Thomas v Southcorp Australia Pty Ltd [2004] VSC 34

6. It is convenient to deal first with the property claim. The relevant provisions of the Act are 75AC, 75AF and 75AG (appearing in Part VA), which read as set out below, so far as relevant. The significant passage, for present purposes, in each provision has been emphasised by underlining.

(1) For the purposes of this Part, goods have a defect if their safety is not such as persons generally are entitled to expect.

(2) In determining the extent of the safety of goods, regard is to be given to all relevant circumstances including:

(a) the manner in which, and the purposes for which, they have been marketed; and

(d) any instructions for, or warnings with respect to, doing, or refraining from doing, anything with or in relation to them; and

(e) what might reasonably be expected to be done with or in relation to them; and

(f) the time when they were supplied by their manufacturer.

75AF Liability for defective goods - loss relating to other goods

(a) a corporation, in trade or commerce, supplies goods manufactured by it; and

(c) because of the defect, goods of a kind ordinarily acquired for personal, domestic or household use (not being the defective goods) are destroyed or damaged; and

the destroyed or damaged goods, suffers loss as a result of the destruction or damage;

(e) the corporation is liable to compensate the person for the amount of the loss; and

(f) the person may recover that amount by action against the corporation.

75AG Liability for defective goods - loss relating to buildings etc.

(a) a corporation, in trade or commerce, supplies goods manufactured by it; and

(c) because of the defect, land, buildings, or fixtures, ordinarily acquired for private use are destroyed or damaged; and

the land, buildings or fixtures, suffers loss as a result of the destruction or damage;

(e) the corporation is liable to compensate the person for the amount of the loss; and

(f) the person may recover that amount by action against the corporation.

42. Accordingly, to succeed in this claim, the plaintiffs must satisfy the court on the balance of probabilities that the fire was caused by the heater and that the defendants, as manufacturers of the heater, were either in breach of one of those legislative provisions, or in breach of a relevant duty of care to the plaintiffs.

46. The heater itself, severely damaged by the fire, and subsequently substantially dismantled, was before the Court, as was a similar but not identical new heater ("the exemplar heater"). A large number of photographs were in evidence, some taken by the plaintiffs and members of their family, others by expert witnesses. Evidence was given by Mr and Mrs Thomas, and by a number of expert witnesses of different technical backgrounds, namely:

Mr Lang, a member of the Institute of Fire Engineers and registered electrical contractor, with experience of thirty years in the Fire Brigade, two years with the Victoria Police Arson Squad, and as a fire cause investigator in private practice since 1984;

Mr Barnes, whose primary qualification is in metallurgy, who is by occupation, a forensic investigator specialising in materials, fire, explosion and comparison tool mark related investigations, who is also a chartered chemist and chartered engineer, with experience in the Defence Department and the Victoria Police Forensic Science Laboratory;

Mr Lee, who is a forensic electrical and mechanical engineer, with considerable experience in investigating the cause of fires, particularly those of electrical origin;

Mr Judd, a combustion engineer, who has many years' experience of working with gas appliances.

Mr Bevitt, the National Technical Manager Service of what he described as "Rheem Australia, formerly known as Southcorp Water Heaters"; it is not in issue that he was referring to the defendants;

Mr Collins, who is an electrical engineer, with a graduate diploma in building, fire safety and risk engineering, and experience in the investigation of fires and of a number of other events;

Mr Marshall, a metallurgical engineer, specialising in fire and explosion investigation.

59. Mr and Mrs Thomas bought and moved into the house at 26 Hedge End Road in late 1978. The street runs north-south and the house is roughly parallel to it, the east side of the house facing the street. The garage is at the north end of that side of the house. An extension was built on to the back of the house (i.e. the west side) some years ago and at the time of the fire construction of a further extension, intended to double the size of the house by putting on a second floor, had been under way for two years, but was not yet completed. The work was being done by subcontractors, and much of it by Mr Thomas, who is an electrical contractor by trade.

63. On the day of the fire, Mr Thomas was away from home. Mrs Thomas left the house at 6.55 pm, to take their three sons, then aged 15, 13 and 5 respectively, and a friend of the fifteen year old to a basketball match in which the eldest son was playing. The Melbourne Fire Brigade was notified of the fire at 7.16 p.m. When Mrs Thomas and the four boys returned, the fire had just been extinguished. The family lived in rented accommodation for some seven months, and then moved back to the house, by which time the renovations were still not completed.

67. The heater was described by Mr Lang as a "Power House Model HZ ducted warm air furnace manufactured by Vulcan Aust Ltd". No issue arises as to the relationship between that company and the defendants. It is agreed that the heater was installed by a plumber on 11 June 1995. Mr Thomas said that after it was installed there were a number of problems. The heater was not responding to the remote control [2] . A number of invoices relating to repairs to the heater were before the Court, and Mr Thomas said that there were other service calls to Vulcan for which invoices were not provided; on some occasions Vulcan had simply said to turn the heater off and turn it back on again. The invoices described below, relating to service calls, were before the Court. There was some uncertainty as to the dates, which were difficult to read, but no issue was made of this.

Adjustment to receiver, check and adjust gas pressure, re-level unit, refer installer as ducting come apart under house.

No heat. Checked complete operation of unit and could not fault.

Replace remote. Test OK. Reprogram. Explain operation.

No heating, traced fault to controller, all lights displaying. Replaced controller.

Mr Bevitt said that "controller" there referred to the main board in the control box of the heater. [3].

72. The heater was made of galvanised sheet metal, and measured approximately 1300 mm long by 550 mm high by 390 mm wide. The operating parts of the heater were effectively contained in a galvanised metal box. Each of the long sides of that box consisted of two parallel metal panels, approximately 20 mm apart. The heater was installed in the sub-floor, underneath the ground floor of the house, resting on two concrete blocks, in such a way that its long side ran parallel to the road, that is, from north (nearer the garage) to south. It is convenient to describe the heater as it was in situ by reference to the points of the compass. It operated in what is largely a closed system, in which cool air is drawn from the living areas of the house into the heater through the return air duct, and after heating is delivered back to the house through the warm air duct at the south end of the heater. Fresh air becomes available to the system through leakages in the house and the duct system. In the heater as installed under the house, the warm air duct travelled to a splitter box, where smaller ducts branched out, some to various vents on the ground floor of the house. The warm air duct also ran roughly horizontally to the riser cavity some 3.8 metres from the heater and then, travelling by way of the riser cavity, took the warm air up through the ground and first floor of the house to the roof space where the duct branched out by way of another splitter box into smaller ducts which heated the rooms upstairs.

