June 12, 2024
155 - New Guideline for PV Fire Safety with Grunde Jomaas
The player is loading ...
Misconceptions in fire science are a strange thing. You present countless proof, publish research papers, and carry conversations, and yet… they live their own lives—spreading with no control and cluttering communication. One space is full of them—the fire safety of photovoltaic installations on flat roofs. In the Fire Science Show, we already had two very powerful episodes on PV—one with Jens Kristensen and one more recently with Reidar Stølen.
In today's episode, we highlight a new guideline document published by the team at FRISSBE led by Prof. Grunde Jomaas. I invited Grunde to walk me through their document, and I took this opportunity to learn the origins, sources and inspirations for all the aspects they covered. The guideline takes you through ignition, fire spread, building roof construction and firefighting, giving a unique holistic overhaul of the topic. Most importantly, it is short, concise, and written in the most approachable way. This is a really good example of how communication in fire science should look.
You can access the guideline directly here: https://www.frissbe.eu/upload/files/FRISSBE-ZAG%20BAPV%20Fire%20Safety%20Guideline%20May%202024%20v3.pdf
To complement this piece, you may also read this article on the background: Rus, N., Jomaas, G. (2024) PV guidelines – are the recommendations sufficiently evidence-based?
As my personal recommendation, Grunde is a top communicator on LinkedIn, and you should follow him. He also runs a weekly newsletter, Burning Matters. I highly recommend this resource!
----
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.
Transcript
WEBVTT
00:00:00.442 --> 00:00:01.806
Hello and welcome to the FireScience Show.
00:00:01.806 --> 00:00:14.212
Whenever I post an episode on photovoltaics and new technologies in the sustainable world, those episodes get a lot of attention and they seem to be very appreciated by the audience.
00:00:14.212 --> 00:00:17.751
So here I am with another episode of that kind.
00:00:17.751 --> 00:00:22.852
It's going to be about photovoltaics, and PVs are quite an interesting story.
00:00:22.852 --> 00:00:35.146
Two years ago I had Jens in the podcast, who I thought cleared most of the misconceptions about PV panels, especially highlighting the role of the roof construction and treating them as a system.
00:00:35.146 --> 00:00:46.831
Then, sometime later, I had Radar from Norway, who was talking about building integrated photovoltaics, also reemphasizing some of the same thoughts as Jens did.
00:00:46.831 --> 00:00:52.067
So with all this information out there, why there would be misconceptions anymore.
00:00:52.067 --> 00:00:53.463
And yet they are.
00:00:53.463 --> 00:01:02.219
Yet you see them every day on LinkedIn and around People not really getting what is the fire issue with the photovoltaic panels.
00:01:02.823 --> 00:01:14.343
So here today let's try once again to clear the misconceptions around the fire safety of PV panels, and for that I've invited Professor Grunde Jumas from Frisbee.
00:01:14.343 --> 00:01:17.230
That's a part of ZAK in Slovenia.
00:01:17.230 --> 00:01:25.311
Grunde is my good friend, we do some research together and, most importantly, we are figuring linkedin out together.
00:01:25.311 --> 00:01:37.412
Since he's 10 times bigger than I am on linkedin, I refuse to call him anything else than grunders and say, but yeah, we're having a lot of fun there trying to get some good communications to the people.
00:01:37.412 --> 00:01:40.825
And yes, communication, that's the point of today's episode.
00:01:40.825 --> 00:02:05.927
I've invited Grunder to talk about PV not just because he has a decade of experience testing and experimenting with photovoltaics and the fire safety, not just because he's a great communicator, but his group, frisbee, has just released a brilliant piece of communication, a guideline on fire safety, photovoltaic panels on flat roofs, and this is the direct reason why we are here today.
00:02:05.927 --> 00:02:11.484
We're going to discuss that guideline in depth, but you know not just what's written inside we're going to talk about.
00:02:11.484 --> 00:02:13.552
Where did all of this come from?
00:02:13.552 --> 00:02:17.284
What is the experimental background for all the claims that we discussed?
00:02:17.284 --> 00:02:24.784
From the book and also other guidelines that exist out there, because that's obvious, that it's not just the sole source of knowledge.
00:02:25.366 --> 00:02:39.786
I think this episode will be very beneficial if you have to deal with PV panels and if you don't, well, perhaps sometime in the future you will, so still worth listening.
00:02:39.786 --> 00:02:41.413
It's very good to listen to this episode having a PV in front of your eyes.
00:02:41.413 --> 00:02:52.205
I know many of you will be jogging or driving, so that's going to be harder, but if you have ability to open up the guideline and just look at the contents as we discuss them, I think it's going to be much better experience for you.
00:02:52.205 --> 00:02:57.644
The link to the guideline is in the show notes and the podcast episode is right behind the intro.
00:02:57.644 --> 00:03:00.406
Let's spin it up and jump episodes.
00:03:00.406 --> 00:03:07.432
Welcome to the Firesize Show.
00:03:07.432 --> 00:03:10.877
My name is Vojtěch Vyngřínský and I will be your host.
00:03:10.877 --> 00:03:27.311
This podcast is brought to you in collaboration with OFR Consultants.
00:03:27.879 --> 00:03:30.179
Ofr is the UK's leading fire risk consultancy.
00:03:30.179 --> 00:03:41.112
Its globally established team has developed a reputation for preeminent fire engineering expertise, with colleagues working across the world to help protect people, property and environment.
