Sept. 10, 2025

217 - Things that go wrong with the smoke control and how we fix them

The player is loading ...

In my personal view, an alarming truth about building fire safety lies in the gap between what's designed and what actually works in a building. After conducting 1000+ hot smoke tests in 200+ buildings, my experience is that most (maybe even 90%) of buildings had deficiencies in their smoke control systems, with 30% experiencing issues significant enough to potentially endanger occupants during a real fire. But it's not just about the problems. Good news - we have solutions.

Hot smoke testing stands as a powerful, yet underappreciated methodology that reveals what standard commissioning simply cannot. By creating controlled fires using methylated spirits and specialized smoke machines, we can observe how an entire building's safety ecosystem responds under fire conditions. The results are often eye-opening: systems operating in the wrong sequence, air flows disrupting smoke layers, pressurization fighting extraction, and critical components failing to activate when needed.

The most dangerous issue we encounter involves systems that don't "lock" to the first activated detector. This programming error causes safety systems to operate in areas far from the actual fire while leaving the fire location unprotected – a potentially life-threatening situation that's surprisingly common but easily fixable. Other frequent problems include excessive air velocity disrupting smoke buoyancy, extraction systems operating out of sequence, and auxiliary systems working against each other rather than in harmony.

What makes hot smoke testing so valuable is that it bridges the gap between aspirational safety (what designers intended) and actual safety (what the building delivers). Almost all identified issues can be corrected during commissioning, making this one of the most cost-effective safety investments possible. While the process may be disruptive and demanding, the alternative – discovering these failures during an actual emergency – is unthinkable.

Connect with me on LinkedIn to discuss implementing this approach in your projects and ensure your buildings aren't just designed for safety on paper, but truly deliver it when it matters most.

Recommended complimentary podcast episodes:

https://www.firescienceshow.com/136-fire-fundamentals-pt-6-the-fire-automation-in-a-building/
https://www.firescienceshow.com/033-science-theatre-or-engineering-polish-take-on-hot-smoke-test-with-piotr-smardz-and-janusz-paliszek/

----
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 - Introduction and Conference Recap

05:42 - The Value of Hot Smoke Testing

12:09 - How Hot Smoke Tests Work

19:10 - Aspirational vs. Real Safety Levels

25:46 - Scenario-Based Errors in Systems

36:02 - Performance Issues in Smoke Control

42:54 - System Cooperation Problems

48:25 - Benefits and Impact of Testing

54:39 - Closing Thoughts and Future Episodes

WEBVTT

00:00:00.642 --> 00:00:02.428
Hello everybody, welcome to the Fire Science Show.

00:00:02.428 --> 00:00:05.389
After gone missing last week, I'm back.

00:00:05.389 --> 00:00:10.172
Last week I've been in Slovenia for the European Symposium on Fire Safety Science, esfs.

00:00:10.172 --> 00:00:33.220
It was a really nice conference, very suited towards early stage career researchers and younger people in fire science, giving them an opportunity to present their work in progress, their research, to gain this important experience in doing conferences and prep themselves for upcoming submissions for IFSS, which are due in a few weeks.

00:00:33.220 --> 00:00:35.789
It was a really pleasurable conference.

00:00:35.789 --> 00:00:40.343
Big thanks to the steering committee for putting out this nice symposium.

00:00:40.343 --> 00:00:48.283
Great thanks to Grunde Jumas and his Frisbee team at Zag in Slovenia for being the local organizers of this event.

00:00:48.283 --> 00:00:51.951
Everything went perfect, so congratulations, big compliments to you all.

00:00:52.052 --> 00:01:08.215
I had a good time, but now it's back to reality and back to my original normal podcast schedule, and this week I have an episode that was supposed to go out last week but I finally finalized it and I'm really eager to share my thoughts with you.

00:01:08.215 --> 00:01:22.572
Today I am again solo on the microphone and I will be talking about some things that go wrong with smoke control in our buildings and actually multiple, multiple ways that fire engineers can investigate them, diagnose them and fix them.

00:01:22.572 --> 00:01:33.227
Them, diagnose them and fix them, and this fixing, I believe, is one of the easiest and cheapest ways to truly influence the safety of a building.

00:01:33.227 --> 00:01:39.924
This is a very strong opinion, but I promise I will argue for it later in the episode.

00:01:39.924 --> 00:01:54.566
This is really things that can improve safety in a measurable, in a quantifiable way that cannot be disregarded, and they are very cost effective if you compare them with other things that you can do to improve safety of your buildings.

00:01:54.566 --> 00:02:02.831
So I guess I'll stop now and just invite you to listen to the episode, because just after the intro you will find much more about the topic.

00:02:02.831 --> 00:02:05.200
Let's spin the intro and jump into the episode, because just after the intro you will find much more about the topic.

00:02:05.200 --> 00:02:07.828
Let's spin the intro and jump into the episode.

00:02:12.020 --> 00:02:13.568
Welcome to the Firesize Show.

00:02:13.568 --> 00:02:17.102
My name is Wojciech Wigrzynski and I will be your host.

00:02:17.102 --> 00:02:46.603
The FireSense Show is into its third year of continued support from its sponsor, ofar Consultants, who are an independent, multi-award-winning fire engineering consultancy with a reputation for delivering innovative safety-driven solutions.

00:02:46.603 --> 00:03:06.883
As the UK-leading independent fire risk consultancy, ofr's globally established team have developed a reputation for pre-eminent fire engineering expertise, with colleagues working across the world to help protect people, property and the plant, established in the UK in 2016 as a startup business by two highly experienced fire engineering consultants.

00:03:06.883 --> 00:03:16.432
The business continues to grow at a phenomenal rate, with offices across the country in eight locations, from Edinburgh to Bath, and plans for future expansions.

00:03:16.432 --> 00:03:24.969
If you're keen to find out more or join OFR Consultants during this exciting period of growth, visit their website at ofrconsultantscom.

00:03:25.981 --> 00:03:27.106
And now back to the episode.

00:03:27.106 --> 00:03:29.105
So hello again.

00:03:29.105 --> 00:03:34.193
It's great to see you after the intro, great to see you after the music.

00:03:34.193 --> 00:03:40.366
Statistically, you already made your mind to listen to this podcast episode, for which I am very thankful to you.

00:03:40.366 --> 00:03:42.687
There's a low chance that you're going to switch right now.

00:03:42.687 --> 00:03:44.426
So I can tell you the truth.

00:03:44.759 --> 00:03:57.461
This is a Hot Smoke Test episode and for some reason, when we mention hot smoke testing to our colleagues outside of Poland, they usually make big eyes and they do not really appreciate the method.

00:03:57.461 --> 00:04:09.848
I sometimes have a feeling this is some sort of alternative medicine of fire or something that people just don't like, or they don't trust it or they don't see big value in it.

00:04:09.848 --> 00:04:17.086
I always hear things like ah, it's theatrical, it's a demonstration, it's nowhere close to real fire.

