249 - PBD of a large car park with EVs (Case study) with Jonathan Hodges, Mark McKinnon and Christian Rippe

Singapore Case Study Kickoff

Christian Rippe 0:00

Hello everybody. Welcome to the Fire Science Show. I've recently returned from Singapore, where I have attended the SFPE Performance Based Design Conference. one of my favorite conferences that happen every two years, I believe. And, one significant aspect of this conference is that. They come up with case studies, with some buildings to be designed and invite teams from all over the world. Actually, it's competition. You can, you can, uh, set a proposal and you may be chosen. So there are teams, uh, from all over the world trying to design the same buildings, showcasing solutions that would perhaps be specific to that country, that region, that, legal framework or a company. And, uh, in this time we had two big case studies at the conference, a car park, and a historical building. As you've seen from the title, this episode is on the car park. Uh. And out of the carpark presentations, one really stood out to me and that was, uh, called Team FSRI. And I have the representatives here with me. Of course. Uh, that was Johnathan Hodges from UL Research Institutes. Mark McKinnon also from ULRI and, uh, Christian Rippe from Jensen Hughes, And these folks have performed very sophisticated, structural analysis for the car park going very probabilistic, very deep on analyzing whether this, uh, concrete structure can collapse or not. They've done tremendous work on the assumptions for their analysis and tremendous work on. Translating those assumptions into scenarios, into simulations, and then into structural analysis that included, uh, for polling. That's, that's actually, it was quite impressive to be honest. And, of course in this podcast episode we will talk about how they got there. What was their assumptions? How did they choose what they were doing? What, what were the steps of their analysis really? And, uh, I recognize it's very difficult to kind of follow if you don't see the building layouts, if you don't see the case study on its own. We are, uh, we're introducing it shortly in the podcast, but really the focus in here is on methodology and framework, not on the particular project that we're dis discussing. So please treat it not as a demonstration of how to design a single car park that was made up for the case study, but rather. Demonstration of how one could work. Designing a car park that they find. I think this is the correct way to look on those case studies. As you'll see in the show notes, there's a ton of resources available up there. There's perhaps a lack of one of the more important ones. The case study report itself, as I said, I've just came back from uh, Singapore. It's very fresh. We are not sure yet if the case study can be or cannot be shared, uh, due to it being a part of the conference. But regardless, I still think, uh, you'll get a lot from what is inside this one hour recording and what's inside the shared material in the show notes. If this changes at one time and I have access to the case study file that I can share with you, I will be sure to make, let you know and it will be. In the show notes as well with all the other resources. Um, bear in mind this was recorded live at the conference in a crazy pace I think, uh, also the atmosphere during the recording was, uh, exceptionally good, which I hope translates into good content. let's be the intro and I really hope you enjoy this.

The Garage And EV Upgrade

Wojciech Wegrzynski 4:01

The Fire Science Show podcast is brought to you in partnership with OFR Consultants. OFR is the UK's leading independent multi-award winning fire engineering consultancy with a reputation for delivering innovative safety driven solutions. we've been on this journey together for three years so far, and here it begins the fourth year of collaboration between the Fire Science Show and the OFR. So far, we've brought you more than 150 episodes, Which translate into nearly 150 hours of educational content available, free, accessible, all over the planet without any paywalls advertisement or hidden agendas. This makes me very proud and I am super thankful to OFR for this long lasting partnership. I'm extremely happy that we've just started the year four, and I hope there will be many years after that to come So big thanks, OFR for your support to the Fire Science show and the support to the fire safety community at large that we can deliver together. And for you, the listener, if you would like to learn more or perhaps even become a part of OR, they always have opportunities awaiting. Check their website@orconsultants.com And now let's head back to the episode. Hello everybody. I am here today in, uh, hot and moist Singapore with an interesting, uh, US-based team, uh, who's done their case study presentation just a few hours ago at the SFP Conference. So here with me, Jonathan Hodges from, uh, UL Research Institute. Hey, Jonathan.

Jonathan Hodges 5:35

Hey, Woj. Good to talk to you. Uh,

Wojciech Wegrzynski 5:37

thank, thanks for taking a very rapid invite. this is like hot still. The laptop has not cooled yet, so thank thanks for that. Uh, together with Jonathan Mark McKinnon from UL Research Institutes. Hey, mark.

Mark McKinnon 5:49

Hey Woj. Thanks for having us.

Wojciech Wegrzynski 5:50

Yeah. Thanks. Thanks for joining us and, uh, Christian Rippe from Jensen Hughes.

Christian Rippe 5:55

Okay, well check. Thanks for the, uh, invite here.

Wojciech Wegrzynski 5:57

Okay guys, so, you've done the, the carpark case study. So SFP for those who, who don't, are not attending the performance based design conferences of SFP, there's a tradition that SFP drops a case studies number of case studies for people to solve around the world. And then we gather here in this beautiful big room and we share how we have approached our case studies this year. There were two, uh, car park and a historical building. You have chosen the car park and you've presented quite an interesting analysis regarding the, the car park. perhaps first we should introduce the listeners to what the case study was like. What, what kind of car park are we talking in here? They, they complicated it as, as they, they usually like to do that. Okay.

Jonathan Hodges 6:40

Yeah. Uh, we can give you, uh, a quick overview of the, the main features for it. Uh, so, uh, big picture, it's, uh, seven stories of parking garage, and then 23 stories of a highrise tower on top of it. Uh, so for the parking garage, there were, I think it was three stories underground, uh, and then four above ground. Uh, the underground had an old sprinkler system with an RTI around 80 or above. They didn't really know. And then the above ground was uns sprinklered. Uh, and then the 23 stories above, there were three stories of, uh, medical or dental. Uh, and then there was mixed use above that with 10 stories of office space and then 10 stories of residential. Uh, we had estimated the total occupancy of the tower to be around 3000 based on IBC, uh, occupancy loads. and, uh, what the owner would like to do in the case study is implement EV chargers in 50% of the 600 spaces in the garage. And then they also wanted to add Uh, stacker, like an automated stacker, exterior to the structure, but adjacent to it on one of the walls. and so what we were looking at with this was really, you know, if you were not really. Considering the difference in vehicles be before and after the upgrades because you've got this mixed use. So, you know, people could be driving their electric vehicles, you know, they, they're, they're big, you know, long range, vehicles and parking them to go to the doctor or to go to the adjacent mall. And there weren't really any operational controls to affect, uh, what types of vehicles would be in the facility. and so the, the question that we had is, well, if you're gonna look at this, the max consequence event is probably the same regardless of whether you have chargers or not. And so we wanted to look at the potential increase in frequency, uh, due to the chargers as another initiating mode, as well as, possibly the potential increase in the likelihood of the fire being more severe should it occur due to having more modern vehicles. And that shifts in population.

Wojciech Wegrzynski 8:36

Yeah. I mean, e even if you try to assume a different population of vehicles, whether they're charged or not, absolutely no means to control that. In Poland, we are very unsuccessful with our no LPG allowed in the car park assumption. And the marking on the wall of the car park doesn't really solve the problem. So I I agree. I I would do the same. Yeah.

Jonathan Hodges 8:57

Yeah. In general, I'm just not a fan of operational controls unless you really have a, a rigorous system if you're enforcement.

