Tenability criteria with Gabriele Vigne

When you studied fire safety engineering, or maybe when you joined your first consultancy company and was trained to work on fire safety engineering projects, at some point you most likely encounter the thing that we call the tenability criteria. The tenability criteria are the conditions in the building or a part of a building, that if they are reached, it means that building became untenable. And that means that the building is not safe anymore. It's a very profound concept in fire safety engineering to use this values to define when the safety ends in the building. And in consequence, most of the design of safety features in the building such as smoke control, for example, is driven by this need to provide the safety for a certain amount of time. Because the tenability criteria are so profound for our buildings and safety in them, we tend to study them in depth. And this is why I brought today my my second guest to the podcast, Dr. Gabriela Vigne from JVVA in Spain, who happens to be a researcher and an engineer interested in tenability criteria from both the scientific and practical perspective. Together with Gabrielle, we have a history of research on visibility in smoke. And now we've opened the new chapter. We've just finished a survey international survey on the use of tenability criteria in fire engineering, which was attended by 250 engineers worldwide, which we're very happy with. And in this episode, we're gonna dive deep into the results of the survey will discuss what surprised us in the survey What did not surprise us, we will then jump into the visibility in smoke, which is something we share the passion about. And in the end, I'll I'll explain Gabriele some of the newest research from my group, that that lays a new light on how we perceive the visibility in smoke. So whatever is your role in fire safety engineering, you're definitely gonna enjoy this one and I hope it will be very beneficial for you. I hope the perspectives given are very wide and will help you see things in a new light. I'm absolutely sure you will enjoy it. So without further do, let's spin the intro and jump into the interview. Welcome to the fire science show. My name is Wojciech Wegrzynski and I will be your host.

Hello, everyone. Today I'm hosting my good friend Gabrielle Veen from GVA fire and risk consultancy, and Gabrielle is an engineer and researcher. And I'm more than happy to have you here. Gabrielle. Hey, Wojciech. It's a pleasure to be here in this fantastic podcast. +

We'll see. We'll see. I hope it is. So I broke up real here. Because together we have a history of cooperation on some interesting topics. And since we are both engineers, and we both deal with design of smoke control systems and and all this stuff related to making building safer for the life of humans. We've both met the same problems with tenability conditions. Gabrielle.

Yeah, well, when we start with that, we identify this visibility as one of the main, the main issue and anything to study more, more deeply.

Yeah, visibility is the culprit of engineering because it's, at the same time, extremely easy to explain to anyone...you can explain the concept of visibility in smoke to your grandmother. I mean, you can see for a given distance, that's it. And yet to capture the value of that parameter in Europe, for high fidelity engineering analysis is is really really painful task. So for for this purpose to learn how we use tenability criteria and to learn what is driving the design of buildings in in today's engineering world. We joined our forces with with some researchers with Professor Lukas Arnold from University of Wuppertal, with Dr. Wolfram Jahn from Chile and our colleague Mateusz Zimny. And we've started a really interesting project that teniability criteria survey. Gabrielle, can you tell us a bit more about the survey?

Yeah, well, actually, we start this survey to understand worldwide how people use this tenability idea in safety, fire modeling. And actually, it's so easy now nowadays, to make a survey and, you know, collect information from all parts of the world. This was just, you know, posted in the in Twitter in, in LinkedIn, we didn't even send messages to people and it spread all over the world! And then there was a bit of a run and the competition between countries as well, because when we post this graph and saying a look, you are pretty low in in the result, some French people posting that..."Hey, come on... man",

that explains the high number of French participants.

exactly. And that was fantastic. Absolutely great. Absolutely great. I think it was a great success. And I am preparing right now, but I will be giving classes to the the municipality of Madrid to that technician about equation modeling, and will be focused on Petfinder action. But there is a strong introduction about the human behavior and, and evaluation modeling itself. And I use the lot the information from the survey that Ronchi and Loverglio did for evacuation modeling. Why that? Because Because what when you have a benchmark of how the model is used all over the world, etc, etc. And so many questions in it, you really have an overview and you understand how those models are used. So I, I really think these survey will highly contribute to this field.