75. In a heater of this type the cool air from the return air duct is drawn into the heater by the circulating fan which in the heater in question was at the north end of the east side of the heater. The fan then pushes the air southward through the air chamber of the heater, in the spaces between the eight heat exchangers which are installed in that chamber. The air being pushed through the air chamber is heated by contact with the surfaces of the heat exchangers, which are in effect thin metal boxes measuring approximately 580 x 370 mm. The four primary heat exchangers, which are some 30 mm in breadth, sit upright in the centre of the air chamber and are flanked on each side by two secondary heat exchangers, approximately 20 mm in breadth. The eight heat exchangers sit side by side on their long edges, running north-south in parallel, save that the primary heat exchangers are tilted vertically so that the north end of each primary heat exchanger is some 100 mm higher than the south end, while the secondary heat exchangers are level. Beyond each of the outer sides of the secondary heater exchangers are the two parallel metal panels referred to in [8] above, so that those panels form the outer side wall of the air chamber.

78. Gas comes into the heater through a valve system and is mixed in the "burning zone" with air brought in through a separate system described below. The mixture in one burner is ignited by the hot surface igniter and there is a cross ignition channel which allows the other three burners to ignite. On this matter Mr Judd and Mr Bevitt were in agreement and I prefer their evidence, particularly that of Mr Bevitt, given his position with the defendants, to the evidence of Mr Collins that ignition took place in a common burning plenum. Each burner leads into the throat of one of the primary heat exchangers at its upper north corner, which, by virtue of the tilting described above, is the highest point of the primary heat exchanger. Thus the hot gases which are the products of the combustion of the gas and air mixture are projected from the burners in the burner chamber into the primary heat exchangers. They pass through the primary heat exchangers and into the secondary heat exchangers through a connecting port some 55 mm in diameter. The products of combustion, principally carbon dioxide, but also including traces of carbon monoxide and other toxic semi-combusted products, transfer the greater part of their heat through the surface of the heat exchangers to the air passing through the air chamber outside the heat exchangers. They are then vented out of the heat exchangers and out of the heater by a flue near the top of the secondary heat exchangers.

81. The other product of combustion is water, which runs down the inside of each secondary heat exchanger and is collected in an open drain running along the bottom of the long side of the secondary heat exchanger to a spigot at the north end of the heater. The four spigots draining the four secondary heat exchangers join together into a single drain tube, which passes under a metal rib so as to form a U-bend and leads to a standard hose connector enabling a hose to be fitted to drain the water away from the heater.

84. The air chamber and the heat exchangers are sealed from each other, because of the risk of the toxic semi-combusted products referred to in [10] above being pushed into the house with the warm air. The air chamber is not designed to deal with the passage of flame. Similarly the burning zone is sealed from the heat exchangers and from the air chamber, save that there is contact between the burning zone and the primary heat exchangers at the throats of the primary heat exchangers where combustion takes place.

87. However, a hole 20 mm in diameter exists in the steel partition between the air chamber and the burning zone. Mr Bevitt stated that that hole had a function in a model of heater which was intended for installation out of doors, but that it had no function in the model of heater with which this proceeding is concerned, which was designed for installation either under a floor (as in this case) or in a ceiling space. The hole was sealed with what was referred to as a grommet, the composition of which was unknown, but which was such that it was destroyed in the course of the fire.

90. Also in the burning zone, at the top of the south end of the heater, is the blower fan, which sucks air into the heater for mixing with the gas, as described in [10] above, to produce a combustible mixture. The vent through which that air is sucked into the heater is placed just above the point where the warm air duct exits the heater as described in [8] above.

93. The power supplied to the heater is mains electricity at 240 volts. However, there is a transformer at the control box on the heater which reduces the current to 24 volts in part and 16 volts in part for functions downstream of the transformer. If the mains power supply is on, a battery operated hand held remote control programmable thermostat ("the remote control"), separate from the heater, sends a message to the central chip in the control box on the heater when the ambient temperature reaches a level preset on the remote control, and ignition then takes place. A code within the remote control is intended to ensure that the control box will not respond to the remote control belonging to any other similarly operated appliance. That arrangement is considered further at [44] below.

96. As well as the normal fuses to cut out electric circuits if there is a flow of electricity beyond the safe capacity of the circuit, there are other measures which are intended to ensure the safe operation of the heater. To begin with, the heater must communicate every five minutes with the central chip in the control box; if it fails to do so twice, the heater shuts down. Then there are four relevant devices, namely the high temperature cutout switch, the warm air thermostat switch, the flame sensor and the pressure switch. All of these, except the high temperature cutout switch, operate at the lower voltage in the secondary system downstream of the transformer.

99. The warm air thermostat switch is situated in the face of the warm air as it leaves the south end of the heater. When the switch senses that the temperature of the warm air passing out is above 90 degrees it opens; that is to say, two filaments part so that current can no longer pass between them. This has the effect that the flow of current ceases and the heater turns off, although the circulating fan continues in operation. When the temperature of the warm air drops below 60 degrees, the switch re-sets itself, in that the two filaments join again, current resumes passing, and the heater resumes operation. The warm air thermostat switch operates on the lower voltage downstream of the transformer.

102. The high temperature cutout switch is situated below the blower fan. It operates on a similar system to the warm air thermostat switch, save first, that the temperature at which it opens is 135 degrees; second, that if it opens it has the same effect as switching off at the power point, in that it cuts off the main current coming into the heater at 240 volts; and third, that if it has opened it does not reset itself; resetting requires the attendance of a mechanic. These two switches, the warm air thermostat switch and the high temperature cutout switch, are designed to prevent air leaving the heater which is hot enough to be dangerous. Mr Bevitt agreed that the provision of those two switches is a recognition by the defendants that temperatures sufficient to be dangerous could in some circumstances be generated in the heater.

105. The flame sensor is placed against the fourth burner, that is the fourth of the burners to be ignited in the ignition sequence [4]. If a flame is present at that burner, the flame sensor causes an electric current to pass, which is detected by the control box. If there is no flame, there is no current, and the control box shuts down the heater**,** which is then required to be reset manually.

108. The pressure switch measures the difference in pressure between two points. One point is at the joining hole at the bottom of one of the outer heat exchangers, and the other at the exhaust flue from the heat exchangers which is referred to in [10] above. If the correct pressure balance between those two points which is necessary for the operation of the appliance is established, the pressure switch closes, allowing current to pass and the heater to operate. If the pressure balance is incorrect, the pressure switch opens so that current cannot pass and the system is ordered to shut down.

112. Mr Lang, Mr Lee, Mr Collins and Mr Marshall all agreed that the correct approach to the investigation of a fire is first to determine the point of origin of the fire, and then, having done so, to seek to determine its cause. Mr Lang attended at the house on the Monday after the fire, which had occurred on Friday, and again a week or so later. Mr Collins conceded that the opportunity for investigation was greatest for a person attending the site immediately after the fire.

