Feb. 18, 2026
239 - Assessing post-fire structural damage in tunnels with Negar Elhami-Khorasani
The player is loading ...
A tunnel can ride out a fire without collapsing (or even critical visible structural damage), but a question whether it is safe for operations, and what is its long-term residual fire resistance remains. With repair bills being in high seven-eight figures, this is more than just a theoretical question... In this episode we dig into the hard middle ground of fire damage post mild/large fires, and cover where modeling and fire science can help reducing the uncertainty and guiding decisions. With Professor Negar Elhami-Khorasani from University at Buffalo, we map how ventilation settings, tunnel slope, and fuel push temperatures into either safe or punishing regimes, and why spalling can turn a survivable event into a structural headache.
We break spalling down to first principles—vapor pressure, thermal gradients, and restraint—then translate that into a practical method: update the section as concrete “disappears” so the thermal boundary moves and heat penetrates realistically. From there, we track damage you can act on: concrete volumes beyond 300°C, steel temperatures that risk incomplete recovery, and bond loss that forces major repairs. Just as important, we model through cooling, when heat keeps migrating and residual capacity sinks. The result isn’t a guess; it’s a bounded map of what to replace and why.
We also take on the tactical questions that matter: How long would an extreme fire need to threaten collapse, given different soils and depths? What’s the real value of polypropylene fibers in high-strength mixes? How should owners structure a fast, post-fire workflow—quick checks for reopening within days, followed by a deeper, simulation-informed durability plan? By pairing observed spalling and known operations with targeted heat transfer and mechanical analysis, you can reconstruct the event, communicate risk clearly, and spend repair budgets where they return the most resilience.
If you care about structural fire engineering, tunnel safety, spalling mitigation, and performance-based design that reduces downtime, this conversation delivers a roadmap you can use.
Further reading - recommended papers by Negar Elhami-Khorasani and her team:
Fire Damage Assessment of Reinforced Concrete Tunnel Linings
----
The Fire Science Show is produced by the Fire Science Media in collaboration with OFR Consultants. Thank you to the podcast sponsor for their continuous support towards our mission.
00:00 - Framing The Post-Fire Problem
02:25 - Partner Message And Mission
03:41 - Meet The Guest And Award
04:46 - Why Tunnels And Post-Fire Decisions
08:37 - Fire Severity, Ventilation, And Slope
12:05 - Temperatures Along The Tunnel
16:10 - Modeling Traveling Fire Effects
18:21 - How Heat Damages Concrete And Steel
20:57 - Spalling Mechanisms Explained
25:10 - Data Gaps And Extreme Fire Curves
28:45 - A Practical Spalling Model
32:22 - Damage Metrics And Repair Triggers
35:02 - Cooling Phase And Residual State
37:22 - Collapse Time For Firefighter Safety
41:30 - Geometry, Soil, And Failure Modes
44:05 - Restraint And Test Limitations
47:15 - Uncertainty And Probabilistic Spalling
50:12 - Quickfire Spalling Q&A
54:02 - Polypropylene Fibers As Mitigation
58:28 - Hidden Damage And Inspection Limits
01:01:05 - Post-Fire Workflow And Timelines
WEBVTT
00:00:00.400 --> 00:00:02.560
Hello everybody, welcome to the Fire Science Show.
00:00:02.560 --> 00:00:23.760
Today we're going into the world of structural fire engineering, and while most of the considerations in structural fire engineering, at least the ones that I usually participate in, they consider whether the fire can destroy, collapse a structure, how safe the structure overall is in case of fire.
00:00:23.760 --> 00:00:39.840
In today's episode, the discussion went to a little different avenue, which I actually quite appreciate because what if the fire was not big enough to destroy a structure, but it's been there and it's done some damage.
00:00:39.840 --> 00:00:46.880
How do we know that the damage is significant and how do we know the damage needs to be repaired?
00:00:46.880 --> 00:00:51.520
That's a really, really, really huge question in a post-fire assessment.
00:00:51.520 --> 00:00:58.240
And uh because we're dealing with tunnels, this question is really valued at many, many millions of dollars.
00:00:58.240 --> 00:01:05.840
To answer this question, uh I have a special guest, Professor Negar Elhami-Khorasani from University at Buffalo.
00:01:05.840 --> 00:01:13.280
She's an expert in structural fire safety, and she's also a recipient of IFSS Magnuson Award.
00:01:13.280 --> 00:01:24.799
That's a very prestigious early career award, and it's very fitting because Professor Sven Magnussen was a pioneer of structural fire engineering, so I'm really happy that uh this award went uh her way.
00:01:24.799 --> 00:01:30.799
In this episode, we cover a lot of things related to structural fire safety of tunnels.
00:01:30.799 --> 00:01:32.480
We cover spalling a lot.
00:01:32.480 --> 00:01:34.159
Finally, I have a spalling episode.
00:01:34.159 --> 00:01:42.560
I've I was planning a spalling episode for such a long time in the podcast, and finally we talk in-depth on spalling, but that's not everything.
00:01:42.560 --> 00:01:52.079
We also talk about probabilistic approaches uh to structural fire safety risk and also the issue of diagnosing uh the structural damage and and fixing it.
00:01:52.079 --> 00:01:57.519
So it's uh packed with great knowledge, I've learned a lot, I hope you will so as well.
00:01:57.519 --> 00:02:00.000
Let's spin the intro and jump into the episode.
00:02:00.000 --> 00:02:06.239
Welcome to the Fire science Show.