Yeah, my goal was to ask two or three very, very particular questions. And to get an honest answer to them. I've hidden them beyond the wall of of many, many other questions. So so people are getting are more comfortable answering and and clicking through. But you know what, I'm really surprised because the data gathered like for almost every single question we got is astounding. And it's like I mentioned before, full of golden nuggets in there went from here, I would like to really thank every single person who participated in the survey, it's, it's worth a lot to us to have your opinion on how the tenability criteria are used. And also huge thanks to everyone who helped us organically spread the information about the survey through the internet. So I'm looking now at the participation chart, it seems we had 252 participants, that's, that's amazing. That's a quarter of 1000s of people. And most of them are engineers, like 81% are engineers who actually do fire modeling. So that's, that's a pretty decent representation.

Yeah, absolutely. And I think it's the right representation. I like this, this answer...Because Can you imagine this 81% if it was a researcher, or lecture, we want and I think the good thing here is that we are both engineer, you are more on the scientific and research path now. But we know that what we are studying and that was the intent, for example of all my research, to contribute to my profession, not just to study something, because, you know, I want to expand the Bougher-Lambert-Beer equation better. Okay, I want the results that can be applied tomorrow into our, our field. And that's a key key thing.

It's fascinating what you say that because I think my path of career was very similar. I was bottom level CFD engineer, a guy who was doing a model and and starting the calculations and then printing pretty plots out of that. And at some point I've started questioning, like the outcomes or the choices are the reasons why we do stuff like like We do and and that pushed me into the researcher career where I seek answers for questions that were unanswered when I was doing the engineering. So I always say it's selfish science because I, I do science for myself, I touched problems that bother me. And I think that it makes the science very useful. Here to close the participation chart, I must mention the top three participating countries. And it seems number one was UK. Number two was Poland. Good job, guys. And number three was Germany. So yeah, very strong European representation in the survey. But we also had participants from Iran, Australia, New Zealand, Chile. Many from USA. So it's definitely a global global survey.

Yeah. And to be honest, if you have a two the three people honestly, in the same countries, you have already, you know, an answer of on how that criteria or many other answers apply to that country. So the result is absolutely great.

We, we also had most of the participants being active users of the CFD tools, like doing the 70% of the participants were doing simulations by the by their own, and 17% of the participants, were ones we have a supervision role on CFD. And one more thing interesting from the, from the scatter of our participants, 94% of them are familiar or use FDS as the tool of CFD. So no, that's kind of not surprising. ,

No, it doesn't surprise me at all. The funny thing is to see some PyroSim.

Oh, yeah, we have a group of people who mentioned PyroSim as CFD model hich is also interesting. I wond r what Brian will say about tha . Well, we are here you have to mention that we have also a v ry healthy mix of, of veterans and newbie participants, because we had 43% of people with experi nce in CFD of more than 10 year , and very similar n mber of of participants who just start with the CFD in their fir t years of engineering careers. So a very, very good mix betwee veterans. A

And also Wojciech. One thing, the professional education is quite interesting as well, because having 56% that study fire safety protection engineering, that's a lot because we know that we have a substantial problem and that we don't have so many university to fire safety engineering.

That is true, especially that we had like a lot of participants from Spain, for example, where you don't have a structure of teaching fire safety engineering as a master's program, right? You have, and there's probably many more countries that do not have a structured way to educate the fire safety engineers, yet more than 50% of, of the participants said that their professional education is, is fire safety engineering, which is great. So the main part of the survey was to ask the participants, what tenability do they use in performing ASET-RSET analysis, life safety analysis? The first two answers are very obvious, because like 80% of people are using temperature 90% of people are using visibility. So that's something that does not shock me. But I was pretty surprised. Like other more complex and abridged criteria are still being fairly well used.

Yeah, well, I think that's all the things we put here, they come from the state of the heart and and best practices. So, you know, as a fire, safety, protection, whatever engineer you know that you have this range of criteria that you need to evaluate, okay. And also, there are codes, international codes or just Local Code required this kind of often obeyed criteria. It doesn't surprise me that the most used are visibility and temperature.

It didn't surprise me there, but I'm still surprised that like 50% of our participants have used mass density of smoke or smoke concentration as as a criterion especially that it takes some effort to plot that in FDS. While visibility is the simplest way to to show that also, we had a good number of people using radiant heat fluxes, gas concentrations, smoke layer height that that was one that surprised me the most. For smoke layer height, we had a combined value of 75% of participants who said that smoke layer height is very often or almost every time used in their analysis, which is kind of surprised to me because I personally would focus on on visibility and temperatures rather than heights. How is it for you Gabrielle and Spanish engineering.

