April 17, 2024
148 - Building Integrated Photovoltaics with Reidar Stølen
In this episode of the Fire Science Show we go in depth on the Building Integrated Photo-Voltaic systems (BIPV). It is a topic relevant to many fire engineers, and one on which it is very difficult to find a lot of information about. For this purpose I’ve invited Reidar Stølen from RISE and a PhD candidate at Norwegian University of Science and Technology – NTNU.
Reidar has hands-on experience with fire testing BIPV façade, as he has performed such experiments with the Swedish test method for a commercial project. The results of the first experiments can be found in this research paper: Large- and small-scale fire test of a building integrated photovoltaic (BIPV) façade system. We go in depth into description of the test sample, the rig and the outcomes of the experiment. Make sure to check the paper for the before and after pictures of the façade!
Another paper worth highlighting is the investigation into the Factors Affecting the Fire Safety Design of Photovoltaic Installations Under Performance-Based Regulations in Norway. Related to this we have an in depth discussion on the factors that play role in PV fires, especially connectors as one of the main sources of the fire.
----
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.
Chapters
00:00 - In-Depth Discussion on Photovoltaic Systems
15:51 - Fire Safety of PV Installations
26:42 - PV Fire Safety Engineering Challenges
41:22 - Fire Safety Engineering Research
Transcript
WEBVTT
00:00:00.600 --> 00:00:01.804
Welcome to the Fire Science Show.
00:00:01.804 --> 00:00:05.652
Today we're having a much more in-depth technical discussion.
00:00:05.652 --> 00:00:21.553
Today we're talking about one particular technical solution that is probably very interesting to a bigger chunk of the audience of the Fire Science Show, and that is photovoltaic systems In particular, mostly about building integrated photovoltaic systems.
00:00:21.553 --> 00:00:27.062
So when you put photovoltaics as a part of your facade or roof structure, what happens then?
00:00:27.062 --> 00:00:30.782
But in this episode we also venture into building attached photovoltaics.
00:00:30.782 --> 00:00:38.427
I'm joined by Radar Stolten from RISE, who has been testing those systems, who has been researching those systems, has some observations.
00:00:38.427 --> 00:00:46.814
He's been also working on investigating best practices at the Scandinavian market on how to put them safely in the building.
00:00:46.814 --> 00:00:54.509
So I think it's a very interesting and in-depth technical conversation, more technical than other episodes of the Fire Reef Science Show recently.
00:00:54.509 --> 00:01:02.551
But yes, this is important to share the in-depth knowledge and I hope it will be interesting to a lot of you, because photovoltaics is something we cannot escape.
00:01:02.551 --> 00:01:18.603
Sustainability requirements require us to have energy generation that is distributed, not just one huge coal-powered power plant like we would have in Poland, but more distributed network, and you can see this popping all over the world.
00:01:18.603 --> 00:01:30.751
Before I invite you to the episode, I also want to let you know that first episodes of my other project, uncovered Witness, are now live, so if you would like to check it out, you can go to uncoveredwitnesscom.
00:01:30.751 --> 00:01:37.703
That's a podcast that also brings fire science perhaps in a little more approachable way, in a more structured way.
00:01:37.703 --> 00:01:43.765
It's not interviews, it's more like a narrated storytelling approach to preaching fire science.
00:01:43.765 --> 00:01:50.346
I have some great experts with me there, so it's not just me rambling about fire safety.
00:01:50.346 --> 00:01:58.751
I hope you will enjoy that one, especially for those of you who've asked me to produce more easy or intermediate level content In that project.
00:01:58.751 --> 00:02:03.748
We start very easily and we ramp up to some quite interesting aspects of fire safety.
00:02:03.748 --> 00:02:06.614
We've started with three episodes on means of escape.
00:02:06.614 --> 00:02:11.451
So how do we design our buildings to make us be able to escape from them?
00:02:11.451 --> 00:02:19.367
I especially highly recommend episode three of that series where we talk about what makes or breaks a good evacuation signage based on theory of ordnance.
00:02:19.367 --> 00:02:20.947
Oh well, I won't spoil it anymore.
00:02:20.947 --> 00:02:22.048
It's a nice project.