115. Mr Lang described in detail the examination of the house which led him, by a process of elimination, to form the firm view that the origin of the fire was in the sub-basement, below the office on the ground floor, and where the heater was located. He said that there was no evidence on the ground floor of any great involvement in the fire. On the first floor, there was a walk-in wardrobe adjoining the master bedroom which was completely burnt out. The fire appeared to him to have started in the corner of the walk-in wardrobe where the riser cavity referred to in [8] above came up from the ground floor. The cavity was lined with chipboard, which was also completely burnt out. A considerable fire which had burnt in the roof of the house seemed to have originated above that cavity. That fire in the roof had escaped into the roof of the garage area and through to the garage floor causing collapse.

118. The patterns of burning in the cavity gave him the impression that the fire had burnt upwards through the cavity. The damage to the timber structure within the cavity led him to the view that the fire had not originated on the ground or first floor. He went down to the sub-floor area where there had been a fire, causing some damage to the floor bearers, but which had been localised (in the sense that there were combustible objects in the fire area which had not actually caught fire). The damage to the timbers in the sub-floor area gave him the impression that there had been a fire there involving the return air and warm air ducts. He looked for possible ignition sources in that area, such as a power point, but the only source he could not eliminate was the heater. Damage to the floor bearers above the heater was visually more appreciable than in the bearers closer to the garage. The overall view of the sub-floor area indicated to him that the fire had moved from the heater to the riser cavity. He formed the view that something had occurred in the heater which had involved an injection of combustible gas into the ducting which had ignited to destroy the ducting, travelling along the ducting to the riser cavity and vertically up the cavity into the roof void above the walk-in wardrobe. It would have moved rapidly up the cavity because of the drawing effect of the chimney-like structure. The ducting from the heater to the cavity rose slightly uphill and that would have assisted the movement of the fire. The ducting in the vicinity of the heater was completely destroyed, but the ducting at a distance from the heater had survived. There was no doubt in his mind that the fire had broken into the house and into the roof via the cavity and had entered the cavity from the sub-floor area.

121. In his report, completed in February 1998, some nine months after the fire, Mr Lang concluded that the fire commenced in the heater, and he said that he had considered on the Monday when he first visited the house that the heater was in some way involved in the fire. However, the conclusion in his report that the fire commenced in the heater was based on reports he had received from other people (by which I took him to be referring to other witnesses called for the plaintiffs). He was, understandably, criticised by Mr Lewis for his failure to take field notes, or to record the directions from which his photographs of the fire scene were taken. However, as Mr Saccardo, for the plaintiffs, submitted, that failure does not detract from his expertise as a fire investigator. Mr Barnes and Mr Lee both proceeded on the basis that Mr Lang's conclusion as to the origin of the fire was correct; neither of them had been instructed to reconsider that conclusion. The issue was not raised with either Mr Judd or Mr Marshall.

124. Mr Collins's evidence on the source of the fire was essentially negative. In his report of 19 September 2002, more than five years after the fire, he expressed the view that the origin of the fire was probably in the house, rather than in the sub-floor, and most probably at the first or mezzanine level. That is an extremely general statement, no more than a bare denial, made in response to the precise determination by Mr Lang, in his report, of the point of origin. It is based on evidence of charred timbers appearing in Mr Lang's photographs, which led Mr Collins to the conclusion that the fire in the upper levels of the house had burnt longer than the fire in the sub-floor. That conclusion may well be correct; but it does not appear to me to justify a finding that the fire could not have originated in the sub-floor. There is no evidence as to the time which elapsed between the calling of the Fire Brigade and the cessation of burning. In a memorandum to his instructing solicitors, dated 12 May 2003, Mr Collins said that the fire had its origin in the walk-in wardrobe or the ceiling space above the walk-in wardrobe or the adjoining en suite. However, his oral evidence was simply that he did not believe that the fire originated in the sub-floor; he believed that it had had its origin elsewhere, but he said that he could not be more specific than that.

127. In the same memorandum, Mr Collins suggested that the fire might have originated in one of the downlights fixed in the ceiling or in a ceiling exhaust fan. It is accepted that a ceiling light must not be installed close to insulation, as the insulation may cause the light to overheat. However, Mr Thomas, who, it will be remembered, [5] is an electrical contractor by trade, had installed the downlights. His evidence was that there was no insulation in the ceiling of the first floor near the downlights, although there was insulation between the tin roof and the trusses. I have no reason not to accept that evidence. No attack was made on the credit of Mr Thomas, save that Mr Collins expressed the view that photographs 20A and 21A on page 243 and 36 on page 9, of the photographs book [6] , demonstrated the presence of "pink batts" insulating material in the upstairs rooms. The only other witness asked to comment on those photographs was Mr Lee, who said that photograph 36 showed that there might have been thermal insulation in the roof or possibly a piece of duct, but it was hard to tell. I am not satisfied, from perusal of those photographs, that what they depict is insulation. There was no evidence of what objects were present in the rooms before the fire. In any case there was no evidence pointing to the conclusion that any of the downlights or exhaust fans had been associated with the origin of the fire, and in oral evidence Mr Collins said that he was not suggesting that a downlight or a fan started the fire. No other suggestion as to a possible cause of the fire was made by the defendants.

130. It is not in issue that, when fire is moving across timber objects such as floor joists and bearers, the side of the timber facing the fire is likely [7] to suffer deeper charring and damage than does the side facing away from the fire. A number of photographs of charred floor joists and bearers under the sub-floor were in evidence. Mr Collins's view was that, in accordance with that principle, those photographs demonstrated that the movement of the fire had been from the riser cavity to the heater, rather than from the heater to the riser cavity. However, I accept the criticism of Mr Barnes, that a number of the photographs showed one side only of the relevant piece of timber, and accordingly were not of assistance in determining the movement of the fire. Others of the photographs before the Court supported Mr Lang's finding that the fire had moved from the heater to the cavity.

133. Mr Barnes said that his observation of the charred joists and bearers in the sub-floor indicated that the fire had moved away from the heater, towards the riser cavity. Mr Lee said the same, adding that in the local area of the riser cavity there was evidence of a fire moving away from that cavity, which was in his belief directly related to material which fell down the duct. If the fire was spreading from one part of the ceiling to the other, and had got to that riser cavity, he would expect that there would be evidence of dropdown from the riser cavity, which there was.

136. It is common ground, I believe, that the tracing of fire movement by the examination of charring is a useful tool, but is regarded by authorities as by no means infallible. In this case, as Mr Barnes pointed out, it is not in issue that the fire burnt outside and around the heater in the sub-floor, and the damage caused by that generalised fire cannot be separated from damage which may have been caused at the time of initiation of the fire. The evidence of charring does not, in my view, establish that the fire moved from the riser cavity to the heater.