00:03:41.112 --> 00:03:56.925
Established in the UK in 2016 as a startup business of two highly experienced fire engineering consultants, the business has grown phenomenally in just seven years, with offices across the country in seven locations, from Edinburgh to Bath, and now employing more than a hundred professionals.
00:03:56.925 --> 00:04:08.586
Colleagues are on a mission to continually explore the challenges that fire creates for clients and society, applying the best research experience and diligence for effective, tailored fire safety solutions.
00:04:08.586 --> 00:04:19.641
In 2024, ofr will grow its team once more and is always keen to hear from industry professionals who would like to collaborate on fire safety futures this year.
00:04:19.641 --> 00:04:22.545
Get in touch at ofrconsultantscom.
00:04:22.966 --> 00:04:28.553
Hello everybody, I'm here today with Professor Grunde Jumas from Frisbee Project in Slovenia.
00:04:28.553 --> 00:04:33.029
Hello, grunde, hey, wojciech, good to see you, sensei, on the podcast.
00:04:33.029 --> 00:04:35.086
Happy that you joined us.
00:04:35.086 --> 00:04:41.490
And the topic is of great importance, that's the photovoltaics on flat roofs.
00:04:41.490 --> 00:04:47.487
The direct reason is a guideline that your group has released something like a month ago.
00:04:47.487 --> 00:04:56.630
Tomorrow there's a big webinar, from what I've heard, and I know there's a lot of interest on PV panels in the fire science community.
00:04:56.630 --> 00:05:03.009
Thanks for getting on this important subject, but what made you do the guidelines?
00:05:03.009 --> 00:05:06.629
What was the inspiration to put the knowledge into a guideline actually?
00:05:06.629 --> 00:05:19.730
Well, I guess the direct answer was, as you can read in the guideline, it came as a result of a meeting that was held in Brussels last year, hosted by NFPA, fm Global and Rockwool
00:05:19.730 --> 00:05:27.086
, where there was also released a document by the NFPA Research Foundation with the slides, some discussions from that meeting.
00:05:27.086 --> 00:05:34.733
But then we realized that we wanted to try to put out what is hopefully a clear communication on the topic.
00:05:34.733 --> 00:06:03.432
And also the second part is that we can't escape this in a discussion between the two of us LinkedIn our friend and foe in a discussion between the two of us linked in our our friend and foe, I could see from a lot of debates there's a lot of things that people don't get that say that I take for granted some key concepts that we try to portray in the guideline, where people think that, oh, it's the bitumen that is the problem, or it's this specific membrane or it's this panel that is the problem.
00:06:03.853 --> 00:06:28.286
so there was a lack of understanding of what one of the main messages in the guideline is that you have to think of it as a system, not just of the individual components or products involved it's funny that you say that, because I also had jens in the podcast like a hundred episodes ago, long time ago, and one of the most popular episodes of the Fire Science Show, jens was your PhD student.
00:06:28.286 --> 00:06:35.125
He'd done his PhD on burning the PV panels, basically on different membranes, different settings and what you just said.
00:06:35.125 --> 00:06:37.519
You know, he kind of cleared that out right.
00:06:37.519 --> 00:06:41.607
There's paper out of that, many papers out of that, there's his PhD thesis.
00:06:41.607 --> 00:06:49.627
You could argue that the information is out there for anyone who is like a Google search away from that information.
00:06:49.627 --> 00:06:52.649
Yet people would still share the misconceptions.
00:06:52.649 --> 00:06:54.466
Yet people would still not seek it.
00:06:55.060 --> 00:07:06.742
We've been talking for some time about having a discussion on your podcast and I said, well, jens already told everything in in terms of pv on the podcast, I don't know.
00:07:06.742 --> 00:07:12.221
And then you have radar, complement that with some slope proof and vipv and other things.
00:07:12.221 --> 00:07:17.512
Then we decide, okay, now the guy and we have a an excuse, uh, to revisit it.
00:07:17.512 --> 00:07:27.675
But it also because the the number of times where I've copied jens's googleolar page and put it as an answer to people in questions on LinkedIn.
00:07:27.675 --> 00:07:31.408
I say just go here, the papers here are telling all of this.
00:07:31.408 --> 00:07:39.809
Or, similarly, check out these papers by Raida, check out these people by the group in Malaysia and their references, of course, and so on.
00:07:39.809 --> 00:07:44.591
But it is amazing how many times you have to say some things.
00:07:44.591 --> 00:07:58.512
And it's actually interesting because you say Jens is PhD, but Jens actually did also a master's on the topic at BTU that started, and before that we did an industry project and that started it all.
00:07:59.000 --> 00:08:04.333
And I talked with another former student just some months ago where I said, oh, can I use this video?
00:08:04.333 --> 00:08:08.923
And he said, well, I don't know if it's really in the public domain, I have to check with some people.
00:08:08.923 --> 00:08:13.988
That probably don't work here, but it's almost 10 years ago and I said because I wanted to show it.
00:08:13.988 --> 00:08:19.821
And he's like, yeah, the thing is he said we already had the answer at that point, we just didn't know it.
00:08:19.821 --> 00:08:28.809
So you know, it just shows how it's a slow process for the messages to get out there and for people to get the details.
00:08:28.809 --> 00:08:37.777
And of course there are many stakeholders that also don't necessarily want to have the debate, that want it to be communicated the way it is.
00:08:37.777 --> 00:08:43.802
So they try to dilute the information pool a little bit.
00:08:43.802 --> 00:08:45.568
And you know there's nothing wrong with that.
00:08:45.568 --> 00:08:52.485
So to speak, everybody has the right to promote their own solutions, as long as they're safe, I guess.
