Fire risk assessment of the chip-pan fire hazard in a large. London hostel

Transcription

1 Fire Science and Technorogy Vol.25 No.2(2006) Fire risk assessment of the chip-pan fire hazard in a large 1. INTRODUCTION London hostel Gordon Cooke International Fire Safety Consultant This paper provides a qualitative fire risk assessment of the contextual hazard arising from the use of electric cooking hobs wrongly placed at the narrow entrances to 80 single-room occupied flats where the occurrence of a chip-pan fire could prevent occupants from escaping. Several options were considered before the decision was made by the author to modify the existing hobs (which each comprised two cast iron hotplates) by adding thermal devices that would allow deep oil frying without the risk of fire from a spillage of oil. After a programme of worst-case scenario cooking trials the approving authorities agreed that the objective had been successfully met with minimal disturbance to the occupants. The hobs were successfully modified and are now in use. 2. THE PROBLEM For many years chip-pan fires (fires caused by frying sliced potatoes in deep oil or fat) have been the cause of disfiguring injuries and deaths and damage to property, and the UK government recognizes this in a current campaign to reduce them. There are approximately 4000 accidents and 130 deaths caused by chip-pan fires in the UK reported by the fire service each year and these casualty statistics do not include the many chip-pan fires not attended by the fire service. Such a fire, which can be a frightening phenomenon, can be made far worse if water is thrown onto it in an attempt to extinguish. An electric hob had been installed near the narrow single entrance/egress doorway in each of the 80 flats in a 5-storey residential hostel in North London managed by a local authority, The planning of the individual rooms did not follow implicit guidance in the UK building regulations (Approved Document B) or the Home Office Guide to Means of Escape and related Measures in certain existing Houses in Multiple Occupation (HMO s) which recommends that areas of highest fire risk (kitchens for example) are placed as far away as possible from the fire exit door the fire exit door is also the flat entrance door in the hostel. It would be difficult and very hazardous for occupants to escape past a hob if a chip-pan fire occurred. The author was commissioned to consider and mitigate the risk. It was paramount that any remedial work should minimize interference as all flats were occupied.

2 106 G. Cooke The layout of each flat is shown in Figure 1. It can be seen that the electric hob is near to the entrance door. Figure 1 Typical room plan showing location of hob The hostel was managed by a local authority and the solution to the problem had to recognize that many of the residents were foreign and could not read or understand the English language, some were often under the influence of drugs and alcohol, some stayed in the hostel for only a short period. Portable equipment ran the risk of disappearing or being meddled with. On the other hand the planning and construction of the 5 storey hostel was generally good, there was 24hr security, and the hostel had an addressable fire detection system with smoke detectors in the rooms and in the escape routes. Figure 2 Test on unmodified hob with both hotplates at maximum power after a small amount of oil was poured onto centre of hotplate. A major spillage could produce a chip-pan fire

3 Fire Risk Assessment of the Chip-pan Fire Hazard in a Large London Hostel OPTIONS Several options were considered as follows. The preferred option would form part of a package of improvements to the fire precautions systems and management in the hostel: 1 Leave the planning and hob installation arrangements as they were. There was a consensus among the approving authorities (building control, environmental, housing and fire officers) that this was unacceptable. 2 Relocate the hob away from the entrance. While this placed a possible chip pan fire away from the entrance it unfortunately introduced the every day hazard of having to carry hot cooking utensils over to the work surface and sink some distance away. If the hob was relocated near the window (an expensive alteration) the authorities were concerned that cooking operations could ignite the curtains 3 Install an automatic powder dump extinguisher immediately above the hob. Problems were cost and space but, more importantly, a chip pan fire so extinguished could spontaneously reignite some time after the powder had done its job. 4 Install a portable powder extinguisher and a fire blanket in each room, and provide instructions to tenants what to do in event of a fire i.e. extinguish the fire with the powder and immediately cover the pan with the fire blanket and leave covered for several minutes so that the oil drops well below its auto ignition temperature - official advice on chip pan fires is to leave the blanket in place for 30 minutes. This option was considered unacceptable as it depends on a high state of consciousness and motivation which occupants may not have due to learning and language difficulties or intake of drugs or alcohol. Also the use of pressurized extinguishers used too close to a fat fire could aggravate the fire. 5 Provide a Class F portable fire extinguisher designed for extinguishing fires from animal fats and vegetable oils. This extinguisher was said to be capable of extinguishing a fire without ejection of burning material and could prevent reignition within 10 minutes. Even so this was not considered acceptable by the authorities who were concerned about misuse by the wide variety of hostel residents. 6 Leave the hob where it was and provide a portable, fully enclosed chip fryer with instructions to tenants to use the fryer, but this would not prevent other frying operations e.g. frying in a shallow frying pan on the hob. Would the enclosed chip fryer ever be used? Would it be mislaid or sold to raise money? 7 Install a low-pressure water mist fire suppression system immediately above the hob connected to the domestic water supply. This would be expensive and the mains pressure would not be high enough (it would probably need to be around 5bar).