139. Much was made of Mr Collins's tertiary diploma in matters including fire safety [8]. However, that would appear to be a more generalised qualification than Mr Lang's membership of the Institute of Fire Engineers, the nature of which he explained. Mr Collins's "Statement of Personal Qualifications and Experience includes a passage headed "Corporate, Insurance and Legal Services: This work has been undertaken since 1982 and has involved many and varied investigations some of which are listed". The list which follows includes a number of types of investigations, many of which are unrelated to fires. They are, as stated, many and varied. Mr Collins could be regarded as a generalist investigator, Mr Lang as a specialist in fire investigation. Mr Collins was not asked how many fires he had investigated, but Mr Lang indicated in evidence that in more than forty years he had investigated possibly a thousand fires for the purpose of trying to determine the source of ignition. That experience must weigh very heavily in the assessment of his evidence as to the source of ignition of the fire with which I am concerned. Further, Mr Lang, it will be remembered, was the only expert witness to visit the house immediately after the fire. By the time any other witnesses attended at the house, much reconstruction had taken place, such that the Thomas family were in residence again.

142. None of the witnesses suggested any source of the fire in the sub-floor other than the heater. Having considered the material to which I have referred, I accept the finding of Mr Lang that the fire had its origin in the sub-floor and specifically in the heater. However, that finding does not, of itself, establish liability in the defendants.

146. Mr Barnes was asked by Mr Lang to examine the heater and determine if possible the cause and origin of the fire. He did so at the Gas Technology Services Workshops at Highett, to which the heater had been removed. His evidence was that he found that the drain tube [9] was blocked, preventing drainage from the heat exchangers of the water which is one of the products of combustion. The blockage consisted of "a conglomerate of particulate matter including burned insect remnants, dirt, rust particles, siliceous particles, pyrolysed paint flakes and foam packaging fragments". It was not suggested that this accumulation resulted in any way from the fire. It is not disputed that blockages have been known to occur in the drain tube of this model of heater.

149. The case for the plaintiffs is that, as the heater operated, the blockage of the drain tube caused water to bank up in the secondary heat exchangers to such an extent that the connecting point [10] was occluded, preventing the passage of the products of combustion from the primary heat exchangers into the secondary heat exchangers, and raising the pressure in the primary heat exchangers. The effect of that raising of pressure was that the products of combustion moved back over the flames at the throat of the primary heat exchangers, rather than down and through the heat exchangers; a process known as "rollback".

152. The burning gases, it is argued, then escaped from the burning chamber into the air chamber through the 20 mm hole [11] , by destroying the grommet plugging that hole, and were pushed by the operation of the circulating fan through the air chamber, past the heat exchangers, and so out through the duct at the warm air exit vent at the south end of the unit, leaving a flow pattern of staining in the air chamber which is characteristic of their being driven by the operation of the heater. Those products of combustion are then said to have travelled in the duct across to the riser cavity and thus by way of the duct into the house, sustained by combustible material in the duct such as dust, fluff and skin scales and by the flammable duct itself, and assisted by the chimney effect created by the riser cavity and the affinity of the flame with the vertical surface of that cavity.

156. Mr Lewis, for the defendants, drew attention to a number of difficulties which he submitted must be overcome if the plaintiffs are to prove that case. To begin with, the heater must have been actually operating for that chain of events to occur, and Mrs Thomas consistently said that she had no recollection of its operating on the day of the fire. However, there are two possible means by which the heater may have been activated.

159. The first of those means is the automatic activation of the remote control described in [15] above by the ambient temperature reaching the level at which the remote control was preset. In a written statement made on 10 February 1998, over the names of Mr and Mrs Thomas, the following appears:

After the unit was installed . . . we received our first gas bill from its use (around $400 for two months) which gave us quite a shock, so we had our slow combustion heater re-installed and only used the ducted heater for dissipating air and heat around the house, and for upstairs for the boys bedrooms (mainly set on 17 degrees to take the chill out of upstairs).

Over the winter of 96, the unit was used to dissipate the heat from the slow combustion fire and to heat the home when morning and evenings were cold, but the days warm.

This carried on over Spring 96 to summer 97. In Autumn 97 the unit wasn't used as April was a reasonably warm month, the day of fire (2 May 1997) the temperature was 20øC, so the unit wasn't in operation.

165. Mrs Thomas said in evidence that her husband wrote that statement and she typed it. When it was written she did not know that there was going to be a court case. She said that there was nothing there to correct, except that the heater was used more often than her husband thought. She said that her husband was concerned not to use the heater because of the expense. However, he would be gone by 7 am and she and the boys would have the heater on to heat up the house because it was a big house and it was cold. The boys would put it on when they came home from school. She thought that was easier than starting up the wood fired Coonara (the slow combustion heater), and she was always frightened of a log rolling out of the Coonara and starting a fire. She would set the remote control at 17 degrees when her husband was at home and after he was gone she would turn the temperature up. After the first bill she paid the bills for the gas.

168. Mr Thomas's evidence was consistent with that of Mrs Thomas. He said that he encouraged the use of the Coonara, and when he was around it was used for preference, because it did not cost anything to run. His wife and the boys found it more convenient to use the gas heater. In the winter of 1996 the gas heater was used on chilly mornings. He was not sure what happened with it after he went to work.

171. Mrs Thomas said that at the time of the fire and afterwards she had not understood that when the heater was not operating but the remote control was set to turn it on at a particular temperature, the remote control was said to be "on standby", and thus the expression "wasn't in operation" in the statement of 10 February 1998 was incorrect. When she left the house on the day of the fire the remote control was in her bedroom, and was on standby. She could not recall the heater operating that day. At that time the remote control was set on 20 or 22 degrees, because the alterations had not been finished on that side of the house and it would be cold in the evenings upstairs. It is common ground that the remote control was not found in the house after the fire. In cross-examination Mrs Thomas said several times that she was sure that the remote control had been set at 22 degrees.

174. She also said that she could not recall the heater coming on that afternoon. It made a noise when it was on, and if you were listening for it you would hear it, but you learnt not to listen to it. Having children, she said, you learn to turn off a lot of noises. I note her evidence that before leaving home she was organising four boys to get ready and leave for the basketball match, and that one of the boys came unwillingly. It is perhaps unlikely that, in that situation, she would have noticed whether or not the heater was operating.

177. While there are obvious difficulties with the evidence concerning the use of the heater and the remote control, I found it convincing, in the context of the evidence of Mr and Mrs Thomas as a whole, and I accept it, noting that Mrs Thomas consistently said that she could not recall the heater operating that day. I am satisfied, on that evidence, that on 2 May 1997 the remote control was on standby and was set at 22 degrees, so as to activate the heater if the ambient temperature fell below that level. It also follows from that evidence that the heater must have been left plugged in and turned on. The power point and switch for the heater were in the sub-floor, access to which was inconvenient, so that it is unlikely, given that evidence, that the Thomases would have been in the habit of switching off the heater at the power point from time to time.

180. A report from the Bureau of Meteorology was before the Court, showing that temperatures on 2 May 1997 at Melbourne Regional Office and at Scoresby Research Institute Automatic Weather Station were as follows:

The report states that in the metropolitan area on that date there was rain at times, which cleared during the evening. No information is given as to the extent of cloud cover.