00:08:53.028 --> 00:08:53.931
Unless they're not fireballs.
00:08:53.931 --> 00:08:56.837
Anyway, it's an interesting momentum.
00:08:56.837 --> 00:09:02.206
When I was entering the profession I was perhaps a little bit naive.
00:09:02.206 --> 00:09:05.284
I thought that you need research, that this regulation.
00:09:05.284 --> 00:09:11.202
But today I'm more aware that there's another step communicating the regulation or communicating the science.
00:09:11.202 --> 00:09:13.648
That's as important as having the science and regulations.
00:09:13.648 --> 00:09:23.263
So and I say that because I first wanted to complement the guideline it's a great piece of communication and we will come back to this point many, many times.
00:09:23.263 --> 00:09:24.923
So we know where it came from.
00:09:24.923 --> 00:09:33.390
Tell me what it's about, what is in the scope and perhaps what important parts are not in the scope of this document.
00:09:33.951 --> 00:09:40.235
So, just first and foremost, right now, you know, there's no names on the guideline.
00:09:40.235 --> 00:09:53.140
We had a bit of a discussion Should we have the names on of authors, but we felt that it's maybe it's better just to have Zag Frisbee as an entity instead of specific names, given that it's a guideline.
00:09:53.140 --> 00:10:08.884
But the authors, the main authors, are Nick Russ, alessi Huguet and myself that have been writing this and, with all due respect to a lot of it comes from compiling existing information and this is one of the main.
00:10:08.884 --> 00:10:13.413
You know, I guess we had the three maybe main goal of the guideline.
00:10:13.413 --> 00:10:18.892
One was it shouldn't be too long, it should be so people will actually read it.
00:10:18.932 --> 00:10:22.024
There are other guidelines that are very good.
00:10:22.024 --> 00:10:35.620
I'm not criticizing them, but we know that people just put them in the drawer because they're too specific, they're too long, too many recommendations, and so we wanted an overview guideline, not an installation guideline.
00:10:35.620 --> 00:10:37.467
It's not an installation guideline.
00:10:37.467 --> 00:10:40.600
We wanted it to be well documented.
00:10:40.600 --> 00:10:57.754
Hence you can see we have two pages of small font references with links and we wanted it to create an overview, holistic overview that touches from sort of like the start.
00:10:57.754 --> 00:11:05.980
So hence we have the figure where we talk about the ignition hazards, the fire dynamics, the roof construction and the firefighting operations.
00:11:05.980 --> 00:11:07.163
And then we go.
00:11:07.163 --> 00:11:12.494
We try to be sort of pedagogical about talking about these four aspects.
00:11:13.320 --> 00:11:17.648
And the scope, if I assume correctly, is for flat roofs.
00:11:17.648 --> 00:11:18.429
That's one point.
00:11:18.429 --> 00:11:39.570
It's for PV panels that are placed on the roof, so not as a part of the roof, and also for fire hazards that would start mainly in the installations or in its vicinity, not as a part of the building fire that eventually spread to roof right yeah, so so that's why you know we start.
00:11:40.030 --> 00:11:43.567
Of course, you know, in the guideline we do have a table of contents.
00:11:43.567 --> 00:11:48.245
Then we we have the summary and then the scope and, as you said, the scope.
00:11:48.245 --> 00:12:06.923
We do have a figure where we explain that you have a fire that can start inside the building and go to the roof and you have a fire that starts on the roof, and predominantly we focus on the ones starting on the roof, because the ones that start in the building technically should be dealt with already through building regulation on the roof, because the ones that start in the building technically should be dealt with already through building regulation.
00:12:06.923 --> 00:12:12.975
That is of course, not always the case, but then you can't always then see.
00:12:12.975 --> 00:12:20.542
You know, it's important also to not blame solar or blame things for things that are are not part of the root cause.
00:12:20.542 --> 00:12:42.965
But at the same time, the frontispiece is a picture from the Asko distribution center in Norway and that fire started inside the building and it spread to the roof and because the fire section walls didn't extend up through the wall, it spread along the wall and then into another fire section, so it became a very big fire.
00:12:42.965 --> 00:12:54.149
So that is a route and that's where, of course, the PV panels played a significant role, because in other circumstances the energy would just be vented up and you wouldn't have spread across the route.
00:12:54.149 --> 00:13:03.572
But then we do focus on this new ignition source, this new system on the roof that can lead to fire scenarios there.
00:13:04.179 --> 00:13:11.173
And yes, we don't talk about building integrated pv vip, it is building applied photovoltaics.
00:13:11.173 --> 00:13:16.669
And when you say flat roofs, some people ask me recently oh, is there a big difference with slope roof?
00:13:16.669 --> 00:13:23.086
There are some many similarities, so it's not like that's a whole new chapter to do.
00:13:23.086 --> 00:13:28.595
You can actually see some of the work that's been done by Rise Fire Research in Norway.
00:13:28.595 --> 00:13:30.186
We mentioned Reidar.
00:13:30.186 --> 00:13:39.285
One thing with the slow proof for example, they did some testing with bitumen was that then you had the flowing of the bitumen, but as a system.
00:13:39.285 --> 00:13:54.750
They've also published, I think, some things in PV Magazine and some other places where the whole thing with the critical gaps, the ignition, the spreading and so on are on a membrane that otherwise doesn't lead to power spread are very similar.
00:13:54.750 --> 00:13:56.046
But there are some nuances.