00:02:22.048 --> 00:02:28.612
If you enjoy fire science, and if you enjoy fire science fundamental series in this podcast, you will love Uncovered Witness as well.
00:02:28.612 --> 00:02:30.062
So that's enough.
00:02:30.062 --> 00:02:35.294
Let's spin the intro and learn from Raider about building integrated photovoltaics.
00:02:35.294 --> 00:02:41.965
Welcome to the Fire Science Show.
00:02:41.965 --> 00:02:45.393
My name is Wojciech Wigrzyński and I will be your host.
00:03:00.300 --> 00:03:03.450
This podcast is brought to you in collaboration with OFR Consultants.
00:03:03.450 --> 00:03:06.299
Ofr is the UK's leading fire risk consultancy.
00:03:06.299 --> 00:03:17.233
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.
00:03:17.233 --> 00:03:33.055
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 100 professionals.
00:03:33.055 --> 00:03:44.715
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.
00:03:44.715 --> 00:03:48.319
Experience and diligence for effective tailored fire safety solutions.
00:03:48.319 --> 00:03:55.753
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.
00:03:55.753 --> 00:03:58.628
Get in touch at ofrconsultantscom.
00:03:59.901 --> 00:04:06.991
Hello Wojciech, good to be here, happy to have you and I was looking for a guest about the subject of photovoltaics for quite a long time.
00:04:06.991 --> 00:04:14.447
We had this episode with Jens a year and a half ago and it's crazy popular and people are looking for more and more information.
00:04:14.447 --> 00:04:23.814
I know you are researching this in Norway, so my first question is you know Norway, country of hydropower, you have a lot of clean energy.
00:04:23.814 --> 00:04:29.004
Is photovoltaics also on the rise in?
00:04:29.084 --> 00:04:29.225
there.
00:04:29.225 --> 00:04:29.726
Pun intended?
00:04:29.726 --> 00:04:32.375
Yeah, I would definitely say it's on the rise.
00:04:32.375 --> 00:04:34.901
We've been lagging a bit behind the rest of Europe.
00:04:34.901 --> 00:04:53.413
We've been very used to cheap electricity from the hydropower and limited capacity of exporting electricity to other countries with generally higher costs of electricity, so historically we've had a very low penetration of PV electricity.
00:04:53.413 --> 00:05:03.367
But over the recent years we've also started to feel the increasing cost of electricity and seen a very sharp increase in PV installations as well.
00:05:03.367 --> 00:05:05.632
Now the recent couple of years.
00:05:06.680 --> 00:05:29.175
It seems that the trends of sustainability are catching everywhere and I assume that the introduction of PV to your country, in Poland, it has been quite wild, especially in residential, you know, because we had quite a good law for a while that made PV as a private person very how to say generous in savings.
00:05:29.175 --> 00:05:33.471
Now it's not that much, but we also see a lot of industrial installations.
00:05:33.471 --> 00:05:49.434
I assume you also face the same challenges as anyone else in the world in integrating them with your buildings, so perhaps you can bring me closer to what types of PV setups are you most interested in?
00:05:49.434 --> 00:05:53.091
All of them or more like roof systems, integrated systems?
00:05:53.091 --> 00:05:54.846
What's in your scope today?
00:05:55.399 --> 00:06:04.800
I'm struggling to narrow it down into the different parts because they have many of the same properties with respect to the fire dynamics.
00:06:06.608 --> 00:06:08.879
They have many of the same properties with respect to the fire dynamics.
00:06:09.319 --> 00:06:18.232
As you know, jens has done a lot of work on flat roofs that are very representative for large industrial buildings where you have very large areas where you can install PV installations.
00:06:18.720 --> 00:06:34.709
But when we look at the PV installations in Norway at least, and probably also other countries, the highest number of installations are the relatively small installations on a single family home and they are mostly on sloped roofs.
00:06:34.709 --> 00:06:41.545
So that's a slightly different approach and has some variations in the fire dynamics and also on vertical facades.
00:06:41.545 --> 00:07:15.786
That I think can be very interesting and also on vertical facades that I think can be very interesting and even though they're not that commonly used yet, I think that particularly in Norway and high up north, where we have a lot of snow on the roofs in the winter, where we in general use the most electricity and the sun is at a very low angle, electricity and the sun is at a very low angle, so the facade is beginning to be, I think, more interesting to install pv installations because you will generate more electricity in the in the winter time I'm not sure if that's the trend.
