Dr Frank Rushbrook, former firemaster of Edinburgh & Lothians Fire Brigade, and one of the leading players in establishing the fire research group in the 1970s, and the 'Rushbrook' fire laboratory in the early 2000s, visited the lab this afternoon. Here is his introduction to the fire tornado experiment.Monday, February 28, 2011
Dr Rushbrook meets the Fire Tornado
Dr Frank Rushbrook, former firemaster of Edinburgh & Lothians Fire Brigade, and one of the leading players in establishing the fire research group in the 1970s, and the 'Rushbrook' fire laboratory in the early 2000s, visited the lab this afternoon. Here is his introduction to the fire tornado experiment.
Friday, February 18, 2011
Princeton - A Learning Experience
I'm typing this post while having coffee at the Woodrow Wilson school on Princeton University's campus. I've been in Princeton over 3 weeks now and I'm absolutely loving it. I get up each morning, eat my breakfast in a ridiculous Hogwarts-esque dining hall then head off to class. I choose what classes I want to sit in on (and I usually try and pack in as many as I can). Last week for example, I attended lectures on Stochastic Calculus of Brownian Motion, Economics of Crime, Democracy in Architecture, Bridge Design and Entrepreneurship. This broad range of classes wouldn't appear unusual to any student studying here - they have a very general view of ‘education’ that I haven't experienced anywhere else.
But, I'm getting ahead of myself. Perhaps I should explain why I'm here in the first place.
I'm studying a PhD in ‘Education in Fire Safety Engineering’. As most of you are aware, Edinburgh already has a very well established fire safety program, but it's not perfect so it can be improved. The question is how? I could spend years trying and testing a range of different teaching styles, or I could just ask a world-class university how they do it.
I chose the latter.
Now, when I sit in on a class I observe student-teacher interactions and take notes on the effects of different teaching styles. Sometimes the students are engaged, other times they're asleep. After studying a diverse range of teachers, I'm beginning to see patterns emerge. It appears there are some fundamental things that if you do/don't do you'll lose your audience, no matter how good you think you are. I'll share some of these with you now.
Probably the most significant observation I've made so far, and from what I can see is the biggest motivator for the students, is choice. In every aspect of what they do here, students are given choice - they choose what classes they do, what subject to specialise in, what topics to present, what groups to work in. This seems to result in a lot less admin which is a bonus, but the real value of this is the positive impact it has on student's self-motivation.
The logic behind it is this: If you want people to be responsible, self-directed individuals, treat them like responsible, self-directed individuals. Having to make choices every day not only promotes accountability and responsibility, but it gives people the opportunity to improve their judgement. And in engineering, particularly in fire safety engineering where reliable data is hard to come by, good judgement is essential.
The second thing that has a huge impact on classroom learning is divergent questioning. From what I've seen/read in my research, most teachers ask convergent questions (i.e. the opposite of what they should be doing). A convergent question is better known as a ‘guess-what-I'm-thinking’ question. The lecturer/tutor may ask something like “what is the main purpose of a roof”. The question is trying to force students towards one answer, so understandably students are reluctant to answer. They realise that there is only one answer and every single other answer they could give...is wrong. So in response to the roof question above, there'll be an awkward silence with students desperately trying to avoid eye-contact, followed by the lecturer answering his/her own question. “It's to keep the water out, obviously. Come on I asked you this question in the test last week remember? I asked, what is the purpose of a raincoat.” There is not really any point in asking convergent questions because asking someone to repeat something is not proof that they have understood it (as almost any teacher will tell you).
My suggestion is that if you want them to memorise a fact, just tell them the fact. Don't ask leading questions.
If you want them to think about something and really understand it, ask them to think. This is where divergent questioning comes in. If you want people to think, ask a question they can't get wrong, one that asks for their opinion. Such a question is phrased like this: “What do you think a roof does?” The students can come up with a range of ideas, and they're all correct. If they don't say what you want them to say, then it tells you something as an educator - namely that the students don't think the information is relevant. At least not yet.
This brings me to my third and final point, purpose. The purpose is the reason why your students are putting in so much of their time and effort to understand new ideas. Every rule in our society, every course in university, every word that's ever been spoken in a lecture has an underlying purpose - a “need-to-know”. Too often lecturers jump into complex methods before creating that “need-to-know”. It's like showing someone a path, without saying where they're going. No two people think the same way, so it's unlikely anyone else apart from you will choose the same path, it just won't make sense to anyone else.
It is often difficult to identify the purpose of a course, despite the fact that many lecturers think they've already done so. There is actually a straightforward test one can use to find a purpose - a reason - for everything. The test was developed by a five year-old - actually, every five year-old - and it consists of asking “Why?”…repeatedly. It's simple, they won't stop until you give them a valid reason, a purpose.
Just yesterday, I had the opportunity to test these ideas, or rather to ask someone else to test these ideas. One of the tutors here was running a tutorial where he was going to ask the students to discuss a lab they'd done the previous week. His idea was to choose a ‘volunteer’ to come up to the board and describe their lab to the rest of the students while he fired convergent questions at them (to make sure they covered all the material). The purpose was to make sure they said everything that might be asked in the exam.
See anything wrong with this picture?
I spent about 10mins explaining the theories above and he left enthusiastic about his upcoming class. Later on I joined him in his tutorial and listened to a full blown discussion going on amongst the students, each one eager to question and challenge each other's point of view. The class finished early, the students said it felt very “comfortable” and the tutor just looked shell-shocked. “That was so much better than last year!” he kept saying. “Thank you!”
I smiled and left. I felt good. After all, I had just improved teaching at Princeton University, even if it was only one small part of it.
How Uncertainty Transforms the Way we Quantify Fire
Posted in the name of Prof Jose Torero.
(related to the previous blog entry "Study or Gamble, but not both - 2nd annual Christmas tree fire test")
It is common practise to use experimental data for many purposes in the analysis of Fire Safety. It can be used as direct input (HRR, flame spread rates, ignition times, etc.) to models (analytical, semi empirical and CFD) as well as to obtain parameters that then can be used as input to other more fundamental models (heat of combustion, thermal properties, etc.). In many cases, due to the complexity of the tests, we rely of single data points to infer the values that we need. We can conduct a detailed analysis of the data and provide output values. In this particular case, the output values were the pHRR and the burn time of the tree. If I was to use this data for modelling, both parameters will be of critical importance and I could define a Q_dot=alpha x t^2 fire on the basis of both parameters. Furthermore, I could divide the HRR curve by the burning rate and obtain a heat of combustion that together with a flame spread model I could convert into another form of Q_dot. I could even use this data as part of a fundamental model that will attempt to predict all processes involved. Much of the research work we do tries to do two things, develop better models and try to make best use of the data we have. Thus this test is a fun example of what we are all about!
So, given the interest that this particular test has created I thought that it will be important to do a little exercise of uncertainty, not to question the winner, or to question the methodology used in defining this winner, simply to establish how important it is to look at these tests with caution and how difficult it is to use them in a manner that is truly representative of the event we are trying to describe via our engineering techniques. Furthermore, it is important to do this analysis to establish one of the key values of apriori estimations coupled with aposteriori explanations.
Apriori estimations have the distinct value of providing predictions that are only biased by the user’s knowledge or experience and not by the knowledge inferred through the observation of the test. Aposteriori estimations always carry the bias associated to having the knowledge of the results of the test. The aposteriori analysis of the apriori predictions reveals the effectiveness of the thought process associated to the apriori predictions. This analysis is extremely valuable in the sense that it can allow to separate the logic that is “user robust” from that that is purely a “guess.” It is also important because it allows establishing which of these “user robust” criteria have large experimental variability. Finally, it allows to identify common errors that can lead you to a “bad guess” but most important to a “good guess.” “User robust” logic with known “variability” is what we want to use to interpret test data and extract this information to introduce into our Fire Safety calculations.
I will do an aposteriori analysis of my estimates not to over-emphasize/or counteract the ridicule of being among the furthest away from the answer or to incontestably establish how my brain seems to have deteriorated with years doing fire research. The objective of this analysis is to encourage you to retrospect on how you achieve your estimate, post it, and let’s see what are the “user robust” criteria, which criteria is not robust, what were purely “guesses” and of these which ones are good or bad.
I know I am taking the risk of taking the joy out of a fun event, but given my role as an educator I find myself compelled to do this. The effort put on the tests and Guillermo’s fantastic statistical analysis encouraged me to do this. In any case, if you do not feel it is important, you will not participate and that is the end of the story!
900 kW and 20 seconds – How did I get there?
The way I reached my estimates, which I tried to qualify, but was told I could not (fair enough), was based on my experience of similar data published in the literature and the previous test conducted in Edinburgh.
When estimating the pHRR I made the following assumptions:
• The variability between tress in the literature was small.
• The pHRR was dominated by upward flame spread (VS) and time to burn out (tBO) of the leaves. Lateral flame spread is negligible compared to upward flame spread of a fuel of such low density, thus the effect of radial spread will happen after the pHRR.
• The base of the fuel burning (A) will be dominated by buoyancy not lateral spread, thus it should be the same for all tests.
• The tBO is very small and the base of the tree tends to have a higher density than the top, thus the pHRR will be generally attained before the flames reach the top of the tree.
• The HRR (given that this is a low density porous medium) will be proportional to the burning volume, so given a constant value of “A” it will be proportional to the height, thus to H=VS.t_b.
• The available data generally estimates a pHRR that ranges between 900 kW and 1100 kW.