00:13:56.960 --> 00:13:59.889
Well, you have to put the frame somewhere, I guess.
00:13:59.889 --> 00:14:03.167
I guess this is how you framed it and that's absolutely fine.
00:14:03.167 --> 00:14:06.249
Let's try to get into the contents of the guide.
00:14:06.249 --> 00:14:16.452
You said it has four specific, not chapters, but areas that it covers the ignition itself, the dynamics, the roof construction and then the firefighting.
00:14:16.452 --> 00:14:19.951
So perhaps let's use this framework also in this discussion.
00:14:19.951 --> 00:14:22.321
Let's start with the ignition hazards.
00:14:22.321 --> 00:14:23.522
Let's start with the ignition hazards.
00:14:23.522 --> 00:14:36.210
So you've been a part of a team that was investigating the probabilities or the statistics of rooftop fires, so let's perhaps start with that how often those fires happen and how did you figure out a number?
00:14:36.750 --> 00:14:44.054
So that was a study that we Jens again as a PhD student and myself got involved with.
00:14:44.054 --> 00:14:45.636
Jens is such a star.
00:14:45.636 --> 00:14:46.636
He's a star.
00:14:46.636 --> 00:14:47.317
I quote him.
00:14:47.317 --> 00:14:55.504
He said at some point during his PhD that I'm a one trick pony, but it's a very, very good trick and it's shown.
00:14:55.504 --> 00:15:04.727
You know, to some extent we've been lucky that it was very timely and we started way back when I was at DTU 2015, 14, almost.
00:15:05.299 --> 00:15:17.547
There's been 10 years of doing some of this and it's not that we were groundbreaking in that sense, because you have the work done by Backstrom and UL and things going back.
00:15:17.547 --> 00:15:24.349
Also, you can see some of the reference we have to Cancellary and SM Global's documents.
00:15:24.349 --> 00:15:31.750
But anyway, so we did this work with a group from Malaysia you know Saputria, and it's difficult to get good numbers.
00:15:31.750 --> 00:15:42.147
Yeah, Because a lot of it's based on media reports and does big graze salts, because what's reported in media is maybe just big fires.
00:15:42.147 --> 00:15:46.760
Nobody will report a small house fire and so on, but they're, you know they're.
00:15:47.101 --> 00:15:57.706
What we ended up with in that paper was to say that to a reasonable extent, we said 29 fires per gigawatt per year, per gigawatt installed.
00:15:57.706 --> 00:16:00.953
And some people say, oh, you know how much is this.
00:16:00.953 --> 00:16:03.765
And, yeah, what does the number mean?
00:16:03.765 --> 00:16:05.450
Actually, possibility?
00:16:05.450 --> 00:16:19.636
Somebody said you know, there was uh, this a few years ago, the fire at uh in bristol that we are the curious museum and somebody going to oh, it's a freak incident, it's only 0.01 chance of a fire.
00:16:19.636 --> 00:16:26.993
And then somebody in the comments said well, there's about I about I don't know like 2 million solar installations in the UK.
00:16:26.993 --> 00:16:37.815
So if you multiply that up, that means that you should have about 200 fires per year, which corresponds quite well to the fire service saying that there are several times a week are called out to PV fires.
00:16:37.815 --> 00:16:48.788
So it is not a small problem because in risk theory you often want to go to 10 to the minus 6, not to 10 to the minus 4, which is 0.01%.
00:16:48.788 --> 00:16:57.052
So it is actually a significant problem in many risk aspects and this is what we also see.
00:16:57.111 --> 00:17:14.984
You know there's a figure there that we took from the Clean Energy Associates, where they had inspected 600 sites worldwide and I think they came with something like 96% of these places had some mistakes and we've shown.
00:17:14.984 --> 00:17:21.441
You know the 10 main things related to grounding issues, damage module, cross-mated connectors, the list.
00:17:21.441 --> 00:17:23.628
You know there's 10 main issues.
00:17:23.628 --> 00:17:28.523
All of them are as high as 50% and more than 20%.
00:17:28.523 --> 00:17:31.529
So combined there's a big, very big chance.
00:17:31.529 --> 00:17:33.320
And people ask you know, but why?
00:17:33.320 --> 00:17:35.567
You know, you know, you just have to do it right.
00:17:35.567 --> 00:17:43.768
But if you think about it, some of these things have 2000, 3000, 4000 panels on a roof, maybe even more.
00:17:43.768 --> 00:17:59.667
You know more than that and imagine the labor to put all of that together, that all those operations, all those connections, that there's not a faulty connection, that there's no poor workmanship or poor maintenance in all of this.
00:18:00.201 --> 00:18:21.869
I was about to ask, and the statistics take where into account, like because we will have a freshly new installations, which I would hope that they are absolutely perfect because they have just been built and commissioned, but you may also have like 10 years old installations that went through many harsh wind events, through many hailstorms, through many thunderstorms.
00:18:22.599 --> 00:18:26.270
I've heard, I mean information, read the information.
00:18:26.270 --> 00:18:42.710
Installations that are maybe a few years old and when there was a fire and people looked at it then they could see, you know if there was a poor cable quality that had been used, that there's already been quite weathered, discolorations of cables and weakening.
00:18:42.710 --> 00:18:47.250
That's if things haven't been installed well.
00:18:47.250 --> 00:18:52.651
You have movements due to wind, snow, load, things like this that can create a strain.
00:18:52.651 --> 00:18:57.907
So there, you know the main, there is a too high of a probability right now.
