Professor Charles Fleischmann talks about the challenges in researching the little-understood phenomenon of smoke explosions, or ‘backdraft’, and why fire safety needs to start in the home.
Professor Fleischmann has been a part of the University of Canterbury Fire Engineering program since it started in 1994. He regularly teaches Fire Dynamics and Advanced Fire Dynamics to the Masters students. His research interests include fire dynamics, compartment fire modelling, performance-based codes, backdrafts, and smoke explosions.
At the Fire NZ Conference, Charley will be speaking on the topic of ‘Smoke explosions, the unexpected and unexplained hazard on the fire ground’.
Smoke explosion is often used synonymously with the more publicly known phenomena of backdraft. Popularised in the 1991 Hollywood blockbuster Backdraft, both smoke explosions and backdrafts, says Prof Fleischmann, are still poorly understood topics.
“We are still in our infancy in terms of understanding and developing guidance concerning these relatively rare phenomena,” he says. “At this point we believe that smoke explosions are separate phenomena that can be classified as “Rapid Fire Progress”, a term used in categorising firefighter fatalities.”
With over 100 publications, he is an accomplished presenter and is regularly invited overseas to teach and present his research. In 2019, he spent three months at the Underwriters Laboratories Fire Safety Research Institute as a visiting scholar researching smoke explosions and teaching fire dynamics.
Read this article in the digital edition:
FNZM: What brought you to New Zealand?
CF: Prior to my arrival, Dr Andy Buchanan at the University of Canterbury secured funding from the Fire Service Commission to start a fire engineering program. This had come about due to the introduction of the Building Code in 1991, which prompted the need for more fire engineers. Andy went looking to see who he might be able to find internationally. One thing led to another, and I ended up taking that position.
My wife (also a fire engineer) and I came here with the idea that this would be an adventure for three to five years, and we stayed.
FNZM: Where did you start your journey professionally?
CF: I grew up in Washington State and became a volunteer firefighter in 1977. In 1982 I went to the University of Maryland, where I finished my undergraduate degrees in fire and civil engineering and lived in a fire station as a resident firefighter. Wanting to get more involved in fire research, I went to the University of California at Berkeley and put myself through a Masters and then PhD. I was in California for about seven years before moving to New Zealand.
Since coming to NZ, I have had the pleasure of helping to build a quality fire engineering program at UC. We’re growing in terms of getting academics here, and the program is internationally recognised in the fire engineering community. When we started, our main goal was that the program survives and thrives as an academic program.
Student numbers are strong, and our graduates are in demand both in NZ and overseas. About 30% tend to go overseas depending on the market and their interest. Yet this is an advantage for NZ because many of our graduates also tend to return home and bring with them a wealth of experience.
FNZM: It’s been 31 years since Hollywood drew our attention to the issues of rapid fire progress in the film Backdraft, so how come our knowledge in this area is still in its infancy?
CF: All Hollywood did was give a backdraft a public reputation. The physics in there are atrocious. In fact, there’s one scene where you see the fire come under a door and roll across the floor, but the problem is that fire is buoyant. You cannot shoot fire across a floor, it doesn’t stay down, it rises. So, what did the filmmakers do? They flipped the room and its contents upside down, flipped the camera upside down, and shot the flame across what was now the ceiling!
In another scene, we see a civilian open a door, there is an instant explosion, and he’s blown across the front yard into the windscreen of his automobile. The problem with a backdraft is that they typically don’t happen immediately. If they did happen immediately, firefighters could just bang open a door and then step out of the way, letting the blast happen before entering. But that’s not the way the phenomenon works. So, Hollywood didn’t really help the situation any by making a movie out of it.
From a research point of view, smoke explosions and backdraft are separate phenomena. For a backdraft to occur, there has to be a change in the ventilation that precedes the explosion event, such as a firefighter opening a door or a window breaking. There is also a measurable time delay for the air to enter the compartment and reach an ignition source. A smoke explosion occurs without any change to the ventilation and the explosion takes place without any obvious warning.
The thing is that backdraft and smoke explosions are rare events. The difficulty is that it’s not like you can look at an incident (postfire) and identify that the space experienced a smoke explosion, backdraft, or flashover. You only know that it occurred when people witnessed it. As such, it’s not something that’s received a lot of attention in terms of research.
If we were to, dare I say it, have a lot of fatalities involving firefighters from smoke explosions or backdrafts, then there’d be more research on it. Fortunately, they are a relatively rare events, in fact a lot of the time, an incident that may be reported as a smoke explosion or backdraft may end up being some other type of explosion.
There was one explosion recently that we started to look into that was reported as a smoke explosion sort of event. But when we started looking into it, what actually happened was that the apartment was being rented out to somebody who was a floor refinisher, and they were using the space to store some of their product. So, it was really a flammable liquid event caused by the improper storage of a flammable liquid that led to an explosion. It was a number of days before investigators pieced everything together, but the media had already reported it as a smoke explosion.