Wojciech Wegrzynski 9:03

Enforcement, yeah.

Jonathan Hodges 9:03

If you're buying security gates and you're actually, you know, or you have assigned spots and you're gonna be towing people away if you're not in 'em, you know, there's, there are things you can do operationally to enforce those rules, but, that's not the typical case in the us.

Christian Rippe 9:16

I also think about over the lifetime of a building. You know, 20 years from now, no one's gonna, you know, no one's gonna try to make changes and no one's gonna think about a performance based strategy that was used 20 years ago to design the building or design certain aspects of the building. Right. It's just not gonna happen. So like, that's where, like these operational controls, right? To think that, oh, like someone's gonna not allow certain buil vehicles in here. Well, that might work for a few years, but five years down the road, someone's gonna go, why are we doing that? I don't know. Mm-hmm. So now they're gonna allow it, right? And no one's gonna look back at a PVD analysis of the building, right?

Mark McKinnon 9:49

That's essentially the case here, right? So we're, the case study, the prompt was, uh, that we have this old building, it was designed with certain assumptions in mind, and we get it and we're asked, you know, how many of these EV chargers can you introduce, uh, without any fire protection upgrades? So it's an interesting case study.

Wojciech Wegrzynski 10:07

Yeah. I, I, I would say, you know, for me, if I got that as a real project What client asks you for is how many chargers you can put in. And I can immediately tell them like, whatever you like. in my first opinion, doesn't really trigger a point that is gonna be a big difference because I can't control for vehicles anyway. But at the same time, it's an opportunity. It it, it gives you a reason to revisit the far strategy for a building and perhaps do some good things for the building because the cars have changed. Right. It's, it's like the, the, the fact that you are now implementing charger is not as big contributor to the hazard as the fact that 20 years ago you had a completely different vehicles in it. Right.

Why Old Assumptions Fail

Jonathan Hodges 10:49

Yeah, that, I think that's a great point since most of the previous guidance was really, uh, there, there's even some, uh, reports that I think they might have been from NFBA that say that, um, all the guidance is based on this assumption that you're not gonna get multi-vehicle spread. Like it's gonna be limited to one vehicle. And there was data in the past to back that up. Um, but now with the, the recent studies by NFPA as well as ULRI, it just seems more likely that you're going to have multi-vehicle spread. Yeah. And so that's why we wanted to, to revisit that and kind of challenge those assumptions that the structure would be safe, because it would be a single vehicle fire.

Christian Rippe 11:22

And to be real, over the last 10 years we've had, you know, a half a dozen major car park tires. Yeah, absolutely. We're like half the, half the building's burned down. Right. You know, so it's not like that, there's not evidence out there in the real world that this doesn't happen. Right.

Jonathan Hodges 11:35

And actually while we were here at the conference, there was a very large, parking garage fire in, in China. Okay. Um, where, uh, I think it spread over the whole, uh, level in Shenzhen. Right. Um, at the BYD factory. and it, uh, I don't know if there was any structural collapse. I'm sure they, they're still gonna do investigations on it, but, uh, you see this huge smoke plume, it's like three, four times the, the height of the, the parking structure.

Wojciech Wegrzynski 11:59

Absolutely. So, uh, we have, uh, a PBD component in here for sure. Because you are applying PBD, uh, methods to design the car park. We have the structural component. Uh, so, so understanding what is the ultimate failure mode of, of the car again, and how likely are we to get it and, uh, to start with it. We also have the design fire component, which perhaps, to, uh, listeners, while they may not be building a seven story, uh. Car park underneath the dentist's office and 30 floors of, of, of residential. They perhaps will have to play with the car park once in, in their life or twice. And, uh, design fire is, uh, when you start working with car parks, you, you, you, you hit the wall. How do I design the fire? So, uh, Jonathan having you here, let's talk design fires. Um, first. So, yeah. May, maybe you can wrap up the philosophy. I, I know you have published the papers and listeners there are in the show notes, of course. Uh, on, on the design of car park design fires, design of the car park, design fires. what's your way of thinking about this problem and how do you simplify this to a design fire?

Jonathan Hodges 13:04

Yeah. Uh, so what we really wanted to understand is based on the, the realistic vehicles on the market, what's, the, the fire size you would see if the vehicle burned at its peak potential. So that's kinda what we had started with and was in our paper from 2024,

Wojciech Wegrzynski 13:18

like maximum hit release rate you'll get from, yeah.

Jonathan Hodges 13:21

So if you, if you put a, a giant, you know, heptane gas burner below this vehicle and just get the whole thing going, what's, what's the peak heat release rate that you're gonna see? so, so that's what we designed to, uh, before in our previous paper. But then when we started to look at the risk based. Approach here, we really wanted to understand is that a likely, is a vehicle likely to burn at its peak potential or not? Mm-hmm. Because that's where you need to start getting into the percentiles rather than the deterministic analysis.

Wojciech Wegrzynski 13:47

So rather than one maximum value you have ever seen in any of the experiments carried out, like what's your 50th percentile, 90th percentile, something like that. Okay. Yeah,

Jonathan Hodges 13:58

exactly. And when you're doing a risk approach, you need to do some sort of bending. So you can say, you could do, you know how many to do, you could do a 25th 50th, 75th 90th, 98, all the way up to 99 point, however many nines you want. Um, but you, you need to decide what those limits are that you're gonna do and what is okay to not mitigate, you know, what's your residual risk? Yeah.

Wojciech Wegrzynski 14:19

Okay. Well, well, I can imagine this is not that difficult for the peak value. Uh, you also introduce a trends in development of that fire in the car park and that, that's the challenging part for me. For example, I, I gave like, you know, I gave up on that at some point because it, it's, it's so predict to me to, to have, maybe not even unpredictable for, for me, because every fire is different. It's impossible to have a generic fire, you know, I, I just don't find it a possibility to have a generic design fire that represents a fire, because every fire will be so vastly different. So, but you need something for design if you wanna have a transient design. So, so what, what's your best, uh, consensus, uh, solution for that?

Jonathan Hodges 15:04

Yeah. So what we did in this analysis is we really had a few factors that we considered. You had the, the size of the vehicle. Yep. Which we know is important. 'cause that's gonna be. driving the overall fuel load and the peak burning behavior. and then you have how the vehicle fires initiated. Um, and you had in your systematic review, uh, you took data from the literature and characterized it by where the fire was started. And so we use that to understand, uh, what percentage of that peak potential do vehicles actually burn at based on where it started.

Christian Rippe 15:35

Mm-hmm.

Jonathan Hodges 15:35

And so we use that to establish these different classes of vehicles. So we basically have a class that is vehicles that burn at their peak potential. Then vehicles that are limited to the engine compartment, vehicles that are limited to the passenger and vehicles that are related just kind of in the body. And so we, we tuned the, the peak e release rate and total energy from those based on statistical analysis of the data that you had collected. And then we pulled data from NFPA, um, in the, previous data sets to characterize based on investigations, you know, where is the. Where do you see these fires contained? Mm-hmm. Um, in the data. And so what we assume there is, if they're able to identify that the fire started in the passenger compartment, then we said, okay, we'll use the passenger compartment bias for that.