00:07:15.966 --> 00:07:30.026
I mean, I mean I understand the reasons for that and then you just explain them, but uh, I'll also, like you know, I'm traumatized by facade fires and everything that's happening in the facade space and, uh, and our research.
00:07:30.026 --> 00:07:35.985
Whenever we put something novel on the facade, it usually ends up not very nice.
00:07:35.985 --> 00:07:40.353
So let's try and talk about this facade photovoltaics.
00:07:40.353 --> 00:07:43.605
How does the solutions on the facades look like?
00:07:43.605 --> 00:07:47.754
Is it the same, like attached panels like you would have on the roof?
00:07:47.754 --> 00:07:49.485
Is the structure different?
00:07:49.485 --> 00:07:52.687
I'm not really familiar with those solutions.
00:07:53.519 --> 00:08:03.964
Yeah, I would say that facades as a facade is much more visible than the roof on a building the aesthetics of the PV installations tend to be more important.
00:08:03.964 --> 00:08:24.274
So, at least up until now, we've seen more of custom-made facades, pv modules that are custom-built for each project and in a way better aesthetically and architecturally integrated in the building than the traditional roof-mounted installations on large flat roofs.
00:08:24.274 --> 00:08:28.211
But then again the cost is relatively high for those systems.
00:08:28.211 --> 00:08:40.769
So I don't have any good statistics on that, but I think that more facades will be integrated with more common and more standardized module sizes to bring the cost down for those systems.
00:08:40.828 --> 00:08:52.428
As well, yeah, they probably also must be quite robust, like I mean they have to carry the wind load because they become the most extra, most layer of the building.
00:08:52.428 --> 00:08:54.466
You mentioned aesthetics.
00:08:54.466 --> 00:08:59.667
We all know that that's a considerable driver for the facade design.
00:08:59.667 --> 00:09:02.649
The substructure of it must be.
00:09:02.649 --> 00:09:05.903
Are they built on aluminum frames, on steel frames?
00:09:05.903 --> 00:09:15.294
Can you tell me a bit more about layers, at least, like what layers would go into such a system so I can imagine, and listeners can imagine, it a little bit better?
00:09:29.748 --> 00:09:37.533
Yeah, I think it can also be integrated as the facade element itself and inside this cavity that we're talking so much about.
00:09:37.533 --> 00:10:02.469
When it comes to facade fires, it can be quite a wide range of different materials in different either external facade claddings behind the building, attached PV modules or wind barriers of different types, and also fixing structures of aluminium, I think is the most common, but steel can also be used and also wood constructions can also be integrated into this cavity.
00:10:02.469 --> 00:10:08.404
So it can be quite a wide range of different construction types and materials in this cavity.
00:10:08.404 --> 00:10:21.160
That will be very interesting to get an even deeper understanding of how these different materials and constructions can determine how the fire will propagate in the cavity.
00:10:22.222 --> 00:10:24.389
I'm going to torture you a little more with the materials.
00:10:24.389 --> 00:10:31.428
So I would assume the exterior has to be some sort of transparent material to let light in and yet look good.
00:10:31.428 --> 00:10:34.208
So it's mostly going to be some sort of glass.
00:10:34.208 --> 00:10:40.587
I don't think any PMMA or stuff like that would be used, so I would assume it's just glass.
00:10:40.587 --> 00:11:06.629
Then you would have the silicone of the of the PV itself, the thing that takes up light and turns it into electricity, and then, to the best of my knowledge, you have some materials that create the back of that panel and that could range from plastics to non-combustible materials, more or less, and then this would be exposed to cavity, which then whatever is on the other side of the building.
00:11:06.629 --> 00:11:08.707
Is that a reasonable summary?
00:11:09.441 --> 00:11:18.192
Yeah, that's the building integrated PV facade that we tested in large scale a couple of years ago.
00:11:18.192 --> 00:11:19.995
That was recently published now.
00:11:19.995 --> 00:11:25.452
That was from a building project and PV modules were made of basically five layers.