00:04:17.086 --> 00:04:23.680
Some of those things are very true about hot smoke testing but it doesn't mean it cannot provide value to your building.

00:04:23.680 --> 00:04:35.987
And I think by doing those hot smoke tests in Poland a lot, doing them really a lot, we have found ways to actually make them in such a way that they bring a lot of value.

00:04:35.987 --> 00:04:42.245
So in the past, my team I think we've done like 200 buildings with hot smoke tests.

00:04:42.245 --> 00:04:48.509
Really there were years where we would be doing like two, three buildings a month with hot smoke tests.

00:04:48.509 --> 00:04:50.206
That was a really intense period.

00:04:50.206 --> 00:04:51.505
Now it's settled down a little bit.

00:04:51.505 --> 00:04:56.228
Now our hot smoke testing is mostly in critical infrastructure and tunnels.

00:04:56.228 --> 00:05:00.206
There are other companies that provide those services to buildings at large.

00:05:00.740 --> 00:05:14.262
But indeed it's a wealth of experience, wealth of observations related to performance of the smoke control and auxiliary devices that work together with smoke control and why I think it's worthy of a podcast episode.

00:05:14.262 --> 00:05:25.355
Because in many buildings, in many cases that we have tested, in many hot smoke tests that we've performed, those systems did not really perform like they were supposed to.

00:05:25.355 --> 00:05:28.165
I had a statistic.

00:05:28.165 --> 00:05:42.923
It's not extremely precise, more like a gut feeling, but in approximately 90% of the buildings 9 out of 10, we found some issues, we found some problems and I would say in 3 out of 10,.

00:05:42.923 --> 00:05:54.684
The issues were quite significant, like if there was a fire that would develop in the same location where we had the hot smoke test and the stuff would go like in the hot smoke test.

00:05:54.684 --> 00:06:02.144
The outcomes for the people in the building would not be great in terms of what the systems provided them with.

00:06:02.144 --> 00:06:09.526
So that's quite a staggering statistics, and when I talk with my colleagues who do hot smoke tests, it's not just my opinion.

00:06:09.526 --> 00:06:18.127
A lot of them share the same view that many of the buildings have different levels of faults in the systems because the systems are so inherently complicated.

00:06:18.720 --> 00:06:24.750
Now, in the beginning of the episode, I also mentioned that it's a cheap way to provide safety.

00:06:24.750 --> 00:06:26.245
I would like to elaborate on that.

00:06:26.245 --> 00:06:38.500
If you think about delivering safety to a building, imagine you have your building which has nothing in it in terms of fire safety systems and you want to spend some money to increase the safety in that building.

00:06:38.500 --> 00:06:43.992
You can purchase some sort of equipment system, smoke alarm systems, smoke control systems.

00:06:43.992 --> 00:06:53.742
Whatever you put into that, most likely spending that money will lead to an increase of safety in the building that you are designing.

00:06:53.742 --> 00:07:00.706
But if you want more safety, well, obviously you can put more systems in it and as the number of systems grows.

00:07:01.247 --> 00:07:07.302
One thing that I think is universally true is that you start to see some diminishing returns on safety.

00:07:07.302 --> 00:07:21.302
Like every another system, every another thing you put into your building is going to provide less and less safety compared to if that solution was used standalone in a building that has nothing.

00:07:21.302 --> 00:07:33.266
So by simply introducing a smoke alarm system, you will tremendously increase the safety of your building, but if you double the number of sensors, you're not going to increase the safety that much.

00:07:33.266 --> 00:07:42.665
And if you triple the number of sensors in your building, I'm not sure if you will even gain much safety at all while you keep spending your money to put new stuff in the building.

00:07:42.665 --> 00:08:08.521
And the other thing that I also think is universally true is that when you design a set of systems, when you design a building, when you design what goes into the building, you probably don't explicitly claim that, but you end up with some level of safety in that building, some level of fire safety in that building, and this is not yet safety that is in the building.

00:08:08.521 --> 00:08:14.401
This is the level of safety that can be in the building if everything works right.

00:08:14.841 --> 00:08:19.452
So I like to call it aspirational level of safety or perhaps design level of safety.

00:08:19.452 --> 00:08:22.826
This is level which the building can get to.

00:08:22.826 --> 00:08:33.393
That's a point at which you can get to if everything in your building works, operates in unison, works as expected and nothing goes wrong with it.

00:08:33.393 --> 00:08:51.272
Now, in reality, as I mentioned, nine out of 10 buildings have some flaws, which means some of those safety features have not operated to the level they were expected to, which means that this aspirational level of safety is very rarely achieved by the systems in the building.

00:08:51.272 --> 00:08:54.746
And here we are talking commissioning of the building.

00:08:54.746 --> 00:08:56.711
Maybe perhaps I should have mentioned that before.

00:08:57.259 --> 00:09:00.668
Here we are talking the final stage of building delivery.

00:09:00.668 --> 00:09:02.952
The building has just been completed.

00:09:02.952 --> 00:09:05.293
It's going through delivery process.

00:09:05.293 --> 00:09:07.261
It's about to be open to public.

00:09:07.261 --> 00:09:12.559
Everything's fresh, new, everything's fine-tuned, everything's started in it.

00:09:12.559 --> 00:09:20.381
So this is perhaps the best level, best quality of all the systems they will ever be in their life cycle.

00:09:20.863 --> 00:09:28.222
So if, at this point, you don't reach the aspirational level of safety, there's a fair chance that you will never do that.

00:09:28.222 --> 00:09:35.986
And also, you know, when you buy a ticket for any concert or venue, you see the tickets start from 20 bucks up.

00:09:35.986 --> 00:09:40.063
It means they may cost a lot more, but they will never cost less than 20 bucks.

00:09:40.063 --> 00:09:43.381
With aspirational level of safety, it's the same, just opposite.

00:09:43.381 --> 00:09:45.469
Like this is the maximum you will get.

00:09:45.469 --> 00:09:51.208
It's not that your smoke extraction systems will spontaneously achieve double the efficiency.

00:09:51.208 --> 00:10:01.110
It's not that your sensors will magically get five times bigger sensitivity to fires and will be able to tell them apart from false alarms in a better way.

00:10:01.110 --> 00:10:04.227
This level of safety will not spontaneously increase.

00:10:04.227 --> 00:10:07.759
So, basically, here we're talking about what you have designed in your building and that's level of safety will not spontaneously increase.

00:10:07.759 --> 00:10:15.660
So, basically, here we're talking about what you have designed in your building and that's the max the building can get to, and there will be faults along the way which will prevent the building from reaching that level.

00:10:16.162 --> 00:10:26.613
Now, the difference between the level of safety you have designed the building for the aspirational one, and the true one, the real level of safety you got your building to, is a gap.

00:10:26.613 --> 00:10:35.807
And here is where host, smoke testing, building commissioning and all the work I'm about to talk to you about is where it takes place this gap.

00:10:35.807 --> 00:10:49.802
Because in this gap, if you are able to identify what causes the gap, if you can identify how big the gap is and what actions take to close the gap.