00:18:57.907 --> 00:19:11.651
So this, this is what we also go all to do, this nfpa 550, fire concepts 3, where we say we need to work on the consequence, we need to work on and on the probability frequency of the fires.
00:19:11.651 --> 00:19:18.230
So the frequency, you know, this is the ignition and aspects of that and that is too high.
00:19:18.309 --> 00:19:22.023
But we in the guideline we tone that down there.
00:19:22.023 --> 00:19:26.840
That's done a lot elsewhere and I'm also no electrical engineer, so.
00:19:26.840 --> 00:19:38.237
So the details of all of these things we defer to other places where people look there are german, you know, from vds and also from fire protection association.
00:19:38.237 --> 00:19:44.912
They have long things with pictures of all kinds of installations and all kinds of faults and errors people can do.
00:19:44.912 --> 00:19:55.330
And if we were going into that detail, I think we would lose the overview, the holistic part, which is the main aspect for us.
00:19:56.185 --> 00:19:59.469
One thing that I wonder how easy is it actually to ignite a PV panel?
00:19:59.469 --> 00:20:05.964
Like to ignite that the fire is is to some extent growing on that panel Like any DCc fold.
00:20:05.964 --> 00:20:09.144
Any arc is is enough to to set it ablaze.
00:20:09.705 --> 00:20:13.621
So this is important here, you know we're not igniting the panel.
00:20:13.621 --> 00:20:17.347
Ah, what is burning is normally what's below the panel.
00:20:17.347 --> 00:20:31.306
Yes, some of the panels you know you have classes of panels but a lot of people say, oh well, this is because you use this bad panel and you're not using glass, glass panel or this panel.
00:20:31.306 --> 00:20:50.931
But in most of the work jens did, we used even a metal plate and and we even in the experiment I mean where the first, very first approximation experiments I did back in den Denmark 10 years ago or so we actually just used some glass wool and we saw the same thing.
00:20:50.931 --> 00:21:04.628
Hence my former students said we knew the answer that it wasn't the panel that was the problem, it's the system, it's this Physics is the problem, the physics, the fire dynamics is the problem.
00:21:04.628 --> 00:21:07.406
So well, so we knew.
00:21:07.406 --> 00:21:12.569
So that's what we chased and that's what ended up with all these very nice papers.
00:21:12.569 --> 00:21:23.690
Also we have to commend he worked with some very good students Ben Jacobs, farah from IMFSC, where they were doing their master's theses.
00:21:23.690 --> 00:21:32.392
Working with Jens, we also had Ming Chang from also I'm a C student who recently completed his PhD now at Ghent University.
00:21:32.392 --> 00:21:35.644
So so we had also other people in in that work.
00:21:35.644 --> 00:21:39.070
Just to highlight that that contributed to the knowledge.
00:21:39.070 --> 00:21:40.784
But that's, you know, the focus.
00:21:40.784 --> 00:21:42.029
There was the fire dynamics.
00:21:42.029 --> 00:21:47.180
So if you move into, you know in the guideline, from the ignition where, yes, we, I think what the ignition, where, yes, we're igniting.
00:21:47.180 --> 00:21:53.673
What we're igniting is typically we're igniting the membrane in the experiments we do with CRIB.
00:21:53.673 --> 00:21:58.073
But some people have been, oh, why aren't you not using an arc?
00:21:58.073 --> 00:22:00.880
And the reason we're not using an arc?
00:22:00.880 --> 00:22:04.667
We have actually done the arc and gotten the same results.
00:22:04.667 --> 00:22:08.665
The wood curve is more repeatable and we get the same end result.
00:22:08.665 --> 00:22:16.749
And also, of course, working with high currents and voltages can imply some other risk assessments in a lab environment.
00:22:16.749 --> 00:22:28.067
And finally, we can see that when people say, oh, but are you sure you're not making it too big, well, empirically we see the fires on the roofs that we see in media.
00:22:28.067 --> 00:22:41.864
So whatever ignited them led to spread to a thousand panels, two thousand panels, so that we are spreading from one panel to another, is empirically shown time and time over in those rooftop fires.
00:22:41.864 --> 00:22:52.442
So we're not making an overly big crib, we're actually quite a small that ignites the membrane, creates this feedback system that is again shown with these nice figures.
00:22:52.595 --> 00:22:57.123
I, you know we have them recreated here, but they're figures made by Jens in his PhD.
00:22:57.123 --> 00:23:00.805
Again, the whole guideline, as you can see, is type sense.
00:23:00.805 --> 00:23:02.102
So all the figures.
00:23:02.102 --> 00:23:12.077
We have references to many other places where we've done it, but we wanted the uniform expression instead of copy and pasting and then using other figures.
00:23:12.077 --> 00:23:23.097
And it's the first time I've done that in the sense to have the luxury of working with the typesetter and getting everything uniform, because I, you know, I'm terrible with graphics myself.
00:23:23.097 --> 00:23:30.082
But so, instead of having this is the figure from this and it looks completely different than the figure from Canceleri.
00:23:30.082 --> 00:23:32.804
And here's the figure from Clean Energy Associates.
00:23:32.804 --> 00:23:34.623
And here's the figure from BRE.
00:23:34.623 --> 00:23:35.941
And here's the figure from there.
00:23:35.941 --> 00:23:40.946
We have the benefit of working with a very good typesetter that put this all together.
00:23:41.455 --> 00:23:46.500
Hence my comment on the good communication in the guide from the start, because it looks like it's done purposefully.
00:23:46.500 --> 00:23:49.523
So yeah, but let's go, let's go back to physics.