00:11:25.452 --> 00:12:05.852
That is more or less the common layers that we see in most PV modules, with a glass front and then the silicon wafers being laminated in between two layers of EVA plastic, and then behind that in our modules there was this plastic film and this plastic film can in some cases be exchanged with a second layer of glass in what you call glass-glass modules and then the cavity behind that and in the system that we tested there was aluminum frames and structures and a 60-millimeter air cavity on top of a gypsum gypsum windbreaker.
00:12:06.813 --> 00:12:27.500
So even if we take the you know photovoltaics away from this system, you perhaps have a exposed or semi-exposed combustible material adjacent to the cavity, so it's kind of in some way resembling the modern facades employing plastic materials.
00:12:27.500 --> 00:12:39.927
Is it common to use cavity breaks or any other cavity solutions to limit the vertical propagation in such systems that, you know, I don't know exactly how common it is.
00:12:40.830 --> 00:12:52.416
It was installed in the facade that we tested in full scale, but I know other buildings are built with the same type of modules and same type of construction without any cavity barriers.
00:12:52.416 --> 00:13:03.533
So I would say that it's definitely not installed in all cases at least solutions how big and how many of cables is there?
00:13:03.572 --> 00:13:18.961
because I assume that every module has to be connected with a cable, because that's the point of having a module, so you have to, like, get the electricity from the facade somewhere, so, so how many cables actually are there and where do they go through?
00:13:19.562 --> 00:14:10.451
yeah, from each module is is well, most of the modules, at least, are connected to the neighboring modules with relatively short cables and for the modules that we tested in the large-scale facade, tried to measure the amount of combustible material in cable installation and junction boxes and the different layers of the plastics inside the PV module, and it was by far dominated by the plastic layers laminated to the PV module and even though that in total is only approximately one millimeter of plastic, that represented the majority of the combustible material in the PV module and the junction boxes and the cables was only a smaller amount of combustible material compared to the plastics laminated to the different layers in the PV modules.
00:14:11.480 --> 00:14:15.011
Okay, perhaps you can guide me through the test that you have carried.
00:14:15.011 --> 00:14:25.033
As you mentioned, there's a paper published in the Fire Safety Journal and I'm going to link this paper and the one on factors affecting the safety of PV installations.
00:14:25.033 --> 00:14:26.325
They will be both in the show notes.
00:14:26.325 --> 00:14:29.330
Let's go and try to discuss the experiment.
00:14:29.330 --> 00:14:39.032
So first you need to bring us in on what is the setup and how does it look like, because I'm not sure if many people are that familiar with Nordic facade test methods.
00:14:39.441 --> 00:14:45.714
Yes, we use the Nordic large-scale facade test method SP-FIRE-105.
00:14:45.714 --> 00:14:58.591
That is one of different large-scale facade test methods that are being used and in the process of being harmonized as we speak, and it's a facade measuring four meters wide and six meters tall.
00:14:58.919 --> 00:15:00.025
Okay, that's a big one.
00:15:00.046 --> 00:15:10.745
Yeah, and it's not a corner, it's a flat facade and two window openings are included in the facade, and below the facade there's a fire chamber where we burn.
00:15:10.745 --> 00:15:18.677
I think it's 60 liters of heptane in a fire generating in the range one to two megawatts.
00:15:19.320 --> 00:15:24.857
Okay, and is there any flames coming out of the windows, or these are just like mock-ups for detail detailing.
00:15:25.178 --> 00:15:26.883
Yeah, the window openings in the facade.
00:15:26.883 --> 00:15:39.342
They are placed higher up in the, at the two stories above the fire compartment, so they are just sealed openings where basically that the detailing of the facade around the windows are the important factor.
00:15:39.342 --> 00:15:50.975
And then the the fire chamber represents a compartment at flashover where you have flames exiting the opening and impinging onto the facades.
00:15:51.380 --> 00:15:56.472
How long does the test last and what are the pass-fail criteria in it?
00:15:57.240 --> 00:16:03.591
The test lasts until the heptane is burned up and in our case it took approximately 30 minutes.
00:16:03.591 --> 00:16:08.412
The first 15 minutes was basically the fire compartment heating up.
00:16:08.412 --> 00:16:15.653
The next 15 minutes we had large flames exposing the facade from the heptane flames.