185. Very roughly, Scoresby is ten kilometres from Mitcham and Melbourne 20 kilometres from Mitcham. I cannot conclude with certainty from that report that the temperature in the Thomases' bedroom on the first floor of a cold house in Mitcham at or around 6.55 pm on 2 May 1997 was or was not below 22 degrees so as to activate the remote control which was in that room. The bedroom was on the north-east corner of the house with a window on the east side and may have been warmed by the sun during the morning and cooled in the afternoon, but this is mere speculation, and indicative only of the range of relevant unknown factors. However, noting in particular the temperatures at Scoresby, the closer to Mitcham of the two recording stations, the report would seem to indicate that it is more probable than not that the temperature in that room at that time was below 22 degrees. It is not, of course, possible to make any estimate as to the time when that level of temperature was reached so as to activate the heater.

188. There is another possibility. In the instruction book provided by the defendants for use with the remote control, the remote control is called the "Comfort Control". Page 43 of that instruction book includes the following passages:

Before the Comfort Control will operate a heating/cooling system, the receiver needs to learn a unique identity code. . . .

Note: If problems occur with neighbouring Comfort Controls inadvertently operating this heater, then the identity code can be changed . . .

It is clear from the second of those passages that the defendants perceived a real possibility that the heater might be inadvertently operated by a remote control connected with a different appliance but containing the same code. Mr Judd, the expert combustion engineer, said of the passage "this to me is saying pretty clearly that it [i.e. the heater] can operate from another radio frequency source.". Mr Bevitt said that theoretically there were one million possible codes, and that he had never known a case of a neighbouring remote control operating a neighbour's heater. Nevertheless, the fact that the defendants have thought it necessary to include that passage in the instruction book means that the possibility of such inadvertent operation cannot be excluded, although it may be regarded as less probable than the activation of the heater by the operation of the remote control.

195. Mr Judd, the expert combustion engineer, was asked about the sooting visible on the heat exchangers and said that in his opinion it was not possible to get any extensive sooting in the heater from external sources if the heater was not operating. There was no doubt in his mind that there was a fire in the heater. The evidence of Mr Barnes was that the effects of heat on the cowling over the circulating fan were not consistent with the heater being exposed only to an external flame. The patterns of products of combustion within the heat exchanger area were consistent only with the effect of the circulating fan moving those products over the air flow deflectors in that chamber. Mr Lee gave similar evidence. He said that the pattern was consistent with an air flow of contaminated material rather than material gradually descending as was submitted by Mr Collins.

198. Mr Lewis relied on evidence indicating that the effects of fire on the outer panels of the sides of the heater [12] were greater than the effects of fire on the inner panels some 20 mm away. This, he submitted, indicated that those effects derived from a source external to the heater rather than from a source in the heater itself such as the claimed rollback. However, it is not in issue that there was burning in the sub-floor, external to the heater, after the actual initiation of the fire, whether caused by drop-down of burning material from above or however otherwise. The effects of fire on the panels could have derived from both the internal and the external fire, with the external fire having a greater effect than the internal fire.

201. Much time was spent on this and other evidence as to whether the effects of heat which were visible inside various parts of the heater derived from the external fire or from fire inside the heater. While clearly there was damage to the heater resulting from the fire in the sub-floor external to the heater, nevertheless that does not demonstrate that there was no fire inside the heater, of the kind for which the plaintiffs contend.

204. One other matter submitted as indicating that there was no fire inside the heater, which requires further consideration, is related to the warm air thermostat switch [13] , which is in the path of the warm air as it leaves the heater. There are two spade connectors attached to that switch, which, after the fire, were more severely affected by heat on the outside, although there was an asymmetric pattern of heat damage on the inner side, from which the warm air comes. Mr Lee agreed that this was consistent with an external rather than an internal heat source. However, he said that there was no evidence to suggest that the airflow out of the heater was even over the whole of the orifice from which it came. He agreed that the switch was presumably intended to be located in the best position for detecting the temperature of the hot air under normal operation. However, the products of combustion on the heat exchangers suggested an airflow during the fire in which the hotter combustion products had exited from the top of the exit duct rather than all at a uniform temperature evenly across the duct. He said that the spade connectors on the exemplar heater were made of a resinous substance which he took to have fire retardant properties. Similarly, Mr Judd said there could be stratification in the air stream that was coming out of the vent, and that this is something which "happens all the time". Thus the fact of the principal damage to the spade connectors being on the outside does not necessarily indicate that there was no fire inside the heater.

208. The safety controls are the requirement that the heater communicate every five minutes with the control box; the warm air thermostat switch; the high temperature cutout switch; the flame sensor; and the pressure switch; and the control box to which they communicate. They are described in [16] to [20] above. The only matter in issue which is relevant to the warm air thermostat switch is considered in the preceding paragraph.

211. The high temperature cutout switch was tested after the fire first by Mr Collins and later by Mr Lee. Mr Collins found it to be in a closed state. He tested it with a cigarette lighter flame at approximately 700 degrees and found that it did not open as it should have done had it been in an operating condition. Mr Lee, on the other hand, found it to be in an open state, and he was unable to reset it. Both witnesses, therefore, found the high temperature cutout switch to be dysfunctional, but counsel agreed that the evidence was irreconcilable. No relevant conclusion can be drawn as to whether the high temperature cutout switch operated during the fire. If it remained closed during the fire, then it had failed in its function of cutting off the power supply to the heater so as to prevent the heater from discharging dangerously hot air.

214. As Mr Bevitt explained in his report, the effect of the blocked drain tube ought to have been that the pick up in the heat exchanger for the pressure switch became submerged, so that the heater would not start to operate. The pressure switch was examined after the fire, but its casing was melted and accordingly it was not in a state where it could have been tested. Mr Bevitt thought it unlikely that it could now do its job. As Mr Judd said, either the pressure switch or the central controller must have been faulty, otherwise the heater would not have started. He said that pressure switches have intermittent failure rates, and there could have been a failure on this occasion. It is to be noted that an earlier pressure switch in the heater had required replacement on 13 September 1995 [14].

217. The flame sensor operates only on the fourth burner to be ignited, so that if there is poor burning on any of the first to third burners, the flame sensor would not detect it. Mr Judd said that it was quite possible to have varying supplies of the gas and air mixture to different burners, perhaps because of dirt in one of the venturis so that less air was drawn in. Mr Judd said that there was a possibility that one burner was burning and the others had not cross-ignited properly, and in those circumstances gases had escaped from the burner chamber into the air chamber.