Well, I think the reason and misunderstanding...global misunderstanding on this smoke layer height. First of all, because it does exist, and in all the large fire test performed, I saw that that existing right, but how is that used? Because normally is just an interpretation of, you know, you have the smoke above your head, and that's done. Okay. So you designed the smoke control system based on that when I when I published my first work and I remember that perfectly It was 2012 in Hong Kong, okay, in the in the SFPE perform based conference, and they discuss the difference approach to plume in time correlation. So the Thomas, Zukowski, McCaffrey, and McCaffrey plumes. Why did that because you're not estimate, estimating what happened at your height, okay, head height, you're just assuming that the smoke is at a certain height into the enclosure. Okay. And that's how you simplify all the issue. The saying, look, I want to smoke layer 2.5 meters, Okay, excellent. Okay, and then you use this very simplified formula to get this smoke layer at a certain height. So I think we have a bit of a problem there. And also, if you do a CFD, so if you run a transient simulation for with tones of CPU hours, okay, to show that this model is stand at three meter or some meters above the floor is not probably the case. Okay? Because you want to really understand what what is happening in your enclosure at a certain time. When you're passing through the smoke.

I think you're touching a very important thing here that in plume correlations where you would calculate the the mass flow of the smoke into a smoke layer, there is this magical parameter of the of the smoke layer height in almost every single model that you can use for for plume calculations. This is sometimes considered to be the smoke layer height that you expect. But in fact, it is not it's it's it's just the length of the part of the plume at which the entrainment happens. And that does not necessarily correlate to the smoke layer height that you will end up in a steady state analysis. And you've done that with your paper about about smoke plumes where you've integrated these calculations over time and and shown how the simple plume models can be used to determine this this parameter over time, right. And it was magical that they actually that they actually met what was measured in many experiments, not just one. I found that fascinating, because it's like, How much time does it take to simulate that huge atrium that you that you used for your experiment that that must be days to run? The Murcia, properly?

Yeah. Absolutely. When we are so many cases, and to be honest, I haven't published that all the work I've done with zone waters as well. Because I run with the CFAST, OZone, Branzfire and risk all the scenarios. Okay, but because the plume correlation where we're there, so I simplify the concept. Again, to get an answer, instead of to have a published paper.

In a way a calculating smoke layer height1 is fascinating, yet, if you consider running CFD analysis, for me, it's kind of the concept. It doesn't click for me because I want the smoke to not be where my humans are. And I want the margin of safety, like above the heads of participants, but that would be project based. I would not evaluate just the height of the layer. But I agree you mentioned that authorities main may require that or firefighters may want that or you want to be in alignment with with the standard that requires that so I assume that that must be.

Yeah, and sometimes you just need to show them because they want that answer. Because for example, in fire dynamics siumlator, FDS, you have the capability to to get the smoke layer. Okay? height. But that is based on temperature and is not that accurate. Okay? If you see the result of from FDS of this smoke layer that is like sort of laser beam, and you measure the the smoke layer between temperature and other coefficient, it doesn't correspond to what you see, for example, if you if you show us nice file with visibility.

Yeah, and I think in one of your papers, I've also read that integration approach for the layer height can make a difference of meters in terms of the evaluated height, depending if you take, like 30% or 10% of the of the of the smoke concentration or temperature as the end of your layer. And it's I mean, it's it's very rarely a very sharp condition, there must be a gradient not not a sharp line.

Exactly. And that it depends a lot of the soot yield of the smoke, if you see them the results. Well, the picture of the test we were performing in ITB, for the most metanol, heptane, toluene and propanol probably.

Oh, man, we're not allowed to burn toluene in that lab after that one...

the toluene as incredible. So you can see hat the most sooty is the smok , the less gradient you have.

Yeah, that's that's true. We, we've seen that. To continue on the survey. A lot of people are very interested, we had 250 people answering the questions. So I assume all of them are interested in the outcomes. And the second important question we asked is, how often do you take engineering decisions based on condition? So so it's not only

Not at all, you should probably define this if you use that, but how often that condition changes the outcomes of your design. And for me here, the only consistent one really is is visibility in smoke, because we had 89% of people saying they use visibility in smoke, and we have 89% people mentioning that they take decisions based on this on this criterion. So it's very consistent. For example, for temperature, it's it's slightly skewed because we had 76% people saying they use it, but just 45% doing design choices based on that temperature. Does this shock you or not? better. Engineering choice. We are talking about what what is the criteria that define your ASET time? Okay, that's one of the most important thing so you have like a carpark. You have a spread of smoke for a fire. Right. Okay. So what is the first criteria? Okay, that appear there and and say, a condition untenable. Okay. So it doesn't surprise me at all that the most use that tenability criteria in the definition that is visibility and temperature appear here in the most used one in determining ASET time. And also, I think we can combine visibility with mass concentration of smoke.