Wojciech Wegrzynski 16:21

Tell me more about that. That's when we're talking about the data I've collected, I, I know a lot about it, but, uh, now you mentioned, uh, you, you took some data to, to see what the fires in real world looked like. Mm-hmm. Okay. Tell me more about that. That's interesting.

Jonathan Hodges 16:34

Yeah. So NFPA has, um, a report, I think it's, uh, McGee's, the author on it, vehicles and Fire. Uh, I think 2024 was the most recent edition of that. and within that there's, there's lots of tables, um, that say like, uh, location that the fire was started, um, for the passenger vehicles. And uh, there's a lot of groups there, but there's, the engine compartment is one passenger compartments one. and then they'll have one like the whole section that's like unknown. And so

Wojciech Wegrzynski 17:01

is that from firefighters report post-fire or investigation insurance?

Jonathan Hodges 17:06

Yeah, it's from the, the N-F-I-R-S data.

Wojciech Wegrzynski 17:09

Okay.

Jonathan Hodges 17:09

Infers data, from incident reports.

Wojciech Wegrzynski 17:12

Okay. But it is post-incident real fire data. Okay. That's nice.

Jonathan Hodges 17:15

And we, one, one of the things we assumed is if they couldn't identify where the fire started, that the whole vehicle was involved.

Wojciech Wegrzynski 17:20

Okay. And

Building Better Car Park Design Fires

Jonathan Hodges 17:21

so that was where we got the full vehicle involvement percentage. Uh, now, uh, with UL we do have the, the nearest data, uh, that the collection now and we're adding more granularity. Mm-hmm. And so hopefully as more data comes in over the next few years, we'll be able to better understand where the fires are starting. 'cause we're actually collecting, um, data that's gonna be. More designed around what we need.

Wojciech Wegrzynski 17:44

Oh man. That, that, that's really awesome. And some baseline statistics like fleet distribution, science distributions. Mm-hmm. Did you took the, like, statistical data of us?

Jonathan Hodges 17:54

Yeah. So what we collected some data on sales of vehicles from, uh, both the US market and globally. And we actually just published a paper on that, I think it came out a, a few weeks ago now. And we'll be presenting on it at IFSS.

Wojciech Wegrzynski 18:08

It's in the show notes.

Jonathan Hodges 18:09

Huh?

Wojciech Wegrzynski 18:10

It's in the show notes.

Jonathan Hodges 18:11

There we go. And, uh, so that one again, that, that model, um, is based purely on, uh, the determinist model we had before, plus the sales data. Um, and then we did start talking about multi-vehicle fire spread, but it was all presuming that a vehicle burns at its peak potential. Uh, but you know, you know, I assessed their few times are long. So that was submitted, you know, what, six, nine months ago? I don't remember at this point. Um, and so. We started the case study after that. Mm-hmm. And so what we wanted to look at the initiating mode as well. And so that was a new feature that we added since that paper. And so we'll have to do it. We're planning on doing follow up paper that, uh, includes some of the additional features in it.

Wojciech Wegrzynski 18:51

Fire Essential brings you knowledge as it happens. that's, that, that, that's great. so you use this model you've described in the previous papers, uh, for each initiating vehicle, for every vehicle, like every vehicle has its own hit release rate curve. Only the first one gets. How do you translate this into the, how do you use it in the project really?

Jonathan Hodges 19:12

Yeah. Uh, so what we did is we, uh, generated a hundred thousand sets of, um, initiating vehicles, and used that to get heat release rate profiles. And then what we did is we said, okay, based on that a hundred thousand, we're now gonna re-sample, uh, for how they were initiated mm-hmm. From the NFPA data that we collected. And so we kind of tied those together, and then we biased the heat release rate down using the statistical analysis that we presented in the case study. So we, we always start with the peak potential and then bias it down based on that initiating load. And then we use that information to predict the heat flux to an adjacent park vehicle, and we randomly sampled from that size distribution, what the size of that vehicle is. Mm-hmm. So then if that vehicle ignites based on our ignition criteria, then we're, we're off to the races. We keep going down the line continuing to predict the heat release rate.

Wojciech Wegrzynski 20:02

One question that always comes to my mind when I see such complicated case studies performed here in SFP is, do people really work like that? Like it, it is unlikely that you do that in, in the, well, maybe it's likely that you do it on a real world project, but I, I, I find it very. Advanced, like if you wanted to incorporate that in the, like, normal PBD framework, like run of the mill car park. Do, do you see any modifications to the process after you went through it? Like,

Jonathan Hodges 20:30

There's definitely areas that we need to do more work, um, to finalize. But my goal would be that we, we we're doing the legwork, we're doing the research, and then we're publishing it. And then when someone's going to design, they can use the curves that we've, um, developed. They don't necessarily need to reinvent the wheel in their exact project. but, you know, someone has to take that first step. And thankfully, uh, with ul, we have the ability to, spend the time to do this investigation.

Mark McKinnon 20:56

uh, you know, the impression that I get, I, I spend a little bit of time in consulting. The impression that I get is. S uh, no owner would pay for this level of analysis, but at ULRI, we have a really diverse stakeholder, a set of stakeholder groups. Mm-hmm. And so, one of the groups that we're really trying to, start to work with and work for are fire protection engineers. And so if we can develop guidance and, and tools and resources for fire protection engineers, you know, to, to make their jobs easier, maybe they can, uh, do this level of analysis, uh, much more cost effective. Yeah. Someone has to do it first. Yeah.

Wojciech Wegrzynski 21:31

So someone has to do it first.

Christian Rippe 21:33

Yeah. And you know, this the first case study I worked on, and I've actually found it kind of fun because there's not an actual life safety implication in the analysis we're doing. Right. Okay. You know, it's because it's all, it's make believe. Right. You know? Mm-hmm. And so you can kind of push that envelope further than you would feel then I would feel comfortable. Right. And, you know, definitely the structural part that, that we did like. There's holes that, like if we were to do this in an actual fire, you know, an actual parking structure, that they would have to be filled a hundred percent right. But we can kind of say, Hey, look, you know, we, we got the ball rolling. And, and I think that this is kind of where these kinds of analyses could end up in the next, three, four, or five years, right. That, that we could be doing this kind of stuff. But that's right. You someone's gotta do it first. And these case studies present a really great opportunity to kind of show, hey, here's where we could go with this. And now maybe Right. Other people right. Can, can take these ideas and run with them and, and kind of a real world application.

Wojciech Wegrzynski 22:26

Yeah. I I, I'm, I'm running this question purposefully because, the design curve or design fired is actually something that can be taken out of this analysis and implemented, but not in a fully performance based risk analysis. You know, like. Uh, in my world, uh, how people design car parks in Poland, they would just take one curve that exists, whatever they, whatever is their preferred curve. and they apply it in one, two locations. I do as well, when I design a car park, I don't, I don't run qa. I, I just, you know, run a demonstration that my sport control works and I just take one design, fire and apply there, and when I apply it, um, that's it. and therefore, my whole risk analysis is at the design fire level really? Because if you take a hundred thousand courses of fire and then sample them by statistical distributions, fully understanding the implications of what you're doing, I think that's awesome. And, and, and, and I think this approach is, is really awesome. But I can also see someone, Misusing this type of, of tool of the design fire, you know, in their single point analysis because Oh yeah. Someone, you know, did that, for their fire. I just took this curve because I like it, and, and here it is.