220. The evidence as to the operation of the safety controls in the fire is thus inconclusive. I note the view of Mr Judd that if a unit has been in a fire of this kind, the results from testing equipment, either good or bad, have not in his experience been relevant. Similarly, Mr Lee expressed the view that the effect of the fire was likely to have rendered the central chip in the control box inoperative and unable to provide data as to its condition beforehand. For that reason he had not tested it. Mr Lee considered that if there was a failure in the heater, the probability was that the failure was in the central chip in the controller, rather than in the individual controls, because for the rollback posited by the plaintiffs to have occurred, all of those controls would need to have failed. A failure in the central chip could mean that both the ability to interpret messages from the safety controls and the decisions made on the basis of that interpretation would be affected. Mr Barnes drew attention to the fact that many complex systems have failures. There was no such thing as something that absolutely would not fail.

223. I note [15] that the controller was reprogrammed on 14 December 1995 and replaced on 19 February 1996; thus, on two earlier occasions, a controller operating in the heater has been found to be defective. Clearly a defective controller is a very real possibility.

Was sufficient water produced to inhibit the passage of the gases?

227. Mr Lewis submitted that the plaintiff's case required findings that:

(i) there was a quantity of water usually standing in the heater - "level X";

(ii) at some point, - "level Y" - if the water level rose, it would occlude the common port to the extent that gases would be prevented from escaping into the secondary heat exchanger and would back up to the throat of the primary heat exchanger; and

(iii) the heater produced enough water in the time available to raise level X to level Y.

Neither level X nor level Y nor the time available for the water to rise from one level to the other can be determined with any degree of certainty.

233. It is not in issue that water is produced in the heater at a maximum rate of approximately five litres per hour. Mr Lewis submitted, on that basis, that there was not time between the activation of the heater, and the call to the Fire Brigade, some twenty minutes after Mrs Thomas left home, for enough water to be produced to attain level Y. He referred to a test performed by Mr Bevitt on the exemplar heater. The exemplar heater is not identical with the principal heater, which has a capacity 30 per cent greater than the exemplar heater. Adjusting the test results for that difference in capacity, an amount corresponding to four to five litres of water in the principal heater had been introduced into the exemplar heater before flame failure occurred; that is, before the flame sensor functioned to shut the heater down. Operating at the rate of five litres per hour, the heater would have taken 48 minutes to produce four litres of water, in order to have the effect claimed by the plaintiffs, and that, in Mr Lewis's submission, far exceeded the available time.

236. In the course of examining the heater, Mr Barnes slit open one of the secondary heat exchangers, revealing a deposit of rust inside both sides along the line of the open drain [16] being some 25 mm deep at one end, overlapping the lower edge of the common port and some 45 mm deep at the other end. It became apparent, and Mr Bevitt agreed, that given the natural propensity of water, under the influence of gravity, to position itself so that its surface is level with the ground, this mark must have been made by debris from the manufacture of the heater which was contained in water lying in the heat exchanger at a time when the heater was not sitting level. That must have been the period between the installation of the heater on 6 June 1995 and 6 September 1995 when the heater was re-levelled [17] . No other cause for the shape of the rust mark was seriously suggested.

239. Mr Bevitt agreed that in order to ensure the proper functioning of the heater, it was important that it be installed on a level base. On 13 October 2003 when several of the witnesses visited the house during the hearing, a spirit level was placed across the two concrete blocks on which the heater had been resting at the time of the fire and which are still under the house, which indicated that the heater had been level at that time**.** It is to be assumed that it had continued level since being re-levelled in 1995.

242. Accordingly, the rust mark is not of assistance in defining level X, save that Mr Bevitt agreed that the rust mark indicated that at the time when it was made there was a lot more water in the heater than one would expect in the course of normal operation. There is no evidence as to the condition of the drainage tube at any time before Mr Barnes found it to be blocked in his investigation after the fire [18] . The description of the material constituting the blockage would indicate that it would have been gradually accumulating over time. It is clear from the evidence of Mr and Mrs Thomas [19] that the heater had been regularly in use, albeit only for short periods, enabling water to build up in the course of its operation as a result of an accumulating blockage.

245. Mr Barnes said in evidence that when he examined the heat exchangers after the fire "a number of litres, it was significant, a large quantity" of water was discharged from them.

248. Mr Bevitt said that, if the heater had been level at the time when the rust mark was made, the quantity of water indicated by the rust mark would have restricted the common port sufficiently to lead to poor combustion, but not to affect the pressure so as to impede the flow of gases. In his view, if the pressure switch and central controller were operating, the minimum quantity of water required to affect the pressure in the heat exchanger so as to cause poor combustion was two litres.

251. Mr Lee and Mr Judd both said that a partial blockage of the common port could be sufficient to inhibit the free passage of gases so as to produce the result for which the plaintiffs contend. Thus that result would not require a complete occlusion of the common port. Given their experience and qualifications, I prefer their evidence to that of Mr Marshall on this point.

254. It can be seen that the evidence indicates that level X might well be substantial, and that level Y need not necessarily require a quantity of water sufficient to occlude the common port entirely. Further, it is not possible to determine at what time the heater was activated by one or other of the mechanisms considered in [36] to [44] above and accordingly it is not possible to say what time was available for the requisite degree of occlusion to take place. Thus I do not find the result of the test described in [56] above to be of assistance in considering the question before me. It is accordingly not necessary for me to consider whether the test would have been affected by the other differences between the principal heater and the exemplar heater which are described by Mr Lee.

257. On balance, it appears to me, on the basis of the material before me, to be probable rather than possible that there was before the fire already a significant accumulation of water in the secondary heat exchangers, and that a relatively small amount of water, and thus a relatively small period of time, was required to achieve the degree of occlusion of the common port which was necessary to inhibit the movement of gases.

Could the ducting have transported the fire to the riser cavity?

261. Mr Lewis submitted that the exit ducting could not have remained intact for long enough to transport the products of combustion from the heater to the riser cavity as described in [23] above. He relied on the evidence of Mr Collins that as the duct began to burn the chimney effect referred to would have diminished, so that there would have been no force drawing the fire towards the riser cavity. However, the video made by Mr Lee of ducting from under the house showed that it burned relatively freely. However, the circulating fan would have been pushing the products of combustion along the ducting towards the riser cavity, as Mr Lee said, very rapidly and with greater intensity than would have been the case without the fan. He said that it would not surprise him if the fire travelled along the ducting, burning the combustible debris to be found in the ducting. I am satisfied that the ducting could have remained sufficiently intact to transport the burning material to the riser cavity.

Could the burning duct carry the fire to the upstairs ceiling?

265. Mr Collins said that in his opinion the burning ducting could not have set fire to the chipboard lining of the riser cavity within the time available between the fire starting in the heater (assuming that it did so) and becoming visible so that the Fire Brigade was called at 7.16 pm [20]. He said that to ignite chipboard requires a radiant incident energy of about 15 kilowatts per square metre over a period of between five and ten minutes. In his view the ducting would produce energy of the order of two kilowatts per square metre "for the duration of the flame which might be ten seconds". However, he did not give any basis for that expressed view. Two videos of duct burning were before the Court. The video referred to in [65] above depicted ducting taken from under the house. Mr Collins may have been relying on the video which he made, using plain aluminium ducting which he later agreed to be not relevant to this matter.