00:10:49.802 --> 00:10:55.100
You start providing a true, real, tangible level of safety in your buildings.

00:10:55.100 --> 00:11:08.565
If a system is not operating and it's not providing a safety to your building occupants, and through commissioning, through your careful expertise, you fix the system and it starts operating.

00:11:08.565 --> 00:11:13.303
You've delivered the level of safety like if you just introduced the system to your building.

00:11:13.303 --> 00:11:22.993
I'm not sure if you follow me, follow my logic, but if a designer designs something and it's broken, it's like they've never achieved their design.

00:11:22.993 --> 00:11:25.037
You know it's like if the system was not there in the building and if you fix it, it's like they've never achieved their design.

00:11:25.037 --> 00:11:29.634
You know it's like if the system was not there in the building, and if you fix it, it's like if you put it in the building for the first time.

00:11:29.634 --> 00:11:31.264
It's the same gain in safety.

00:11:31.264 --> 00:11:36.856
I'm not saying that commissioner is the only one who did that, it's the design team and everyone else.

00:11:36.856 --> 00:11:43.230
Of course, I'm not speaking about credit, but I'm speaking about what has finally been delivered in the building, what has finally been delivered in the building.

00:11:43.230 --> 00:11:55.332
And this is why I so believe in those methods, because this is the point where I can make sure that my buildings are as safe as they have always intended to be.

00:11:56.061 --> 00:11:57.365
Now, how do we do that?

00:11:57.365 --> 00:11:59.105
We do them with hotspot tests.

00:11:59.105 --> 00:12:04.772
As I mentioned, this is our tool of choice in the commissioning stage of the building.

00:12:04.772 --> 00:12:09.211
We have done our own twist on this methodology.

00:12:09.211 --> 00:12:12.671
A lot of people in different countries use this method in different ways.

00:12:12.671 --> 00:12:15.107
We have our own twist of it.

00:12:15.107 --> 00:12:24.493
Actually, there was an episode early in the podcast about hot smoke testing already where I talked with Janusz and Kjotr from InBeppo on how to perform hot smoke tests.

00:12:24.493 --> 00:12:27.229
You're very welcome to revisit that one.

00:12:27.229 --> 00:12:28.566
That was like 200 episodes ago.

00:12:28.566 --> 00:12:42.048
Anyway, poland has its own twist on hot smoke testing and our method is largely focused on qualitative assessments of smoke control system and all the systems that interconnect with smoke control.

00:12:42.048 --> 00:12:54.561
So stuff that detects fires, stuff that issues electrical signals tos fire, stuff that issues electrical signals to the network, stuff that powers up ventilators and other devices in the building, everything interconnected with smoke control is actually undergoing the hot smoke test.

00:12:54.561 --> 00:12:57.006
It's not just the test of a fence in your building.

00:12:57.006 --> 00:12:58.269
No, it's much more.

00:12:58.269 --> 00:13:02.626
It's everything that consists of the safety layer of the building.

00:13:03.008 --> 00:13:21.753
To start the test, you first have to plan it out, which means you have to build your understanding of the building and the systems that you will be dealing with, which is already stage one of the assessment, because you look into the drawings, you look into technical designs and you can already understand what to expect from the building.

00:13:21.753 --> 00:13:26.125
Often, in the building, it's very different than what you've expected, but here you can build up some expectation towards how the building.

00:13:26.125 --> 00:13:31.447
Often in the building, it's very different than what you've expected, but here you can build up some expectation towards how the building will look like and what you will be dealing with.

00:13:31.447 --> 00:13:42.606
This is also a point in time where you can capture some little errors in the design, if they are still present in the design and perhaps they're fixable by the designer even before you get into the building.

00:13:42.606 --> 00:13:46.754
Of course, if you went to a solid third party, there should be less and less of those.

00:13:46.754 --> 00:13:53.121
In Poland we don't have that culture of third party, so perhaps this is the moment when we go through that, perhaps that's we experienced this.

00:13:53.121 --> 00:14:06.201
But again, this stage allows you to prep for the hot smoke test, plan it, the safety of it, and it's critical, in my opinion, to really do the hot smoke testing well if you have to prepare well for it.

00:14:06.201 --> 00:14:06.844
As for the setup.

00:14:07.245 --> 00:14:09.693
There are two components to the hot smoke testing.

00:14:09.693 --> 00:14:12.341
One is your hot smoke generators.

00:14:12.341 --> 00:14:18.461
So we are using commercial devices produced in the UK by a company called Concept Smoke.

00:14:18.461 --> 00:14:19.524
They're not an affiliate.

00:14:19.524 --> 00:14:22.152
I'm not an affiliate, they're not sponsoring this podcast.

00:14:22.152 --> 00:14:31.912
I am simply using the devices of those people for more than 16 years and I'm very happy with how they work and how reliable they are.

00:14:31.912 --> 00:14:34.828
We use machines called Vulcan.

00:14:34.980 --> 00:14:43.412
There was another machine called V-Count that we have used in the past and this, basically, is a big industrial generator of smoke.

00:14:43.412 --> 00:14:47.250
It is a different smoke than you would have from your theatrical generator.

00:14:47.250 --> 00:14:49.609
So this one is not based on paraffin.

00:14:49.609 --> 00:15:01.159
This one is based on material which is named smoke mineral oil 180, if I'm not wrong and apparently this is a fraction of oil which gets evaporated.

00:15:01.159 --> 00:15:13.298
It's turned into aerosol in the machine and it is released as a white smoke, and the benefits of this particular aerosol is that it is very stable in the air.

00:15:13.298 --> 00:15:22.701
It has a very long life cycle, so you can do your hot smoke test and it's going to stay for hours in the air until it is removed.

00:15:22.701 --> 00:15:31.128
Therefore, it makes it very suitable for those types of tests which may take an hour, two, three, depends on some of the test data.

00:15:31.961 --> 00:15:33.147
You need a stable source.

00:15:33.147 --> 00:15:34.666
Paraffin will not give you that.

00:15:34.666 --> 00:15:35.804
It will disappear in the air.

00:15:35.804 --> 00:15:39.471
The particles will deteriorate as they age.

00:15:39.471 --> 00:15:43.080
The smoke produced by the machine is not also very hot itself.

00:15:43.080 --> 00:15:45.208
It's maybe 30 degrees centigrade.

00:15:45.208 --> 00:15:50.091
So it's not enough to create this strong effect of buoyancy.

00:15:50.091 --> 00:15:53.240
But still it's hot enough to basically fly upwards.

00:15:53.240 --> 00:15:58.788
So you have a chance to control it and the smoke itself is not toxic.

00:15:58.788 --> 00:16:03.211
This oil is apparently neutral to the human.

00:16:03.211 --> 00:16:08.322
It has some certificates that you can refer to when the client asks.

00:16:08.383 --> 00:16:11.071
I know that health and safety is sometimes a concern.