So, given that the real height of the tree should not matter, then the pHRR should be similar to that of the literature. Because the tree was small, it was not so dry and it did not seem that dense I decided to opt for the lower bound value and estimate 900 kW.
Now, that being said, generally, literature values tend to be corrected by the time delay of the calorimeter. Our calorimeter has a time delay of about 10 seconds. What does that mean? Basically, it means that oxygen consumption measurements lag by 10 seconds the mass burning rate measurements. This generally makes no difference for events where things do not change within that period. If the event time scale is of the same order of magnitude of the time delay, then the measured value is somewhere between the measurement and that 10 seconds later. So, if I was to take the HRR curve measured by the calorimeter, then the value will be somewhere between what was measured 697 +/- 25 kW and 1000 kW.
An important lesson to learn is that the pHRR of a fast event (actually, even a slow event, but for different reasons) is a very difficult quantity to estimate precisely, thus the +/-25 kW stated as the error is truly only the direct measurement error. The true error will have to include the variability associated to the burn out time, the buoyantly driven upward spread, the global density and the comprehensive experimental error which is a parameter that is relevant in this case because the times are so short. So, any estimates within +/- 200 kW will probably have exactly the same value if the variable used is the pHRR. Thus 11/28 of you truly guessed the same answer.
If a different variable, such as the average HRR, or the Heat of Combustion was to be used as the “estimate,” then once all corrections due to time delay were made, would have probably delivered a smaller error bar.
The second variable to be estimated was the burning time. My estimate was 20 seconds and was based on a simple calculation of a typical upward flame spread rate of 10 cm a second. This had nothing to do with trees but with a fuel I know better (polyurethane foam). I estimated that the global value of “krhoC” is dominated by the density and I assumed that the density was more or less the same for both fuels. Thus I took that number. The tree was about 1.5 m, this gave about 15 seconds, time to burnout is so short that once the flame spread to the top, I could assume the fire was over.
Now, here is where I tried (unsuccessfully) to introduce a qualifier, I could not engage to estimate the initiation time (from the moment of ignition to the moment when the fire truly takes off). Furthermore, after the pHRR, what is left is lateral spread, then the branches and finally the trunk. The trunk will extinguish as soon as the branches die (bulk wood does not burn unless assisted!), but the lateral spread (being dominated by the shape of the tree) and the branches (being dominated by their individual shape and size) are impossible to predict. So at the end I gave up and simply estimated the time that it will take to achieve the pHRR from the moment the fire truly takes off. I reluctantly added a 5 second buffer for the slow initiation. While not a good estimate for what I was being asked, there is something to be said for the accuracy of the estimate! From the HRR curve we can establish that the primary burning will be somewhere between 10-20 sec (considering the instrument delay).
Now, what have I learnt, buoyancy is such a strong driving force that the estimate of the upward flame spread is a very robust one. The estimate of total burning time is one that carries a massive error bar, thus I will be reluctant to dismiss any of your estimates. From my perspective 28/28 gave estimates that I will consider within the expected error bars. Needless to say, last year’s Christmas tree was the proof to this point; the initial time could have been infinite if I did not decide to push the candle towards the denser part of the tree!
A final point, did I think of all of this in the 2 minutes that passed between the moment I learnt of the bet and the moment I provided my estimates? Obviously not! Most of this knowledge resides within your experience, and the estimate is an “educated guess.” Nevertheless, for the estimate to be adequate we need to carefully assess the question being asked (which I unfortunately decided to ignore) and the question needs to be posed correctly (meaning that what is being asked needs to have error bars that are smaller than the discrimination we are seeking). Otherwise, our guess will not be educated, nor it will be an estimate, it will just be a guess. If the error bars are small your chances of being the closest answer are very small (the educated estimate will have a much greater chance), but if the error bars are large you have as much of a chance to get it right as the most educated of estimates.
So citing Guillermo Rein: “while many stories can be told aposteriori,” and 3 hours of rationalizing my estimates can lead to this story, the stories need to be told and the discussion needs to follow. It is within the aposteriori 3 hours of introspection that I have truly managed to gain some insight into what happened not within the 2 minutes it took me to “guess.”
Congratulations to the winner!
Prof Jose Torero.
(related to the previous blog entry "Study or Gamble, but not both - 2nd annual Christmas tree fire test")
It is common practise to use experimental data for many purposes in the analysis of Fire Safety. It can be used as direct input (HRR, flame spread rates, ignition times, etc.) to models (analytical, semi empirical and CFD) as well as to obtain parameters that then can be used as input to other more fundamental models (heat of combustion, thermal properties, etc.). In many cases, due to the complexity of the tests, we rely of single data points to infer the values that we need. We can conduct a detailed analysis of the data and provide output values. In this particular case, the output values were the pHRR and the burn time of the tree. If I was to use this data for modelling, both parameters will be of critical importance and I could define a Q_dot=alpha x t^2 fire on the basis of both parameters. Furthermore, I could divide the HRR curve by the burning rate and obtain a heat of combustion that together with a flame spread model I could convert into another form of Q_dot. I could even use this data as part of a fundamental model that will attempt to predict all processes involved. Much of the research work we do tries to do two things, develop better models and try to make best use of the data we have. Thus this test is a fun example of what we are all about!
So, given the interest that this particular test has created I thought that it will be important to do a little exercise of uncertainty, not to question the winner, or to question the methodology used in defining this winner, simply to establish how important it is to look at these tests with caution and how difficult it is to use them in a manner that is truly representative of the event we are trying to describe via our engineering techniques. Furthermore, it is important to do this analysis to establish one of the key values of apriori estimations coupled with aposteriori explanations.
Apriori estimations have the distinct value of providing predictions that are only biased by the user’s knowledge or experience and not by the knowledge inferred through the observation of the test. Aposteriori estimations always carry the bias associated to having the knowledge of the results of the test. The aposteriori analysis of the apriori predictions reveals the effectiveness of the thought process associated to the apriori predictions. This analysis is extremely valuable in the sense that it can allow to separate the logic that is “user robust” from that that is purely a “guess.” It is also important because it allows establishing which of these “user robust” criteria have large experimental variability. Finally, it allows to identify common errors that can lead you to a “bad guess” but most important to a “good guess.” “User robust” logic with known “variability” is what we want to use to interpret test data and extract this information to introduce into our Fire Safety calculations.
I will do an aposteriori analysis of my estimates not to over-emphasize/or counteract the ridicule of being among the furthest away from the answer or to incontestably establish how my brain seems to have deteriorated with years doing fire research. The objective of this analysis is to encourage you to retrospect on how you achieve your estimate, post it, and let’s see what are the “user robust” criteria, which criteria is not robust, what were purely “guesses” and of these which ones are good or bad.
I know I am taking the risk of taking the joy out of a fun event, but given my role as an educator I find myself compelled to do this. The effort put on the tests and Guillermo’s fantastic statistical analysis encouraged me to do this. In any case, if you do not feel it is important, you will not participate and that is the end of the story!
900 kW and 20 seconds – How did I get there?
The way I reached my estimates, which I tried to qualify, but was told I could not (fair enough), was based on my experience of similar data published in the literature and the previous test conducted in Edinburgh.
When estimating the pHRR I made the following assumptions:
• The variability between tress in the literature was small.
• The pHRR was dominated by upward flame spread (VS) and time to burn out (tBO) of the leaves. Lateral flame spread is negligible compared to upward flame spread of a fuel of such low density, thus the effect of radial spread will happen after the pHRR.
• The base of the fuel burning (A) will be dominated by buoyancy not lateral spread, thus it should be the same for all tests.
• The tBO is very small and the base of the tree tends to have a higher density than the top, thus the pHRR will be generally attained before the flames reach the top of the tree.
• The HRR (given that this is a low density porous medium) will be proportional to the burning volume, so given a constant value of “A” it will be proportional to the height, thus to H=VS.t_b.
• The available data generally estimates a pHRR that ranges between 900 kW and 1100 kW.
So, given that the real height of the tree should not matter, then the pHRR should be similar to that of the literature. Because the tree was small, it was not so dry and it did not seem that dense I decided to opt for the lower bound value and estimate 900 kW.
Now, that being said, generally, literature values tend to be corrected by the time delay of the calorimeter. Our calorimeter has a time delay of about 10 seconds. What does that mean? Basically, it means that oxygen consumption measurements lag by 10 seconds the mass burning rate measurements. This generally makes no difference for events where things do not change within that period. If the event time scale is of the same order of magnitude of the time delay, then the measured value is somewhere between the measurement and that 10 seconds later. So, if I was to take the HRR curve measured by the calorimeter, then the value will be somewhere between what was measured 697 +/- 25 kW and 1000 kW.
An important lesson to learn is that the pHRR of a fast event (actually, even a slow event, but for different reasons) is a very difficult quantity to estimate precisely, thus the +/-25 kW stated as the error is truly only the direct measurement error. The true error will have to include the variability associated to the burn out time, the buoyantly driven upward spread, the global density and the comprehensive experimental error which is a parameter that is relevant in this case because the times are so short. So, any estimates within +/- 200 kW will probably have exactly the same value if the variable used is the pHRR. Thus 11/28 of you truly guessed the same answer.
If a different variable, such as the average HRR, or the Heat of Combustion was to be used as the “estimate,” then once all corrections due to time delay were made, would have probably delivered a smaller error bar.
The second variable to be estimated was the burning time. My estimate was 20 seconds and was based on a simple calculation of a typical upward flame spread rate of 10 cm a second. This had nothing to do with trees but with a fuel I know better (polyurethane foam). I estimated that the global value of “krhoC” is dominated by the density and I assumed that the density was more or less the same for both fuels. Thus I took that number. The tree was about 1.5 m, this gave about 15 seconds, time to burnout is so short that once the flame spread to the top, I could assume the fire was over.