Christian Rippe 23:39

Yeah. And there's, there's been some discussion here at the conference of, you know, I, in the design fire development, that's really the heart of the engineering that we do, right? Mm-hmm. You know, that's, that's the judgment and it's such a hard thing to teach. You know, it's something that experience is really like the only way to, to get it right. And, we can teach people to run models very easily, right? You can give 'em all the inputs, they can build an FDS model and, and they can go, right? But, and so it's great for new, you know, people come, coming out of university, stuff like that. But to tell someone, Hey, can you figure out how big of a fire to use for a car in a car park? You know, it's people who have years of experience and they're still like, oh, there's so much unknown. Right? Like, it's such a hard question.

Wojciech Wegrzynski 24:19

And we're back to the, is there a generic fire? Because I, I would argue that the same goes to just using experimental data. You can as well just abuse experimental data. You choose a calorimeter output that you liked, you put it directly in your, in your simulation, and it's, it's there. Mark, you, you look like someone who could comment on, on using, uh, experimental data.

Mark McKinnon 24:40

I think what's, what's really important with, experiments and, determining which experimental data to use is, that the researchers themselves have to kind of define their limitations. And, you know, I guess, uh, that's something that you don't often see. In reports and, and peer-reviewed journals is, uh, people defining exactly how things should be used. And so I think one thing that we, have tried to do here and we try to do in our work, at UL is, you know, provide guidance for how these things are gonna be used. and I guess in this case study, you know, this is, it is a case study. It's a, it's a way to show, our potential stakeholders how you should use the data. Mm-hmm.

Jonathan Hodges 25:18

And if you limit, you know, if you're using an experimental curve, you can always find an experimental curve that'll match your Exactly. That'll, that'll look better. and so that's the value of really doing the engineering analysis, like Christian was mentioning, to look at the statistics on the types of vehicles and the data that we have to come up with these distributions. And then I can recommend to you what's a, percentile to use, or we can codify what percentile needs to be used. And then we've got a curve. It's a single curve. And now we've removed a lot of that, um, subject matter judgment, from it. Mm-hmm.

Wojciech Wegrzynski 25:50

so some parameters of those curves also, uh, the, the growth rate

Jonathan Hodges 25:55

mm-hmm.

Wojciech Wegrzynski 25:55

That's also part of the istic assessment, right?

Jonathan Hodges 25:58

That's correct. And, the growth rate, we, we did analysis based on the data that you had collected in your systematic review. And, uh, basically we looked at the, um, using some subject matter judgment. We looked at the, the first peak, and then it kind of extrapolated it down until it crossed the zero, uh, neglecting the incipient phase. Um,

Wojciech Wegrzynski 26:17

how, how did you decide when it ended the incipient phase?

Jonathan Hodges 26:20

I think we did 10% of the peak, or maybe it was one megawatt. Okay. It, it some criteria,

Wojciech Wegrzynski 26:25

just a rule for a cutoff.

Jonathan Hodges 26:27

Yeah. I

Mark McKinnon 26:27

think it was fairly subjective. I think, you know, typically 10, around 10%.

Wojciech Wegrzynski 26:31

Okay.

Jonathan Hodges 26:31

But In most of the data, it's pretty obvious when you look at it. The ones where it's not, were really like the body ones where it kind of you body

Wojciech Wegrzynski 26:39

has in ignition in on the body of the car.

Modeling Multi-Vehicle Fire Spread

Jonathan Hodges 26:41

Yeah. Okay. Those ones, it, it grew fairly slowly, maybe almost linear up to a megawatt after like 20, it's a mud 30 minutes. And so for those and for those ones, again, those weren't the ones we were as worried about. It was more because those are not usually gonna be the founding scenarios, but in the risk analysis, they are important. Yeah. but we did that analysis and basically came up with if the whole vehicle's involved and it's really growing at a fast or an ultra fast growth rate.

Mark McKinnon 27:07

Mm-hmm.

Jonathan Hodges 27:08

And then if it's in the passenger or the engine compartment, then the medium fast that we used in the previous was pretty reasonable. That's the 2 25 second critical time.

Wojciech Wegrzynski 27:18

That's really awesome because there used to be a discussion that I know that some people would be preaching, oh yeah. Medium curve is what you should use for a car. And there would be others who would be going like fast, ultra fast. I, I think I would be in the camp of, of fast growing fires. 'cause I'm, if I, my purpose is to do smoke control and if there's a medium fire growth in my car park, I don't need smoke control at all. I have not proven anything. Like I, I don't feel I've tested the system. It's, again, subject,

Jonathan Hodges 27:46

again, if you're only using one curve, you need to use a curve that is at that upper level of design.

Wojciech Wegrzynski 27:52

Yeah, yeah. Yeah. I, I think I agree with that. I agree with that. Totally. And, uh, what, what about the experiments that go weird? Like have two peaks or that have very late peak? How did you account those?

Jonathan Hodges 28:02

Yeah, so for those, the, the growth rate, we, based on the first beak. Okay. Um, but then in terms of the, the. the peak, uh, heat release rate and the total energy. Um, we did our statistical analysis there to basically say how much lower is it?

Wojciech Wegrzynski 28:16

Yeah.

Jonathan Hodges 28:17

And so if it's in the engine compartment or the passenger compartment and it's limited to that, it's about a 50% of the full potential for the peak. And so when we're using the exponential form for the heat release rate, did you get about the same total energy? It extends that duration out and flattens the curve a bit.

Wojciech Wegrzynski 28:34

Okay. How about the chargers? Because it was like, the point of the study was that 50% of the spots will have now a charger. So this charge charger changed Anything from your perspective at this point?

Jonathan Hodges 28:45

So that's a, that's a really complicated question, uh, where there's not great data. You know, some of the, the really high speed chargers, now you're, you're charging from another battery, right? Mm-hmm. And so it's possible that that battery could be part of the system, and so you could be increasing the fuel load within the car park even. Um, the assumption that we made in the case study, however, was that they weren't contributing to the fuel load. So either you have a separate energy storage system if you've got batteries, or it's just coming off of, um, the, the power. There's no fixed, um, energy with the charger. so within that framework, the way the chargers are contributing is as a new initiating mode for the electric vehicle fires. And there's not great data on that. the data that we did use, uh, was basically that 18% of EV fires, occurred either while the vehicle was on the charger or within 60 minutes of being on the charger. But that doesn't tell you that the charger started the fire, it just tells you that it was charging. But it's also, typically when you have an ev, people are gonna park it and charge it. but that doesn't, doesn't give you that answer, but it's at least something that we can use as a starting point.

Mark McKinnon 29:52

Yeah, I think, uh, one of the major takeaways that we had as a team was, you know, this, this kind of acted like a gap analysis to see, you know, what's available for practitioners out there. And this is one major gap we just don't have the data to support some of these, uh, analyses. And I think the assumption that we made is probably a pretty good one, but, in fire. Science, fire research. We're fairly reactive and you know, this is a, a fairly new technology, so I think in the coming years, this is something that's gonna get refined really well.