Yeah, we will, we will do that when processing the survey results properly. Because it's a lot of data that needs to be scientifically processed. And we will do that. Definitely. Here. It's just a quick look, because we are so excited with the results. I mean, we had 250 colleagues share their opinion with us. And then that makes us very happy. Sorry, the other ones were smoke concentration, which define ASET in 40% of cases. And I also saw the smoke layer hide was actually quite high because it was like 60% of the cases. And the other ones other conditions like radiant heat flux, toxic gases, they were usually below 25%. So less than 25% engineers say they this this criterion is very impactful for the ASET analysis.

Yeah, just one one thing Wojciech. I think this is absolutely consistent with the science. Okay, off course. With the mention here, if the building is an airport, a tunnel a smaller room, or if we look into fire intervention or the evacuation... Well, it's mostly evacuation because it is ASET-RSET. But when it happened to me for example, in remember that in Riyadh I did some analysis for one of the biggest station of the metro there and the authority as gas to measure to provide as a result of radiative heat flux, CO concentration, what more here

we are very often asked for oxygen concentration,

Oxygen concentration etc, etc, okay. So, at the end, we did provide that, but it was absolutely useless, okay. Because, if you have your visibility at like 10 meters, for example, most of the toxic gases, we are talking again about a normal enclosure with an celulosic or plastic fire, okay, something that we normally deal with. And when you have a visibility of 10 meters, normally all the toxic gases are far down in the limit and the concentration. And the same thing happen, for example, in the in the radiation, because radiation is temperature to the times of four. So, when your temperature is so low, your radiation consequently is low, unless you are very close to the fire.

And the emissivity of the smoke will depend on the optical thickness of the smoke layer, which again is the same thing as the visibility. So, if you have very thick, optically, very thick smoke, it means there was a lot of soot around and if there was a lot of soot around, you have most likely failed your visibility before you've reached this temperature and critical emissivity, to break the radiative heat flux so. It all aligns. It's, it's, I mean, for this, I'm very happy because this thing's when we discussed this as as two people with with experience in that and then we can call ourselves researchers, it is in a way obvious to us. But then again, you have a junior engineer who will face a project in the Rijadh, and the authorities will ask him for this and they will not be able to tell that it's pointless to do that, let's focus on the important things and important design choices. Now they will waste one or two days, processing meaningless data just to show the accurate number of of colorful plots to the to the AHJ. And that's that's one thing. The good isn't that

A Junior engineer is there to learn. A fire authority should be there to teach. So probably more. There are plenty of fire authority there that should put themselves like junior engineers and learn more.

Yeah, I hope that many of them will listen to this podcast and I hope this will be a very nice and easy way to improve your qualifications and learn something new. To roundup the survey. I mean, the survey had a lot more questions like how do you represent your your outcomes? What height do you measure them? These are all very interesting, but I just cannot go over the visibility factor. Like the scatter in the visibility factor question is just amazing. First, Gabrielle, can you bring us to the point what is the visibility factor?

Yeah, well, we recently published post on LinkedIn in because it was the easiest ways to reach engineers on on that. So basically the visibility estimated in in safety calculation. The one that when we mostly use is very simple. So the visibility is this visibility factor divided by the light extinction coefficient. Okay, this visibility factor has plenty of coefficients inside, that everything is the gcomes from the Bougher Lambert Beer equation or Beer-Lamber...or just Beer... so the simplification is there. So you have this special number that is this visibility factor, that is normally used the in the engineering world, is divided into reflecting sign that is equal to three and a light emiting sign for it is eight. And that's pretty much that that's it. This is far more complicated because the plenty of coefficients, but you know, as an empirical coefficient, it is just the number. And the results on the surveys quite interesting.