Now, here is where I tried (unsuccessfully) to introduce a qualifier, I could not engage to estimate the initiation time (from the moment of ignition to the moment when the fire truly takes off). Furthermore, after the pHRR, what is left is lateral spread, then the branches and finally the trunk. The trunk will extinguish as soon as the branches die (bulk wood does not burn unless assisted!), but the lateral spread (being dominated by the shape of the tree) and the branches (being dominated by their individual shape and size) are impossible to predict. So at the end I gave up and simply estimated the time that it will take to achieve the pHRR from the moment the fire truly takes off. I reluctantly added a 5 second buffer for the slow initiation. While not a good estimate for what I was being asked, there is something to be said for the accuracy of the estimate! From the HRR curve we can establish that the primary burning will be somewhere between 10-20 sec (considering the instrument delay).
Now, what have I learnt, buoyancy is such a strong driving force that the estimate of the upward flame spread is a very robust one. The estimate of total burning time is one that carries a massive error bar, thus I will be reluctant to dismiss any of your estimates. From my perspective 28/28 gave estimates that I will consider within the expected error bars. Needless to say, last year’s Christmas tree was the proof to this point; the initial time could have been infinite if I did not decide to push the candle towards the denser part of the tree!
A final point, did I think of all of this in the 2 minutes that passed between the moment I learnt of the bet and the moment I provided my estimates? Obviously not! Most of this knowledge resides within your experience, and the estimate is an “educated guess.” Nevertheless, for the estimate to be adequate we need to carefully assess the question being asked (which I unfortunately decided to ignore) and the question needs to be posed correctly (meaning that what is being asked needs to have error bars that are smaller than the discrimination we are seeking). Otherwise, our guess will not be educated, nor it will be an estimate, it will just be a guess. If the error bars are small your chances of being the closest answer are very small (the educated estimate will have a much greater chance), but if the error bars are large you have as much of a chance to get it right as the most educated of estimates.
So citing Guillermo Rein: “while many stories can be told aposteriori,” and 3 hours of rationalizing my estimates can lead to this story, the stories need to be told and the discussion needs to follow. It is within the aposteriori 3 hours of introspection that I have truly managed to gain some insight into what happened not within the 2 minutes it took me to “guess.”
Congratulations to the winner!
Prof Jose Torero.
Saturday, February 12, 2011
Study or Gamble, but not both - 2nd annual Christmas tree fire test
An esteemed colleague had generously donated a Christmas tree to the scientific cause for the 2nd annual Christmas tree fire test. It had been used in the living room during the winter celebrations.
The tree was a Nordmann Fir of conical shape, 1.5 m tall and 0.9 m diameter at the bottom. It weighted 4.74 kg and was in dry conditions (measured in the oven at ~8% moisture content in dry base) after having spent one month not watered inside a warm living room.
Before conducting the experiment, fire experts were asked to bet on the peak heat release rate (pHRR) and the burning time (t_b). We recorded 28 guesses (£1 was collected per guess). A person with no research experience and no previous knowledge on fire dynamics (an international lawyer) was asked to provide a guess and act as control. NOTE: This required explaining the concept of HRR in layman terms, after which the control quantified the pHRR in terms of the equivalent number of burning matches.
Significant spread was recorded in the guesses. pHHR guesses ranged from 400 and 2300 kW, with average at 1173 kW. Guesses for t_b ranged from 15 to 377 s, with an average of 120 s. Two people provided guesses for pHRR but not for t_b, so they were assigned the average t_b value from the other participants.
The tree was ignited putting a small household candle next to the tree trunk at 1/3 of the height from the base. The HRR was measured using oxygen consumption calorimetry (corrected for CO and CO2 production). Figure 2 shows the HRR as a function of time. The growth of the fire is very fast, reaching a peak near 700 kW, 45 s after candle ignition. The decay is also fast, and reduces the fire to 50 kW 60 s after the peak. The peak value (pHRR) was 697 kW ± 25 kW. And the burning time t_b was 146 s ± 24 s. This was measured by visual observation using the video of the test and defined as the period going from first observed ignition of a tree element (between 0 and 24 s after candle ignition) to the end of significant flaming (between 146 s and 170 s).
A short video of the test can be seen below (NOTE: it starts 30 s after ignition and lasts for 55 s):
Measurements and guesses are plotted in Figure 3. There was only one guess falling within the measured result range. This person won the bet. For the quantification of how close a guess was to the measurements, the Euclidean distance was calculated, nondimensionalizing each guess by the measurement. The resulting average distance is 0.97, with minimum 0.1 and maximum 2.12. The control was at a distance of 0.26, well below the average and closer to the result than 89% of the participants.
The participants were grouped in three sets: Academics, Postdocs and Students. The years each participant has been researching fire was estimated and plotted against the distance of each guess (see Figure 4). There is a positive correlation of distance with experience. Students and postdocs show a similar large slope, but Academics are a distinct group from the rest and have a smaller slope.
Upon seeing this data, one could conclude that the longer you stay in research, the less you earn. And, study or gamble, but not both!
The tree was a Nordmann Fir of conical shape, 1.5 m tall and 0.9 m diameter at the bottom. It weighted 4.74 kg and was in dry conditions (measured in the oven at ~8% moisture content in dry base) after having spent one month not watered inside a warm living room.
Before conducting the experiment, fire experts were asked to bet on the peak heat release rate (pHRR) and the burning time (t_b). We recorded 28 guesses (£1 was collected per guess). A person with no research experience and no previous knowledge on fire dynamics (an international lawyer) was asked to provide a guess and act as control. NOTE: This required explaining the concept of HRR in layman terms, after which the control quantified the pHRR in terms of the equivalent number of burning matches.
Significant spread was recorded in the guesses. pHHR guesses ranged from 400 and 2300 kW, with average at 1173 kW. Guesses for t_b ranged from 15 to 377 s, with an average of 120 s. Two people provided guesses for pHRR but not for t_b, so they were assigned the average t_b value from the other participants.
 |
| Figure 1. Sequence of images, from left to right: The first day (early December 2010) when it was brought to the living room. Just seconds before ignition when the tree was inside the medium scale calorimeter. Fire spread over the tree about 40 s after ignition. Remains left after the test. |
The tree was ignited putting a small household candle next to the tree trunk at 1/3 of the height from the base. The HRR was measured using oxygen consumption calorimetry (corrected for CO and CO2 production). Figure 2 shows the HRR as a function of time. The growth of the fire is very fast, reaching a peak near 700 kW, 45 s after candle ignition. The decay is also fast, and reduces the fire to 50 kW 60 s after the peak. The peak value (pHRR) was 697 kW ± 25 kW. And the burning time t_b was 146 s ± 24 s. This was measured by visual observation using the video of the test and defined as the period going from first observed ignition of a tree element (between 0 and 24 s after candle ignition) to the end of significant flaming (between 146 s and 170 s).
 |
| Figure 2. Evolution of the HRR (power) as a function of time measured by oxygen consumption calorimetry.The ranges of observed times for the ignition of first tree element and end of flaming are indicated. |
A short video of the test can be seen below (NOTE: it starts 30 s after ignition and lasts for 55 s):
Measurements and guesses are plotted in Figure 3. There was only one guess falling within the measured result range. This person won the bet. For the quantification of how close a guess was to the measurements, the Euclidean distance was calculated, nondimensionalizing each guess by the measurement. The resulting average distance is 0.97, with minimum 0.1 and maximum 2.12. The control was at a distance of 0.26, well below the average and closer to the result than 89% of the participants.
 | |
|
The participants were grouped in three sets: Academics, Postdocs and Students. The years each participant has been researching fire was estimated and plotted against the distance of each guess (see Figure 4). There is a positive correlation of distance with experience. Students and postdocs show a similar large slope, but Academics are a distinct group from the rest and have a smaller slope.
 |
| Figure 4. Non-dimensional Euclidean distance from guess to measurements vs. years in fire research of each participants. Blue line is the trend of the Student and Postdoc populations. |
Upon seeing this data, one could conclude that the longer you stay in research, the less you earn. And, study or gamble, but not both!
Wednesday, February 09, 2011
EU Project on Aircraft Fire Safety starts
The University is one of 13 partners collaborating on a three year, EU funded research project in Aircraft Fire Safety. Below is a photo of the delegates who attended the 'kick-off' meeting in Poitiers, France, in January this year.
Monday, January 31, 2011
Wilde, mask and peer review
I just read that once Oscar Wilde wrote:
"Man is least himself when he talks in his own person. Give him a mask, and he will tell you the truth".
This might describe part of the rational on which the blind peer-review system stands? :)
"Man is least himself when he talks in his own person. Give him a mask, and he will tell you the truth".
This might describe part of the rational on which the blind peer-review system stands? :)
Sunday, December 26, 2010
Forecasting Fire on Scottish TV News
On 29 Nov 2010 Dr Guillermo Rein was interviewed by Scottish TV about a recent research paper published about "Forecasting Fire Growth".
On the same day he was interviewed for BBC Radio Scotland and The Scotsman.
On the same day he was interviewed for BBC Radio Scotland and The Scotsman.
Monday, December 20, 2010
Fertilizer fire aboard cargo ship
A recent journal paper titled "Small-scale experiments of self-sustaining decomposition of NPK fertilizer and application to the events aboard the Ostedijk in 2007" has published in Journal of Hazardous Materials. Its content is presented here.