Wojciech Wegrzynski 30:25

Okay. Let, let, let's move. I mean, it's already awesome, but we haven't got to the good part yet. So Yeah, let, let's perhaps move towards the, uh, the structural analysis and, and the general QA process that you went through. So we just covered extensively how you got into the design fire, but now tell me what you intended to do with that, with those, once you, once you figured them out,

Jonathan Hodges 30:47

Uh, so I think I mentioned this earlier in the podcast, but whenever you're doing a risk analysis, you have to decide on how to bend things. Um, because if we're, if we have a hundred thousand, scenarios for the different heat release rates, that's great, but we're not gonna run cf d simulations for all of those. so in terms of the bins. Maybe, maybe with AI future episode.

Wojciech Wegrzynski 31:07

Yeah, I, I don't want to do that with ai. Jesus, you can just ask Chad, but will my car park collapse as well at this point?

Jonathan Hodges 31:15

You know, I didn't actually ask it that, but

Wojciech Wegrzynski 31:18

there's

Jonathan Hodges 31:18

All right. Uh, anyway, so we've been, uh, the bins we came up with were, you know, sprinklers on sprinkler's not, um, and then three different bins for the heat release rate for an individual vehicle. So 50th percentile, 90th, or 98th percentile.

Wojciech Wegrzynski 31:32

There's been three curves of heat release rate that you would put into analysis.

Jonathan Hodges 31:35

That's correct.

Wojciech Wegrzynski 31:36

Okay.

Jonathan Hodges 31:36

And then, we had single vehicle or multi-vehicle

Wojciech Wegrzynski 31:40

uhhuh,

Jonathan Hodges 31:40

and then we had the mitigation result, which was either free burning, sprinkler control, or suppressed.

Wojciech Wegrzynski 31:46

Mm-hmm.

Jonathan Hodges 31:47

And so each of those had different heat release rate curves that we were considering. Um, and then each of these had different probabilities that we tune based on engineering data and some of the modeling results.

Wojciech Wegrzynski 31:57

So, so it's like fuel QA,

Jonathan Hodges 31:59

That's correct. It's a full QRA and there were some things that we didn't vary. Okay. There were some things that we just assumed. So, uh, again, you have to, when you're doing a risk analysis, you have to draw some bounds on it or else you'll never be able to get an answer.

Mark McKinnon 32:10

Mm-hmm.

Jonathan Hodges 32:11

but pretty early on when we started doing the analysis, uh, we realized that in terms of occupant, tenability, the R sets are generally pretty low. And so as long as you have that automated detection and notification, people are able to get out.

Wojciech Wegrzynski 32:25

Okay. So, uh, so you, you have your scenarios, you said it's full probabilistic. So how, how do you, put frequencies or probabilities on those scenarios?

Jonathan Hodges 32:35

Yeah. Uh, so for our base frequency of just the events per year, we pull data from, the same NFPA report on the number of vehicle fires per year. And so it came up to about six 50, a little bit of variability from year to year. and that was a,

Wojciech Wegrzynski 32:49

is that fires in car parks or fires of vehicles in total?

Jonathan Hodges 32:53

That's, uh, fires of vehicles. In parking areas.

Wojciech Wegrzynski 32:57

In parking areas, okay.

Jonathan Hodges 32:58

Is the way it's described. And so that could be. A, uh, car park or it could be just in a, an in an open, you know, asphalt, uh, area, uh, with no building. the data collection doesn't break that down, which is one of those things that we roping to improve upon with narrows where we're gonna do better data collection.

Mark McKinnon 33:17

Mm-hmm.

Jonathan Hodges 33:18

so once we had the number of fires, uh, per year, we needed to estimate or in parking areas, we needed to estimate the number of parking garages. and so there was, uh, some work done, by, few researchers, uh, out of New Zealand. I So once we knew how many fires or we had an estimate of fires in parking areas, we then needed to know the denominator Yeah. For the number of parking structures. And so there was an article, it'll be in the show notes, I'm sure.

Wojciech Wegrzynski 33:41

Yes,

Jonathan Hodges 33:41

it is. Where they estimated the number of parking structures in the United States. Uh, so we used that number. It ended up being about 122,000. there were a few other sources we tried to look at. We looked at like registered parking businesses and things like that. but you know, the numbers you get are pretty different depending on how you try and characterize it.

Wojciech Wegrzynski 34:01

Yeah, I can imagine.

Jonathan Hodges 34:02

Um, that gets us our base frequency, and that was 0.005 fires per year, per garage. So that'd be like one fire every 200 years in a garage

Wojciech Wegrzynski 34:12

that's actually low. That's actually, I would expect the number, like one in 30, one in 50 maybe. But yeah. Fine. And now per scenario.

Jonathan Hodges 34:19

Yeah. so for the scenarios, uh, kinda the first break point we had was whether sprinklers, um, were active or not, uh, for the existing structure in the case study, where there weren't sprinklers, we just assumed there weren't sprinklers. Okay.

Wojciech Wegrzynski 34:31

So weren't on no active.

Jonathan Hodges 34:32

Yeah. And so there

Wojciech Wegrzynski 34:33

very robust.

Jonathan Hodges 34:34

um, in the case with the upgrade where we're recommending sprinklers throughout the structure, uh, we use data from NFPA, uh, where sprinklers are observed to operate in 92% of scenarios. Okay. Where the fire is considered large enough to activate them. and then we have another break point later on in the entry that is sprinklers are, you know, how effective are they when they do operate at controlling the fire? And so their FBA lists, the sprinklers are being 97% effective at controlling fire when they operate.

Wojciech Wegrzynski 35:02

So the sprinklers that are present, they operate at 97% of cases they were successful in controlling the size of, okay.

Jonathan Hodges 35:08

Yeah. And so we took that to be mean that 3% of fires when sprinklers were active, um, still burn as free burn.

Wojciech Wegrzynski 35:14

and, and the, does the free burn mean they're, they're, they're gonna transition into multi-vehicle fire or?

Jonathan Hodges 35:20

Yes.

Wojciech Wegrzynski 35:21

Okay.

Jonathan Hodges 35:22

So, we could always try and say maybe you get some benefit from the sprinklers and reducing the likelihood of it transitioning to multi-vehicle so you get some cooling. Uh, but that wasn't something we considered as part of the case study. We just assumed if it was uncontrolled then, um, the sprinklers were providing no benefit. but then the remaining 97%, uh, we had to decide how to distribute that between fire suppression versus fire, fire control, where suppression means that heat release rates is going all the way to zero and fire control is, um, being limited either at the time of activation or, um, at some, you know, minimum level that, uh, for a shielded fire or something.

Wojciech Wegrzynski 36:02

Okay. And then to translate those results into, input for further, thought process, good old FDS simulations. How did you did?

Jonathan Hodges 36:10

Yep. Uh, so we, we modeled the, the scenarios. Uh, so

Wojciech Wegrzynski 36:14

36 of them,

Jonathan Hodges 36:15

yeah, we modeled 36 scenarios. That's

Wojciech Wegrzynski 36:17

actually impressive

Binning Scenarios For Risk Analysis

Jonathan Hodges 36:18

for, for the occupant. Tenability, we did that. And then, uh, for the most part we really just looked at the free burning scenarios because with the sprinklers, we didn't really see occupant um, impacts. Mm-hmm. but we ran, um, more simulations in FDS looking at the heat transfer to the ceiling. Um, and the columns, and that's where we buried the heat release rate and a quasi steady form, to get that exposure. But particularly with these multi-vehicle fires, we were just seeing such high potential for exposure and we wanted to understand the likelihood of, uh, failing the structure. so that's when we, we brought, uh, Jensen Hughes in as a partner to, uh, help us, uh, evaluate the structural analysis

Christian Rippe 36:58

at the very last minute. I'll add, but uh, so we ended up for the structural, it was, it was I think an interesting approach. Um, and really I think what's novel that we did is with the elephant in the room for car parks, at least in the US for modern car parks, they're all concrete structures. Um, okay. You know, you get some old stuff that might be steel. but, you know, everything's concrete for car parks. Especially in the Oh, that's interesting.