00:02:06.239 --> 00:02:09.840
My name is Wojciech Wegrzynski, and I will be your host.
00:02:09.840 --> 00:02:29.680
The Firest Show podcast is brought to you in partnership with OFR Consultants.
00:02:29.680 --> 00:02:39.599
OFR is the UK's leading independent multi-award-winning fire engineering consultancy with a reputation for delivering innovative safety-driven solutions.
00:02:39.599 --> 00:02:48.719
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.
00:02:48.719 --> 00:03:06.159
So far, we've brought through more than 150 episodes which translate into nearly 150 hours of educational content available, free, accessible all over the planet without any paywalls, advertisements, or hidden agendas.
00:03:06.159 --> 00:03:13.039
This makes me very proud and I am super thankful to OFR for this long-lasting partnership.
00:03:13.039 --> 00:03:20.319
I'm extremely happy that we've just started the year 4, and I hope there will be many years after that to come.
00:03:20.319 --> 00:03:29.039
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.
00:03:29.039 --> 00:03:36.479
And for you, the listener, if you would like to learn more or perhaps even become a part of OFR, they always have opportunities awaiting.
00:03:36.479 --> 00:03:39.120
Check their website at OFRconsultants.com.
00:03:39.120 --> 00:03:41.039
And now let's head back to the episode.
00:03:41.039 --> 00:03:42.080
Hello, everybody.
00:03:42.080 --> 00:03:46.240
I am joined today by Negar Elhami-Khorasani from University at Buffalo.
00:03:46.240 --> 00:03:47.039
Hey Negar.
00:03:47.360 --> 00:03:49.599
Hello, thank you so much for having me.
00:03:49.599 --> 00:03:51.520
It's a real pleasure to be here.
00:03:51.520 --> 00:03:59.199
And I just mentioned this very quickly that I've been a longtime fan of your podcast and the conversations you've been bringing to the community.
00:03:59.199 --> 00:04:01.680
And I often recommend episodes to my students.
00:04:01.759 --> 00:04:04.639
So that explains some spikes in the US recently.
00:04:04.639 --> 00:04:05.439
Thank you, Negar.
00:04:05.439 --> 00:04:06.400
Well, but but thank you.
00:04:06.400 --> 00:04:07.039
Thank you very much.
00:04:07.039 --> 00:04:08.159
I appreciate that.
00:04:08.159 --> 00:04:13.439
And uh hopefully today we create some great content uh for for those who come after us.
00:04:13.439 --> 00:04:21.040
And uh I cannot start the episode with anything else than congratulating you on the Magnuson Award from the ISS.
00:04:21.040 --> 00:04:23.519
Wow, amazing! Like great job there.
00:04:23.519 --> 00:04:25.920
I'm really, really happy that it went your way.
00:04:26.319 --> 00:04:27.279
Thank you also.
00:04:27.279 --> 00:04:28.240
Thank you, thank you.
00:04:28.240 --> 00:04:42.720
And I should say that these are collective efforts, you know, my collaborators, my students, and I really appreciate everybody in my network basically um contributing all these years to the research we're doing.
00:04:43.040 --> 00:04:49.519
Yeah, we're we're all blessed with people around us who allow us to do great science and great science, let's talk about it.
00:04:49.519 --> 00:04:58.079
Uh, we're about to talk about post-fire structural assessments, and as I understand, we're gonna mostly focus on tunnels.
00:04:58.079 --> 00:05:02.079
So maybe uh let's start with how how did you get to work on this?
00:05:02.079 --> 00:05:04.399
What triggered this interest uh for you?
00:05:04.720 --> 00:05:05.040
Right.
00:05:05.040 --> 00:05:13.759
So basically, a couple of years ago, there was this research question that we know tunnel fires, you know, it potentially could happen.
00:05:13.759 --> 00:05:16.639
And when they happen, there could be you know extreme events.
00:05:16.639 --> 00:05:22.959
We have had historical cases, but uh not all of these fires are extreme.
00:05:22.959 --> 00:05:32.959
And when we looked at the literature, we couldn't find much um solid guidance on post-fire damage assessment.
00:05:32.959 --> 00:05:45.839
Uh, there are um you know qualitative guidance, but uh we wanted more uh quantitative assessment, a little bit more solid step-by-step approach, and that was something that was missing.
00:05:45.839 --> 00:05:55.040
And in parallel, so that's something that fire happens, then an engineer goes into the tunnel looking around, inspecting, and trying to decide, okay, is it safe to reopen?
00:05:55.040 --> 00:05:59.759
What's the decision, especially for mild fires, something that is like a gray area?
00:05:59.759 --> 00:06:08.079
You don't know if because you're looking, if there's not much damage, you're like, okay, if there's extreme damage, there's no way that you can reopen, but somewhere in the gray area.
00:06:08.079 --> 00:06:14.079
And then in parallel, in my research group, we've been working on in general performance-based design of structures.
00:06:14.079 --> 00:06:20.399
And in this case, we thought that okay, if we can frame the research question in a way that, okay, what do you do afterwards?
00:06:20.399 --> 00:06:36.000
But then use the same procedure to quantify the potential downtime and functionality of the structure during the design phase, then you can also uh design for basically cases where you minimize losses because of that downtime.
00:06:36.560 --> 00:06:42.879
So that's actually quite an interesting question now that you pose that, because obviously if the tunnel collapsed, you have your answer.
00:06:42.879 --> 00:06:45.199
Uh the post fire damage is huge.