00:16:11.071 --> 00:16:18.467
A big concern in health and safety is the second element of the hot smoke test apparatus, which is the source of heat.

00:16:18.467 --> 00:16:32.167
So a very efficient source of heat in fire safety are fires and in here to generate a lot of heat, to generate strong, buoyant thermal plumes through which we inject the aerosol.

00:16:32.167 --> 00:16:35.706
We use open flames, we use pool fires.

00:16:35.706 --> 00:16:40.091
Of course it's not that we spill anything on the floor and set it on fire.

00:16:40.519 --> 00:16:42.607
We have trays, steel trays.

00:16:42.607 --> 00:17:03.138
Ours are dimension 0.5 by 0.5 meters, so a quarter of a square meter, and you fill them up with a few liters of methylated spirit methanol, sometimes ethanol, sometimes isopropanol, depends on the scenario that we're testing Basically low-rank alcohols.

00:17:03.138 --> 00:17:11.948
The benefit of them is that they don't generate tremendous heat release rate per unit area and they are very clean.

00:17:11.948 --> 00:17:14.709
They lead to very clean combustion, very little soot.

00:17:14.709 --> 00:17:23.769
If you burn methanol, you basically get no soot at all pretty much in the test, which is very convenient because you do not make the building dirty.

00:17:23.769 --> 00:17:34.469
With the hot smoke test you get more or less 100, maybe 200 kilowatts per tray, depending on the fuel, and that's the size of the fire that you will end up using them.

00:17:34.469 --> 00:17:39.768
The caveat is that if you use more altogether, they start interacting with each other.

00:17:39.768 --> 00:17:49.842
Therefore, it means each of them is going to generate more heat release rate than it would if they were used independently, which you have to be kind of aware of.

00:17:50.241 --> 00:17:54.069
That can be problematic and you don't want to be surprised by your own fire.

00:17:54.069 --> 00:17:59.287
This, of course, comes with extreme health and safety challenge.

00:17:59.287 --> 00:18:05.826
To run open flame in a building that's about to be delivered is quite a hazardous thing.

00:18:05.826 --> 00:18:16.079
So you really need to be mindful of what you are doing, how you are approaching it, how you are protecting things around you from direct flame contact, from heat radiation.

00:18:16.079 --> 00:18:17.707
You have to shield things up.

00:18:17.707 --> 00:18:22.929
You have to be very mindful of what's above you, because you are creating a thermal band plume.

00:18:22.929 --> 00:18:39.833
You may not have flames touching your ceiling you should not have that but you may have a thermal stream of 80, 90, 100 degrees above your source of the fire, which can already be enough to cause damage, especially if you have some fragile plastic elements above you.

00:18:39.833 --> 00:18:49.068
So you have to be extremely mindful about the space in which you are performing the hot smoke test and you have to prepare that space for those hot smoke tests.

00:18:49.068 --> 00:18:56.102
In the end, this may not be a real fire, but it's very close to what a real fire will do in the building.

00:18:56.102 --> 00:19:02.332
So, yeah, we need to be considerate about that.

00:19:03.713 --> 00:19:06.363
The hot smoke test starts by initiating the fire.

00:19:06.363 --> 00:19:14.015
You release the smoke and basically what happens from that point is something you should not play too much manually.

00:19:14.015 --> 00:19:21.653
What I mean is that you want to test the building automation, so you want to let the building automation act respond to the fire you've done.

00:19:21.653 --> 00:19:30.041
You don't want to start a hot smoke test and then trigger a manual alarm and then have a guy run stuff from their computer and operate the things for you.

00:19:30.041 --> 00:19:33.861
No, you want the building to respond to this, because that's the point of the Hot Smoke Desk.

00:19:34.342 --> 00:19:41.846
You basically sit down and watch, you take photos, you take notes, you look into operation of different devices.

00:19:41.846 --> 00:19:43.308
You walk around the building.

00:19:43.308 --> 00:19:45.904
You basically try to see as much as you can.

00:19:45.904 --> 00:19:49.781
You record it to get additional layers of information from spaces.

00:19:49.781 --> 00:19:53.029
You were not physically present when you were walking around the room.

00:19:53.029 --> 00:20:02.284
And after the test finishes, after you conclude the test, after some minutes, you take the logs of the system and you look into the logs.

00:20:02.284 --> 00:20:05.212
What has operated at what point of time?

00:20:05.212 --> 00:20:09.921
This is a very critical information that you have to extract after the fire test.

00:20:09.921 --> 00:20:16.702
So you need to be aware what happened and when to really give an interpretation of the outcomes of the test.

00:20:16.702 --> 00:20:17.645
This is very important.

00:20:17.806 --> 00:20:20.271
We also do some measurements along the fire tests.

00:20:20.271 --> 00:20:30.009
We usually would measure extraction rates and at the openings that supply air extraction rate, at openings that extract air and smoke from your building.

00:20:30.009 --> 00:20:33.028
So those are fundamentally important.

00:20:33.028 --> 00:20:38.730
We would perhaps measure pressure differences, we would perhaps measure forces on the doors.

00:20:38.730 --> 00:20:51.559
Those are things that you can do while doing hot smoke tests and they provide you additional level of information and in the end, once you have all of that, you run your analysis level of information and in the end, once you have all of that, you run your analysis.

00:20:51.559 --> 00:21:05.989
By analysis I mean you look into the plan, you look into the design and you compare the design state of the operation of the systems, the performance, this aspirational performance that was designed for the building, and the performance you really got on your building in the hot smoke test indicating where the safety gap is.

00:21:06.009 --> 00:21:07.895
If you find anything, sometimes they're very obvious when you find an is.

00:21:07.895 --> 00:21:18.430
If you find anything, sometimes they're very obvious when you find an error, sometimes you have to go through the logs to see that something has not operated like it should have went.

00:21:18.430 --> 00:21:23.825
So in here we do a lot of work to prepare a final analysis.

00:21:23.825 --> 00:21:29.866
Unfortunately, this work has to be done very quickly because it's a building undergoing a commissioning stage, so there are no months to write a report.

00:21:29.866 --> 00:21:32.997
Work has to be done very quickly because it's a building undergoing a commissioning stage, so there are no months to write a report.

00:21:32.997 --> 00:21:35.182
It has to happen very quickly, almost overnight.

00:21:35.182 --> 00:21:39.770
You have to deliver those reports extremely robustly and at speed.

00:21:39.770 --> 00:21:52.191
But we can do that and this is where we can provide value to the building by identifying the gap between the real level of safety and the aspirational level of safety in the building.

00:21:52.980 --> 00:22:04.569
Now, an important thing to understand about hot smoke tests and perhaps this is the source of the confusion why a lot of people don't like them, why they think they're useless is what you cannot get from a hot smoke test.

00:22:04.569 --> 00:22:16.481
And one thing that you cannot confirm with a hot smoke test is that the performance of extraction, the smoke control system in your building, will be sufficient for your design fire.

