BCD 2017

76 T H E B U I L D I N G C O N S E R VAT I O N D I R E C T O R Y 2 0 1 7 A series of tests has been carried out to: • investigate the operational parameters of wood-burning stoves using traditional fuel types and a range of air venting levels throughout the life of the fire including initial ignition and later refuelling • investigate the effects of various materials, such as paper, card and kindling, used to help light the fire • study flue gas properties and methods of spark/ember transport and ejection • measure heat transfer by conduction through brickwork, with a variety of flue configurations, including damaged and partially blocked flues • measure the transfer of heat by convection (movement of hot gases) via imperfect brickwork and damaged flue liners • assess the risk of thatch ignition due to a chimney fire. RESEARCH FINDINGS TO DATE The research is ongoing, and further tests will be carried out to investigate other means of thatch ignition related to wood-burning stoves and open fires. However, some significant findings have already come to light. Spark ejection Sparks were ejected from the flue during lighting and the early life of the fire, during re-fuelling and poking, and randomly during normal operation, perhaps associated with collapse of burning logs. The size and frequency of sparks were affected by the type of fuel and the volume of air used for venting. Some sparks were ejected into the air at high speed and flared momentarily, but others stayed alight while floating down towards the notional roofline. Sparks from cardboard and paper were larger and glowed for longer than those from kindling and logs, and some retained their energy during their descent. More sparks were produced when the stove was being well-vented and therefore burning at high temperature. Measurement showed a direct relationship between stove temperature and flue gas velocity: the higher the velocity, the greater the risk of burning material being transported upwards and ejected from the chimney. Some wood burners allow aggressive ventilation, and are therefore potentially more dangerous than other models. For example, models in which all controlled ventilation occurs under the fuel bed may be operated in such a way that they generate high temperatures and flue gas velocities, thereby increasing the risk of lighter material in the fire bed being carried up and ejected from the flue. Stoves that have a separate ash-pan door below the main door which can be opened while the stove is in use are of particular concern, and manufacturers warn against venting them in this way. Temperature of flue gases and chimney brickwork During typical wood burner operation, the temperature within the fire box was generally between 500 to 800 o C, depending on the refuelling rate and degree of ventilation. However, temperatures dropped significantly with height. At the top of the stove/base of the flue, the temperature was approximately 200 o C lower than in the fire box, and dropped by approximately 50 o C per metre up the flue pipe. At the height corresponding to the ridge level (where the chimney would pass through the thatch), the outside surface of the metal flue during normal wood burner operation was generally between 75 and 125 o C. During aggressive wood burner operation, especially when the fire was vented beneath the fuel bed, temperatures in the wood burner could reach 900 o C. However, to maintain a temperature of 800 to 900 o C typically required 4½ hours of extremely aggressive venting and constant refuelling (during which time it became impossible to approach the word burner without wearing protective clothing). The result was that, at ridge height, the external surface of the metal flue pipe reached 275 o C, but the internal temperature of the chimney brickwork was 100 o C lower. Moreover, the external surface of the bricks (that is, the surface that would be in contact with the thatch) never reached more than 80 o C, regardless of whether the wood burner was operated with or without a flue liner. This is well below the temperature required to ignite thatch. This suggests that conduction of heat from the flue into the thatch via sound chimney brickwork and mortar is, on its own, unlikely to be a cause of fire in thatched buildings. Ignition of thatch by convection Cutting out a mortar joint or even a complete brick in the unlined experimental chimney, leaving the underside of the thatch exposed to the interior of the flue, did not result in ignition of the thatch so long as the chimney was unobstructed. However, once a partial blockage (such as a bird’s nest or soot accumulation) further up the chimney was simulated, hot flue gases were diverted through the defective brickwork and into contact with the thatch. Ignition soon followed. Nesting birds A bird’s nest significantly increased the risk of setting light to the thatch. When the hot flue gases came into contact with the nest, the twigs did not burn because there was insufficient oxygen inside the flue. Instead they were converted to charcoal in a process known as pyrolysis. As charring continued, the nest lost its structural stability and collapsed. Charcoal is very light, so some of the smaller fragments were lifted up by the rising flue gases, and ejected from the top of the chimney. Immediately on contact with the oxygen in the air they burst into flames and rained down on the thatch surface. These burning fragments were much larger than the sparks emitted during normal or even aggressive wood burner operation, and had sufficient energy to ignite the thatch very quickly. There is often a spate of thatch fires if there is a cold snap in late spring, and it is possible that this may be related to nest building at this time of year. CONCLUSIONS The research has demonstrated that ‘heat transfer’ (conduction of heat from flue gases via sound brickwork) is very unlikely to be a prevalent mechanism of thatch ignition. It is clear that aggressive venting of a wood- burning stove increases the temperature and velocity of flue gases, promoting ejection of sparks and burning brands from the chimney, some of which may have sufficient energy to ignite the thatch when they land on it. Stoves that allow aggressive venting can achieve higher operating temperatures and flue gas velocities and therefore can potentially increase this risk. Partial blockages in the chimney, whether due to soot accumulation or a bird’s nest, increase the risk of hot gases penetrating chimney defects and igniting the thatch. A bird’s nest in the chimney can also increase the risk of fire due to pyrolysis and ejection of burning nest material. The two testing rigs: the one on the left was used to establish the general operating parameters of wood-burning stoves, including flue gas temperatures and speeds. The rig on the right was used to evaluate the effects of heat conduction through brickwork, and movement of hot flue gases via defects in the brickwork or flue liner. (Photo: Alison Henry).