268. Mr Lang's estimate, expressly based on "visual observations of seeing chipboard catch fire within buildings" was that it would probably take about five minutes for the escaping cloud of burning gas to set fire to the chipboard.

271. Mr Lee considered that if a piece of ducting was placed beside chipboard and set alight, the ducting would consume itself and would effect superficial burns on the chipboard, but it was likely that the chipboard would not have been set alight sufficiently to keep going. However, Mr Lee went on to say that a test of that kind would not provide any indication of what was likely to happen in the riser cavity, and the test which he had done was not intended to demonstrate how the fire would advance within the riser duct. Fire has an affinity for vertical surfaces, and the fire would tend to move up the cavity against the vertical walls of the cavity. There would be some turbulence from the effect of the cross-pieces known as noggins placed against the walls to stop the vertical studs from bowing. The noggins would present sharp edges, which are easier to ignite than are smooth surfaces. It was most likely that internal surfaces within the riser cavity would be ignited from flames from the burning duct and that that fire would persist.

274. It is important to note that the only areas of fire damage (as opposed to smoke or water damage) on the ground floor of the house were to a desk in the office which stood over a heating vent, an area where a television set was over a heating vent and the place where the return air vent opened into the riser cavity. This supports the submission of the plaintiffs that the fire travelled with the ducting up the riser cavity.

278. The submissions of the defendants were, in large part, devoted to criticism of the expert witnesses called by the plaintiffs. I have already dealt [21] with the criticism of Mr Lang. Much of the criticism related to the preparation and presentation, rather than the substance, of the evidence of those witnesses. The value of their evidence, however, derives not only from their expertise but also from the long experience in the investigation or fires (or in the case of Mr Judd, in combustion engineering) on which their opinions were based. In contrast, I would say of Mr Collins, in addition to what I have said in [30] above, that his evidence tended to be more academic and theoretical than that of the witnesses with more specialised experience of the investigation of fires. The evidence of Mr Marshall added little. Mr Bevitt was precise and helpful.

281. Counsel for the defendants called no evidence as to any possible cause of the fire apart from that described in the submission of the plaintiffs. The earlier suggestion of Mr Collins that it originated in a downlight or an exhaust fan was expressly abandoned [22] .

284. On consideration of the material before me, and appreciating that there are many difficulties with the factual evidence, I find myself nevertheless to be satisfied, on the balance of probabilities, that the fire originated in the heater, and was caused by the chain of events within the heater for which the plaintiffs contend [23] . The question then is whether that finding renders the defendants liable to compensate the plaintiffs for the results of the fire.

Liability under the Act

288. It is not in issue that the defendants are corporations which manufactured the heater and supplied it in trade or commerce. Mr Saccardo submitted that I should find liability in the defendants, in that the heater had a defect, in terms of the definition in section 75AC of the Act [24] , in that its safety was not such as persons generally were entitled to expect, and that the fire occurred because of that defect. That being so, he submitted, the defendants were liable to compensate the plaintiffs under section 75AF for the loss which they have suffered because of the defect as a result of the destruction of or damage to their goods, being goods of the kind described in section 75AF(c); and are liable to compensate the plaintiffs under section 75AG for the loss which they have suffered because of the defect as a result of the destruction of or damage to the house. Mr Lewis made no submission in relation to sections 75AC, 75AF or 75AG. The claim of the plaintiffs under those provisions appears in paragraph 22 of the statement of claim.

291. I note that section 75AC(2) requires me, in determining the extent of the safety of goods, to have regard to "all relevant circumstances", including a number of circumstances specifically set out in that sub-section. In the absence of any submissions by counsel for the defendants I must infer that they did not consider that any of the specified circumstances, or any other relevant circumstances, would have affected my finding as to the extent of the safety of the heater. I note in particular that in the consideration of matters relevant to the cause of the fire no submissions were made by either party in respect of any of the matters described in paragraphs 75AC(2)(e) and (f). I assume, given the extensive and thorough attention given by both sides to the evidence, that if there had been any relevant action or inaction by either party which could be described under those headings, my attention would have been drawn to them in the context of causation.

294. In ACCC v Glendale Chemical Products Pty Ltd [25] Emmett J said [26]:

The standard to be adopted in respect of the definition of "defect" in section 75AC is an objective one based upon what the public at large, rather than any particular individual, is entitled to expect. On the other hand, one cannot foresee all possible uses to which consumers will put the goods. The explanatory memorandum published in connection with the Trade Practices Amendment Bill 1992 (Cth) which inserted Part VA says that section 75AC(1) does not require goods to be absolutely free from risk. The level of safety required is that which the community is entitled to expect. It is thus the objective knowledge and expectations of the community which are to be assessed, not the subjective knowledge and expectations of an injured party.

298. The finding that I have made as the cause of the fire leads me to conclude that the safety of the heater was not such as persons generally are entitled to expect; and thus that there was a defect, as defined, in the heater. In my view, persons generally - "the public at large", to adopt the words of Emmett J - are entitled to expect that a gas heater will not operate in the way in which I have found the heater did operate, so as to cause a significantly destructive fire.

301. That being so, as Mr Saccardo submitted, the defendants are liable to compensate the plaintiffs under section 75AF for the loss which they have suffered because of the defect as a result of the destruction of or damage to those of their goods, which are goods of the kind described in section 75AF(c); and are liable to compensate the plaintiffs under section 75AG for the loss which they have suffered because of the defect as a result of the destruction of or damage to the house.

304. Counsel agreed at the outset of the hearing that should the plaintiffs succeed in their claim, as they have done, the compensation payable should be assessed by a Master pursuant to Order 51 of the Supreme Court (General Civil Procedure) Rules 1996 ("the Rules").

308. It is common ground that although Mrs Thomas pleaded a claim under the Act in respect of psychological injury, that claim was brought out of time, and accordingly it is only her claim in negligence which is to be considered. The first question is whether my finding as to the cause of the fire set out in [72] above establishes a breach of the relevant duty of the defendants, as manufacturers of the heater, which is a duty to take reasonable care to avoid a reasonably foreseeable and real risk of injury to Mrs Thomas, as a user of the heater [27] .

311. Mr Saccardo drew attention to the hole between the burning chamber and the air chamber, and the plugging of that hole with the combustible grommet, with the result that if there were a fire in or around the heater the burning gases would, by destroying the grommet, be able to pass from the burning chamber to the air chamber and hence into the house, as I have found did in fact occur. In his submission the existence of that situation justified a finding that the defendants were in breach of their duty of care. The evidence of Mr Bevitt was that in that model of heater the hole served no purpose; and that in any case it could be patched with a piece of steel, similar to one appearing on the front of the exemplar heater. Mr Saccardo submitted that, while the risk of injury to Mrs Thomas arising from the destruction of the grommet might be small, it was reasonably foreseeable, not far-fetched or fanciful. Patching the hole with steel rather than with combustible material would, in his submission, have been a cheap, easy and effective remedy.

