed no dispute between the experts that in very general terms battery chargers could cause fires. However, there was no evidence of any examination of these chargers by anyone with expertise to indicate that there was or was not a specific problem with any of them. In any event, had some malfunction occurred in a battery charger such as to start a fire, that fire would have started against a wall which, on the evidence, was highly flammable and about 10 to 12 metres to the west of the grading line. The inference to be drawn is that, had the fire started there, it would have ignited the wall against which the chargers were sitting, that is the eastern wall of the cool store, first and then travelled to the grading area. That direction of travel of fire was inconsistent with the views expressed by both Mr Muench and Mr Thomas, and the observations of attending fire fighters and Mr Langmaid. For those reasons, I would discount the battery chargers as a cause of the fire.

As to the switchboard, the evidence was that it was found on the ground below where it was believed to have been on the wall. The electrician Mr Dutton had not worked on that switchboard on 28 March. It was Mr Muench's view that the damage to that switchboard had started at the top of it and that the state of its components did not indicate a level of damage consistent with the fire starting in it. The same comments about the consequences of a fire starting where it was, as I made in relation to the battery charges, are apposite here. As a consequence I would discount the switchboard as being a source of the fire.

That leaves the possibility that by-products of hot work caused the fire. At about 4pm on 28 March, Michael Cooper, an employee of the first plaintiff, performed hot work on a conveyor at the northern end of the grading line. He used an angle grinder to cut a section of metal from the side of the conveyor. He performed these works from the eastern side of the grading line. To his right, a little over a half a metre away was a large wooden crate which he had put in place earlier in the day to prop up the conveyor he was working on. To his left, he thought there was a PVC bucket, within about the same range. The works he undertook were done in a few short minutes. Sparks from his grinding, he said, largely fell to the ground. He estimated he finished working in that area about 4pm.

Rodney Maine then moved into the area on the western side of the same conveyor. His work involved the fitting of a chain guard. In the period up to, but finishing at about 7pm on 28 March 2006, Maine performed hot work near the same conveyor and then immediately beside a large roller at the end of a conveyor near the northern end of the grading line. There was a rubber belt on that conveyor. He placed a guard between his final work and the rubber conveyor to prevent damage from hot work products to the conveyor belt. The wooden box which had been on Cooper's right was to Maine's left although a little further away. He used a MIG welder. When he did his work he was wearing a mask to protect himself and he accepted his vision was restricted by that. He believed that any spatter from his welding fell to the ground.

It is apparent from the video of the testing done by AWAT Product Testing that sparks from grinding and spatter from welding do not always fall straight down. There was evidence that each could fall at least five metres away from where work was being done. In the case of the work being done by both Cooper and Maine, the wooden crate, the PVC bucket, rubber on the conveyor and the surface of the roller were all potentially combustible materials to various degrees, and were all within a five metre radius of the work done. Sparks from Cooper's grinding and spatter from Maine's welding could have fallen into the wooden crate or on to the PVC bucket, conveyor belt or roller.

Apart from the guard placed by Maine between the conveyor belt and where he was welding, neither Maine nor Cooper took any other steps to protect potential combustibles from sparks or spatter. None of the possible combustibles was covered, and the wooden crate was not wet down. There was no suggestion on the evidence that the floor underneath where the men were working had any sort of fuel load on it which could have been ignited.

Could therefore sparks and/or spatter from their work have caused or contributed to the fire? It was the view of both Mr Muench and Mr Thomas that by-products of hot work were the likely cause of the fire. It was Mr Barnes' view that it was unlikely. Mr Thomas described this as a slow build-up fire, and suggested that by-products of hot work could have fallen into a crevice of some description in the wooden crate. Mr Barnes disagreed with this view. His view was that it was simply not possible for the fire to have smouldered for as long as was suggested, assuming that the flames might not have begun in earnest until, say, 11pm at the earliest.

If Mr Thomas is correct, then the last possible time at which a by-product of hot work could have fallen into the wooden crate was about 7pm. This means that any such by-product would have had to ignite the crate to the extent that smouldering would begin, and that smouldering would then have had to continue for several hours before flames were produced. That initial stage would have had to occur in circumstances where, some 45 minutes after the process was started, it was undetectable in any way by Maine, Nalder or Andrew Langmaid

Counsel for the plaintiffs referred to the extract from Kirks which was exhibit P17 at 198 where the authors said:

"Fire watches after hot-work operations are typically only 30 minutes long, but some smoldering ignitions from hot work operations have been known to take hours, instead of minutes, to manifest in a flaming fire."

Unfortunately, the circumstances in which that phenomenon might occur were not explored in any evidence, nor was there any evidence about the number of hours over which this type of smouldering pre-flames might occur.