00:23:49.743 --> 00:23:51.417
Yeah, so then you know we ignite it.
00:23:51.417 --> 00:23:52.019
But you know.
00:23:52.019 --> 00:24:20.701
So we look at this figure in the guideline, figure six, where, if you do a woodgrip, so that you know, let's backtrack a bit when you build a roof you have some insulation typically, and then you have a roofing membrane and before PVs were put on roofs and European Commission told us that we need to put them on the roof, or you know, there's not just the European Commission saying many people want to do it because energy prices are costing an arm and a leg, as they say.
00:24:20.701 --> 00:24:31.944
So the tests have been done, that the roofing membranes, if you have a burning brand, if you have a small fire attack, they shouldn't spread the fire across there.
00:24:31.944 --> 00:24:36.747
So you have these roof tests with the T1, t2, t3, t4.
00:24:36.747 --> 00:24:41.266
Commonly, if you pass T2, you will also pass all the other ones.
00:24:41.266 --> 00:24:43.442
It's the one that's most stringent in that.
00:24:43.442 --> 00:24:48.292
So we show this that, yeah, we use the membrane, we use the builder.
00:24:48.313 --> 00:24:53.724
With a crib it doesn't show, but then you put the panel above it and then they have an edge.
00:24:53.724 --> 00:24:55.502
So you gather some small.
00:24:55.502 --> 00:25:01.900
There may be a little bit of combustible material on the panel, but you get the heat feedback.
00:25:01.900 --> 00:25:06.001
You get re-radiation you on the panel, but you get the heat feedback.
00:25:06.001 --> 00:25:06.424
You get re-radiation.
00:25:06.424 --> 00:25:07.755
You have a flame that extends also up against a slow panel.
00:25:07.755 --> 00:25:13.606
So you get a bigger flame area, a bigger radiating area, and that preheats the membrane.
00:25:13.606 --> 00:25:20.429
That is then releasing enough paralysis gases that you can have a thermal runaway.
00:25:20.429 --> 00:25:24.727
You have a progressive set of ignitions, which is fire spread.
00:25:25.375 --> 00:25:31.820
One thing to clarify, because I guess not every fire science listener is an expert on the roof construction.
00:25:31.820 --> 00:25:36.010
The membrane, the uppermost layer of the roof, is usually combustible.
00:25:36.010 --> 00:25:36.914
Right, that's over.
00:25:37.055 --> 00:25:38.757
So the membranes will pass the B roof test, which is a horizontal test.
00:25:38.757 --> 00:25:43.541
The membranes will pass the B roof test, which is a horizontal test.
00:25:43.541 --> 00:26:01.500
But if you test it vertically, as you would do in EN 13501 reaction to fire tests, you will test it vertically and these tests will then be classified typically as an E, you know, a flammable material.
00:26:01.500 --> 00:26:09.502
Yeah, so there's no, there's no surprise that they burn per se, but they're constructed that when they're put flat they don't spread flame.
00:26:09.502 --> 00:26:17.385
I think similar with carpets if you put the car particularly and ignite it at the bottom, many of those would burn.
00:26:17.385 --> 00:26:21.361
That don't really spread the fire while lying flat on the on.
00:26:21.642 --> 00:26:53.707
It's really interesting because you're taking, like technically, a solution that's certified, that's fit for the purpose the flame would not spread on that surface if it was just the roof and then you kind of change the setting, you know, because suddenly you put something over it which is also perhaps certified, perhaps even non-combustible If you and the ends have put a bunch of steel and then the wool on top of that, which is non-combustible by definition, and suddenly your B-roof approved membrane becomes a hazard.
00:26:53.707 --> 00:27:04.766
Precisely say that the intended use of that membrane has changed, because it was intended to use in an open air and now it's used under a roof as a part of a mini compartment.
00:27:04.766 --> 00:27:06.978
To be honest, exactly so.
00:27:07.019 --> 00:27:23.904
This is one of the main technical messages that we try to portray in the guideline is that because we change the fire dynamics, then we need to revisit the roof construction, because the ignition wasn't used to be a problem but now it is.
00:27:23.904 --> 00:27:25.663
So it leads to a change for aerodynamics.
00:27:25.663 --> 00:27:31.681
Hence we need to go to a roof construction and eventually maybe also have different firefighting provisions.
00:27:31.681 --> 00:27:40.105
But so when we go there it is a key point that you're saying that I don't think you know in that sense that a membrane is a culprit.
00:27:40.105 --> 00:27:41.657
It's not a panel, that's a go.
00:27:41.657 --> 00:27:42.902
I said that before.
00:27:43.001 --> 00:27:47.221
It's important to point out that it's not this or that or this, it's a system.
00:27:47.221 --> 00:27:49.445
We just changed the rules of the game.
00:27:49.445 --> 00:27:52.578
Take a, you know french open is going on.
00:27:52.578 --> 00:28:07.701
If we all of a sudden change the height of the net or the length of the of the playing field, or if we say shorten it even more, so they would miss the field all the time because they're programmed to hit this and that's sort of the same.
00:28:07.701 --> 00:28:15.664
It's not something you can say blame them for, and that's because you can see also in if we go to figure seven.