00:06:45.199 --> 00:06:52.000
If the if it was like a single passenger vehicle, I would guess it's just dirty, water spray it, and then you're done.
00:06:52.000 --> 00:06:57.920
But uh, we had a fire in in Warsaw which involved uh heavy goods vehicle.
00:06:57.920 --> 00:07:09.199
Well, it was like uh a smaller type of truck, and I would estimate the the hit release rate of that fire in the range of 10, 20-ish megawatts, more probably closer to 10-ish megawatts.
00:07:09.199 --> 00:07:14.240
It was not part of the assessment after the fire, but they reopened the tunnel very quickly.
00:07:14.240 --> 00:07:25.759
But I must imagine if the fire was a little bit larger, and if there was like some obvious discoloring damage to the concrete, I don't think the decision would be that easy to be made.
00:07:25.759 --> 00:07:29.680
So that that that's a quite a valid research question.
00:07:29.680 --> 00:07:33.120
How is how big is the scatter of the fires in the tunnels?
00:07:33.120 --> 00:07:39.920
Because I assume the extreme fires are extremely rare, but are they a big part of all fires or fires are just rare in tunnels?
00:07:40.319 --> 00:07:53.839
I think uh working with the US stakeholders, and in the US, the small fires happen, but again, they're quick and very small, smaller than, for example, the example you mentioned.
00:07:53.839 --> 00:08:02.800
Um, and then we haven't had a major extreme event, but they are also building new tunnels.
00:08:02.800 --> 00:08:06.959
Uh, and they did exactly have this question for us.
00:08:06.959 --> 00:08:20.319
So statistics maybe doesn't necessarily show that there's so many moderate level uh fires, but the question that authorities have here is that what if it happens, what do we do?
00:08:21.120 --> 00:08:22.800
Preemptively, uh before another.
00:08:23.360 --> 00:08:23.680
Right.
00:08:24.079 --> 00:08:26.000
Which is not a great management strategy.
00:08:26.000 --> 00:08:31.040
Arguably, like you worry when it happens, but uh okay, for question to be asked.
00:08:31.040 --> 00:08:42.240
Um, in terms of what what okay, fire is a fire, but uh what what kind of temperature heat fluxes those tunnel structures are exposed to?
00:08:42.240 --> 00:08:50.080
Like maybe you can introduce the listeners to the range of fires and their consequences in the tunnels, and maybe how do you figure that out, how bad they will be.
00:08:50.399 --> 00:08:50.639
Right.
00:08:50.639 --> 00:09:00.639
So we specifically in one case looked at uh, and I'll expand on it, but we specifically looked at a tunnel which was for passenger trains.
00:09:00.639 --> 00:09:22.240
So temperature inside the tunnel eventually, before I get to the uh heat release rate, and temperature will be a function of number of parameters, not just the heat release rate, but also even the slope of the tunnel and the diameter of the tunnel and the material properties um of you know the bounding basically here's concrete.
00:09:22.240 --> 00:09:24.960
So all of the and ventilation, ventilation also.
00:09:24.960 --> 00:09:39.519
If it's um there are fans, mechanical fans in the tunnel, all of this um makes a difference in terms of the type of temperatures you get and the way that the the temperature spreads, basically the fire spreads inside the tunnel.
00:09:39.519 --> 00:09:53.360
So we wanted because we were looking at damage, we didn't want to just look at the temperature, let's say in one cross section of the tunnel, but we also wanted to know how that profile changes along the length of the tunnel.
00:09:53.360 --> 00:09:56.639
So time and space, they were two variables.
00:09:56.639 --> 00:10:08.720
And when we started looking at the literature of potentially experiments for passenger trains that were set on fire so that we can get heat release rate, we actually found some data.
00:10:08.720 --> 00:10:12.159
And then we found some bounding values.
00:10:12.159 --> 00:10:24.720
And from there, we said, okay, this is the input fuel, and then we have different ventilation values, and the ventilation values are typically so the fans, if it's a passenger train, it means that people need to evacuate.
00:10:24.720 --> 00:10:27.360
So there's this concept of critical velocity.
00:10:27.360 --> 00:10:39.039
Basically, it means that if a fire happens, the fans should be set in a way that you do not get the fire spread in one direction, you don't get this back layering so that people can evacuate from the other side.
00:10:39.039 --> 00:10:57.759
So assuming that the operator is setting the ventilation in the range that is pushing the fire on one side, and then consider different slopes, we ended up uh running hundreds of simulations in FTS, and we had HRR values in the order of 40 megawatt as input.
00:10:57.759 --> 00:11:02.480
And then we had the other extreme, even something less than 10 megawatt.
00:11:02.480 --> 00:11:04.639
What happened in that whole range?
00:11:04.639 --> 00:11:17.120
Any fire, I would say, by combining then with the slope and ventilation and everything else, then we ended up having a completely bimodal type of temperatures.
00:11:17.120 --> 00:11:26.080
So anything more than I would say 30 megawatt fire would end up having temperatures in the order of, you know, median was maybe in the order of 1000 degrees Celsius.
00:11:26.080 --> 00:11:27.840
So we were pushing like very high temperatures.
00:11:27.840 --> 00:11:35.360
And then fires less than maybe 20 megawatt, we had low temperatures, like really low temperatures.
00:11:35.360 --> 00:11:37.279
Structurally wouldn't impact.
00:11:37.279 --> 00:11:41.120
I mean, for humans still need to evacuate, but structurally, we wouldn't be concerned.
00:11:41.120 --> 00:11:44.960
But then in between, that we would call it intermediate fires.
