15 February 2010 – Tracy Wakefield, managing director of Appalachian Log Homes responded to our article on timber, Timber: complex, sustainable and good in bushfires in a recent letter published here. Timber is highly resilient to fire, Dr Wakefield says in this article, which develops the theme in greater depth.
With the recent drought and extreme weather in Australia, much focus has been brought to global climate change and what we can do to reduce carbon emissions to address it. It is possible that climate change is responsible for a series of disastrous bushfires, and as a result has necessitated significant changes to our building standards for bushfire prone areas.
While timber as a building material is one of the most sustainable, renewable and lowest carbon emitting materials we can use to address climate change, it is severely restricted for use in bushfire prone areas, especially in the extreme category of bushfire attack (BAL-FZ or “Flame Zone”). Testing to Australian Standard 1530.8.2 – Methods for fire tests on building materials, components and structures has demonstrated that the fire resistance of heavy timber walls is an answer to this problem and, in fact, heavy timber is an appropriate building material to be used in extreme fire risk areas.
Sustainability of Timber as a Building Material
Removing carbon from the atmosphere, rather than simply reducing our additions to the atmosphere, is the ultimate way of addressing climate change. Growing trees is the obvious answer to this, which is why timber is considered carbon positive as the carbon is stored in the wood. The Building Code focuses on operational energy of buildings but as they become more energy efficient, the embodied or life-cyle energy of materials used in construction takes on more meaning. For example, using solid timber (logs) instead of conventional building materials can result in an 88 per cent reduction in greenhouse gas emissions. The difference between constructing a residential home in solid timber instead of brick and steel frame is the equivalent of running a car for 17 years, or heating and lighting your home for 25 years.
Timber is not just a renewable resource – young actively growing trees remove carbon from the atmosphere and timber in long lived constructions stores the carbon removed from the atmosphere. By using more wood (storing carbon) and growing more trees (removing carbon from the atmosphere) the construction industry can make a significant impact on carbon emissions.
Performance of Heavy Timber in Bushfires
The use of timber in bushfire prone areas has recently been highlighted following the Victorian Bushfires and the ensuing adoption of Australian Standard 3959-2009 Construction of Building in Bushfire-Prone Areas. Property owners, particularly in the Marysville region, wishing to rebuild their timber homes destroyed during the bushfires have found that on the surface the new regulations appear to preclude the use of exterior timber applications in the “Flame Zone”.
Timber is much maligned for its fire performance as a cladding on “stick framing”; while it’s exceptional fire resistance in heavy sections is often overlooked. Even the image evoked of a log home in Australia is that of CCA treated-pine clad style of 20-30 years ago, not the well sealed new-age solid timber of log homes of today.
The fire performance of heavy timber has been well researched over many years, but not specifically with reference to bushfire until recently. Experimental and case study work has shown that heavy timber performs exceptionally well under Australian bushfire conditions.
For example, fire testing was undertaken on a solid cypress wall section (1/3 scale) to the then draft Australian Standard 1580.8.1. At a radiant heat flux of 40kW a square metre the panel took between 45 and 58 seconds to ignite. The results of this experimental work showed remarkable performance, with the panel unable to sustain flaming with a burning crib when imposed radiant heat was lowered to 24 kW a squ m. The panel self-extinguished at 16kW a sq m even with a burning crib present, resulting in less than 4 per cent through thickness charring. Insulation properties were excellent keeping the internal face at room temperature, despite reaching over 600°C (873K) on the fireside.
Case studies of two homes (one being the author’s home) supported this particular testing with both homes subjected to burning adjacent combustibles in a real bushfire scenario. One of these homes was in an area currently rated Flame Zone (Warrimoo, NSW). Charring was less than 4 per cent of the through thickness, the walls apparently self-extinguished after adjacent combustibles were consumed, and both homes survived unoccupied with minimal intervention (no water available) to extinguish smouldering.
To the author’s knowledge there have been a total of six solid timber homes that have survived in extreme fire situations when other houses have been lost. One survived the 1994 fire in Hawkesbury Heights (NSW), two survived the Boxing Day 2001 fires in Warrimoo/Yellowrock (NSW), and now three more solid radiata log homes have survived fires in Buxton, Marysville, and Callignee (VIC).
Testing Regimes of AS3959-2009 for Flame Zone
With the release of AS3959-2009 there are now performance provisions for building elements in the extreme level of bushfire attack. These provisions require a Fire Rating Level (FRL) of -/30/30 with a minimum setback of 10m. Where this is not achievable testing to AS1530.8.2 is required. These are both fire endurance tests (enclosure fire situation), requiring the building element to withstand full flame contact and high radiant heat flux for 30 minutes with integrity and insulation performance criteria.
Testing to AS1530.4 subjects a 3m x 3m specimen to a standard furnace heating regime for various durations. In the case of a required FRL of -/30/30, the test is 30 minutes in duration, with the temperature of the furnace increasing to 576oC (849K) within 5 minutes, 679oC (952K) within 10 minutes, 738oC (1011K) within 15 minutes and reaching 841oC (1114K) at 30 minutes.
Integrity must be maintained with no sustained flaming on the non-fire side of the specimen for the duration of the test. Insulation must be maintained with a maximum temperature rise of 140oC (K) in specified locations or 180oC (K) in any location (measured with a roving thermocouple).
Testing to 1530.8.2 test is essentially an extension of the AS1530.4 test with added performance criteria including (i) the specimen must cease flaming within 30 minutes after the end of the heating regime, (ii) the specimen must be monitored for a full hour following heating for any re-ignition, and (iii) the radiant heat flux on the fire exposed side must be less than 3kW/m2 at a distance of 250mm, within 30 minutes of end of heating.