00:00:00.442 --> 00:00:01.806
00:00:01.806 --> 00:00:14.212
00:00:14.212 --> 00:00:17.751
00:00:17.751 --> 00:00:22.852
00:00:22.852 --> 00:00:35.146
00:00:35.146 --> 00:00:46.831
00:00:46.831 --> 00:00:52.067
00:00:52.067 --> 00:00:53.463
00:00:53.463 --> 00:01:02.219
00:01:02.823 --> 00:01:14.343
00:01:14.343 --> 00:01:17.230
00:01:17.230 --> 00:01:25.311
00:01:25.311 --> 00:01:37.412
00:01:37.412 --> 00:01:40.825
00:01:40.825 --> 00:02:05.927
00:02:05.927 --> 00:02:11.484
00:02:11.484 --> 00:02:13.552
00:02:13.552 --> 00:02:17.284
00:02:17.284 --> 00:02:24.784
00:02:25.366 --> 00:02:39.786
00:02:39.786 --> 00:02:41.413
00:02:41.413 --> 00:02:52.205
00:02:52.205 --> 00:02:57.644
00:02:57.644 --> 00:03:00.406
00:03:00.406 --> 00:03:07.432
00:03:07.432 --> 00:03:10.877
00:03:10.877 --> 00:03:27.311
00:03:27.879 --> 00:03:30.179
00:03:30.179 --> 00:03:41.112
00:03:41.112 --> 00:03:56.925
00:03:56.925 --> 00:04:08.586
00:04:08.586 --> 00:04:19.641
00:04:19.641 --> 00:04:22.545
00:04:22.966 --> 00:04:28.553
00:04:28.553 --> 00:04:33.029
00:04:33.029 --> 00:04:35.086
00:04:35.086 --> 00:04:41.490
00:04:41.490 --> 00:04:47.487
00:04:47.487 --> 00:04:56.630
00:04:56.630 --> 00:05:03.009
00:05:03.009 --> 00:05:06.629
00:05:06.629 --> 00:05:19.730
00:05:19.730 --> 00:05:27.086
00:05:27.086 --> 00:05:34.733
00:05:34.733 --> 00:06:03.432
00:06:03.853 --> 00:06:28.286
00:06:28.286 --> 00:06:35.125
00:06:35.125 --> 00:06:37.519
00:06:37.519 --> 00:06:41.607
00:06:41.607 --> 00:06:49.627
00:06:49.627 --> 00:06:52.649
00:06:52.649 --> 00:06:54.466
00:06:55.060 --> 00:07:06.742
00:07:06.742 --> 00:07:12.221
00:07:12.221 --> 00:07:17.512
00:07:17.512 --> 00:07:27.675
00:07:27.675 --> 00:07:31.408
00:07:31.408 --> 00:07:39.809
00:07:39.809 --> 00:07:44.591
00:07:44.591 --> 00:07:58.512
00:07:59.000 --> 00:08:04.333
00:08:04.333 --> 00:08:08.923
00:08:08.923 --> 00:08:13.988
00:08:13.988 --> 00:08:19.821
00:08:19.821 --> 00:08:28.809
00:08:28.809 --> 00:08:37.777
00:08:37.777 --> 00:08:43.802
00:08:43.802 --> 00:08:45.568
00:08:45.568 --> 00:08:52.485
00:08:53.028 --> 00:08:53.931
00:08:53.931 --> 00:08:56.837
00:08:56.837 --> 00:09:02.206
00:09:02.206 --> 00:09:05.284
00:09:05.284 --> 00:09:11.202
00:09:11.202 --> 00:09:13.648
00:09:13.648 --> 00:09:23.263
00:09:23.263 --> 00:09:24.923
00:09:24.923 --> 00:09:33.390
00:09:33.951 --> 00:09:40.235
00:09:40.235 --> 00:09:53.140
00:09:53.140 --> 00:10:08.884
00:10:08.884 --> 00:10:13.413
00:10:13.413 --> 00:10:18.892
00:10:18.932 --> 00:10:22.024
00:10:22.024 --> 00:10:35.620
00:10:35.620 --> 00:10:37.467
00:10:37.467 --> 00:10:40.600
00:10:40.600 --> 00:10:57.754
00:10:57.754 --> 00:11:05.980
00:11:05.980 --> 00:11:07.163
00:11:07.163 --> 00:11:12.494
00:11:13.320 --> 00:11:17.648
00:11:17.648 --> 00:11:18.429
00:11:18.429 --> 00:11:39.570
00:11:40.030 --> 00:11:43.567
00:11:43.567 --> 00:11:48.245
00:11:48.245 --> 00:12:06.923
00:12:06.923 --> 00:12:12.975
00:12:12.975 --> 00:12:20.542
00:12:20.542 --> 00:12:42.965
00:12:42.965 --> 00:12:54.149
00:12:54.149 --> 00:13:03.572
00:13:04.179 --> 00:13:11.173
00:13:11.173 --> 00:13:16.669
00:13:16.669 --> 00:13:23.086
00:13:23.086 --> 00:13:28.595
00:13:28.595 --> 00:13:30.186
00:13:30.186 --> 00:13:39.285
00:13:39.285 --> 00:13:54.750
00:13:54.750 --> 00:13:56.046
00:13:56.960 --> 00:13:59.889
00:13:59.889 --> 00:14:03.167
00:14:03.167 --> 00:14:06.249
00:14:06.249 --> 00:14:16.452
00:14:16.452 --> 00:14:19.951
00:14:19.951 --> 00:14:22.321
00:14:22.321 --> 00:14:23.522
00:14:23.522 --> 00:14:36.210
00:14:36.750 --> 00:14:44.054