Yeah, we had 34% people say they use value of three, four, like the reflecting signs, and 10%, say they use value eight, which means there is just 44% of people using the values three, or eight as we would like I would personally usually do, pick one of them, there's a great group of people saying the 30-31% is they choose individual value of the C factor on based on the project requirements. And there's a good group of people like 10% of the participants who use values from within a range, because in the original paper in Japanese, from 1970, by Jin, this, this value was actually determined, as linearly correlated with the background brightness and contrast. So there's a possibility to use a range of these factors. So yeah, what surprised me is that I thought that we'll have like half of the survey participants choosing three half choosing eight, and just few percent saying they do otherwise. And I'm quite surprised. And actually, that's pretty cool. Because it means people try to learn and and really use these advanced methods as they were conceived, because it's engineering. Yeah, I hope.

I hope so. Because if not, we have a trouble.

Oh, yeah. From for that side, it's it could be kind of bland,

And to be honest. In my point of view, people are using that. Well, so and so... This article that we published together, raised the problem of understanding, because if you use three or eight, then are you able to understand the result of your of your CFD model, okay, if you use like eight, for example, for emitting sign, so you see the smoke layer, through the visibility, you might be particularly wrong. And under conservative,

I personally do my judgment based on mass concentration of smoke, I'm in the minority, that does not really use visibility as visibility, I have to translate my engineering into visibility, because of peer pressure. To explain this concept to firefighters or someone else, I usually have to use visibility in smoke. But the engineering judgment is usually based on the critical value of mass density of smoke, which which was devised actually, by by the Jin's correlation and specific mass extinction coefficient of Mulholland. And we just calculated what's the mass coresponding to the 10 meter visibility for light emitting science for C factor eight. So in a way, it's hidden in there. But I rather prefer working with mass smoke densities, because that's what the solver solves for. I mean, the physics of smoke is there, not physics of visibility. And that's why I prefer that. In past you, you've actually touched that scientifically. The sensitivity of CFD analysis to these factors. Actually, that's where we met in in Copenhagen. I think it was,

Yeah, then where everything started with you, the academic relationship. So yeah, basically, I was supervisor to Michael Alonso, that did his master thesis in here in Madrid. And the aim of the thesis was to play a bit with input on fds analysis to get an understanding of what a change in the input that would determine a change in the output. So we run like few cases, based on the Murcia atrium, we make a matrix changing the manifestation coefficient, the C factor and the soot yield. And actually, between the range of applicability, okay, so using a value of three and eighton the visibility factor, and mass extinction coefficien between, I think, is 7.6k to 8.7k (8700 m2/kg), a range between that and the visibility (soot yield) between 0.01 up to 0.2. I guess. We we came out with the visibility itself as a an outpu and and actually the results were quite, quite interesting.

I remember that that research and because I was sitting there I was watching. And when I saw that, I came to realization that you've modeled that with a computer like with the CFD code. And I figured out I have a lab that we can model that. And you know what it was a coincidence. Like few weeks before this conference, we have hosted a group of politicians, of Polish politicians in the lab. And we were asked to do some nice presentation for them what we're what we are doing at the ITB. So we figured, okay, let's let's do a fire in the ISO chamber. And we did the fire at the ISO chamber. And during this presentation, I realized that for the mass of fuel that we have used and after we turned of ventilation of the chamber, because we wanted to show them a very nice layer of smoke, we, we came to the realization that we've reached a pretty decent steady state on the smoke layer. And because our approaches to measure the extinction coefficient of the smoke layer was running, we saw it, it just flattened out. So by accident, we have turned transient problem of, of the evolution of visibility into steady state problem of having a stable layer of the gas in, in our lab, and I just liked that. But then a few weeks later, I met your talk, and it immediately caught my attention. Okay, we can actually do this in the lab and turn this problem into something researchable just let's use fuels with different soot yields. And then we can measure how big the scatter is from an experiment, not the simulation. And that's what that's what he did with with with experiments with methane, ethanol, sorry, ethanol, propanol and toluene. And yeah, the outcome was interesting. Because we have we have founded, we can call you the magic number, but that there is this, this this point, above which it barely matters, if you change the soot yield, and below it, it matters a lot. That was 0.1 gram per gram, by the way,