00:22:16.481 --> 00:22:23.720
This is something you are unable to quantify and confirm with a hot smoke test.

00:22:23.720 --> 00:22:29.333
The reason is that, as I mentioned, the source of heat would be a few hundred kilowatts per a tray that we use.

00:22:29.333 --> 00:22:31.587
We may use four trays in a large shopping mall.

00:22:31.587 --> 00:23:01.545
We may use six, eight trays in a tunnel, so our fires could go up to one or two megawatts even, but that's still far below the usual design fires you would use for those spaces in your design stage, which means you're not really exposing your system to the maximum size of the fire you have envisioned in that building for those devices to work at, and it's simply impossible to reach those levels until you make a real big fire in the building, and you don't want to do that.

00:23:01.545 --> 00:23:02.704
You don't want to damage stuff.

00:23:02.704 --> 00:23:16.816
So if your system works in a hot smoke test, it doesn't necessarily mean it's going to achieve the level of performance you expect in it in terms of providing sufficient extraction capacity, maintaining smoke layer height, etc.

00:23:16.816 --> 00:23:19.366
In a full scale fire.

00:23:19.839 --> 00:23:35.369
There were attempts to scale up hot smoke tests to say, okay, if I have smoke layer height of this much in a hot smoke test of this size, what does it mean for a fire of much larger size and a particular hot smoke layer height?

00:23:35.369 --> 00:23:37.171
What is going to be in the real case?

00:23:37.171 --> 00:23:40.733
The thing is they are very difficult to apply.

00:23:40.733 --> 00:23:47.278
I don't really appreciate the scaling with fruit number that people try to do for those.

00:23:47.278 --> 00:23:57.172
I see too many challenges with this scaling approach, so I don't really feel there's a real value to do this type of confirmation.

00:23:57.172 --> 00:23:57.914
There are simulations for that.

00:23:57.914 --> 00:24:46.707
Cfds is a tool that gives you this kind of answer and also, if you want to do that, you are required to run a lot of additional measurements in your hot smoke test, a lot of additional things that are necessary to give you this quantification, which takes a lot of your time, even days of your time, and my choice is that, if I have to spend a few days setting a single hot smoke test to give me this quantification, which I don't even think I need at that point, versus a very robust and quick hot smoke testing routine which allows me to run multiple hot smoke tests in a day of work in a building, I take the second one because it allows me to test so much more and unravel so much more gaps in the safety compared to the precise measurements and attempt to extrapolate the results of the hot smoke test to a full-scale scenario.

00:24:46.707 --> 00:24:49.003
So, yes, I am absolutely convinced.

00:24:49.124 --> 00:24:53.290
I cannot judge the system performance fully.

00:24:53.290 --> 00:25:02.640
I cannot tell you, yes, your system is absolutely sufficient for the design fire you have seen, because the capacity is enough in the hot smoke test.

00:25:02.640 --> 00:25:03.480
No, I cannot say that.

00:25:03.480 --> 00:25:06.423
However, the opposite is true.

00:25:06.423 --> 00:25:24.998
If I make a hot smoke test which is a smaller fire than you have designed the building for and the system fails to operate in that hot smoke test, so it does not manage the hot smoke test fire which is smaller, it is to some extent unlikely it's going to work in a full scale.

00:25:24.998 --> 00:25:31.286
Going to work in a full scale, perhaps in natural ventilator systems, where buoyancy plays a lot of role.

00:25:31.286 --> 00:25:33.250
This is not 100% true, but in most cases it is true.

00:25:33.250 --> 00:25:40.633
If the system is insufficient to clean the hot smoke test, it's probably insufficient to clean up a real fire in the real world.

00:25:40.633 --> 00:25:43.586
So, yeah, these are the things.

00:25:43.645 --> 00:25:46.962
That number one you have to be aware of what you cannot do with hot smoke test.

00:25:46.962 --> 00:25:48.315
Now let's talk about things that you cannot do with hot smoke test.

00:25:48.315 --> 00:25:50.346
Now let's talk about things that you can do with hot smoke test.

00:25:50.346 --> 00:25:55.969
So what kind of errors do we find in buildings and how do we approach fixing them?

00:25:55.969 --> 00:26:05.651
I would group them into three groups scenarios, performance and maybe ill cooperation between the systems.

00:26:05.651 --> 00:26:08.128
Yeah, I think that those will be good groups to start with.

00:26:08.128 --> 00:26:26.763
So we will start with scenarios and the design, operation of the systems in the building, and here we immediately jump into number one biggest issue we find during hotspot testing, which is potentially life-threatening, and unfortunately this issue has been identified in multiple buildings that we were working with.

00:26:26.763 --> 00:26:30.502
So before I tell you what it is, I need to give you a context.

00:26:31.244 --> 00:26:38.830
When you, as a designer, design a fire safety system in your building, you have to design some way of operation.

00:26:38.830 --> 00:26:51.903
Basically, we write things in our fire strategies like when the fire is detected in a compartment, number, this, this system should activate, this door should shut, this dampers should close, etc.

00:26:51.903 --> 00:26:54.450
We give a list of things that shall happen in the fire.

00:26:54.450 --> 00:26:58.724
This is, let's say, a write-up of what is supposed to happen.

00:26:58.724 --> 00:27:26.627
Now this has to get translated into technical language, which means if a detector, number A52, operates in zone number 1.7, it means that the dampers on the duct number 1.7.1, 1.7.2 will have to shut and the HVAC will have to operate in this way and the extraction fan number 7.1.A will have to extract at 50 Hz and the other fan will have to extract at 20 Hz.

00:27:26.627 --> 00:27:30.986
Extract at 50 hertz and the other fan will have to extract at 20 hertz.

00:27:30.986 --> 00:27:39.570
You know a very technical list of things that precisely define you the state of every single device in the building in case of a fire in particular location.

00:27:39.951 --> 00:27:47.588
We called it the matrix of operation and it's usually a gigantic Excel spreadsheet which contains all of this information.

00:27:47.588 --> 00:28:01.628
Now, this is not yet the delivery of the system in the building, because now this matrix has to be programmed into the smoke control panel or fire alarm panel or an integration device.

00:28:01.628 --> 00:28:10.773
Whatever is steering, driving the devices in your building, it has to be programmed to reflect this matrix of operation.

00:28:10.773 --> 00:28:19.368
So there are multiple steps, you know, from the designer, from the fire engineer, creating the strategy, into building this matrix, into programming in the building.

00:28:19.368 --> 00:28:24.044
There are multiple steps on the way and during those steps then there could be errors.

00:28:24.044 --> 00:28:44.800
Unfortunately, one of such errors is that the last person that does it, or perhaps the earlier, but usually it's the programmers, but from my experience it's usually at the programming stage, at the last stage, people do not lock the order of operations to the first instance of fire alarm.

00:28:44.800 --> 00:28:46.023
What I mean by that?

00:28:46.203 --> 00:28:49.152
If you have a fire is detected by its nearest sensor.