00:11:44.960 --> 00:11:48.879
That's where the ventilation could push the fire either go high or low.
00:11:48.879 --> 00:11:57.200
So we would get either we could get like 800 degrees Celsius or we could get like 400 degrees Celsius, depending on the condition of the slope and everything else.
00:11:57.840 --> 00:12:02.399
How much of that is like direct flame impingement near the fire source?
00:12:02.399 --> 00:12:09.919
And how much of that is just you know an average smoke temperature downstream, exposing a bigger part of the structure?
00:12:10.159 --> 00:12:10.320
Right.
00:12:10.320 --> 00:12:11.360
So that's a great question.
00:12:11.360 --> 00:12:16.159
And I think um numbers that I reported in terms of temperature, those were the maximum temperatures.
00:12:16.159 --> 00:12:19.440
Maximum that would be right in the cross section where you have the source.
00:12:19.440 --> 00:12:37.519
When we were looking at this, we actually assumed that there will be multiple train cars, and then the ignition happens in one train car, but the actual fire spreads across multiple train cars, and then we had to also set an ignition threshold in FTS, and we did consider a range, it wasn't just a value.
00:12:37.519 --> 00:12:40.240
We said, you know, average is 300, but let's assume.
00:12:40.240 --> 00:12:43.759
So all of these values, all the parameters actually came with a distribution.
00:12:43.759 --> 00:12:55.679
If we didn't have any information, we would assume like a uniform distribution between a you know range for 300 to you know 400 degrees Celsius where an ignition could happen.
00:12:55.679 --> 00:13:00.799
So we did consider that potential for fire spit inside the tunnel.
00:13:00.799 --> 00:13:21.759
So those values I reported tells them that was the maximum T max in the cross section, but then there were hundreds of meters of the tunnel length that would be impacted, but the temperature profile could be actually uh the temperature could be a lower value as you go further away.
00:13:21.759 --> 00:13:30.879
So another thing though, I would mention that you may get so the fire starts at a car and then starts spreading.
00:13:30.879 --> 00:13:35.440
At some point, your highest temperature may not be right at where the ignition started.
00:13:35.440 --> 00:13:38.799
It may be like a car, you know, uh nearby.
00:13:38.799 --> 00:13:40.720
And that comes with a delay, right?
00:13:40.720 --> 00:13:51.279
So I'm not gonna deviate, but we did some experiments as part of the second part of this project, which we tested just concrete slap, uh tunnel slabs.
00:13:51.279 --> 00:13:55.679
And then one of the fire scenarios we considered, we actually had a delay.
00:13:55.679 --> 00:13:58.960
Like the maximum temperature was 850 degrees Celsius.
00:13:58.960 --> 00:14:07.039
It was one of the fire curves that we got out of the FDS, but it didn't reach 850 immediately.
00:14:07.039 --> 00:14:12.799
It had temperatures in the order of 300, 400 degrees Celsius for some time, and then it reached 500.
00:14:12.799 --> 00:14:20.080
So it's it's basically a section downstream of the ignition that sees the maximum temperature with a delay.
00:14:20.320 --> 00:14:21.919
Yeah, kind of like a traveling fire.
00:14:22.480 --> 00:14:22.799
Exactly.
00:14:22.799 --> 00:14:23.519
Yeah, exactly.
00:14:23.840 --> 00:14:25.679
Nice, nice, very interesting.
00:14:25.679 --> 00:14:33.200
I ballpark number, I got to similar numbers when I was doing my right railway projects, but they were usually capped at 20 megawatts for some.
00:14:33.200 --> 00:14:46.559
Okay, yeah, and and we also were observing surprisingly low temperatures, and and probably that's because you know this the perhaps there's a like a turning point where the flames engulf the entire cross section of the tunnel.
00:14:46.559 --> 00:14:53.039
Maybe there's some kind of like inflation point where where the physics kind of changes a little bit, uh yeah, right.
00:14:53.120 --> 00:15:02.720
But right, and I think the slope, like we had we had steep, we considered, and it was we we have like, for example, an example tunnel in the US, so two percent.
00:15:02.720 --> 00:15:06.000
Uh, you know, so the slope uh made a difference.
00:15:06.320 --> 00:15:08.559
Wouldn't ventilation cancel out slope in your case?
00:15:08.799 --> 00:15:14.639
So that's the thing, but if your ventilation is also on uh so we did consider there could be an error by the operator.
00:15:14.720 --> 00:15:15.840
We're not okay.
00:15:16.000 --> 00:15:26.399
So if you're not doing exactly what you know the fans are supposed to do, so everything basically what could go wrong, everything that could go wrong, let's assume one of those cases.
00:15:26.480 --> 00:15:34.240
Yeah, you're proper structural engineers testing all all all possibilities and running hundreds of simulations, yeah.
00:15:34.240 --> 00:15:35.279
Typical stuff.
00:15:35.279 --> 00:15:41.759
Okay, so how does high temperature damage the tunnels exactly?
00:15:41.759 --> 00:15:44.320
Like what are we worried about in here?
00:15:44.720 --> 00:15:48.000
So um a couple of cases, uh, and we've been expanding.
00:15:48.000 --> 00:15:56.559
So when we started first, we said, okay, um, it's a reinforced concrete, let's say uh tunnel liner, and it will be exposed to high temperatures.
00:15:56.559 --> 00:15:57.840
What is going to happen?