The global fertilizer industry produces 170 million tonnes of fertilizer annually. As the global population increases and countries develop, this is expected to rise. Production sites are limited to locations with good availability of key raw materials. Therefore, large quantities are required to be shipped to the point of use.
Fertilizers contain three main ingredients essential for plant growth: nitrogen, phosphorous and potassium (NPK). These are present in various forms, however it is the presence of ammonium nitrate that constitutes the biggest risk. Ammonium nitrate is classified as a Dangerous Good by the UN Recommendations on the Transport of Dangerous Goods. This is because in the presence of an initiating event, ammonium nitrate will undergo self-sustaining decomposition. This is a chain reaction that occurs when a molecule of ammonium nitrate breaks down and releases heat which allows the decomposition of further molecules. In the presence of organic material this may result in explosion as in Texas City (1947) in which 581 people were killed.
Figure: The Ostedijk on 21st February (the 5th day) after the hold was opened and before specialized fire-fighting activities had commenced. Derived from photograph courtesy of Agencia EFE.
The research presented here gives an experimental insight into the decomposition of NPK fertilizers, highlights some of the limitations of the current UN Recommendations and applies the results to the events aboard the cargo ship Ostedijk in 2007.
The Ostedijk was carrying a cargo on NPK fertilizer from Norway to Spain when an accidental decomposition reaction occurred. The decomposition continued for seven days before it was stopped by partial flooding of the cargo hold as previous attempts to cool the cargo had been unsuccessful. During this time, a large plume of toxic gases formed and the crew had to be evacuated from the ship.
This unique set of experiments was performed in the laboratory using NPK 16.16.16, an industrially available fertilizer, and three different apparatus. The propagation behaviour was studied in an apparatus similar to that proposed by the UN test. Thermo-gravimetric analysis was performed to identify the reactions occurring and investigate the reaction mechanism. Finally, the state of the art for testing reactive materials, the Fire Propagation Apparatus, was used to find the conditions under which the reaction would become self-sustaining and to measure the heat of reaction.
The experiments showed beyond doubt that NPK 16.16.16 can undergo a self-sustaining decomposition reaction. This results in temperatures up to 350°C and releases heat at a rate of 1.8 MJ/kg of reacting fertilizer. This is in contradiction to the UN classification that the material is free from the hazard of self-sustaining decomposition. The paper allows us to understand and quantify some of the observations during the accidental event aboard the Ostedijk.
These experiments are important as there is very little research in the open literature regarding decomposition of ammonium nitrate containing fertilizers and this is the first time such measurements have been applied to a real scenario. They also provide an insight into this complex risk and the controlling mechanisms. The data and experimental methods can be used to further investigations into other incidents which may help in identifying causes of, and reduce losses from, this phenomenon.
The global fertilizer industry produces 170 million tonnes of fertilizer annually. As the global population increases and countries develop, this is expected to rise. Production sites are limited to locations with good availability of key raw materials. Therefore, large quantities are required to be shipped to the point of use.
Fertilizers contain three main ingredients essential for plant growth: nitrogen, phosphorous and potassium (NPK). These are present in various forms, however it is the presence of ammonium nitrate that constitutes the biggest risk. Ammonium nitrate is classified as a Dangerous Good by the UN Recommendations on the Transport of Dangerous Goods. This is because in the presence of an initiating event, ammonium nitrate will undergo self-sustaining decomposition. This is a chain reaction that occurs when a molecule of ammonium nitrate breaks down and releases heat which allows the decomposition of further molecules. In the presence of organic material this may result in explosion as in Texas City (1947) in which 581 people were killed.
The research presented here gives an experimental insight into the decomposition of NPK fertilizers, highlights some of the limitations of the current UN Recommendations and applies the results to the events aboard the cargo ship Ostedijk in 2007.
The Ostedijk was carrying a cargo on NPK fertilizer from Norway to Spain when an accidental decomposition reaction occurred. The decomposition continued for seven days before it was stopped by partial flooding of the cargo hold as previous attempts to cool the cargo had been unsuccessful. During this time, a large plume of toxic gases formed and the crew had to be evacuated from the ship.
This unique set of experiments was performed in the laboratory using NPK 16.16.16, an industrially available fertilizer, and three different apparatus. The propagation behaviour was studied in an apparatus similar to that proposed by the UN test. Thermo-gravimetric analysis was performed to identify the reactions occurring and investigate the reaction mechanism. Finally, the state of the art for testing reactive materials, the Fire Propagation Apparatus, was used to find the conditions under which the reaction would become self-sustaining and to measure the heat of reaction.
The experiments showed beyond doubt that NPK 16.16.16 can undergo a self-sustaining decomposition reaction. This results in temperatures up to 350°C and releases heat at a rate of 1.8 MJ/kg of reacting fertilizer. This is in contradiction to the UN classification that the material is free from the hazard of self-sustaining decomposition. The paper allows us to understand and quantify some of the observations during the accidental event aboard the Ostedijk.
Figure: (a) Unreacted fertilizer granules and (b) cross section showing partially reacted sample with 4 phases visible.
These experiments are important as there is very little research in the open literature regarding decomposition of ammonium nitrate containing fertilizers and this is the first time such measurements have been applied to a real scenario. They also provide an insight into this complex risk and the controlling mechanisms. The data and experimental methods can be used to further investigations into other incidents which may help in identifying causes of, and reduce losses from, this phenomenon.
Saturday, December 11, 2010
Prof Jose Torero's Christmas Lecture
Fire: A story of fascination, familiarity and fear
University of Edinburgh Christmas Lecture 2010
Presented by Prof Jose Torero
Recorded Wednesday 8th December 2010
Prof Jose Torero with the Tam Dalyell medal.
Wednesday, December 08, 2010
FireGrid: An e-infrastructure for next-generation emergency response support
by Dr Sung-Han Koo
A recent journal paper titled "FireGrid: An e-infrastructure for next-generation emergency response support" has been published in the Journal of Parallel and Distributed Computing. Its content is presented here.
The costs of fire are great, commonly estimated in the range of 1-2% of GDP. Despite this, emergency service intervention at fires is often reliant upon very basic information (i.e. fire alarm panel information) or simple “gut instinct” of experienced fire officers. This need not be the case in the modern era, when a range of technologies are available which, if effectively harnessed, could transform the way in which fire emergencies are tackled, thereby significantly impacting the costs associated with failures. Here we describe development and demonstration of a novel concept which integrates sensor technologies, fire simulation, High Performance Computing (HPC) and knowledge-based reasoning, to provide an “intelligent” emergency response system known as FireGrid.
The heart of the system is the sensor-linked fire model (described in more detail in reference 17). While fire simulation has found wide application historically for design purposes, the uncertainties of fire development defeat any attempt to provide a true predictive capability of hazard evolution, generally precluding real-time use. We bypass these uncertainties by continually updating our model with a flow of sensor-derived information regarding conditions in the building. The modelling strategy exploits Monte-Carlo techniques in combination with Bayesian inference for “steering”; being “embarrassingly parallel” in nature it is ideal for implementation on multiprocessor HPC systems. The output contains embedded probabilistic information about the likelihoods of various future hazard conditions, encompassing both threat to humans (i.e. escaping occupants, and incoming fire and rescue personnel) and to the building itself (in terms of structural weaknesses, or collapse potential). The interpreted information is conveyed rapidly to the end user, i.e. the “incident commander”, to provide decision support information that can effectively assist their intervention strategies.
Initial application of a system such as FireGrid would be most relevant to high-risk and critical infrastructures, including tall buildings. It is readily apparent that better information to incident commanders could be vital in avoiding scenarios comparable to the World Trade Center tragedies, where emergency responders continued intervention operations totally oblivious to the impending
collapse of the towers. FireGrid is an ambitious vision, and its success also depends upon an effective partnership and engagement with potential end users. Our initial project was undertaken in conjunction with various members of the UK fire and rescue services, culminating in a live fullscale demonstration test attended by a broad audience including a senior fire officer. The complex evolution of the fire, with unexpected behaviours and ultimate transition to “flashover”, was an ideal test of the sensor-linked model running on the grid, and the system capabilities were effectively demonstrated. Further development of such systems extends a genuine hope that some of the chronic and long-standing problems associated with accidental fires might be eventually be overcome, with wide–ranging benefits to all relevant stakeholders.
Editor note: A related paper is discussed in "Towards the forecast of fire dynamics to assist the emergency response"
A recent journal paper titled "FireGrid: An e-infrastructure for next-generation emergency response support" has been published in the Journal of Parallel and Distributed Computing. Its content is presented here.
The costs of fire are great, commonly estimated in the range of 1-2% of GDP. Despite this, emergency service intervention at fires is often reliant upon very basic information (i.e. fire alarm panel information) or simple “gut instinct” of experienced fire officers. This need not be the case in the modern era, when a range of technologies are available which, if effectively harnessed, could transform the way in which fire emergencies are tackled, thereby significantly impacting the costs associated with failures. Here we describe development and demonstration of a novel concept which integrates sensor technologies, fire simulation, High Performance Computing (HPC) and knowledge-based reasoning, to provide an “intelligent” emergency response system known as FireGrid.