00:28:49.152 --> 00:28:59.432
Therefore, the first device that finds the fire, the first detector that is triggered by the fire, is usually the one that will be closest to the fire.

00:28:59.432 --> 00:29:05.405
By definition, it has to be the closest sensor to the fire that is going to operate.

00:29:05.405 --> 00:29:08.971
So this is your most reliable information.

00:29:08.971 --> 00:29:20.030
You will ever get in where the fire is located in your building because as more and more sensors get triggered, the information will be happening further and further away from the source of the fire.

00:29:20.030 --> 00:29:32.548
So you really want those devices to operate at the seat of the fire, and that's why you have to lock your alarm to the first location indicated and then operate for this location and this location only.

00:29:32.548 --> 00:29:42.784
The issue is we often find that the smoke travels through the building, it triggers different devices and eventually the alarm state could switch to a newly triggered device.

00:29:42.784 --> 00:30:03.671
So perhaps the smoke has traveled across the boundary between my smoke detection zones and suddenly the operations start for the second zone and the third zone and the fifth zone and eventually I have a building which operates the safety systems far away from the fire and does not operate the systems in the location of the fire.

00:30:03.671 --> 00:30:10.220
And this is, in my opinion, life threatening because, one, you do not provide any safety at the location of the fire.

00:30:10.220 --> 00:30:16.693
Two, actually, the performance of those devices for different zones could be adversely affecting your fire.

00:30:16.693 --> 00:30:23.334
So it's not just you're not providing the safety, you're actively making it worse than if the systems have not operated at all.

00:30:23.940 --> 00:30:25.988
This is a very major flaw.

00:30:25.988 --> 00:30:27.851
I've talked about it in the communications episode.

00:30:27.851 --> 00:30:28.750
I attribute this to miscommunication in fire engineering, which is a very major flaw.

00:30:28.750 --> 00:30:29.371
I've talked about it in the communications episode.

00:30:29.371 --> 00:30:35.708
I attribute this to miscommunication in fire engineering, which is a very difficult thing, very difficult skill.

00:30:35.708 --> 00:30:37.426
I don't blame the programmers.

00:30:37.426 --> 00:30:50.041
I see a reason why they do that and I also understand that fire engineers have to be vigilant about this and have to pick it up and have to fix it, because this is one of the biggest threats you have in your building.

00:30:50.202 --> 00:30:52.367
And it's easily fixable you just apply the lock.

00:30:52.367 --> 00:30:56.008
Sometimes it's super easy, it's just one click and it's there.

00:30:56.008 --> 00:31:01.949
Sometimes it takes you to write some complicated scripts and algorithms to fix it, but it is fixable.

00:31:01.949 --> 00:31:10.800
Another thing is when your scenarios involve a lot of things that have to happen simultaneously at the exact same second.

00:31:10.800 --> 00:31:14.148
Like it's easy to write after two minutes everything starts.

00:31:14.148 --> 00:31:22.681
But when we're talking about issuing thousands of signals to thousand different devices, it actually takes time to deliver those signals.

00:31:23.142 --> 00:31:45.289
We had a case where we were commissioning an extremely large car park under a very big building, a rectangular car park, tens of thousands of square meters, four big extraction inlet points in the corners of the car park, and when we were doing those tests the performance of the system was very random, like we really could not understand what the hell is happening, why the system sometimes works, sometimes does not work.

00:31:45.289 --> 00:31:50.048
It felt very, very odd and on paper everything seemed fine.

00:31:50.048 --> 00:31:59.065
So as an act of desperation, we've sent people to each of extraction rooms with a stopwatch and they were measuring what happens and when.

00:31:59.065 --> 00:32:09.787
And then from this exercise we found out that in one test, let's say, northeastern extraction point would activate immediately, south one would activate after two minutes.

00:32:09.787 --> 00:32:13.946
In another test, the northern would take five minutes and the southern would activate in one minute.

00:32:13.946 --> 00:32:26.593
The reason was that the system was absolutely overwhelmed by the number of signals to be issued and it took it minutes to send signals to particular elements of the system.

00:32:26.593 --> 00:32:37.691
And eventually those minutes have created this, you know mismatch in when the devices were operating in the car park, creating an extremely dangerous situation of the system not working.

00:32:37.691 --> 00:32:39.987
It was good on the paper.

00:32:39.987 --> 00:32:47.319
It's just that electronics that were used to deliver that have not matched, you know, the expectations.

00:32:47.882 --> 00:32:50.210
The solution was again fairly simple.

00:32:50.210 --> 00:33:02.460
We had to spread apart sending out the signals, we had to give some priorities and once this was done we could make sure that the critical elements start immediately and then other elements follow.

00:33:02.460 --> 00:33:05.349
So it was absolutely fixable.

00:33:05.349 --> 00:33:19.990
But again, I believe it would be not possible to capture that if it was not during the hot smoke deaths, because you could immediately see from the behavior of hot smoke layer and when the hot smoke went that something is very wrong with the system.

00:33:19.990 --> 00:33:23.446
That looks good on the paper and the measurements confirm that it should be good.

00:33:23.446 --> 00:33:41.012
Another scenario-based errors that we sometimes capture in hot smokes is that you probably want to subdivide your space into some sort of smoke control compartments where not everything operates at the same time in your building, and we sometimes see issues with those compartments in terms of how they are designed.

00:33:41.012 --> 00:33:45.632
Sometimes the real building behaves a little bit different than what the designer expected.

00:33:45.632 --> 00:33:48.925
Maybe there's a mismatch between the CFD and what has been built.

00:33:48.925 --> 00:33:50.710
In the end it's fixable.

00:33:50.829 --> 00:33:52.834
Usually you can move those boundaries a little bit.

00:33:52.834 --> 00:33:56.346
You can use different devices to establish new boundaries.

00:33:56.346 --> 00:33:59.241
You can sometimes split a bigger zone into two smaller ones.

00:33:59.241 --> 00:34:02.634
You can sometimes combine two smaller zones into one larger one.

00:34:02.634 --> 00:34:07.769
Those actions are helpful in fine-tuning the system performance.

00:34:07.769 --> 00:34:16.184
We once had, for example, a very wide balcony in a shopping mall which created an extremely wide spill plume which was problematic for the system.

00:34:16.184 --> 00:34:31.606
We've cut the balcony into half with a small curtain and it magically fixed the operations of the system because suddenly the spill plume was much narrower, much more narrow and easier to maintain, to contain, by the extraction system, for example.

00:34:31.606 --> 00:34:32.250
Things like that.

00:34:32.773 --> 00:34:37.371
Sometimes with jet fan systems, people don't appreciate how much momentum the jet fans introduce into your car park.

00:34:37.371 --> 00:34:53.711
If there is discrepancy between how much momentum you input into the car park and the extraction rate of your system, the jet fans will start moving your smoke around the building beyond the point where it's extracted, which creates some problematic situations where the smoke is just spilled into the building.

00:34:53.711 --> 00:34:57.840
And you can fix that by changing the zones in which the jet fans operate.