4 108 G. Cooke 8 Install fusible-link-actuated fibreglass roller fire curtains to the front and sides of the hob so as to enclose it. There was no space and, anyway, the curtains would be costly to install and maintain. Further, in descending, they could tip over a pan. They would only provide a flame barrier and have no fire suppressing function. 9 Rely on automatic smoke detection to warn the occupants before a flaming fire could occur, but this option could cause many general false alarms because of detector sensitivity unless a stand-alone smoke detector was fitted in each room. A stand-alone detector would require the battery to be regularly checked for charge or presence (it might be removed because the alarm was causing a nuisance or because the battery might be useful for some other purpose). Heat detection could not be relied to give an alarm early enough. 10 Leave the hob where it was and modify the heating controls to both hot plates so that it was impossible to produce a chip-pan fire but still be able to cook chips and undertake water-based cooking. Option 10 was eventually chosen as it was thought technically and commercially feasible using an arrangement of carefully positioned thermostats and fuses frequently used in other cooking appliances. At this stage it seemed appropriate to review the ignition temperatures of cooking oils and specify some fire scenarios. 4. IGNITION PROPERTIES OF COOKING OILS The most authoritative and comprehensive survey on the subject of ignition was compiled by Babrauskas in the 2003 Ignition Handbook [1]. This referred, among other works, to a UK Fire Research Station Note [2] dealing with research work in 1965 using a solid iron hotplate (like the hostel hotplate) and a range of oils and fats. Conclusions in the Note may be summarized as follows: The hot surface of the hotplate had to be raised to a dull red heat for a drop of oil to ignite. The maximum temperature of oil needed for cooking purposes is 205 o C. The spontaneous ignition temperature and the fire point are reached when the oil temperature is increased to 310 to C. There is therefore an immediate danger of fire at temperatures above 310 o C. There is no substantial difference in the fire hazard of different oils, nor does the repeated use of oil increase the hazard. Interestingly the Note stated that the substantial interval between maximum cooking temperature and the temperature at which there is an imminent fire hazard indicates that the presence of a suitable thermostatic control on cookers would greatly reduce the risk of dangerous overheating. The data on ignition temperatures of drops of fat dropped onto a hotplate, when the latter is at the temperatures shown in Figure3, is relevant.

5 Fire Risk Assessment of the Chip-pan Fire Hazard in a Large London Hostel 109 Oil/fat Fresh Used 8 times Corn oil Ground nut oil Olive oil Lard Dripping Hydrogenated cooking fat Figure 3 Ignition temperature ( o C) of fats on a hotplate (Table 5 from Ref 2) 5. THERMOSTATIC CONTROL VERSUS ENERGY CONTROL It was clearly unacceptable to have a hob which, if left switched on, continued to heat the pan contents as shown in curve A in Figure 4, for once the temperature went above the danger temperature, point C, fire could occur. The existing hob operated in this way. Thermostatic control, indicated by curve B, could, on the other hand, prevent ignition irrespective of duration of cooking. It was surprising to discover that in the UK no therm ostatically-controlled domestic hobs were on the market though there were thermostatic ally-controlled hotplates for laboratory use. Figure 4 Difference between energy and thermostatic control Before developing a system of thermostatic control it was necessary to specify some fire test scenarios to be used in the proving trials for cooking.

6 110 G. Cooke 6. FIRE SCENARIOS Worst case scenarios were identified. It was considered that a fire should not result from leaving a small or large amount of oil in a pan on either hotplate for several hours dumping wet or dry chips into the oil when the oil in the pan was at its highest temperature such that overflow onto the hotplate occurred spillage of hot oil onto the hotplate immediately upon removing the pan from the hotplate any of the above conditions when both hotplates were in use and one hotplate was bare or in use placing a wad of folded newspaper or dropping a piece of paper towel or a linen tea towel or a plastic cup onto a bare hotplate when at its highest temperature these were included at the suggestion of the fire authority although it was accepted that this was a less important scenario as the hostel had a multi-sensor fire detector (smoke or heat) installed in each room and this would be expected to warn occupants of a smouldering fire during the daytime when the sensors were programmed to sense smoke, and massive flames sufficient to bar the exit could not result from this scenario. 7. HEATING TESTS Tests were first carried out on one of the energy-regulated hobs already in use in the hostel. This indicated that a chip-pan fire could be easily achieved, Figure 2, as the hob continued to heat the oil well above its ignition temperature. For safety and environmental reasons the tests were done by spilling small amounts of oil onto either of the two hotplates when at maximum setting on the hob controls. The oil spilt was hot. Temperatures of the oil were measured with inconel sheathed K type thermocouples suspended in the oil. An infrared digital thermometer was used to measure the temperature of the bare hotplates. Thermostatic control was clearly needed and the hob manufacturer (based in Italy) was asked, via the importer, to produce a prototype hob with this form of control. Three months later the new hob arrived, it was tested and it was then discovered that the hob simply had hotplates with lower power ratings fitted but did not have the thermostatic control requested, Figure 5. A search was made for a UK specialist design and development company with in-house test facilities to help take the work forward and, as a result, the author recommended to the client that Destech UK should become involved in the project.