A lot of the time all you see is a thirty second video from one camera angle and somebody says gee can you tell us what happened. There’s so much missing information. The incidents just don’t get well investigated in most cases unless somebody gets seriously injured or worse.
FNZM: How do you approach research into a phenomenon that’s not so common, unexpected when it happens? To what extent can you model a smoke explosion scenario?
CF: In the first instance, the approach we’re taking is to be able to consistently recreate a smoke explosion in a laboratory so that we understand what leads up to the event. We focus mostly on wood-based fires in a non-combustible enclosure, where we understand the burning reasonably well.
At the Fire Safety Research Institute in the US, we’ve studied plywood compartments using a small crib to get the fire going. We know that when we work with certain compartment configurations, we’re likely to get a smoke explosion. We’re at the point now where we’re saying, ‘let’s play with it a bit more to understand this configuration and see when we get a smoke explosion or when we don’t get it.’
We’re starting to see, for example, a potential for problems if the ventilation is interrupted, but we’re not at a point yet where we can draw definitive conclusions. One of the difficulties is that when we see a smoke explosion type event on the fire ground, we don’t know what somebody may have been doing on the other side of the building. Did they open a door? Did they close a door? Did they close and open a door?
We’ve got to a point where we can recreate a smoke explosion in the laboratory under certain conditions, but it’s not as consistent as we’d like it to be. The next step is to do a few more experiments so that we can recreate it consistently. We think we know what we’re doing, but it’s a matter of consistently repeating it inside the lab so we can take away some of the uncertainties. Then we’ll start looking at what actions could make it worse or more likely to occur.
In firefighter training, they take a compartment and let the fire build up really well. Then they close off the ventilation and leave it for long enough to stop the flaming. The fire then transitions into a smouldering state. This results in the build-up of unburned fuel gases which ignite and explode when they open the door. This helps train firefighters for backdrafts, but it’s not what typically happens in a real event. In the real world, the fire service doesn’t just sit there and let a fire build up, close the door, and then open it and watch it go boom.
So, although we can create training scenarios, the situation is very much contrived. What we’re working towards is identifying the realistic conditions that may lead to a smoke explosion. Then we can experiment with the actions that can be taken in certain circumstances to prevent the event from happening.
That’s where we’re heading. I want to say that in my career, I’ll be able to write a manual about how to stop smoke explosions, but I’m not sure we can necessarily do that. There are a number of challenges still to work on!
FNZM: You spent three months at the Underwriters Laboratory Fire Safety Research Institute as a visiting scholar. How valuable was your time there in terms of researching smoke explosions?
CF: They have some excellent facilities. Having started in 2013 with a few people, they now have more than 50 people in research, training and amplification. The Director of Research at FSRI, Dan Madrzykowski, is a former PhD student of mine from UC.
In addition to their facility in Columbia, Maryland, which has a small-scale laboratory, they’ve got a very good relationship with Delaware County Emergency Services Training Center in Pennsylvania, just outside of Philadelphia, where they conduct various large scale fire experiments.
It’s not unusual for them to build something on the scale of a house. When I was there in July, they had two identical houses side-by-side that had different heating and ventilation systems. They’d been studying the effect the two heating, and ventilation systems can have on fire development, the survivability of the occupants, and the resulting fire patterns.
I had time during my sabbatical to go over there and work with them on smoke explosions. It was a great experience working with talented people on large-scale fire phenomena. Right now, research on smoke explosions fits around everything else they’re doing. I’m hoping to get it a bit more ‘on the schedule’ and potentially do more at UC with their feedback and funding, and I hope to get another PhD student here to focus on smoke explosions.
The FSRI has all this tremendous expertise and funding, and hopefully, smoke explosions will bubble up to the top and become a more active area of research. The challenge is that they have more ideas than time and personnel.
FNZM: What do you see as the big fire safety priorities for NZ on the horizon?
CF: By and large, our buildings in New Zealand are of a reasonable standard for fire safety, but if you look at where fire fatalities occur, it’s still in their own home. People say “safe as houses” but those places are pretty dangerous – not only from fires but other hazards such as slips, trips, falls, etc. If you’re going to die in a building from a fire, it will most likely be in your own home.
Smoke detectors help. It also helps that we are replacing the annual replacement of the nine-volt battery, and we’re now putting in long-lasting battery devices. A homeowner can now be expected to pay attention to the useful lifespan of the smoke alarm, and it’s less likely that a resident will remove the battery because they can use it for another device.
Since the late 1970’s we’ve made huge strides in home fire deaths, reducing fire fatalities by more than 50%. However, from a socioeconomic perspective, there’s still sectors of our society and building stock we need to work on.
From an operational point of view, we can always do more in terms of firefighter training. In the last decade, there’s been a lot done on the firefighting operational side, personal protective equipment, and prevention. There’s more training available on best practices around PPE, such as laundering, and best practices for removing PPE, which is an area of significant improvement.
The challenge is getting the message out there and motivating people to do the training. For example, FSRI offers free on-line courses on most aspects of fire firefighting and cancer exposure reduction. These courses are available from https://training.fsri.org/