00:15:57.840 --> 00:16:07.440
So concrete has low thermal conductivity, but still with these fires, uh, it will pick up the temperature and there will be a uh thermal gradient inside the cross section.
00:16:07.440 --> 00:16:08.960
So a couple of things happen.
00:16:08.960 --> 00:16:16.639
A, the area of concrete that is reaching high temperatures, and so far we've been using a 300 degree Celsius threshold.
00:16:16.639 --> 00:16:23.519
So any concrete that reaches beyond 300 degrees Celsius should probably be replaced.
00:16:23.519 --> 00:16:28.559
That's one criteria for damage without looking at it structurally, just material-wise.
00:16:28.559 --> 00:16:34.080
And then there is the reinforcement, if you have you know conventional reinforcement.
00:16:34.080 --> 00:16:38.639
Uh tunnel sections are typically in design mostly because of their shape.
00:16:38.639 --> 00:16:51.200
It's the if they're circular, if it's like even this horseshoe, so you have the count, they're mostly in compression, they're taking the load in compression, but they are still against the soil.
00:16:51.200 --> 00:16:58.720
And if you want a structure, if structure is heating up, so it wants to expand, it can't, depending on the stiffness of the soil.
00:16:58.720 --> 00:17:01.120
So the behavior actually changes.
00:17:01.120 --> 00:17:15.039
If your tunnel is in a soft soil or is in rock, you're gonna see some differences because if it's in soft soil, you're gonna see more deflections, not huge, but you're gonna see more deflections compared to rock.
00:17:15.039 --> 00:17:23.200
Whereas in the in rock, you're gonna see more stresses building up because the tunnel doesn't really have the ability to move or that displacement.
00:17:23.200 --> 00:17:37.279
And therefore, you will, depending on how long this fire and how hot this fire is, you will start building up obviously stresses and then potentially yield your rebar, again, depending on how long that fire is.
00:17:37.279 --> 00:17:46.160
And then another primary, and this is something that we spent some time trying to incorporate in the model, although it's very complex, is spalling.
00:17:46.160 --> 00:17:53.759
So when we started looking at this, we said, okay, if realistically we want to analyze and evaluate damage, we need to be able to capture spalling.
00:17:53.759 --> 00:18:00.559
Spalling is basically you have these um when concrete heats up, you have these pieces of concrete that falls off.
00:18:00.559 --> 00:18:05.200
And it can continuously, you know, you're gonna you're losing your concrete basically.
00:18:05.200 --> 00:18:08.880
And it's a pretty complex process.
00:18:08.880 --> 00:18:17.119
If I go into a little bit of detail in terms of what happens, there is the basically pore pressure buildup.
00:18:17.119 --> 00:18:22.400
I'm not a material scientist, but I know enough in terms of what um what happens during spalling.
00:18:22.400 --> 00:18:24.960
So basically, concrete contains moisture.
00:18:24.960 --> 00:18:38.079
So when you heat up concrete, basically that moisture turns into steam, vapor, and then at some point you have this vapor pressure that builds up, uh, which exceeds the tensile capacity of concrete.
00:18:38.079 --> 00:18:40.160
And the tensile capacity of concrete is not much.
00:18:40.160 --> 00:18:42.640
So then you start having these pieces fall off.
00:18:42.640 --> 00:18:45.519
That's not the only mechanism that drives balling.
00:18:45.519 --> 00:18:47.599
There's also that thermal stress built up.
00:18:47.599 --> 00:18:57.039
So you get a gradient, and because of that gradient, you build you also start experiencing tensile stresses inside concrete, which again exceeds tensile capacity.
00:18:57.039 --> 00:19:04.160
So a combination of these mechanisms and how fast concrete heats up, and what's the restrain.
00:19:04.160 --> 00:19:08.480
So the amount of restraint it has, it will again make an impact in terms of the stress built-up.
00:19:08.480 --> 00:19:14.880
All of those, so the boundary conditions, they all make a difference in terms of when and how much you're gonna get smolen.
00:19:14.880 --> 00:19:20.160
Modeling this, so there's a lot of literature out there.
00:19:20.160 --> 00:19:25.039
I would classify them as experimental and numerical modeling.
00:19:25.039 --> 00:19:30.319
So there's experimental lot of you know, different scales, small scale, larger scale.
00:19:30.319 --> 00:19:39.519
And then with numerical, I would say there are these micro-level modeling where they really go into modeling the physics of the problem.
00:19:40.000 --> 00:19:42.319
There's also Nasser who just drops AI on it.
00:19:42.559 --> 00:19:43.279
Right, right, right.
00:19:44.240 --> 00:19:45.839
Which I actually find brilliant.
00:19:45.839 --> 00:19:48.559
I seen that in that, but that's another topic.
00:19:49.279 --> 00:19:51.440
And and his paper came after we were working.
00:19:51.440 --> 00:19:57.039
So I I wish his paper would have been published while we were working on this topic.
00:19:57.039 --> 00:20:06.319
When we started looking at this, we said we have these models for heat transfer and mechanical analysis, thermo mechanical analysis, finite element analysis.
00:20:06.319 --> 00:20:08.799
We're doing all these structural analysis at high temperatures.
00:20:08.799 --> 00:20:10.799
How do we actually incorporate spelling?
00:20:10.799 --> 00:20:15.200
And we couldn't really find the straightforward way to do it in the literature.
00:20:15.200 --> 00:20:20.240
So we came up with an approach, it's very simplistic, and I'll explain what it is.