Wojciech Wegrzynski 37:22

Twist because in, in Europe and UK. You would find a lot of steel structure cardboards

Christian Rippe 37:27

and in concrete, the, the elephant is sping, right? you know, we, our, our understanding of sping is really still in its infancy. You know, we're just starting to like really build the theoretical backgrounds for what's like, what is sping, like, the mechanisms of sping and things like that. But then in terms of practical application of some of those, of those mechanisms, you know, we're not there yet. and so, you know, we, we kind of gave it our stab here, right? and we, you know, got some kind of interesting approaches. There's been some, some work out in the literature in the last couple years about kind of people putting in some interesting approaches. so for us, what we ended up doing, you kind have two things, right? It's how you implement the actual mechanisms of spalling. Mm-hmm. Or in the, and the kind of theory behind that, but then how you put them into an engineering model as well. 'cause you, our engineering models aren't set up to handle spalling. And so. On the kind of more theoretical side, you know, you look at the most common mechanism spine that people are familiar with, right? Is the, what's called a thermo hydro spalling, where you got the, the moisture in the concrete and it's evaporating with the heat, and that vapor is trying to move through the concrete and it can't move very well. And so the pressure start building up and eventually get this, you know, explosive sping mechanism going. and there's some, there's some great work coming out the last couple years and the last few years about, and I'm trying to understand the theory behind that. And it's awesome for a research perspective, right? But when you go to even this, this case study, right? Things like concrete porosity and permeability and things like that, we don't measure those things in, you know, when something builds a car park, no one measures the concrete porosity or the permeability, right? Um, and so, you know, let alone a historical, you know, a historical car park and trying to analyze something like that for an upgrade. You're never gonna know. but even a new structure, you're not gonna know. And so we, one of the challenges we had was to try to kind of convert what's done for the research scale where in a lab you can go, yeah, let's, let's cast some concrete, let's do all these really cool measurements with get all these cool, these properties that we need to try to implement these theoretical models in, in some way. and the challenge we had was to try to scale that back into what could we really kind of apply this to in a real world situation. and then second to that is then you come up with a method. It's how do you put that into an actual, you know, a, a solver, right? And so in this case, we're doing thermal finite element in Abacus. and, you know, the idea of removing material that's in, in an FEA model, that's already a very advanced concept. Right? You know, if you're, if you're kind of doing any sort of element deletion in a, in an FE model, you, you're really. Kind of in the advanced user category for that model. Right.

Wojciech Wegrzynski 39:59

So, so, so in practically, I, I mean, I have an idea because I see, saw the presentation, but how, how, how did you apply that? Because as you say, it's not that you turn on a button that says includes polling and, and your model starts popping. But, uh, how did you do that?

Christian Rippe 40:16

Yeah, it was, I've actually very recently Abacus has a, a Python scripting interface that, um, not just interfacing with the input files, but you actually essentially interface with the Abacus gui. Mm-hmm. Um, and you, you know, every single command you, you, you do in the Abacus Gooey has an associated Python command that essentially is, is under the hood. And you can actually then write Python scripts and, and people write macros all the time. Right? And so we ended up kind of creating this, master python script that then edited templates for these kind of, essentially like Abacus macro scripts to build these models. And so then it would. You kind of build this, the first model of the, of the, of kind of the, the virgin material that you have to start, you would run that analysis, look at when you first have a spalling event, kind of freeze the analysis there, re remove the geometry, then from the spa, kind of rebuild the model, pick up the analysis where it left off, and you keep doing that over and over again. Um, you know, in some of these analysis, you know, we could have 10 or 15 sping events over the duration, you know? Mm-hmm. You're looking at like multi-vehicle fire, three hour durations. You have a lot of Sping events. but it was all done kind of in a, a, a procedural scripting way. Right. So I didn't have to sit there and click, you know, over and over again and measure and results. Right. But it's, it's not, it's not that straightforward process in terms of like an actual, like commercial viability. It's definitely right on the edge. Right.

Wojciech Wegrzynski 41:37

so now you have this model that allows you to, to understand how the, the transient thermal, uh, exposure changes with, with, with polling included. How, how do you go from here to answering the question whether the residents of the tower above are alive or not safe or not?

Christian Rippe 41:56

Yeah. And, and we did a more, I would say a traditional PBD idea there. We, you know, took kind of some reinforcement and prestressing steel temperatures, right? With a concrete structure.

Wojciech Wegrzynski 42:07

Mm.

Christian Rippe 42:07

You know. If you look at even like the prescriptive codes, like a as tmv one 19 test, right? You know, a pass fail on the thermal criteria for something that's a, a structural member, you know, you measure the temperatures at the reinforcing steel or the pre stressing steel and you'll have temperature requirements there. And so we, we stole those shamelessly from a, the E one 19 standards and, and it's, uh, there's definitely kind of a, a bit of a brute force and, and way, or a very simplified way to, to apply those, right? 'cause you do kind of say that you hit those criteria and your structural member quote unquote fails, which in reality that's not really the case, right? You know, they, they correspond to about 50% strength, um, for the, for the steel where you could be, you know, either losing your pre stressing compression or losing confinement in a reinforced member. And it's not that, you know, these things, you know, lose all their strength when their steel hits 50%. Right. You know, you do get a gradual reduction so.

Wojciech Wegrzynski 43:01

Hmm.

Christian Rippe 43:02

You know, but we took that kind of simplified route, which I think is more traditional in PBD. You'll find a lot of people kind of doing that. Um, using like something like an E one 19 temperature criteria, as a PBD performance objective. Right. Or, or performance criteria.

Wojciech Wegrzynski 43:15

I probably should have asked that just at the beginning of the episode, but part of the car park was open from what I remember from the brief, but there was an underground part. Were they equipped with any sort of smoke control in it?

Jonathan Hodges 43:27

Uh, so the case study did not have any smoke exhaust. Uh, there was a code compliant, uh, recert system for just mechanical, the managing co and ambient levels. Um, we could have recommended a mechanical exhaust system, and that would help mainly on the, the occupant tenability side. Uh, but for the structural, we wouldn't expect that to be a huge impact.

Wojciech Wegrzynski 43:49

And did you see any differences in the consequences of the fire between the underground part an open part that was above ground or,

Jonathan Hodges 43:58

uh, so we didn't really focus on that.

Wojciech Wegrzynski 44:00

Okay.

Jonathan Hodges 44:00

you know, I think we, we only modeled the one scenario down at the underground, uh, from an o tenability that was gonna be the worst scenario. and on the structural side, we, uh, did the, the smaller domain kinda representative model.