We defined that as a cutoff point that is not actually cutoff point, because you have this hyperbolic behavior that is obviously based on the formula of the visibility. But the interesting thing is that you have the formula in so researcher can play with that fantastic. But then, as an engineer's, we have values, okay, to put into that formula. And normally, as fire safety engineers would play in the range of zero to 0.1. That's or 0.2. Actually, that's the range when we move into, okay, so if you have, for example, pine wood, you can like have like 0.03 or 0.04, it's quite difficult to measure that soot yield. So it's not straightforward. Okay. It depends if you are over ventilated, well ventilated. You know, it's quite difficult to measure that. But anyhow, quite, it is quite clear that said, change it from 0.1 to 0.2 or 0.3. Okay, it's not changing almost anything. Okay. Because we are very, you're on the right side of hyperbolic trend. But if you're playing between zero and 0.1, well, your result can change dramatically. In probably still FDS5, and definitely in FDS4, the default reaction was propane with a soot yield of 0.01 gram per gram. Okay, so if you see where 0.01 is there, in the graph, it's a massive change if you 0.01 or 0.1.

Yeah, exactly. It's also difficult to get scientifically accurate number, as you mentioned, is difficult to measure that, but you also do not have homogeneous materials, we have very complex materials, yeah, and and mixes of materials. So you'd have to estimate this value for your whole mix. And then to add complexity on top of that, you also should include the efficiency of the combustion. So the ventilation conditions at which your fire occurs, because I mean, if we measure that on a simple device like a cone calorimeter these are like fairly small samples in the very well ventilated spaces. But in reality in fires, you'll have smoke generated in in a very complex turbulent combustion non premixed. And that will lead to a completely different efficiency of combustion. And with the worst efficiency of combustion, the more products that can create soot will be produced. So this number is probably impossible to reach a value that would describe everything from a physics point of view. And we did that backwards from the user point of view, like what would be the value that would be still reasonable, so between 0.01 and 0.2, and yet leads to the least user error, while while doing that, if you do not have a better scientifically sound value,. And this value of soot yield 0.1 gram per gram was this kind of magical value, if, if you use the larger number, there would be barely any difference in your analysis, if you use lesser value, your impact is considerable on the outcomes on the results of your study. So, we figured out, let's give engineers 0.1. It's easy, simple to remember and, and give safe results. And if someone is competent enough to, to play with soot yield and explain the choice of a different value or find a proof of a different value, they can do it regardless and probably will do that.

Yeah. Also, one of the outcome of this is, do you know what you have into that space? Do you know what is actually burning? Because if I do have, like, paper storage, okay, so I can use paper? Of course, that makes sense. Okay, so cellulosic material now there will be like, 0.05, I don't know, okay. And that's fine. But it and if I have like, plastic, or, you know, I know what I have inside, I put what I have inside, but if I don't know, if I have a mixture of everything, so be careful.

Yeah, and you know, it's also trap, because if you say you store a paper or you have a shop that sells paper, like books or something, and your assumption of a very low soot yield could be justified. And the next day someone pulls an armchair inside filled with polyurethane foam. And they will not consider is this fuel representative for the CFD analysis that were used to design the smoke control system of the building, because they absolutely have no idea about, I mean, it's safer to use this safe value, because it covers problems like that. Because even if you use 0.1, and someone puts an archer that would generate, let's say 0.17 or something for very heavy polyurethane foams, it doesn't make a huge difference for the outcome of your analysis, it would change your available save evacuation time by by two or three seconds, which means you have not done a major mistake estimating that at the first place. So yeah, I think this was I mean, I'm very proud of that study. I know, it's been cited a lot. And I've met many people who who are using our value and we're very happy with the fact that someone has published that and and gave it to them. So I think that was a great, great piece of work.

And a final thing. 0.1 gram per gram corresponds to a normal plastic. Okay, so it's not like magic number No, it correspond to something that exists. And makes sense to us.

Yeah, that I mean, this research was was pretty interesting. I wanted to share with you one, one last final item. On this itenirary. We've also recently done a very interesting research on wavelength sensitivity of visibility in smoke, and I was wondering if that ever caught your attention. Basically, when you consider the smoke from a science point of view, it's a bunch of particles hanging in the air, and these particles will have different sizes shapes, actually, you can most likely distinguish what was burning from just knowing what is the distribution of particles in your smoke. And because these particles will have different optical properties related to their size, you will end up with absorbing or scattering different bands of the visible light in a different manner on the same smoke. So we know that the blue light will be the most absorbed in the in the fire smoke, while the red one will will pass it the easiest. So I mean, we knew that that it's that there's gonna be a difference between them. In our engineering, we usually use just the general concept of white lights and, and one value of specific mass extinction coefficient for everything. We never go into wavelength sensitivity. However, we did that experimentally. We've built a device that measures light extinction coefficient, at different wavelengths. It's just a bunch of different colored lasers. That are separate from each other and they just measure the extinction simultaneously. And what we what we ended up with is a scatter of visibility in smoke from one single source of heptane from 10 meters or even less seven meters for for the blue light, while you get something like 15 meters for for the far end spectrum of the red light. So, almost a factor of two difference in in the visibility measured by that. I was wondering if you ever caught your attention in like professional analysis or is it something yet exotic for for this community?