The heart of the system is the sensor-linked fire model (described in more detail in reference 17). While fire simulation has found wide application historically for design purposes, the uncertainties of fire development defeat any attempt to provide a true predictive capability of hazard evolution, generally precluding real-time use. We bypass these uncertainties by continually updating our model with a flow of sensor-derived information regarding conditions in the building. The modelling strategy exploits Monte-Carlo techniques in combination with Bayesian inference for “steering”; being “embarrassingly parallel” in nature it is ideal for implementation on multiprocessor HPC systems. The output contains embedded probabilistic information about the likelihoods of various future hazard conditions, encompassing both threat to humans (i.e. escaping occupants, and incoming fire and rescue personnel) and to the building itself (in terms of structural weaknesses, or collapse potential). The interpreted information is conveyed rapidly to the end user, i.e. the “incident commander”, to provide decision support information that can effectively assist their intervention strategies.
Initial application of a system such as FireGrid would be most relevant to high-risk and critical infrastructures, including tall buildings. It is readily apparent that better information to incident commanders could be vital in avoiding scenarios comparable to the World Trade Center tragedies, where emergency responders continued intervention operations totally oblivious to the impending
collapse of the towers. FireGrid is an ambitious vision, and its success also depends upon an effective partnership and engagement with potential end users. Our initial project was undertaken in conjunction with various members of the UK fire and rescue services, culminating in a live fullscale demonstration test attended by a broad audience including a senior fire officer. The complex evolution of the fire, with unexpected behaviours and ultimate transition to “flashover”, was an ideal test of the sensor-linked model running on the grid, and the system capabilities were effectively demonstrated. Further development of such systems extends a genuine hope that some of the chronic and long-standing problems associated with accidental fires might be eventually be overcome, with wide–ranging benefits to all relevant stakeholders.Editor note: A related paper is discussed in "Towards the forecast of fire dynamics to assist the emergency response"
Thursday, November 25, 2010
Lloyd’s Science of Risk Prize goes to Fire Technology
Congratulations to Dr Francesco Colella for winning the Lloyd’s Science of Risk Prize in the Technology Category.
The prize was for his research paper "A Novel Multiscale Methodology for Simulating Tunnel Ventilation Flows During Fires" (published in Fire Technology). He led this work as a Research Associate at The School of Engineering from 2007 to 2010.
Dr Richard Ward, CEO of Lloyds told Francesco "The judging panel, comprising experts from academia and insurance felt your paper illustrated how novel computational methods can be used to reduce fire risk in the future. The panel were particularly impressed with how you reduced model run-time by concentrating on what is critical and by coupling fast and slower models".
This is Lloyd’s research prize for academics and aims at keeping the world’s leading specialist insurance market with the pace of academic knowledge and cutting edge thinking.
On top of this winning paper, The University of Edinburgh had three more papers short-listed as the top of each category (two of them from the fire group as well):
* Mr Craig Poland, short-listed in Technology Risk (best runner up), from the School of Medicine for his paper "Carbon nanotubes introduced into the abdominal cavity of mice show asbestos-like pathogenicity" (published in Nature Nanotechnology).
* Dr Wolfran Jahn, short-listed in Technology Risk, from the School of Engineering for his paper "Forecasting Fire Growth using an Inverse Zone Modelling Approach" (published in Fire Safety Journal).
* Dr Claire Belcher, short-listed in Climate Change Risk, from the School of Geosciences for her paper "Increased fire activity at the Triassic/Jurassic boundary in Greenland due to climate-driven floral change" (published in Nature Geoscience).
See related article Hot talent in risk research in the Staff Bulletin of the University of Edinburgh.
See press release by Springer.
Dr Richard Ward, CEO of Lloyds told Francesco "The judging panel, comprising experts from academia and insurance felt your paper illustrated how novel computational methods can be used to reduce fire risk in the future. The panel were particularly impressed with how you reduced model run-time by concentrating on what is critical and by coupling fast and slower models".
This is Lloyd’s research prize for academics and aims at keeping the world’s leading specialist insurance market with the pace of academic knowledge and cutting edge thinking.
On top of this winning paper, The University of Edinburgh had three more papers short-listed as the top of each category (two of them from the fire group as well):
* Mr Craig Poland, short-listed in Technology Risk (best runner up), from the School of Medicine for his paper "Carbon nanotubes introduced into the abdominal cavity of mice show asbestos-like pathogenicity" (published in Nature Nanotechnology).
* Dr Wolfran Jahn, short-listed in Technology Risk, from the School of Engineering for his paper "Forecasting Fire Growth using an Inverse Zone Modelling Approach" (published in Fire Safety Journal).
* Dr Claire Belcher, short-listed in Climate Change Risk, from the School of Geosciences for her paper "Increased fire activity at the Triassic/Jurassic boundary in Greenland due to climate-driven floral change" (published in Nature Geoscience).
See related article Hot talent in risk research in the Staff Bulletin of the University of Edinburgh.
See press release by Springer.
Tuesday, November 23, 2010
The domesticated animals sometimes turn back into wild beasts…
As somebody very wisely defined, fire is like a wild animal domesticated by humans: we not only learned how to use it push our vehicles, to do industrial hot work, or to produce glass, but also we taught and trained the pet to help in our daily housework with the cooking or the heating of our homes. Certainly, nobody can deny its usefulness, but the domesticated animal is always waiting for the opportunity to turn back into a wild beast… And, once this animal is on his runaway escape, it tends to climb like a primate, growling as a wild dog, searching for food in this desperate rush trying to satisfy its appetite for transformation which, far from the wild environment, mutates into an appetite for destruction instead…
In a dwelling, we typically have a few small domesticated fires; for example, those from the hobs in the kitchen, that from the boiler, and those from candles or even lighters. Very basically, these fires are kept small and dominated by controlling their supply of fuel and –sometimes– air; i.e. their basic menu. But just let these apparently harmless pets taste the flavor of the combustibles surrounding them, to see them –like vampires once they taste blood for the first time– switch into wild in an attempt to keep feeding and growing drastically powerful. Following this line, residential high-rise buildings with tens or even hundreds of apartments and figuratively countless combustibles are an awfully-high risky combination and “temptation” for these domesticated fires to break free and initiate a drastically fatal outcome eating everything in their reach.
Inevitable? Let’s say we can’t avoid the pet to turn wild every now and then, but we can definitely stop it from its fugitive run, keeping it fenced in its room of origin. Lately with the sighted rampaging beasts consuming all in front of them, can we conclude that the will to control the beast is all but extinguished, or will the human learn to capture the beast within science and engineered fields?
Thursday, November 18, 2010
Christmas Lecture
This years University of Edinburgh Christmas Lecture will be given by the winner of the Tam Dalyell Prize 2010 - Professor Jose Torero.
The lecture is entitled: Fire: A story of fascination, fear and familiarity. Prof Torero will examine how fire can provide welcome warmth in everyday life but, on a bigger scale, the unpredictability of fire can be terrifying.
More information on the talk and the prize here.
Wednesday, December 08, 2010 from 6:00 PM - 7:15 PM
at George Square Lecture Theatre, EH8 9LK
http://www.ed.ac.uk/maps/buildings/george-square-lecture-theatre
Book free tickets online at: http://www.ed.ac.uk/news/all-news/dalyell-171110
Note that tickets have run out fast for this event in previous years.
Monday, November 08, 2010
Fire Scholarships from The Lloyd’s Register Educational Trust
New Fire Safety Engineering scholarships from The Lloyd’s Register Educational Trust aim to make buildings safer from fire.
Modern buildings and the people who live and work in them will be better protected from the risk and consequences of fire, thanks to new education and research initiatives within the BRE Centre for Fire Safety Engineering at the University of Edinburgh.
Researchers at the University of Edinburgh are aiming for a better understanding of how contemporary building features – such as lighter construction materials and open-plan interiors – can influence how fires take hold and how fast they spread.
More than £200K in new student scholarships supported by The Lloyd’s Register Educational Trust will help to create a core of leaders who will use new understanding to bring change to the field.
Research and teaching programmes will seek to influence safety planning and design such as building evacuation procedures, fire-safe construction, and guidance for firefighters.
Top-flight undergraduate and postgraduate scholarship students will be recruited to create a cohort of fire safety specialists with expertise in all aspects of modern fire safety techniques.
Three LRET international MSc scholars will be sponsored through a new two-year International MSc in Fire Safety Engineering (IMFSE). The degree, the first multi-institution course of its kind globally, is operated by the Universities of Edinburgh, Lund and Ghent and funded by the European Commission’s Erasmus Mundus programme.
A further six LRET International MEng scholars will be supported in their final two years of the existing degree in Structural and Fire Safety Engineering at the University of Edinburgh.
Dr Luke Bisby, a researcher at the University of Edinburgh’s BRE Centre for Fire Safety Engineering, said: “Building design has changed radically in recent decades – we need a pioneering approach to developing fire safety solutions. We have to ensure that the chances of fire are as low as possible and that if a fire should occur, it will have little chance to spread, everyone inside can be evacuated safely, and economic and environmental losses can be minimised. Only through research linked to innovative educational programs can new approaches to fire safety take hold.”
Michael Franklin, Director of The LRET commented: “The Lloyd’s Register Educational Trust funds exceptional students studying science, engineering and technology throughout the world. We want to encourage and help them to become the future leaders in their chosen field. We hope The LRET scholarships at the University of Edinburgh will help to increase fire safety significantly in the years to come.”
The 2010 winners of the LRET Scholarships are (from left in the photo below):
• Ieuan Rickard, LRET MEng Scholar in Fire Safety Engineering
• Sarah Higginson, LRET MEng Scholar in Fire Safety Engineering
• Eduardo Maciel, LRET International MSc Scholar in Fire Safety Engineering
Congratulations to all three of the winners!