7 Fire Risk Assessment of the Chip-pan Fire Hazard in a Large London Hostel 111 Figure 5 Typical heating test arrangement. The strategy was to modify the existing hobs by adding thermostats and fuses which resulted in the required temperatures of the cooking medium (oil or water), hotplates and other parts of the appliance which just enabled chips to be cooked while also allowing normal water-based cooking. The choice of thermostat had to take account of phase lags in heating and cooling processes, and also by what thermostats were available in the market, and this choice was not easy. This had to recognize that these devices had manufacturing tolerance on the temperature of on/off operation, typically plus or minus 8-10 o C. Further they could only be positioned where they would not invalidate warranties and continue to satisfy the relevant British Standards. Although the heating rate up to thermostat cut off point was within control using the energy regulator fitted (represented by six heat input switches) it was not possible to influence the rate of cooling in each on/off cycle as it was a natural phenomenon and depended on what, if anything, was resting on the hotplate. Furthermore the choice of operating temperature of the thermostats could not be tuned too finely as there was a lag between the heating of the heating element embedded in the cast steel hotplate and the heating of the oil. The heat transfer processes were complicated, as shown in Figure 6, and of course the figure does not indicate time-based phenomena such as phase lag. The proving tests were done on a spare hob both in the test laboratory and in the hostel.

8 112 G. Cooke Figure 6 Heat transfer processes in the hotplate and pan The thermostatically-controlled hob can now boil water in a reasonable time, cook chips and yet will not cause a fire using sunflower oil or lard (an animal fat). This is despite hotplate temperatures which when bare can peak at degC as shown in Figure 7. Note that the infrared temperatures are safely below the hot surface temperatures cited in Figure 3 Figure 7 Variation of oil temperature with time for preferred thermostats. Additional tests, made at the request of the Fire Authority, also showed that bits of paper and a disposable plastic cup dropped onto the hotplate did not catch fire. However

9 Fire Risk Assessment of the Chip-pan Fire Hazard in a Large London Hostel 113 it was possible to ignite a thick bundle of folded newspaper but only after several minutes of smouldering by which time the room was full of smoke - smoke that would have triggered the smoke detection and alarm at a far earlier stage. Figure 8 Inside of inverted hob with bottom cover hinged-over. The white energy regulators are at bottom of picture. The inside of the hob with the thermostats and fuses fitted is shown in Figure 8 8. CONCLUSIONS An electric hob has been successfully modified, using commercially available parts, to produce a fire-safe hob. It proved possible to boil water within a reasonable time on both hotplates though vigorous boiling was not maintained over the short period when the hotplate was naturally cooling in each on/off cycle. More importantly it proved possible to cook chips safely and for hot animal fat or vegetable oil to be spilt onto either hotplate when at is maximum temperature without causing a fire. However a hotplate, after it had reached its maximum temperature, could cause smouldering ignition of a thick bundle of newspaper but it was agreed this was acceptable. Ignition of thin paper or a plastic cup did not occur. All 80 hobs in the hostel have now been replaced with the modified hob. There is a management issue associated with the hobs in hostel-like situations. It appears that if a tenant uses the hob as a space heater to supplement the conventional room heating, the hob will fuse after several hours of continuous operation at maximum setting on both hotplates when left bare. The hob will not fuse when used for normal cooking operations. Such misuse of the hob would shorten the life of the hob and new users need to be warned against this malpractice. There are nominally 11,000 chip pan fires in the UK each year and these result in more than 4,000 injuries and 30 fatalities a year. It does not need much imagination

10 114 G. Cooke to see that the use of a fire-safe hob in hostels and other residential buildings could do much to reduce the casualty toll, especially where reliance is on electric cooking devices to avoid gas-related problems such as explosion due to leaking gas. Such a hob would also reduce costs of repairing heat and smoke damage. REFERENCES 1. Babrauskas V, Ignition Handbook, co-published by Fire Science Publishers and Society of Fire Protection Engineers, USA, 2003, ISBN: , pp Stark G V W and Mulliner Wendy, The fire properties of cooking fats, FR Note 610, Fire Research Station, November 1965