00:34:57.840 --> 00:35:02.211
You can trigger them off and perhaps fix that.

00:35:02.211 --> 00:35:09.052
Of course, this is not something that an engineer that runs hot smoke test has a sole, you know ability to do.

00:35:09.512 --> 00:35:13.250
In this stage you really need to work with the designer of the building and the third party.

00:35:13.250 --> 00:35:21.923
They have to be present at the hot smoke test, they have to be part of the process and if you find a gap, it's up to them to find a solution.

00:35:21.923 --> 00:35:32.673
You can assist them, you can advise them, but it's their them who have to fix their system and and, together with the designer, you can do those choices If the designer agrees yes, this is better.

00:35:32.673 --> 00:35:34.842
You can perhaps implement them in the building.

00:35:34.842 --> 00:35:37.869
Sometimes you have to run some CFD ad hoc for some changes.

00:35:37.869 --> 00:35:57.615
Sometimes you have to get some other opinions to support the change, but usually it's manageable and we have found that those designers, those people who design the building, are very helpful in the process because they are also very interested that their systems reach this aspirational level of safety they wanted them to reach in the first place.

00:35:57.615 --> 00:35:59.967
So that would be the scenario problems.

00:36:00.099 --> 00:36:02.108
Now let's talk about issues with performance.

00:36:02.108 --> 00:36:05.769
Let's perhaps start with performance of passive fire protection.

00:36:05.769 --> 00:36:09.650
So in general, walls, doors, gates, etc.

00:36:09.650 --> 00:36:11.543
Hot smoke testing is brutal.

00:36:11.543 --> 00:36:16.922
If you are supposed to have a fire compartment, which means nothing exits the compartment.

00:36:16.922 --> 00:36:24.380
If there's a hole, hot smoke test will exit the hole and you will see it on the other side of the wall and you will immediately be able to point out there's leakage in this wall.

00:36:24.380 --> 00:36:31.010
The fact that there is no leakage does not mean that the wall is perfect or it achieved some class of integrity or whatever.

00:36:31.010 --> 00:36:38.150
But if there is a hole, the hot smoke will pass through that hole and it will create a problem on the other side of the wall.

00:36:38.150 --> 00:36:40.882
You can rely on hot smoke test for that.

00:36:40.882 --> 00:36:42.266
You can trust me on that.

00:36:42.266 --> 00:36:48.750
It's very good to verify if we really sealed the fire compartments like we say.

00:36:48.750 --> 00:36:55.119
We are Other problems with performance, perhaps more about the smoke control systems.

00:36:55.621 --> 00:37:06.342
Well, you could say there's either too much or too less of the flow or extraction capacity that you want to have, and both will be problematic For me.

00:37:06.342 --> 00:37:16.431
If I had to rank the number one cause of failure of smoke control systems, I would say the inlet air supply is the most challenging one.

00:37:16.431 --> 00:37:20.427
Extraction not that often, but inlets very, very often.

00:37:20.427 --> 00:37:37.096
And I think people just do not appreciate how much momentum is within the air that you input into your building and how little is needed to disturb why our smoke control systems work in the evacuation phase.

00:37:37.096 --> 00:37:38.817
How can we protect people from smoke?

00:37:38.817 --> 00:37:41.373
You want the smoke to be away from the people, right.

00:37:41.373 --> 00:37:51.295
And because the smoke is hot, it has buoyancy, it will fly up, it will accumulate in a smoke reservoir, which means it is far from people and they can evacuate.

00:37:51.295 --> 00:37:59.054
This is the basic strategy how our smoke control systems operate, how we want them to operate and how we achieve safety through smoke control systems.

00:37:59.644 --> 00:38:10.918
Now, if you disturb this buoyancy, if you do not allow this layer to form or, even worse, you disturb that layer, you suddenly bring smoke to the people, which is a hazard and you do not want that.

00:38:10.918 --> 00:38:26.757
So you have to be very mindful of where you inject air so it does not mix with the buoyant hot smoke in a way that the smoke will lose its buoyancy and the smoke will fall, descend into the space where people are and cause hazard.

00:38:26.757 --> 00:38:32.606
We've seen that a lot of times because it's actually very little that is needed to disturb this.

00:38:32.606 --> 00:38:35.838
If you have one meter per second flows, you're pretty much safe.

00:38:35.838 --> 00:38:45.188
But if you have anything above two, two and a half meters per second that's already very strong flows that will cause disturbance in buoyancy.

00:38:45.188 --> 00:38:57.306
If you have five meters per second, that's a lot, that's really a lot of velocity in the compartment and it will interfere with your layers even far away from the point of injection.

00:38:57.306 --> 00:39:07.474
You have to be extremely mindful and when you find problems like that there are sometimes fixes to be applied so you sometimes can operate a different inlet point far away.

00:39:07.693 --> 00:39:13.416
The further away from the fire you can supply air and the closer to fire you can extract air.

00:39:13.416 --> 00:39:17.476
Those are your rules of TAMP, where you want them to be.

00:39:17.476 --> 00:39:21.516
So sometimes you can move the inlet point far away and it's solved.

00:39:21.516 --> 00:39:28.518
Sometimes you cannot do that and as a last resort, we sometimes set up ramp up functions on the devices.

00:39:28.518 --> 00:39:41.596
So basically, the device starts to operate at very low capacity and slowly, with time, builds up the capacity until it reaches the full potential, because the fires grow it's also a transient phenomenon.

00:39:41.596 --> 00:39:46.416
You probably do not need all of your extraction capacity very early.

00:39:46.416 --> 00:39:58.436
Therefore, you can build up this capacity within minutes and actually create space for evacuation to happen and yet reach the full potential when you expect the fire to be fully grown.

00:39:58.905 --> 00:40:01.735
Again, this is not the responsibility of the hot smoke engineer.

00:40:01.735 --> 00:40:15.692
It's the responsibility of the designer of the building, who participates in the process along with you and together you may find out what is suitable, delay time for this to make space for evacuation and get the performance you want.

00:40:15.692 --> 00:40:23.364
Sometimes you end up with not enough flows and we, as I said, you don't see that through hot smoke test directly, but unless it failed.

00:40:23.364 --> 00:40:31.748
But you can measure the velocities at your grills and estimate how much air is flowing into and out of your compartment.

00:40:31.748 --> 00:40:37.429
This is very efficient and you can indicate where the system is lacking the capacity which it should have.

00:40:38.052 --> 00:40:40.849
Sometimes there are issues with how the ducts are made.

00:40:40.849 --> 00:40:46.032
Sometimes there are programmable issues, like perhaps a wrong frequency on the frequency inverter.

00:40:46.032 --> 00:41:03.177
Sometimes, actually, in a very sealed spaces, like corridors of high-rise buildings, you may not have sufficient inlet air, which reduces the amount of extraction air you can have because it's a sealed space and it will also manifest itself with a high pressure gradient in that case.

00:41:03.177 --> 00:41:05.347
So you have to be mindful of that.