For further information, please contact:
Dr Luke Bisby, School of Engineering, tel 0131 650 5710; email Luke.Bisby@ed.ac.uk.
Notes:
The Lloyd’s Register Educational Trust is an independent charity that was established in 2004. Its principal purpose is to support advances in transportation, science, engineering and technology education, training and research worldwide for the benefit of all. It also funds work that enhances the safety of life and property at sea, on land and in the air.
Modern buildings and the people who live and work in them will be better protected from the risk and consequences of fire, thanks to new education and research initiatives within the BRE Centre for Fire Safety Engineering at the University of Edinburgh.
Researchers at the University of Edinburgh are aiming for a better understanding of how contemporary building features – such as lighter construction materials and open-plan interiors – can influence how fires take hold and how fast they spread.
More than £200K in new student scholarships supported by The Lloyd’s Register Educational Trust will help to create a core of leaders who will use new understanding to bring change to the field.
Research and teaching programmes will seek to influence safety planning and design such as building evacuation procedures, fire-safe construction, and guidance for firefighters.
Top-flight undergraduate and postgraduate scholarship students will be recruited to create a cohort of fire safety specialists with expertise in all aspects of modern fire safety techniques.
Three LRET international MSc scholars will be sponsored through a new two-year International MSc in Fire Safety Engineering (IMFSE). The degree, the first multi-institution course of its kind globally, is operated by the Universities of Edinburgh, Lund and Ghent and funded by the European Commission’s Erasmus Mundus programme.
A further six LRET International MEng scholars will be supported in their final two years of the existing degree in Structural and Fire Safety Engineering at the University of Edinburgh.
Dr Luke Bisby, a researcher at the University of Edinburgh’s BRE Centre for Fire Safety Engineering, said: “Building design has changed radically in recent decades – we need a pioneering approach to developing fire safety solutions. We have to ensure that the chances of fire are as low as possible and that if a fire should occur, it will have little chance to spread, everyone inside can be evacuated safely, and economic and environmental losses can be minimised. Only through research linked to innovative educational programs can new approaches to fire safety take hold.”
Michael Franklin, Director of The LRET commented: “The Lloyd’s Register Educational Trust funds exceptional students studying science, engineering and technology throughout the world. We want to encourage and help them to become the future leaders in their chosen field. We hope The LRET scholarships at the University of Edinburgh will help to increase fire safety significantly in the years to come.”
The 2010 winners of the LRET Scholarships are (from left in the photo below):
• Ieuan Rickard, LRET MEng Scholar in Fire Safety Engineering
• Sarah Higginson, LRET MEng Scholar in Fire Safety Engineering
• Eduardo Maciel, LRET International MSc Scholar in Fire Safety Engineering
Congratulations to all three of the winners!For further information, please contact:
Dr Luke Bisby, School of Engineering, tel 0131 650 5710; email Luke.Bisby@ed.ac.uk.
Notes:
The Lloyd’s Register Educational Trust is an independent charity that was established in 2004. Its principal purpose is to support advances in transportation, science, engineering and technology education, training and research worldwide for the benefit of all. It also funds work that enhances the safety of life and property at sea, on land and in the air.
Sunday, November 07, 2010
A Note on the Philosophy of Engineering Research
Foreword to the PhD Thesis of Dr Cecilia Abecassis Empis.
A Note on the Philosophy of Engineering Research
With the arrival of the computer era came a desperate frenzy of research in all fields with an ever increasing urge to quantify, discretise and explicitly pick apart nature enabling its eloquent description using the languages of mathematics and physics.
This very urge appears to be our largest limitation in attaining a precise representation of nature. Nature is, by nature, a continuum with an infinity that can not be quantified as much in the infinite immensity of the universe’s expanse as in the infinite minuteness into which things can be dissected and in the natural continuum of anything in between, exemplified by the naturally recurring but non-recurrent irrational numbers of Pi, Euler and Fibonacci.
Nevertheless intrinsic to human nature is a desire to group things, categorise, to box knowledge into entities we can comprehend and computers have allowed us to do this more quickly. Part of this process requires an evaluation of what is to be done and what it is to be used for. Be it an equation that represents the physics of electricity, the theories that describe types of intelligence or music that depicts the dance of the bees, the limits of its “accuracy” always lie within the bounds of the assumed scale, an agreement of the axioms of compliance.
Engineering is precisely the art and craft of deciphering such problems. The skill lies in evaluating the scope of the conundrum and identifying the critical players. In outlining the discrete pieces of this puzzle, engineers have to untangle the fundamentals from the peripheral fillers. They then stand back and reason the rules of the game using them to discard unnecessary detail and weave back together the key pieces creating an optimal solution. Engineering is a mere translation tool that allows for the interpretation of nature in a way we can fathom.
It is important however to distinguish a “solution” from “natural reality”. With the computing world fast-appealing to more and more of our senses, it is often tempting to indulge in smaller and smaller dissections of our problems. As we become increasingly obsessed with intricate dependencies we run the risk of creating a solution that is self-fulfilling without realising it has departed so far from its application that it has become a mere representation of the human ego with little or no use beyond the amusement of a select few curious minds. Detail can lead to a false sense of proximity to nature whereas the very nature of engineering is to accept that any attempt to model nature will always fall short of perfect. Instead engineering embraces the asymptotic nature of complex solutions and opts for providing simple and effective shortcuts that are perfect if they solve the particular problem at hand within the scope of its axioms. Hence an engineer must be humble and not lose sight of the problem objectives, the initial assumptions and the scale delineating the limitations and applications of engineering work.
Engineering research aims to provide rational solutions that make daily life just a little bit easier in order to make time for sitting back, relaxing and to enjoy the awesomeness of the irrational, chaotic magnificence of nature.
In this light it is hoped this work will make a useful contribution.
by Cecilia Abecassis Empis
A Note on the Philosophy of Engineering Research
With the arrival of the computer era came a desperate frenzy of research in all fields with an ever increasing urge to quantify, discretise and explicitly pick apart nature enabling its eloquent description using the languages of mathematics and physics.
This very urge appears to be our largest limitation in attaining a precise representation of nature. Nature is, by nature, a continuum with an infinity that can not be quantified as much in the infinite immensity of the universe’s expanse as in the infinite minuteness into which things can be dissected and in the natural continuum of anything in between, exemplified by the naturally recurring but non-recurrent irrational numbers of Pi, Euler and Fibonacci.
Nevertheless intrinsic to human nature is a desire to group things, categorise, to box knowledge into entities we can comprehend and computers have allowed us to do this more quickly. Part of this process requires an evaluation of what is to be done and what it is to be used for. Be it an equation that represents the physics of electricity, the theories that describe types of intelligence or music that depicts the dance of the bees, the limits of its “accuracy” always lie within the bounds of the assumed scale, an agreement of the axioms of compliance.
Engineering is precisely the art and craft of deciphering such problems. The skill lies in evaluating the scope of the conundrum and identifying the critical players. In outlining the discrete pieces of this puzzle, engineers have to untangle the fundamentals from the peripheral fillers. They then stand back and reason the rules of the game using them to discard unnecessary detail and weave back together the key pieces creating an optimal solution. Engineering is a mere translation tool that allows for the interpretation of nature in a way we can fathom.
It is important however to distinguish a “solution” from “natural reality”. With the computing world fast-appealing to more and more of our senses, it is often tempting to indulge in smaller and smaller dissections of our problems. As we become increasingly obsessed with intricate dependencies we run the risk of creating a solution that is self-fulfilling without realising it has departed so far from its application that it has become a mere representation of the human ego with little or no use beyond the amusement of a select few curious minds. Detail can lead to a false sense of proximity to nature whereas the very nature of engineering is to accept that any attempt to model nature will always fall short of perfect. Instead engineering embraces the asymptotic nature of complex solutions and opts for providing simple and effective shortcuts that are perfect if they solve the particular problem at hand within the scope of its axioms. Hence an engineer must be humble and not lose sight of the problem objectives, the initial assumptions and the scale delineating the limitations and applications of engineering work.
Engineering research aims to provide rational solutions that make daily life just a little bit easier in order to make time for sitting back, relaxing and to enjoy the awesomeness of the irrational, chaotic magnificence of nature.
In this light it is hoped this work will make a useful contribution.
by Cecilia Abecassis Empis
Monday, November 01, 2010
Towards the forecast of fire dynamics to assist the emergency response
A recent journal paper titled "Forecasting Fire Growth using an Inverse Zone Modelling Approach" has published in Fire Safety Journal. We are happy that the work has been widely featured in the media and many people is being exposed to the novel idea:
Effective control of a compartment fire saves lives and money. When fire fighters manage to put out a fire before it grows out of proportions, live safety is greatly increased and significant damage can be avoided. Moreover, the affected building can be re-occupied without major investment of resources. But when a fire passes a certain size, the building might collapses as a consequence of the fire damage to the structure (eg, 2001 WTC or 2005 Windsor Tower) or might have to be demolished due to irreversible damages.
Due to a lack of the required technology to support emergency response, fire fighters often have to follow their intuition when it comes to attacking the fire instead of basing their decisions on knowledge of the actual fire. This lack of information can lead to lost opportunities or unnecessary risks.
Prediction of the ongoing fire development ahead of time under different possible conditions based on the current events taking place would give fire fighters insight into the dynamics of the particular fire being flighted. With this extra knowledge, they could weight other options and feed more information into the emergency management. However, fire dynamics follow complex physical processes closely coupled to one another, which makes current tools not able to accurately forecast fire development in real time.