There was no direct evidence as to when flames first appeared. The fire was not seen until just before 3am, by which stage it had already caused significant damage to the grading area, the roof in that area had collapsed, that area was no longer engulfed in flames, the computer room was glowing with heat, and there were significant flames in the western side of the buildings. There was also no direct evidence to either include or exclude the possibility that the unidentified pile of burnt remains in the extreme north-eastern corner of the shed, which must have been very close to what had become the northern end of the grading line, might have caused the fire. I mention that only because it was an obvious low point of burning and not significantly far away from the area of severe overhead damage identified.

Against that background, am I able to be affirmatively satisfied that by-products of hot work carried out by Maine were the cause of the fire? Despite the conclusions of Mr Muench and Mr Thomas, this issue must be informed by the evidence of the AWTA testing and Mr Allan. The testing done by Mr Bassett's firm would clearly suggest that the type and duration of hot work undertaken by both Cooper and Maine are unlikely to have resulted in this fire. The aggressive nature of those tests did not produce flames which escalated beyond a matter of a few minutes. Further, there would have been a smell associated with the combustion process and no-one in the shed noticed any such smell despite three people being in there between 7.45pm and about 8pm, that is about 45 minutes after the last hot work was done. Counsel for the plaintiffs position in essence was, well what else could possibly have caused the fire? It can only have been the hot work, and therefore this fire is down to the hot work of Maine.

That is a very simple way to look at this case and, I accept, is an outcome which seems very logical. However what it ignores is that, even if by-products of hot work carried out on 28 March caused this fire, both Cooper and Maine did such work, albeit about three hours apart, and there was no evidence which might clearly inform the Court about how long it might take for hot work by-products falling on combustible materials to go undetected before erupting into flame. For example, hot work by Cooper could have been the offending activity with the fire erupting into flame within a couple of hours after the last person left the shed. That explanation is as likely as that Maine's work was the cause, particularly given no smell identified with fire was noticed by anyone at least until about 8pm.

While, therefore, it is quite possible that hot work caused the fire, I cannot be affirmatively satisfied that it did. More particularly, if hot work did result in the fire, I cannot be satisfied that it was the hot work of the defendant's employees which caused the fire.

I have not dealt with the specific causes of action pleaded because it is not necessary to do so. Absent the underlying causal connection between asserted activities or failures of the defendant and the fire, it is irrelevant whether some actions or failures may have amounted to a breach of statutory duty, a term of a contract, or duty of care. The plaintiffs claim should therefore fail, and there should be judgment for the defendant.

Test	Test Location	Ignition Time (s)	Sample Flame Time (s)	Observations
Weld	Outside of chassis	5	No ignition
Grind	Outside of chassis	5	No ignition
Weld	Outside of chassis – scrap metal barrier in place	5	No ignition
Grind	Outside of chassis –scrap metal barrier in place	5	No ignition
Weld	Above conveyer belt	40	No ignition	Pitting and scorching of belt
Grind	Above conveyer belt	60	No ignition
Weld	Adjacent to edge of roller / conveyer belt – Test 1	20	2	Self extinguished
Grind	Adjacent to edge of roller / conveyer belt – Test 1	40	No ignition
Weld	Adjacent to edge of roller / conveyer belt – Test 2	20	25	Self extinguished
Grind	Adjacent to edge of roller / conveyer belt – Test 2	No ignition
Oxy-acetylene Torch	Conveyer belt – above roller	1	1	Self extinguished
8	2	Self extinguished
Oxy-acetylene Torch	Conveyer belt – edge	2	1	Self extinguished
10	2	Self extinguished
Oxy-acetylene Torch	Edge of roller / conveyer belt	3	3	Self extinguished
10	58	Self extinguished
Oxy-acetylene Torch	Face of roller cover	5	5	Self extinguished
10	10	Self extinguished
20	20	Self extinguished
Oxy-acetylene Torch	Edge of roller cover	2	5	Self extinguished
12	10	Self extinguished
30	22	Self extinguished
Hot nut	Centre of conveyer belt	140	Self extinguished
Conveyer belt above roller	135	Self extinguished
Roller cover centre	50	Self extinguished
Centre of roller cover, with conveyer belt over the top	No ignition – smouldering/smoking for 300 seconds. Hole in conveyer belt after 10 seconds of placing belt over nut
Edge of roller cover, with conveyer belt over the top	No ignition – smouldering/smoking for 420 seconds. Hole in conveyer belt after 20 seconds of placing belt over nut
Edge of roller cover adjacent to metal chassis, with conveyer belt over the top	No ignition – smouldering/smoking for 450 seconds. Hole in conveyer belt after 10 seconds of placing belt over nut