Well, definitely yet exotic for for this community, I can tell you and I think you're doing a great job in mixing this fire science aproach to what is called spectrophotometry. That is another sort of sign that needs to be mixed to understand what is actually happening when evaluating visibility through smoke, because we are really relying on a very simple equation. That Well, according to my my study, the first equation of Bougher-Lambert-Beer was discovered by Pierre Bougher in 1729. Okay, so we are talking about something that have a lot of history. Okay. And we are still stuck on that. And I think what you're doing is a great job in getting this updated. And, and improved.

You know, the funny part is that, what I, what I'm touching in here with my research, is something that Jin has already mentioned in like, third paper in the 70s. It just never never went into mainstream, because he already recognized that the light will be consumed by the smoke in different manner for four different wavelengths. So it's something that has been there always, it's just, it was probably omitted for simplicity. I mean, I mean, this theory was never deceived for CFD, it was it was built for simple analysis, where you have a homogeneous smoke in one huge room and you want to estimate what would be the visibility in that room? Not for someone using that the 50 years later with very advanced numerical models, right?

Yeah, exactly. Exactly. You know, we are considering the light come into this piece of something has this particular particles inside and then to get out of that, okay, and we measure how much light is staying there on getting out absorbing or, okay. But we are not actually considering that. Well, what is happening into that box.

Okay, Gabrielle, I am mindful of your time. And I know, you have a very busy schedule, and I'm very thankful that you found a spot for me in it. It was great chatting with you and even greater to do the research with you. And this is not the last word we've, we've placed on visibility in smoke. I'm sure everyone enjoyed this. Long chat on on the tenability criteria and visibility in smoke. So yeah, man, thank you. Thank you so much for being here and see you around, man.

Thank you very much Wojciech. That was fantastic. I'm sure this podcast will be success. So thanks for inviting me.

Yep, that's it. I told you, it's gonna be great. I really enjoyed talking to Gabriele and it's always fun to recap our previous studies together. It's, it's been so many years we've been researching this topic together and every now and then it's fun to chat about it. I hope you enjoyed it a lot. I think Gabriell is a fantastic engineer and it's worth staying in touch with him. So if you'd like to do that Gabrielle is very active on Twitter, with his handle @Vigne_Gabriele. And his company @JVVAFIRE is also very active, and probably his most active on LinkedIn. So you want to follow him there. And if you meet him on the conference, make sure you go for a beer with him. And he also besides visibility and smoke can tell you some quite interesting rock and roll stories from a different career, which is absolutely worth it. I highly recommend doing that. Anyway, I hope you've enjoyed this episode. And that's that's it for the first week of the podcast. I'm more than happy to finally share this content with you. And there's gonna be so much more coming in in the next weeks. So yeah, look up to the north, every Wednesday morning, because there will be new podcast episode waiting for you starting next week with Dr. Matthew Boehner, on facades, and then, so many interesting guests and topics. I'm sure you will enjoy them a lot. Make sure you subscribe to the podcast on your favorite podcast app, make sure you sign up to the newsletter. And as always, I'll drop all the links to the papers and interesting things that were mentioned in the episode in the show notes. So make sure you check them out. And yeah, go on with your research on tenability criterion visibility in smoke. Oh, and look forward to the paper from the survey. I hope we can manage writing it before, like September or something. I think that's that's the goal. So I hope we we meet that and once the paper is ready, and the data is scientifically worked out. I'm sure I'm gonna bring more people into the podcast. I'm gonna bring Lucas and Wolfram and Gabriele once again, and t en will give a proper scienti ic discussion on the results of he survey. For now, thank you v ry much for being here. An I really appreciate you and ye h, see you around. Thank you v ry mu