This emerging technology has been called Sensor Assisted Fire Fighting. The FireGrid project, to which this paper belongs together with the recent PhD thesis of the lead author, aims at providing physics-based forecasts of fire development by combining measurements from sensors in the fire compartment with a range of computational modelling tools. The sensor measurements can provide essential lacking information and compensate the accuracy lost, and thus overcome the shortcomings of current modelling tools and speed them up. The proposed methodology is to collect measurements in the fire compartment, and to assimilate this data into the computational model.
When enough measurements are available to characterize the current fire, a forecast is made. This forecast is then constantly updated with new incoming data. If, for example, a door is opened or glazing breaks, and the ventilation conditions change drastically, the sensor measurements will steer the computational model towards capturing the new conditions. With this technology, fire fighters could act upon forecast behaviour.
This paper presents one of the first steps in this direction. Data is assimilated into a simple zone model, and forecasts of the fire development are made. Positive lead times are reported here for the first time. These results are an important step towards the forecast of fire dynamics to assist the emergency response. Together with the application to CFD within the same PhD thesis, the previous thesis of Cowlard on flame spread predictions and the most recent paper by Koo et al. on probabilistic zone models, these establish the basis for technology for sensor assisted fire fighting. The envisioned system is not yet fit for operational purposes and further research is needed. The investigation of the effects of adding further realism in the fire scenarios will be the focus of future studies.
The paper can now be read at the website of Fire Safety Journal.
Note: A related paper is discussed in "FireGrid: An e-infrastructure for next-generation emergency response support"
- Interview for Scottish TV News (go to minute 19 here). Aired on 29 Nov 2010.
- Interview for BBC Radio Scotland (or go minute 42.20 here). Aired on 29 Nov 2010.
- Articles in The Scotsman, CORDIS-EU and Xinhuanet (in Chinese).
Effective control of a compartment fire saves lives and money. When fire fighters manage to put out a fire before it grows out of proportions, live safety is greatly increased and significant damage can be avoided. Moreover, the affected building can be re-occupied without major investment of resources. But when a fire passes a certain size, the building might collapses as a consequence of the fire damage to the structure (eg, 2001 WTC or 2005 Windsor Tower) or might have to be demolished due to irreversible damages.
Due to a lack of the required technology to support emergency response, fire fighters often have to follow their intuition when it comes to attacking the fire instead of basing their decisions on knowledge of the actual fire. This lack of information can lead to lost opportunities or unnecessary risks.
Prediction of the ongoing fire development ahead of time under different possible conditions based on the current events taking place would give fire fighters insight into the dynamics of the particular fire being flighted. With this extra knowledge, they could weight other options and feed more information into the emergency management. However, fire dynamics follow complex physical processes closely coupled to one another, which makes current tools not able to accurately forecast fire development in real time.
Figure: Conceptual representation of the data assimilation process and the sensor
steering of model predictions even when critical events take place in an evolving fire scenario.
steering of model predictions even when critical events take place in an evolving fire scenario.
This emerging technology has been called Sensor Assisted Fire Fighting. The FireGrid project, to which this paper belongs together with the recent PhD thesis of the lead author, aims at providing physics-based forecasts of fire development by combining measurements from sensors in the fire compartment with a range of computational modelling tools. The sensor measurements can provide essential lacking information and compensate the accuracy lost, and thus overcome the shortcomings of current modelling tools and speed them up. The proposed methodology is to collect measurements in the fire compartment, and to assimilate this data into the computational model.
When enough measurements are available to characterize the current fire, a forecast is made. This forecast is then constantly updated with new incoming data. If, for example, a door is opened or glazing breaks, and the ventilation conditions change drastically, the sensor measurements will steer the computational model towards capturing the new conditions. With this technology, fire fighters could act upon forecast behaviour.
This paper presents one of the first steps in this direction. Data is assimilated into a simple zone model, and forecasts of the fire development are made. Positive lead times are reported here for the first time. These results are an important step towards the forecast of fire dynamics to assist the emergency response. Together with the application to CFD within the same PhD thesis, the previous thesis of Cowlard on flame spread predictions and the most recent paper by Koo et al. on probabilistic zone models, these establish the basis for technology for sensor assisted fire fighting. The envisioned system is not yet fit for operational purposes and further research is needed. The investigation of the effects of adding further realism in the fire scenarios will be the focus of future studies.
The paper can now be read at the website of Fire Safety Journal.
Note: A related paper is discussed in "FireGrid: An e-infrastructure for next-generation emergency response support"
Tuesday, October 26, 2010
10 January 2011 is the Application Deadline for the International Master of Science in Fire Safety Engineering Program
The blog of the SFPE remind us that 10 January 2011 is the application deadline for the International Master of Science in Fire Safety Engineering Program (IMFSE).
The IMFSE is commonly organized by the universities of Ghent (Belgium - coordinator), Edinburgh (UK) and Lund (Sweden). This two-year educational program in the Erasmus Mundus framework provides the required knowledge for a professional fire safety engineer in a Performance Based Design environment.
The application forms, basic requirements and all other information are found on the website: http://www.imfse.ugent.be.
The IMFSE is commonly organized by the universities of Ghent (Belgium - coordinator), Edinburgh (UK) and Lund (Sweden). This two-year educational program in the Erasmus Mundus framework provides the required knowledge for a professional fire safety engineer in a Performance Based Design environment.
The application forms, basic requirements and all other information are found on the website: http://www.imfse.ugent.be.
Monday, October 18, 2010
A novel methodology for simulating tunnel fires
A recent journal paper titled "A Novel Multiscale Methodology for Simulating Tunnel Ventilation Flows During Fires" has recently been published in the journal Fire Technology. Its content is presented here. This is a joint research effort between Politecnico di Torino and University of Edinburgh.
PD NOTE: This paper won this year’s Lloyd’s Science of Risk Prize in the Technology Category. The prize is awarded to academics and aims to keep the world’s leading specialist insurance market abreast of the latest academic knowledge and cutting-edge thinking. See press release by Springer.
In the past decade over four hundred people worldwide have died as a result of fires in road, rail and metro tunnels. In Europe alone, fires in tunnels have destroyed over a hundred vehicles, brought vital parts of the road network to a standstill - in some instances for years - and have cost the European economy billions of euros. Disasters like the Mont Blanc tunnel fire (1999) and the three Channel Tunnel fires (2008, 2006 and 1996) show that fire poses a serious threat.
Comprehensive risk assessments for tunnel fires are not easy to conduct. The development of the possible emergency scenarios is dependent on the combined influence of fire detection technologies, ventilation system, tunnel layout, atmospheric conditions at the portals and the presence of vehicles. Nowadays, the analysis of such complex phenomena is performed using numerical computational fluid-dynamics (CFD) tools. But CFD has a significant drawback: its requires very large computational resources (e.g., weeks or months of computing time). This limitation affects the completeness of the risk analyses because they can only be based on a limited number of possible scenarios but do not explore the wide range of possible events.
This recent paper proposes a novel multiscale modelling approach generated by coupling a three dimensional CFD model with a simple one-dimensional model. This allows for a more rational use of the computational resources. The methodology has been applied to a modern tunnel of 7 m diameter section and 1.2 km in length (similar layout to the Dartford Tunnels in London). Different ventilation scenarios are investigated involving fire sizes ranging from 10MW to 100MW.
The multiscale model is proved to be as accurate as the traditional time consuming CFD techniques but provides a reduction of two orders of magnitude in the computational time. This greatly widens the number of scenarios that can be efficiently explored. The much lower computational cost is of great engineering value, especially when conducting comprehensive risk analyses, parametric, sensitivity and redundancy studies, required in the design or assessment of ventilation and fire safety systems.
The multiscale methodology is the latest contribution to the state-of-the-art in computational methods for tunnel flow simulations. The model has been validated against experimental data of cold flow ventilation and shown to be accurate. This work was published in Building and Environment in 2009. It has also been used to provide the tunnel operator with a comprehensive assessment of the ventilation in the Dartford Tunnels, located under the River Thames about 15 miles east of London. This work was published in Tunnelling and Underground Space Technology in 2010 (open access version).
PD NOTE: This paper won this year’s Lloyd’s Science of Risk Prize in the Technology Category. The prize is awarded to academics and aims to keep the world’s leading specialist insurance market abreast of the latest academic knowledge and cutting-edge thinking. See press release by Springer.
In the past decade over four hundred people worldwide have died as a result of fires in road, rail and metro tunnels. In Europe alone, fires in tunnels have destroyed over a hundred vehicles, brought vital parts of the road network to a standstill - in some instances for years - and have cost the European economy billions of euros. Disasters like the Mont Blanc tunnel fire (1999) and the three Channel Tunnel fires (2008, 2006 and 1996) show that fire poses a serious threat. Comprehensive risk assessments for tunnel fires are not easy to conduct. The development of the possible emergency scenarios is dependent on the combined influence of fire detection technologies, ventilation system, tunnel layout, atmospheric conditions at the portals and the presence of vehicles. Nowadays, the analysis of such complex phenomena is performed using numerical computational fluid-dynamics (CFD) tools. But CFD has a significant drawback: its requires very large computational resources (e.g., weeks or months of computing time). This limitation affects the completeness of the risk analyses because they can only be based on a limited number of possible scenarios but do not explore the wide range of possible events.
This recent paper proposes a novel multiscale modelling approach generated by coupling a three dimensional CFD model with a simple one-dimensional model. This allows for a more rational use of the computational resources. The methodology has been applied to a modern tunnel of 7 m diameter section and 1.2 km in length (similar layout to the Dartford Tunnels in London). Different ventilation scenarios are investigated involving fire sizes ranging from 10MW to 100MW.