00:41:05.347 --> 00:41:10.748
Usually, if there's a problem like that, it's necessary to investigate and find the reason why the air is missing.

00:41:10.748 --> 00:41:21.369
Many times, like in the road tunnels when we had long docks, we have found like someone left a hatch open to some space and we lost like 30% of volume to that hatch.

00:41:21.369 --> 00:41:33.246
So if you measure the capacity and it's not meeting the design, there must be a reason why it is like that and you absolutely have to investigate why this mismatch is present.

00:41:33.809 --> 00:41:35.873
Finally, you may have temporal mismatch.

00:41:35.873 --> 00:41:43.780
Like I mentioned before, your systems will take time to operate, but sometimes the temporal mismatch, like I mentioned before, your systems will take time to operate, but sometimes the temporal mismatch is also with the sequence of operations.

00:41:43.780 --> 00:41:48.315
So you must open a damper before you start the fan.

00:41:48.315 --> 00:41:49.757
That's fundamental.

00:41:49.757 --> 00:41:59.746
You must close the HVAC dampers before you start the extraction in a very big building, because if your extraction runs to the maximum, it may be very difficult to close them.

00:41:59.746 --> 00:42:04.505
At that point, if you don't open the damper and operate the fan, you may break the damper.

00:42:04.505 --> 00:42:16.496
Trust me, we've seen doors being ripped apart from ventilation rooms, mechanical rooms, because something did not work and the damper has not opened when it was supposed to.

00:42:16.496 --> 00:42:27.469
So this mismatch may cause real damage in the building and, in the case of a real fire, may cause a very big problems in operation of the safety features of the building.

00:42:27.469 --> 00:42:28.913
You have to observe that.

00:42:28.913 --> 00:42:35.804
That's why you go through the logs after you've done your hot smoke test and that's how you indicate if everything happened.

00:42:35.804 --> 00:42:37.067
But it doesn't have to happen.

00:42:37.067 --> 00:42:46.831
It doesn't only need to happen, it has to happen in the very correct, precise order that you have designed the things to happen in, and this is absolutely critical.

00:42:47.592 --> 00:42:56.346
Finally, we reached the third group of issues ill cooperation of the system, and I would say it's mostly between pressurization systems and smoke control.

00:42:56.346 --> 00:43:02.016
That's a big challenge to me, the reason for that being, in Poland we use a lot of pressurization PDS systems.

00:43:02.016 --> 00:43:04.952
I know they are not that popular in the world.

00:43:04.952 --> 00:43:10.409
Perhaps the things will change in the future, but we use them very commonly in anything taller than 25 meters.

00:43:10.409 --> 00:43:12.637
I would probably have a pressurization system there.

00:43:12.637 --> 00:43:15.032
So yeah, it's my bread and butter.

00:43:15.032 --> 00:43:24.038
The problem is that those systems are very specific and therefore they are kind of designed in a silo independently.

00:43:24.038 --> 00:43:28.072
So smoke control is being designed independently, pds is designed independently.

00:43:28.144 --> 00:43:37.496
Of course there is some sort of coordination between the projects but it's different design concept, different goals of operation and sometimes those lead to some issues in the building.

00:43:37.496 --> 00:43:42.492
Perhaps you will have too much airflow from the PDS that extraction cannot take away.

00:43:42.492 --> 00:43:51.132
Perhaps the extraction will be too strong and it will increase the pressure difference for the PDS system beyond the region in which the system can operate.

00:43:51.132 --> 00:43:56.630
Both things can lead to excessive forces on the door handles which may prevent you opening the doors.

00:43:56.630 --> 00:44:02.335
It can cause an extreme air streams like too much air velocity in your building.

00:44:02.335 --> 00:44:03.356
You lose the buoyancy.

00:44:03.356 --> 00:44:07.432
You can have an adverse effect of the pressurization.

00:44:08.014 --> 00:44:34.329
In skyscrapers we often have pressurization serving as your main air inlet, as the makeup air, which is a brilliant strategy, very robust, very safe, but it needs to be very mindful how you do that, how you achieve that state, and in that case you are unable to design those systems in silos because they together have to work in a way that provides the performance level that you expect.

00:44:34.329 --> 00:44:41.840
This is a very big challenge actually, and it sounds simple, but in so many buildings we find problems with that.

00:44:41.840 --> 00:44:48.614
You indicate those problems by measuring the forces on the doors, by measuring the pressures, by measuring the flows, by observing the response.

00:44:48.614 --> 00:44:53.416
We've also developed a methodology, eventually, that we not only measure this.

00:44:53.416 --> 00:44:58.016
You know, you put an anemometer to the grill and you just have a reading.

00:44:58.016 --> 00:45:02.137
No, we have systems that continuously measure those parameters.

00:45:02.137 --> 00:45:12.340
We set up the devices in the building and we measure that as a function of time, to observe the temporal evolution of those flows and how they interact in also the time space.

00:45:12.340 --> 00:45:16.817
It's very important in the very sealed spaces in which you rely on those.

00:45:16.817 --> 00:45:19.233
So that's one way of ill cooperation.

00:45:20.085 --> 00:45:39.344
Another way is between the passive fire protection and the ventilation systems, especially when we talk about fire doors, which are normally open and they are released close to when the fire is observed, and fire gates that shut down the connections between big fire compartments, especially in car parks.

00:45:39.344 --> 00:45:47.512
So when you design a system for your fire compartment, what you would do, you would run a simulation for that compartment and that compartment only.

00:45:47.512 --> 00:45:58.597
The assumption is that the fire cannot exit that compartment, so it's useless to account for what's outside right, and I would agree that this is the standard procedure for fire simulations.

00:45:58.597 --> 00:46:15.994
However, in reality, there are things that can go wrong with those closing devices so they can be physically inactive, they can be turned off, they can be unpowered, they can be broken, there can be a physical object blocking the operation of the device, there can be a wrong signal sent to the device.

00:46:15.994 --> 00:46:32.534
There's a ton of things that can go wrong, leading to an opening not be shut when it's supposed to, and in that case you will have a mischievous airstream in your building that may actually completely change the way your smoke control system operates.

00:46:32.534 --> 00:46:40.748
So this is a very big thing and, again, something very easily identified and found during hot smoke testing and perhaps not that easy.

00:46:40.748 --> 00:46:47.449
When you are just testing your systems by hand, you know running scenarios manually, etc.

00:46:47.449 --> 00:46:51.405
Much easier to pick those up in hot smoke tests and those things are big.

00:46:52.086 --> 00:46:58.478
Another thing is sometimes when especially again when the premises are sealed.

00:46:58.478 --> 00:47:16.175
And sometimes, especially again when the premises are sealed, if you have a sealed compartment in your car park, for example, and you extract too much air from it, you create a big pressure difference between this fire compartment and another fire compartment.

00:47:16.175 --> 00:47:22.672

217 - Things that go wrong with the smoke control and how we fix them

Listen On

Recent Episodes

Support On