00:14:44.054 --> 00:14:45.636
00:14:45.636 --> 00:14:46.636
00:14:46.636 --> 00:14:47.317
00:14:47.317 --> 00:14:55.504
00:14:55.504 --> 00:15:04.727
00:15:05.299 --> 00:15:17.547
00:15:17.547 --> 00:15:24.349
00:15:24.349 --> 00:15:31.750
00:15:31.750 --> 00:15:42.147
00:15:42.147 --> 00:15:46.760
00:15:47.101 --> 00:15:57.706
00:15:57.706 --> 00:16:00.953
00:16:00.953 --> 00:16:03.765
00:16:03.765 --> 00:16:05.450
00:16:05.450 --> 00:16:19.636
00:16:19.636 --> 00:16:26.993
00:16:26.993 --> 00:16:37.815
00:16:37.815 --> 00:16:48.788
00:16:48.788 --> 00:16:57.052
00:16:57.111 --> 00:17:14.984
00:17:14.984 --> 00:17:21.441
00:17:21.441 --> 00:17:23.628
00:17:23.628 --> 00:17:28.523
00:17:28.523 --> 00:17:31.529
00:17:31.529 --> 00:17:33.320
00:17:33.320 --> 00:17:35.567
00:17:35.567 --> 00:17:43.768
00:17:43.768 --> 00:17:59.667
00:18:00.201 --> 00:18:21.869
00:18:22.599 --> 00:18:26.270
00:18:26.270 --> 00:18:42.710
00:18:42.710 --> 00:18:47.250
00:18:47.250 --> 00:18:52.651
00:18:52.651 --> 00:18:57.907
00:18:57.907 --> 00:19:11.651
00:19:11.651 --> 00:19:18.230
00:19:18.309 --> 00:19:22.023
00:19:22.023 --> 00:19:26.840
00:19:26.840 --> 00:19:38.237
00:19:38.237 --> 00:19:44.912
00:19:44.912 --> 00:19:55.330
00:19:56.185 --> 00:19:59.469
00:19:59.469 --> 00:20:05.964
00:20:05.964 --> 00:20:09.144
00:20:09.705 --> 00:20:13.621
00:20:13.621 --> 00:20:17.347
00:20:17.347 --> 00:20:31.306
00:20:31.306 --> 00:20:50.931
00:20:50.931 --> 00:21:04.628
00:21:04.628 --> 00:21:07.406
00:21:07.406 --> 00:21:12.569
00:21:12.569 --> 00:21:23.690
00:21:23.690 --> 00:21:32.392
00:21:32.392 --> 00:21:35.644
00:21:35.644 --> 00:21:39.070
00:21:39.070 --> 00:21:40.784
00:21:40.784 --> 00:21:42.029
00:21:42.029 --> 00:21:47.180
00:21:47.180 --> 00:21:53.673
00:21:53.673 --> 00:21:58.073
00:21:58.073 --> 00:22:00.880
00:22:00.880 --> 00:22:04.667
00:22:04.667 --> 00:22:08.665
00:22:08.665 --> 00:22:16.749
00:22:16.749 --> 00:22:28.067
00:22:28.067 --> 00:22:41.864
00:22:41.864 --> 00:22:52.442
00:22:52.595 --> 00:22:57.123
00:22:57.123 --> 00:23:00.805
00:23:00.805 --> 00:23:02.102
00:23:02.102 --> 00:23:12.077
00:23:12.077 --> 00:23:23.097
00:23:23.097 --> 00:23:30.082
00:23:30.082 --> 00:23:32.804
00:23:32.804 --> 00:23:34.623
00:23:34.623 --> 00:23:35.941
00:23:35.941 --> 00:23:40.946
00:23:41.455 --> 00:23:46.500
00:23:46.500 --> 00:23:49.523
00:23:49.743 --> 00:23:51.417
00:23:51.417 --> 00:23:52.019
00:23:52.019 --> 00:24:20.701
00:24:20.701 --> 00:24:31.944
00:24:31.944 --> 00:24:36.747
00:24:36.747 --> 00:24:41.266
00:24:41.266 --> 00:24:43.442
00:24:43.442 --> 00:24:48.292
00:24:48.313 --> 00:24:53.724
00:24:53.724 --> 00:24:55.502
00:24:55.502 --> 00:25:01.900
00:25:01.900 --> 00:25:06.001
00:25:06.001 --> 00:25:06.424
00:25:06.424 --> 00:25:07.755
00:25:07.755 --> 00:25:13.606
00:25:13.606 --> 00:25:20.429
00:25:20.429 --> 00:25:24.727
00:25:25.375 --> 00:25:31.820
00:25:31.820 --> 00:25:36.010
00:25:36.010 --> 00:25:36.914
00:25:37.055 --> 00:25:38.757
00:25:38.757 --> 00:25:43.541
00:25:43.541 --> 00:26:01.500
00:26:01.500 --> 00:26:09.502
00:26:09.502 --> 00:26:17.385
00:26:17.385 --> 00:26:21.361
00:26:21.642 --> 00:26:53.707
00:26:53.707 --> 00:27:04.766
00:27:04.766 --> 00:27:06.978
00:27:07.019 --> 00:27:23.904
00:27:23.904 --> 00:27:25.663
00:27:25.663 --> 00:27:31.681
00:27:31.681 --> 00:27:40.105
00:27:40.105 --> 00:27:41.657
00:27:41.657 --> 00:27:42.902
00:27:43.001 --> 00:27:47.221
00:27:47.221 --> 00:27:49.445
00:27:49.445 --> 00:27:52.578
00:27:52.578 --> 00:28:07.701
00:28:07.701 --> 00:28:15.664