Wojciech Wegrzynski 44:12

Okay. Underst un understood. Understood. so, so what kind of conclusions did, did, did you go, is it safe to do that for stories on sprinklers?

Jonathan Hodges 44:21

Uh, in general, uh, sprinklers do a great job at controlling the fire and preventing fire spread. Uh, we found that for this particular case, it was very important. Um, we saw, I think it was, uh, something like a 90% reduction in the residual risk by adding sprinklers, uh, throughout the structure. And, uh, there was still some residual risk, but, it was, uh, very impactful.

Christian Rippe 44:42

Mm-hmm.

Mark McKinnon 44:43

Yeah, I think, uh, you know, you can't argue with the data. Um, it's obvious that sprinklers are, are really important. I think one of the things that, you know, that looking back we all kind of agree on is that, this sort of building just would not, it would, would've never been built this way in the us you know, with a, uh, large, uh, multi-use tower on top of a parking structure that's uns sprinkled.

FDS Results And Heat Exposure

Wojciech Wegrzynski 45:06

I mean, this isn't, and I, I'm venturing now beyond the, the case study, but in general, when I think about car parks, I just feel if you have a, a, a car park. Where the likelihood of a fire is quite high. I mean, even if it's one every 200 years, this means like one in the four lifetimes of the building, there will be a fire. If in Poland we use the value of one per 50 years, it means you will have a fire in the car park. That's it. You will, it's, it's not, it's subject of a question and now you have a residential tower above it. This is a very high consequence fire. People are losing their homes. If something goes, it doesn't even have to collapse. We had exactly that happen in Poland. We had the fire in the car park in Poland. There was a cutoff point of the fire compartment area, which was 1500 square meters above which you have to put smoke detection, smoke control, various different things. So there was a time in my country where every car park was 1499 just to not have to have, you know, detections and everything. And, and unfortunately this car park was built underneath, uh, medium sized, but still a tower building residential. There was a massive fire in that car park, and if I'm not wrong, it took them two and a half years until people could return to their homes. They fixed the car park, which is a, a, a miracle. There's a technical research paper by, uh, professor Kowski, uh, which is likely in the show notes, uh, that that describes the fixing process. But it was madness to fix it. And, and they fought for the building because there was so many families who lost their homes and they just find it like unjust to lose your living, place your residence because of a likely event. You know, and I find this even like, I, I just don't feel this is right. I don't if, if you find safety engineering tells you you should be comfortable with that. There's something fundamentally wrong with it, I think.

Jonathan Hodges 47:11

yeah, I think, I think that's one of the main things we learned, uh, or really confirmed through the case study is that, when the consequences are as potentially severe as. Um, having this tower collapse, or as you mentioned, having people outta their homes, it was just

Mark McKinnon 47:25

damage.

Jonathan Hodges 47:26

Yeah.

Wojciech Wegrzynski 47:26

It didn't collapse, but the damage was so big they were not, allowed to go back.

Jonathan Hodges 47:30

so if you have consequences that severe, then it's an unacceptable solution. And so, we need to decrease that likelihood, uh, to where it's, you know, once in 10,000 years Mm. Or whatever the acceptable limit is. And what we found in the case study was the best way to do that was to put the sprinklers on.

Mark McKinnon 47:48

Mm-hmm.

Jonathan Hodges 47:49

You could do compartmentation or, limiting how many vehicles can be parked next to each other. Like there are other strategies that could be investigated for that. Uh, but it's very important to keep it to, you know, as few vehicles as possible. Now when you start looking at, you know, in the US at least, there's a lot of parking structures that aren't below a high rise. They're a separate building entirely. And so the, the risk, um, profile for that is different.

Mark McKinnon 48:14

Yep.

Jonathan Hodges 48:15

And so once you, once the consequences are not nearly as severe, then it becomes much more of an open question on how much remediation should be done and the cost benefit to society and things like that.

Wojciech Wegrzynski 48:28

And also on the underground, it's a different story. Man, that was a bad pun. Mike.

Mark McKinnon 48:35

Yeah, I, so I, I mean, I think, you know, the adding sprinklers to this, uh, this structure, it's not controversial at all. I think if I'm

Wojciech Wegrzynski 48:42

Well it is. It is in Poland. It is in Poland,

Mark McKinnon 48:44

yeah. Yeah. I guess from ice water, the perspective of all these teams that have, have done this case study, everyone agreed that it needed to be sprinkled.

Wojciech Wegrzynski 48:53

Yeah.

Mark McKinnon 48:54

You know, that there was no. Controversy at all, uh, here at the, at the conference

Wojciech Wegrzynski 48:59

as if engineering led you to, to good designs, right? Well, no, but, while I agree with you, that is essential and it's proven and proven beyond doubt, yet in my country it's controversial for some reason. But I also would like to comment that what you said, uh, is very important. John, you said that we need to lower the, the, the probability or frequency of those events to, to whatever number is. Now, if I think about the design process that gets me there, if I am in my traditional hat of far engineer doing my normal job for a normal client for a moderate pay, I'm not there because I, I do a design fire, I do a single case, uh, you know, investigation here. What you have shown was a really neat probabilistic, uh, pathway. There. We, we did something similar with OR or OFR, did something similar with our small, um. Small part in it for, for car parks in uk. And God, the show notes are really long right now, but if you like car parks, you have a lot of reading my friends, but, uh, I, I think, uh, getting those methods mainstream for these types of structures is something that, that we should work towards.

Jonathan Hodges 50:05

And uh, something we haven't talked about today but kind of ties in nicely with this is, all the teams that participated in this case study each had their own unique twist on the, the problem. And you know, I think we, we did a lot of work on the design, fire and the structural analysis. Um, but a lot of the teams had, uh, interesting creativity on how to have different mitigations. And it would be interesting to try and tie together, you know, our, frequency and consequence analysis with the other mitigation strategies, uh, to see, how does that affect the overall risk if you use some of those more creative strategies. and so that's may, maybe something we can look at over the next, you know, few months. But, you know, there's a lot of really, uh, creative fire protection engineers out there. And we do have options with the, the compartmentation, the ventilation and the sprinkler

Wojciech Wegrzynski 50:54

and yeah, shout out to teams. SFP Canada Team East meets West. That was Boston and Macau team. Oh, that was an unexpected, uh, combination. And Team SFP Korea, who did their, their takes

Jonathan Hodges 51:08

on the same, uh, the same, uh, case study

Wojciech Wegrzynski 51:11

to finalize it. Like for each of you, like what's lesson out of it for you as, as a fire safety engineer? Like, uh, what, what did you, what new things that opened for you, uh, as an engineer?

Mark McKinnon 51:22

So I think, uh, I'd mentioned that, we saw this as a bit of a gap analysis and. I think about the research that's currently being done around the world, uh, but also at ULRI, um, specifically around modern vehicles and how we had, had we had three more months, what we could potentially do, with this experimental data that we're, we're starting to roll out. Um, I think recently, uh, we published a paper on, uh, the toxic products associated with EVs. we've got, data on, the, um, sorry, the, the heat flux, uh, coming from these, fully involved, vehicle fires.

Wojciech Wegrzynski 52:05

Ah, the spread one.