00:20:20.240 --> 00:20:40.319
And we're aware of how simplistic it is, but it actually provides us, and we do not necessarily look at it as this is exactly what's going to happen, but it's more of it will give us bounding, like a lower bound of damage, upper bound of damage, to get an idea of what could be the worst case scenario.
00:20:40.319 --> 00:20:43.119
So no spalling, fine, that's like one extreme.
00:20:43.519 --> 00:20:43.759
Perfect.
00:20:43.759 --> 00:20:44.000
Yeah.
00:20:44.480 --> 00:20:48.640
And then we went and said, hey, look, let's look at the data in the literature.
00:20:48.640 --> 00:20:52.400
And we limit it, we did again, there's a lot of data.
00:20:52.400 --> 00:21:02.160
So we just looked at larger scale tests because the shape and the size of the specimen also makes a difference.
00:21:02.160 --> 00:21:05.039
If you, for example, have a reinforced concrete column.
00:21:05.039 --> 00:21:11.680
I know we're talking in the context of tunnels, but if you have a reinforced concrete column, you can have spalling in the corners because of the shape.
00:21:11.680 --> 00:21:15.039
But this is this reinforced concrete slab for a tunnel section.
00:21:15.039 --> 00:21:26.960
So we said, let's look at relevant shapes and larger scale sizes to have a to get a better feeling of how this thing behaves, and potentially data on spalling.
00:21:26.960 --> 00:21:46.720
So we looked at the test measurements, and we found that okay, a lot of these tests they use either hydrocarbon fire curves, extreme cases, and then uh so there's RWS in the experiments, there is a wrapped ZTV fire curve, and then hydrocarbon or modified hydrocarbon from URCO.
00:21:46.720 --> 00:21:52.400
So we found those, and then we saw that in the test they record the time.
00:21:52.400 --> 00:21:59.359
So they're probably like listening to when you can hear in a furnace test, for example, that there's like sound.
00:21:59.359 --> 00:22:02.480
So They're reporting the time that something happens.
00:22:02.480 --> 00:22:04.240
We're like, okay, this is when it started.
00:22:04.240 --> 00:22:13.200
And because you know the fire curve, you can actually get the fire temperature, not the material temperature, which is another thing that we know that you want the material temperature.
00:22:13.200 --> 00:22:16.880
But you get the temperature of the fi at that point because it's a standard curve.
00:22:16.880 --> 00:22:24.640
And then also they typically report the depth of spalling, which is post-fire.
00:22:24.640 --> 00:22:26.079
Which is post-fire, right?
00:22:26.079 --> 00:22:30.640
After it's done, the simulate the experiment is done, cool down, they go and measure this.
00:22:30.640 --> 00:22:41.519
We said, okay, number one, we had to pause here because the whole motivation where we started, we said these are moderate fires, we're not necessarily looking at all these like very extreme fires.
00:22:41.759 --> 00:22:45.200
So because the curves that you've mentioned, they are all ridiculous.
00:22:45.200 --> 00:22:48.160
Like they are like uh RWS.
00:22:48.160 --> 00:22:51.680
It's like I might maybe wrong, but it's like 1200 in five minutes.
00:22:52.000 --> 00:22:54.880
Yes, and then the maximum is 1350.
00:22:55.200 --> 00:22:57.279
It's it's it's absolutely ridiculous.
00:22:57.279 --> 00:23:01.119
Like we're doing those experiments in our furnaces in ITB.
00:23:01.119 --> 00:23:07.519
We had to buy a special set of extremely expensive platinum thermocouples for those tests.
00:23:07.519 --> 00:23:11.759
Because they though those curves destroy your normal equipment in the furnace.
00:23:11.759 --> 00:23:16.720
That's how ridiculously high uh temperatures we're talking about for fire resistance testing.
00:23:16.720 --> 00:23:23.920
Like in the standard fire curve, you get like 830 degrees in the first 30 minutes, if I'm not wrong, something like that, maybe 900 in an hour.
00:23:23.920 --> 00:23:24.480
That's it.
00:23:24.480 --> 00:23:26.960
And then here, like bam, 1200 five minutes.
00:23:26.960 --> 00:23:28.559
Wow, it's it's it's insane.
00:23:28.640 --> 00:23:30.960
Like exactly, totally agree.
00:23:30.960 --> 00:23:33.759
So we worked with what we could find.
00:23:33.759 --> 00:23:37.119
So we ended up expanding our data set again.
00:23:37.119 --> 00:23:54.720
Although it's not a tunnel fire curve or relevant perhaps, but because we couldn't find anything else, we ended up looking at slab tests exposed to ISO or ASTM, which has this lower again, um, heating rate.
00:23:54.720 --> 00:23:57.279
And we said, okay, this at least has a slower heating rate.
00:23:57.279 --> 00:23:58.559
So it gives us an idea.
00:23:58.559 --> 00:24:06.240
Then, so we are now talking about less than 100 tests collected, uh, sometimes maybe 60.
00:24:06.240 --> 00:24:11.599
And uh we said, okay, we have an idea of the when we again we've been expanding.
00:24:11.599 --> 00:24:20.160
So when we started, we said, let's just look at the mean or the median of the time or and corresponding temperature where the spalling happens.