The multiscale model is proved to be as accurate as the traditional time consuming CFD techniques but provides a reduction of two orders of magnitude in the computational time. This greatly widens the number of scenarios that can be efficiently explored. The much lower computational cost is of great engineering value, especially when conducting comprehensive risk analyses, parametric, sensitivity and redundancy studies, required in the design or assessment of ventilation and fire safety systems.
The multiscale methodology is the latest contribution to the state-of-the-art in computational methods for tunnel flow simulations. The model has been validated against experimental data of cold flow ventilation and shown to be accurate. This work was published in Building and Environment in 2009. It has also been used to provide the tunnel operator with a comprehensive assessment of the ventilation in the Dartford Tunnels, located under the River Thames about 15 miles east of London. This work was published in Tunnelling and Underground Space Technology in 2010 (open access version).
Thursday, October 14, 2010
Heron Tower and the begining of the concept of travelling fires in design
A recent article about Arup's fire design of Heron Tower (among the tallest buildings in London) appeared in Info4fire.com. The Heron Tower project won the Fire Safety Engineering design category at the Fire Excellence Awards in May 2009.
Heron Tower is a landmark in our collaboration with Arup since it led to a joint PhD thesis and a series of papers. Since 2007 we are working together to define novel design fires in similar large spaces to that in Heron Tower. We came up with the concept of "travelling fires". The initial work was presented at Interflam 2007.
Figure: Snapshot from the fire model using FDS published here. Temperature map for a 500 kW/m2 well-distributed fire on the bottom floor with top and bottom floor ventilation. The atrium acts as a chimney, linking the bottom and the top floors.
Since then, the research has advanced significantly and led to several other papers and case studies. We recently published an overview and a building survey in the magazine Fire Risk Management. The key element behind this research is the need to provide design solutions to the large parts of modern buildings that fall outside the limits set out in the Eurocodes.
The two articles published in Fire Risk Management led to an unusual number of Letters to the Editor. Letters from Mike Wood, Pilkington Group, and from Dr Kirby, Sirius Fire Safety Consultants, were received. This and this were our respective replies (our reply to Dr Kirby is also attached below).
NOTE: Thanks to Chris for mentioning the article.
---
On Fire Risk Management Feb 2010, Dr Kirby from Sirius Fire Safety Consultants commented on our article "Out of Range".
Our reply, Beyond Limits, in Fire Risk Management March 2010 read:
We are pleased to read the letter that Dr Kirby, from Sirius Fire Safety Consultants, wrote in response to our December cover article, "Out of range". In our article, we reported a survey of 3,080 compartments on the campus of the University of Edinburgh buildings underlining the compartment volume that falls inside of the design fire specifications of current Eurocode 1 (66 % of the older buildings, but only 8% of the most modern one). Instead of volume, Dr Kirby prefers to quote our results as % of the number of compartments (95 % of the older buildings, but only 63% of the most modern one), assuming perhaps that all compartments are equally important regardless of the very large differences in size (e.g., atria vs. single desk office). But the main conclusion of our article, that the modern building contains a very large portion of built environment outside the limits of the Eurocode, stands true no matter what survey quantity is quoted.
Dr Kirby also refers to the UK National Application Document which extends beyond the Eurocode 1 range and without limit, the use of these post-flashover design fires. We consulted this document while investigating the technical origins of the Eurocode, but after two years of searching and requests, we have not been able to find a copy of the validation work it cites. If Dr Kirby or any reader of the FRM magazine could kindly send us a copy of the validation work, we would be grateful. We hope that full details of these studies are made available to the fire research community at large for the benefit of all.
We agree that Eurocode 1 is a good document and a first step putting fire engineering into a codified form. We appreciate Dr Kirby's kind words of support for research in alternative design fires. His comments on fuel-control fires in large compartments resonate very well with our previous article in this publication ("Travel guide", November 2009, pp.12-16 by J Stern-Gottfried, G Rein and J Torero). In that article, we highlight that in large compartments, a post flashover fire is not likely to occur, but that a travelling fire spreading across the floor plate should be considered instead. We think that in the future travelling fires should also be considered as design fires and compliment the current Eurocode. Work conducted to date is available and easily accessible to the fire research community at large for the benefit of all.
Dr Guillermo Rein, BRE Centre for Fire Safety Engineering, The University of Edinburgh
14 September 2009, Info4fire.com
Heron Tower is a landmark in our collaboration with Arup since it led to a joint PhD thesis and a series of papers. Since 2007 we are working together to define novel design fires in similar large spaces to that in Heron Tower. We came up with the concept of "travelling fires". The initial work was presented at Interflam 2007.
Figure: Snapshot from the fire model using FDS published here. Temperature map for a 500 kW/m2 well-distributed fire on the bottom floor with top and bottom floor ventilation. The atrium acts as a chimney, linking the bottom and the top floors.
Since then, the research has advanced significantly and led to several other papers and case studies. We recently published an overview and a building survey in the magazine Fire Risk Management. The key element behind this research is the need to provide design solutions to the large parts of modern buildings that fall outside the limits set out in the Eurocodes.
The two articles published in Fire Risk Management led to an unusual number of Letters to the Editor. Letters from Mike Wood, Pilkington Group, and from Dr Kirby, Sirius Fire Safety Consultants, were received. This and this were our respective replies (our reply to Dr Kirby is also attached below).
NOTE: Thanks to Chris for mentioning the article.
---
On Fire Risk Management Feb 2010, Dr Kirby from Sirius Fire Safety Consultants commented on our article "Out of Range".
Our reply, Beyond Limits, in Fire Risk Management March 2010 read:
We are pleased to read the letter that Dr Kirby, from Sirius Fire Safety Consultants, wrote in response to our December cover article, "Out of range". In our article, we reported a survey of 3,080 compartments on the campus of the University of Edinburgh buildings underlining the compartment volume that falls inside of the design fire specifications of current Eurocode 1 (66 % of the older buildings, but only 8% of the most modern one). Instead of volume, Dr Kirby prefers to quote our results as % of the number of compartments (95 % of the older buildings, but only 63% of the most modern one), assuming perhaps that all compartments are equally important regardless of the very large differences in size (e.g., atria vs. single desk office). But the main conclusion of our article, that the modern building contains a very large portion of built environment outside the limits of the Eurocode, stands true no matter what survey quantity is quoted.
Dr Kirby also refers to the UK National Application Document which extends beyond the Eurocode 1 range and without limit, the use of these post-flashover design fires. We consulted this document while investigating the technical origins of the Eurocode, but after two years of searching and requests, we have not been able to find a copy of the validation work it cites. If Dr Kirby or any reader of the FRM magazine could kindly send us a copy of the validation work, we would be grateful. We hope that full details of these studies are made available to the fire research community at large for the benefit of all.
We agree that Eurocode 1 is a good document and a first step putting fire engineering into a codified form. We appreciate Dr Kirby's kind words of support for research in alternative design fires. His comments on fuel-control fires in large compartments resonate very well with our previous article in this publication ("Travel guide", November 2009, pp.12-16 by J Stern-Gottfried, G Rein and J Torero). In that article, we highlight that in large compartments, a post flashover fire is not likely to occur, but that a travelling fire spreading across the floor plate should be considered instead. We think that in the future travelling fires should also be considered as design fires and compliment the current Eurocode. Work conducted to date is available and easily accessible to the fire research community at large for the benefit of all.
Dr Guillermo Rein, BRE Centre for Fire Safety Engineering, The University of Edinburgh
Tuesday, September 28, 2010
Flashover Training
And yet again the Lothian & Borders Fire & Rescue Service has played host to a couple of University of Edinburgh ‘boffins’. This time however, they would be crawling into a steel shipping container to watch an indoor bonfire…I somehow managed to video the event, however some parts I couldn't film properly without my camera melting.
For example, at one point you hear the instructor shouting “close the vent!” at which point the chimney above us shuts and flames shoot overhead, just about cooking me and my camera. I hit the deck and accidentally hit the zoom button. When I recover there is a firefighter ready to “knock back” the fire with a hose. Good thing someone knew what they were doing!
Shortly after I ditched the camera out the back of the unit and shuffled forward towards the flames. When I got there the instructor called me forward and handed me the hose. “Close the vent!” (Flames immediately start shooting overhead) “Wait...” he says. “Let it get going…Now!”
I started spraying water around. Too much. The atmosphere immediately filled with steam, the temperature increased, the smoke layer dropped and the pressurise in the compartment squeezed my head. Oops. Guess I should've listened to the briefing a bit more carefully. ‘Use as little water as possible’…Got it.
I was lucky I got to shuffle to the back again and ‘cool’ off. The temperature inside the compartment was about 750C at head height and a mere 250C down where we were crouching. I don't know how else to say this, but it was hot. (I mean, just imagine sitting inside an oven on full heat, while wearing a ski-suit).
The day was meant as a trial run for November, where I plan to
organise a repeat of the demonstration/experience with a larger group from UoE. Whether this goes ahead or not will depend largely on whether LBFRS will be willing to commit any more of their time and resources. Although from what I gather they are more than happy to demonstrate their knowledge of compartment fire dynamics!
I believe these days have been very effective in strengthening the links between academia and firefighters, and in breaking (and confirming :) stereotypes from both sides. Many thanks again to Kenny, John and Des from the SIFTC. Cheers guys.
Watch the full video on YouTube.
Mike.
Subscribe to:
Posts (Atom)