Mark McKinnon 52:06

Yeah, yeah, yeah. And then, you know, I think, we, uh, we used a fairly simplistic model for ignition of. Of these, vehicles that are adjacent. And, you know, I think we've been talking about how we could characterize these materials a little bit better, uh, make sure that they're more modern. Some of these, some of these products were measured, you know, 20, 30 years ago. They just don't represent modern vehicles anymore. So I think there's, there's a lot of opportunity to apply some, contemporary research to, to get even better results here.

Concrete Spalling In Abaqus

Jonathan Hodges 52:36

Uh, I think what's, uh, stuck out to me with this, and I think it's been a topic of the conference a lot as well, is that, in practice we don't do a lot of fire risk assessments in the industry. Um, part of that is, you know, lack of time. Some of it's lack of education. Um, and I think we talked earlier in the podcast about just the level of complexity in coming up with these design fires. And so, uh, but what sticks out to me is, you know, we have this ability to provide that to fire protection engineers. Be more involved in, you know, creating webinars, uh, to do these training, uh, to be more involved in, uh, teaching the classes at universities and giving guest lectures to, to help educate people so that they're aware of these methods and, uh, try and increase the, the quality across the industry. Just a way that as an organization, uh, we can give back, uh, to the, the next generation. And hopefully, you know, in 20 years we can look at what they're doing and I'd just be surprised about how much progress has been made.

Christian Rippe 53:35

Might be a little more selfish with my response. Yes, please. Go.

Jonathan Hodges 53:38

I've learned little spike

Wojciech Wegrzynski 53:40

in

Jonathan Hodges 53:40

coding.

Christian Rippe 53:41

No, but so, so as you know, as someone who is normally right, as, as a consultant, I'm, you know, dealing with real world stuff. Right. And, and

Wojciech Wegrzynski 53:48

yeah.

Christian Rippe 53:48

You know, my, my colleagues here, John, Jonathan, and Mark right. Get to get to deal more on the research side of things, and I've really enjoyed. You know, being able to push the envelope more than we could really be allowed to in a real setting. Right. Um, and you know, I, I personally really enjoy kind of bridging that gap from, from research grade topics and ideas and concepts and methods and, can we really apply these to, you know, and, you know, what are the problems with things like convincing AHJs that this is, you know, whether we come in with the right answer. Right. you know, 'cause to put, uh, something like this, like what we did in front of an a hj, you'd have to have either a very knowledgeable A HJ or maybe a very not knowledgeable A HJI don't know, maybe that could be the, uh, the bad way to get or very

Wojciech Wegrzynski 54:31

trusting aj.

Christian Rippe 54:32

Right. you know, it would be so hard right. To, to actually implement this in today's world. 'cause it's, it's not just about, you know, our knowledge of some of these more advanced, like the risk-based approaches, but it's like the knowledge of, of AHJs and, and, and even the public. Right. And trusting that, when we do these sorts of analyses that we are coming up with. Of valid engineering solutions. Um, you know, that, that do prioritize life safety, right? And, and provide an, an acceptable level of safety. Um, but, but yeah, so I really enjoy kind of being able to, in a situation like this, kind of reach further into the, in, into the research side of things and, and try to claws some into the, uh, kinda more like the Commercialability

Wojciech Wegrzynski 55:13

side. Actually, I'll pull you on on, on that because it is interesting for, for Jonathan from Mark at the URI, it's, it's your normal world that you do research. You, you support people. For me, I'm a consultant, I'm a researcher. For me, it's normal. I do that. I, I love it. But actually it is quite interesting. Would you recommend to, to your fire engineering colleagues, consultants, to, to do experiments like that where you're uncon I mean, it, it's obviously difficult and it comes with the budget, it comes with the cost of time, et cetera. But the value in, in personal growth and development as a fire safety engineer, I think that's in invaluable. Is, is I, I know by myself, right. But

Christian Rippe 55:51

it, it's always, you know, it's always worth learning. Right. You know, and we can, we can never stop learning, right. That's just what we, what we have to do. And, and maybe I'll shamelessly even plug the podcast here 'cause it's, it's a, it's a great vehicle for even people who are practitioners, right? They're, they're not researchers. Right. You know, I tell, I I, I would say that I am much more a consumer of fire science than a producer of fire science. Right. but I listen to the podcast and I go, that's a cool idea. Let me spend, you know, just like an hour just digging into some literature and, maybe in the months or years down the road, I can start to start to think about how that can be incorporated. Into an actual, like commercial design or, or you know, an actual building or something like that. Um, and, and so I've, I've personally just found the podcast to be a great vehicle for that,

Wojciech Wegrzynski 56:36

that that's why Fire safety Engineering. I'm happy we, we've created, I had, uh, David Stacy in the podcast in the, in the past who, who talked about exactly that years ago. He's in the conference. Maybe I'll chase him if he changed his mind. But anyway, guys, uh, thank you so much for, for, for doing the case study, for showing us this interesting concept. You must have taken a lot of your time, so I really appreciate that. and, uh, thanks for, for. Spending this hour with me and the listeners and sharing this further. Yeah. Thank you, EK.

Mark McKinnon 57:04

Thank you. Thanks a lot Ji. I'm, I'm really excited about your case study. I wanna see what you guys came up with, please.

Wojciech Wegrzynski 57:10

You really have to control your, your expectations. You've not helped us. You're still like 12 more hours. How many CCF d scenarios I can pull off in 12? No. Anyway, thanks guys.

Sprinklers And Residual Risk

Lessons And Conference Wrap

Christian Rippe 57:24

And that's it. Thank you for listening. In the end, our uh, case study on the historical building went, went also quite, uh, well, which I am very happy with. And as for the car park case study in the conference, we've seen for different attempts on providing fire safety strategy to the building at hand. I guess it could be. Little bit difficult to follow on the podcast episode without having seen the presentation or the case study itself. But I, I tried to make, uh, the best out of it and figure out the most framework ish thinking, uh, I could get outta the guys in front of me. I'm very thankful they've agreed To this episode being jumped on a corridor straight after they finished, uh, their presentation. Uh, so as I said, it was delivered hot to you. Out of all the stuff they've, uh, they've done, I really admire their work on, on getting the. Data. Right? And getting, you know, the assumptions for the analysis, right? It's a lot of work. Really a lot of work to get through. And I'm a person who's done a lot of car parks and I've done a lot of papers on car parks. I've done a lot of research in car parks, and I fully comprehend how difficult it is to get all those puzzles in place. And I, I think actually, if we. Joint efforts and perhaps we will in the future. The work that Jonathan and his team done and the work that, uh, Danny Hopkins and myself and Mike, uh, Spearpoint and OFR has done on the car parks in the uk, if we joined efforts and we've combined them, I would be a pretty neat analysis and, uh, perhaps this is something to be done. For the future. for now, I can only recommend you going to SFP conferences because those case studies were really brilliant. I've been in some SFP conferences and, uh, I, I wouldn't be expecting that the case studies will be my favorite part of the conference, but in this, in this one, they absolutely were. And so thanks SFP for inviting, uh, John and his team. Thanks, SFP for inviting us and my team to the, uh, case study presentation, I believe the next one is in Toronto. So I hope we'll see more interesting case studies there and I'll be sure to be there and report on the most interesting ones I find there thanks for being here with me. and uh, yeah, see you around in the next episode. Cheers. Bye. I.