This testing regime does not replicate conditions normally found in bushfires, but is the standard test for fire separation from enclosure fires. Enclosure fires experience a much longer duration of fire attack and radiant heat flux and as a result this testing can be considered extremely conservative.
For instance, in an enclosure the power output of burning contents is contained by the ceiling, creating a furnace effect which a separating element must resist. In the case of bushfires, the fire front moves through very quickly (within minutes) and in an extremely chaotic manner due to the high winds and non-homogeneous nature of the forest. The radiant heat from a bushfire will fluctuate significantly, whereas the testing regime creates a constant heat flux for 30 minutes.
While the radiant heat flux from the flame front is extreme, it must be considered from the point of view of the materials subjected to this short duration of fire attack and not from the human perspective. Humans are 100 times more susceptible to radiant heat than timber. What will take five seconds to cause a second degree burn on exposed skin will take over 27 minutes to ignite soft wood timber.
This human perspective incorrectly focuses on the “big flames” or the massive peaks in radiant heat flux from a fire, even though there has been very little evidence that radiant heat or flame impingement causes housing loss.
The main mechanism of house loss is ember attack, or the small flames which enter a building and cause an enclosure fire. While these little flames seem insignificant from a human perspective, embers can easily ignite the multitude of highly combustible materials found inside buildings creating an enclosure fire. An enclosure fire is extremely dangerous and can grow within minutes to a full flashover fire involving all combustibles in a building. This is why most photos of homes burning following a bushfire show flames shooting through windows, roofs, and skylights.
Fire Testing Results to AS1530.8.2
Recently a 3m x 3m test panel was designed and constructed using 90mm thick tongue and groove Australian White Cypress logs mortised into a heavy cypress post and beam structure specifically for testing. This panel was tested to AS1530.8.2 at Bodycote Warrington Fire Laboratory in Dandenong (NATA accredited to perform this test) in late April 2009. The results were fascinating, although not so surprising given the well known fire resistance properties of heavy timber.
Australian White Cypress is not considered to be a fire resistant timber, in fact it is considered to be very combustible by most lay persons. This is partly because it is often referred to as cypress pine. This leads to the incorrect assumption that it is a Cupressus, an ornamental tree thought to have caused many housing losses in the Canberra fires. It is also commonly mistaken for a Pinus, which is a low density softwood. Australian White Cypress is in fact Callitris Glauca (one of the few indigenous softwood species in Australia); it is a softwood but with a medium density and hardness approaching many hardwoods.
After 30 minutes of the standard heating regime for the test the insulation of the panel was excellent with less than 20K increase in temperature on the non-fire side, achieving the integrity criteria. A series of photos shows the next stage of the testing (specifically AS1530.8.2).
At the end of the heating cycle the panel was pulled away from the furnace with intense flaming. Within 8 seconds, before the panel was pulled completely away and still subjected to radiant heat from the furnace, there was a significant reduction in flaming . After 18 seconds, before the radiant heat shield was moved into place, flaming had decreased considerably. After 54 seconds the radiant heat shield was in place and flaming to over 90 per cent of the panel had self-extinguished . After 2 minutes and 50 seconds the panel was essentially extinguished with the exception of some glowing combustion.
Radiant heat flux measurements from the fire side were measured after 15 minutes with less than 1 kW radiant heat flux 250mm from the burnt surface. The panel was then monitored for a full hour by several observers; where even tiny flames from spitting char were considered to be flaming (the tiny jet of flame as char blows off and air gets to the glowing combustion under the char). Cypress is a very resinous timber and spits and considerably crackles as it burns, however, the standard does not define “flaming” so all flames were considered to be failure.
Charring measurements have not been completely confirmed as yet, but less than a third of the through thickness was charred, and mainly in the corner formed by the post and the wall timbers. The steel flashing used to protect the sill plate did not stop the timber from igniting under the flashing, which can be observed in Figure 3 at the very bottom of the panel.
One source of difficulty was another small Colorbond flashing that was used as edge protection on the top beam. The fixings were spaced too widely or at insufficient depth and the metal expanded considerably during the heating regime (Figure 2 and 3 at the top of the panel), causing the flashing to buckle approximately 100mm away from the top beam. A small enclosure fire formed in the pocket created. Within 25 minutes the metal contracted somewhat and the small pocket of flame finally extinguished. The ensuing further 35 minutes of observations did not result in any further “flames”. The result of the testing was a BAL-FZ rating for this solid timber wall system.
The testing to AS1530.8.2 shows that a solid timber wall system can achieve BAL-FZ and can be used in the most extreme level of bushfire attack referred to as “Flame Zone” in the recently published AS3959-2009.
Testing showed that the wall system could not sustain flaming without a high imposed radiant heat flux, and within a few minutes the panel was safe to walk near with exceptionally low radiant heat flux from the fire side of the panel. This means that in the unlikely event that the wall was subjected to these extremes of imposed radiant heat for a long enough duration to ignite, it would self extinguish without intervention, and would be safe to evacuate within minutes. The insulation properties of timber also make it an excellent radiant heat shield to shelter behind during the passage of a fire front.
Those living in bushfire prone areas have recently witnessed the environmental devastation that climate change can cause. As an industry we should be focusing our efforts on developing and implementing appropriate materials available in an effort to reduce our environmental impact and carbon footprint. It is patently clear that heavy timber can perform as an excellent wall system for bushfire prone areas. Timber is also a sustainable, carbon positive material. By utilising timber in bushfire prone areas, the construction industry is making a safe and environmentally friendly choice.
This article was first published in Fire Australia Magazine