00:00:00.400 --> 00:00:02.560
00:00:02.560 --> 00:00:23.760
00:00:23.760 --> 00:00:39.840
00:00:39.840 --> 00:00:46.880
00:00:46.880 --> 00:00:51.520
00:00:51.520 --> 00:00:58.240
00:00:58.240 --> 00:01:05.840
00:01:05.840 --> 00:01:13.280
00:01:13.280 --> 00:01:24.799
00:01:24.799 --> 00:01:30.799
00:01:30.799 --> 00:01:32.480
00:01:32.480 --> 00:01:34.159
00:01:34.159 --> 00:01:42.560
00:01:42.560 --> 00:01:52.079
00:01:52.079 --> 00:01:57.519
00:01:57.519 --> 00:02:00.000
00:02:00.000 --> 00:02:06.239
00:02:06.239 --> 00:02:09.840
00:02:09.840 --> 00:02:29.680
00:02:29.680 --> 00:02:39.599
00:02:39.599 --> 00:02:48.719
00:02:48.719 --> 00:03:06.159
00:03:06.159 --> 00:03:13.039
00:03:13.039 --> 00:03:20.319
00:03:20.319 --> 00:03:29.039
00:03:29.039 --> 00:03:36.479
00:03:36.479 --> 00:03:39.120
00:03:39.120 --> 00:03:41.039
00:03:41.039 --> 00:03:42.080
00:03:42.080 --> 00:03:46.240
00:03:46.240 --> 00:03:47.039
00:03:47.360 --> 00:03:49.599
00:03:49.599 --> 00:03:51.520
00:03:51.520 --> 00:03:59.199
00:03:59.199 --> 00:04:01.680
00:04:01.759 --> 00:04:04.639
00:04:04.639 --> 00:04:05.439
00:04:05.439 --> 00:04:06.400
00:04:06.400 --> 00:04:07.039
00:04:07.039 --> 00:04:08.159
00:04:08.159 --> 00:04:13.439
00:04:13.439 --> 00:04:21.040
00:04:21.040 --> 00:04:23.519
00:04:23.519 --> 00:04:25.920
00:04:26.319 --> 00:04:27.279
00:04:27.279 --> 00:04:28.240
00:04:28.240 --> 00:04:42.720
00:04:43.040 --> 00:04:49.519
00:04:49.519 --> 00:04:58.079
00:04:58.079 --> 00:05:02.079
00:05:02.079 --> 00:05:04.399
00:05:04.720 --> 00:05:05.040
00:05:05.040 --> 00:05:13.759
00:05:13.759 --> 00:05:16.639
00:05:16.639 --> 00:05:22.959
00:05:22.959 --> 00:05:32.959
00:05:32.959 --> 00:05:45.839
00:05:45.839 --> 00:05:55.040
00:05:55.040 --> 00:05:59.759
00:05:59.759 --> 00:06:08.079
00:06:08.079 --> 00:06:14.079
00:06:14.079 --> 00:06:20.399
00:06:20.399 --> 00:06:36.000
00:06:36.560 --> 00:06:42.879
00:06:42.879 --> 00:06:45.199
00:06:45.199 --> 00:06:52.000
00:06:52.000 --> 00:06:57.920
00:06:57.920 --> 00:07:09.199
00:07:09.199 --> 00:07:14.240
00:07:14.240 --> 00:07:25.759
00:07:25.759 --> 00:07:29.680
00:07:29.680 --> 00:07:33.120
00:07:33.120 --> 00:07:39.920
00:07:40.319 --> 00:07:53.839
00:07:53.839 --> 00:08:02.800
00:08:02.800 --> 00:08:06.959
00:08:06.959 --> 00:08:20.319
00:08:21.120 --> 00:08:22.800
00:08:23.360 --> 00:08:23.680
00:08:24.079 --> 00:08:26.000
00:08:26.000 --> 00:08:31.040
00:08:31.040 --> 00:08:42.240
00:08:42.240 --> 00:08:50.080
00:08:50.399 --> 00:08:50.639
00:08:50.639 --> 00:09:00.639
00:09:00.639 --> 00:09:22.240
00:09:22.240 --> 00:09:24.960
00:09:24.960 --> 00:09:39.519
00:09:39.519 --> 00:09:53.360
00:09:53.360 --> 00:09:56.639
00:09:56.639 --> 00:10:08.720
00:10:08.720 --> 00:10:12.159
00:10:12.159 --> 00:10:24.720
00:10:24.720 --> 00:10:27.360
00:10:27.360 --> 00:10:39.039
00:10:39.039 --> 00:10:57.759
00:10:57.759 --> 00:11:02.480
00:11:02.480 --> 00:11:04.639
00:11:04.639 --> 00:11:17.120
00:11:17.120 --> 00:11:26.080
00:11:26.080 --> 00:11:27.840
00:11:27.840 --> 00:11:35.360
00:11:35.360 --> 00:11:37.279
00:11:37.279 --> 00:11:41.120
00:11:41.120 --> 00:11:44.960
00:11:44.960 --> 00:11:48.879
00:11:48.879 --> 00:11:57.200
00:11:57.840 --> 00:12:02.399
00:12:02.399 --> 00:12:09.919
00:12:10.159 --> 00:12:10.320
00:12:10.320 --> 00:12:11.360
00:12:11.360 --> 00:12:16.159
00:12:16.159 --> 00:12:19.440
00:12:19.440 --> 00:12:37.519
00:12:37.519 --> 00:12:40.240
00:12:40.240 --> 00:12:43.759
00:12:43.759 --> 00:12:55.679
00:12:55.679 --> 00:13:00.799
00:13:00.799 --> 00:13:21.759