314. McHugh J in Tame v New South Wales [28] said, as to reasonable foreseeability:

102. So far as possible, the issue of reasonable foreseeability of risk in breach of duty situations should no longer be determined in isolation from the issue of reasonable preventability and the ultimate issue of what reasonable care requires. Indeed at the breach stage, it is better to avoid the question of reasonable foreseeability. Instead, courts should see their task as that of deciding whether the defendant knew or ought to have recognised that he or she had created an unreasonable risk of harm to others. Whether the creation of the risk was unreasonable must depend on whether reasonable members of the community in the defendant's position would think the risk sufficiently great to require preventative action. This is a matter for judgment after taking into account the probability of the risk occurring, the gravity of the damage that might arise if the risk occurs, the expense, difficulty and inconvenience of avoiding the risk and any other responsibilities that the defendant must discharge.

His Honour earlier [29] cited the following passage from the judgment of Mason J in Wyong Shire Council v Shirt [30] :

The perception of the reasonable man's response calls for a consideration of the magnitude of the risk and the degree of the probability of its occurrence, along with the expense, difficulty and inconvenience of taking alleviating action and any other conflicting responsibilities which the defendant may have. It is only when these matters are balanced out that the tribunal of fact can confidently assert what is the standard of response to be ascribed to the reasonable man placed in the defendant's position.

320. Mr Lewis relied on the statement of Lush J in Todman v Victa Ltd [31] where His Honour said of a motor mower manufactured by the defendant:

It . . . appears to serve no purpose to classify such a machine as dangerous in itself. The question is whether the machine has been designed and constructed with a degree of care appropriate to the dangers to minimize the risks.

324. It was reasonably foreseeable that if a fire emanated from the heater, significant damage to Mrs Thomas would result from that fire. However, the degree of probability of its occurrence was slight; Mr Bevitt said that 6000 of these heaters have been sold and there is no evidence before the Court that what occurred on this occasion had ever occurred before. The measures which the defendants took, in the manufacture of the heater, which were intended to ensure its safe operation, are described in [16] to [20] above. In my view, by taking those measures, the defendants acted reasonably to prevent injury to Mrs Thomas arising from the operation of the heater. The fact that those measures appear to have failed in this case does not demonstrate that that they were so inadequate as to be unreasonable. Taking those matters into account, I find that, in all the circumstances of this case, the defendants were not in breach of their duty of care towards

330. That being so, the claim of Mrs Thomas must fail.

334. The trial of this matter was listed to commence on 19 May 2003 but was not reached on that date. On 25 September 2002 Master Kings ordered that any Order 44 material be served on or before 2 December 2002. That date was extended to 15 January 2003 by a further order made on 9 December 2002. In August 2003 the solicitors for the defendants sought permission for Mr Collins to visit the house and take photographs. After some exchange of correspondence, permission was given and Mr Collins visited the house on 10 September 2003 and took the photographs which are considered at [27] above. They were duly served on the plaintiffs, but Mr Collins's report on the photographs and a memorandum to his instructing solicitors dated 22 September 2003 were not included in the materials served. It was not suggested that the omission was other than an oversight.

337. On 3 October 2003 the solicitors for the plaintiff wrote to the solicitors for the defendant seeking confirmation that there were no further reports, and when no reply was received to that letter, they presumed that that was the case.

340. When the oversight was discovered on the second day of the trial, during the cross-examination of Mr Lang**,** Mr Saccardo stated that, had he had that material before the trial commenced, he would have asked Mr Lang, Mr Barnes and Mr Lee to revisit the property and obtained detailed instructions from them on the matters contained in it. He submitted that neither the photographs nor the report and memorandum should not be allowed to be adduced on behalf of the defendants, but that if they were admitted into evidence, he would need time to deal with it.

343. After hearing submissions from both parties I ruled that the material should be admitted, and arrangements were made for one day's adjournment to enable the plaintiff's witnesses to visit the house**,** with submissions as to the costs of the adjournment to be made at a later date. The visit took place on 13 October 2003.

346. In the course of final addresses, Mr Lewis submitted that no order for costs should be made against his clients on the ground that very little was achieved at the visit and that it should have been made earlier in any case. However, I am satisfied that the adjournment was necessary and that good use was made of it. Given that the unchallenged cause of the adjournment was the inadvertence of the defendants' advisers, I am satisfied that it is appropriate that the defendants bear the costs thrown away by the plaintiffs as a result of the adjournment.

349. Mr Saccardo submitted that those costs should be borne by the solicitors for the defendants and that they should be indemnity costs. I do not propose to make an order against the solicitors for the defendants. Rule 63.23(3) of the Rules requires that the solicitors be given a reasonable opportunity to be heard before the making of such an order, and the giving of that opportunity would involve the incurring of further costs by both parties, which I do not consider to be justified in all the circumstances. I do not consider that the admitted inadvertence which gave rise to the need for the adjournment justifies an order for indemnity costs. See Colgate Palmolive Co v Cussons Pty Ltd [32] where the principles relevant to the exercise of the discretion to order costs on a solicitor-client or indemnity basis are considered at length.

352. Accordingly, there will be an order that the plaintiffs' costs thrown away as a result of the adjournment in respect of 13 October 2003 be paid by the defendants on a party-party basis

355. For these reasons there will be the following orders:

(a) under section 75AF of the Trade Practices Act 1974 for the loss which the plaintiffs have suffered because of the defect in the heater as a result of the destruction of or damage to their goods, being goods of the kind described in section 75AF(c); and

(b) under section 75AG of that Act for the loss which the plaintiffs have suffered because of the defect as a result of the destruction of or damage to the house.

2. That the compensation be assessed by a Master pursuant to Order 51.

That the plaintiffs' costs thrown away as a result of the adjournment in respect of 13 October 2003 be paid by the defendants on a party-party basis.

367. Counsel may wish to make submissions as to the details of the orders and as to costs.

[3] as to which see [15] and following below

[6] in which many of the photographs in evidence were brought together for ease of reference

[24] the relevant provisions of the Act are set out in [2] above

[27] per Mason, Wilson, Deane and Dawson JJ in Cook v Cook [1986] HCA 73; (1986) 162 CLR 376 at 382

[28] [2002] HCA 35; (2002) 191 ALR 449 at [102]

[30] [1980] HCA 12; (1980) 146 CLR 40 at 47-8

[31] [1982] VicRp 85; [1982] VR 849 at 851

[32] [1993] FCA 536; (1993) 46 FCR 225 at 229 ff