00:13:21.759 --> 00:13:30.879
00:13:30.879 --> 00:13:35.440
00:13:35.440 --> 00:13:38.799
00:13:38.799 --> 00:13:40.720
00:13:40.720 --> 00:13:51.279
00:13:51.279 --> 00:13:55.679
00:13:55.679 --> 00:13:58.960
00:13:58.960 --> 00:14:07.039
00:14:07.039 --> 00:14:12.799
00:14:12.799 --> 00:14:20.080
00:14:20.320 --> 00:14:21.919
00:14:22.480 --> 00:14:22.799
00:14:22.799 --> 00:14:23.519
00:14:23.840 --> 00:14:25.679
00:14:25.679 --> 00:14:33.200
00:14:33.200 --> 00:14:46.559
00:14:46.559 --> 00:14:53.039
00:14:53.120 --> 00:15:02.720
00:15:02.720 --> 00:15:06.000
00:15:06.320 --> 00:15:08.559
00:15:08.799 --> 00:15:14.639
00:15:14.720 --> 00:15:15.840
00:15:16.000 --> 00:15:26.399
00:15:26.480 --> 00:15:34.240
00:15:34.240 --> 00:15:35.279
00:15:35.279 --> 00:15:41.759
00:15:41.759 --> 00:15:44.320
00:15:44.720 --> 00:15:48.000
00:15:48.000 --> 00:15:56.559
00:15:56.559 --> 00:15:57.840
00:15:57.840 --> 00:16:07.440
00:16:07.440 --> 00:16:08.960
00:16:08.960 --> 00:16:16.639
00:16:16.639 --> 00:16:23.519
00:16:23.519 --> 00:16:28.559
00:16:28.559 --> 00:16:34.080
00:16:34.080 --> 00:16:38.639
00:16:38.639 --> 00:16:51.200
00:16:51.200 --> 00:16:58.720
00:16:58.720 --> 00:17:01.120
00:17:01.120 --> 00:17:15.039
00:17:15.039 --> 00:17:23.200
00:17:23.200 --> 00:17:37.279
00:17:37.279 --> 00:17:46.160
00:17:46.160 --> 00:17:53.759
00:17:53.759 --> 00:18:00.559
00:18:00.559 --> 00:18:05.200
00:18:05.200 --> 00:18:08.880
00:18:08.880 --> 00:18:17.119
00:18:17.119 --> 00:18:22.400
00:18:22.400 --> 00:18:24.960
00:18:24.960 --> 00:18:38.079
00:18:38.079 --> 00:18:40.160
00:18:40.160 --> 00:18:42.640
00:18:42.640 --> 00:18:45.519
00:18:45.519 --> 00:18:47.599
00:18:47.599 --> 00:18:57.039
00:18:57.039 --> 00:19:04.160
00:19:04.160 --> 00:19:08.480
00:19:08.480 --> 00:19:14.880
00:19:14.880 --> 00:19:20.160
00:19:20.160 --> 00:19:25.039
00:19:25.039 --> 00:19:30.319
00:19:30.319 --> 00:19:39.519
00:19:40.000 --> 00:19:42.319
00:19:42.559 --> 00:19:43.279
00:19:44.240 --> 00:19:45.839
00:19:45.839 --> 00:19:48.559
00:19:49.279 --> 00:19:51.440
00:19:51.440 --> 00:19:57.039
00:19:57.039 --> 00:20:06.319
00:20:06.319 --> 00:20:08.799
00:20:08.799 --> 00:20:10.799
00:20:10.799 --> 00:20:15.200
00:20:15.200 --> 00:20:20.240
00:20:20.240 --> 00:20:40.319
00:20:40.319 --> 00:20:43.119
00:20:43.519 --> 00:20:43.759
00:20:43.759 --> 00:20:44.000
00:20:44.480 --> 00:20:48.640
00:20:48.640 --> 00:20:52.400
00:20:52.400 --> 00:21:02.160
00:21:02.160 --> 00:21:05.039
00:21:05.039 --> 00:21:11.680
00:21:11.680 --> 00:21:15.039
00:21:15.039 --> 00:21:26.960
00:21:26.960 --> 00:21:46.720
00:21:46.720 --> 00:21:52.400
00:21:52.400 --> 00:21:59.359
00:21:59.359 --> 00:22:02.480
00:22:02.480 --> 00:22:04.240
00:22:04.240 --> 00:22:13.200
00:22:13.200 --> 00:22:16.880
00:22:16.880 --> 00:22:24.640
00:22:24.640 --> 00:22:26.079
00:22:26.079 --> 00:22:30.640
00:22:30.640 --> 00:22:41.519
00:22:41.759 --> 00:22:45.200
00:22:45.200 --> 00:22:48.160
00:22:48.160 --> 00:22:51.680
00:22:52.000 --> 00:22:54.880
00:22:55.200 --> 00:22:57.279
00:22:57.279 --> 00:23:01.119
00:23:01.119 --> 00:23:07.519
00:23:07.519 --> 00:23:11.759
00:23:11.759 --> 00:23:16.720
00:23:16.720 --> 00:23:23.920
00:23:23.920 --> 00:23:24.480
00:23:24.480 --> 00:23:26.960
00:23:26.960 --> 00:23:28.559
00:23:28.640 --> 00:23:30.960
00:23:30.960 --> 00:23:33.759
00:23:33.759 --> 00:23:37.119
00:23:37.119 --> 00:23:54.720
00:23:54.720 --> 00:23:57.279
00:23:57.279 --> 00:23:58.559
00:23:58.559 --> 00:24:06.240
00:24:06.240 --> 00:24:11.599
00:24:11.599 --> 00:24:20.160