00:16:15.653 --> 00:16:34.030
And then the criteria is that the fire should not propagate above a certain level in the facade and also temperature requirements in the eave on top of the facade are the basic two criteria and also it shouldn't fall down large objects on the facade.
00:16:34.721 --> 00:16:35.840
How do you measure the objects?
00:16:35.840 --> 00:16:38.990
You capture them underneath and then wave them.
00:16:40.601 --> 00:16:42.187
No, we didn't do that.
00:16:42.187 --> 00:16:43.442
And the test method?
00:16:43.442 --> 00:16:58.812
I don't remember if there is a specific definition of what a large object is, but assuming that it's okay to have smaller debris falling down, but larger, heavier objects are not allowed to fall down.
00:16:58.812 --> 00:17:03.533
At least that should be noted in the test report should be noted in the test report.
00:17:04.519 --> 00:17:14.046
I'm asking because you've previously mentioned that this is a commonly aluminum-structured facade and the source that you've mentioned, the 60 liters of heptane.
00:17:14.046 --> 00:17:17.946
That sounds like quite large fire loads and I saw pictures from the test.
00:17:17.946 --> 00:17:19.866
It seems like quite a serious fire.
00:17:19.866 --> 00:17:24.784
So that's definitely something that can damage an aluminum substructure of a facade.
00:17:24.784 --> 00:17:30.518
So you could actually expectructure of a facade, so you could actually expect some of them falling.
00:17:30.518 --> 00:17:44.589
Can you tell me how did the test go on this PV facade and perhaps some differences that you've observed versus commonly used non-electricity generating solutions that you perhaps have also tested before?
00:17:44.589 --> 00:17:44.691
Sure?
00:17:45.595 --> 00:18:07.704
Yeah, we made a number of interesting observations and some of them maybe not that surprisingly that the high heat from the large fire was severely damaging both the glass and the aluminum and the glue that basically held the PV installation together and in place.
00:18:07.704 --> 00:18:32.767
So in the lower part of the facade, where you had direct impingement of the large heptane flames, we saw that most of the aluminum frames were just melted and destroyed and the glass panels were holding out from the frames and the glass was shattering and breaking and basically quite severe structural damage to the lower part of the.
00:18:33.596 --> 00:18:36.065
How quickly that happened after the ignition.
00:18:36.065 --> 00:18:38.963
Was it very soon into the fire or took, I know, 20 minutes?
00:18:39.476 --> 00:18:52.263
Modules started falling down quite early after the flame started impinging out from the flame compartment, so it was a matter of a few minutes the first module started falling down.
00:18:52.914 --> 00:18:56.345
It seems that there's quite a lot of structural work to be undertaken.
00:18:56.345 --> 00:19:00.665
Perhaps aluminum is not the best material, but that's for another discussion.
00:19:00.665 --> 00:19:03.182
Okay, and what happened later?
00:19:03.182 --> 00:19:05.795
What happened with the upper panels that did not fall out immediately?
00:19:06.316 --> 00:19:06.876
Yeah.
00:19:06.876 --> 00:19:27.682
The second and maybe most crucial observation that we made was that, even after the heptane had burned out, we saw the fire was able to propagate up through the cavity, even though there were installed cavity barriers, and what we saw was that the cavity barrier appeared to function as intended and after the tests, it remains.
00:19:27.682 --> 00:19:30.731
It appeared to function as intended and after the tests it remains.
00:19:30.731 --> 00:19:34.345
It appeared to be sealed and be intact.
00:19:34.345 --> 00:19:46.625
As the glass and aluminum structure that the cavity barrier sealed against was damaged, the fire was able to pass the barrier and propagate up in the cavity self-sustained.
00:19:46.625 --> 00:19:56.741
Despite that, there was no other combustible material, only the gypsum boards and the aluminium in addition to the plastics in the PV modules.
00:19:56.741 --> 00:20:12.684
The fire was able to propagate up without the additional heat from the heptane fire and the modules were still falling down as the fire propagated upwards it's relevant information.
00:20:12.704 --> 00:20:14.157
I think we didn't say that earlier.
00:20:14.157 --> 00:20:15.821
How big was the cavity?
00:20:15.821 --> 00:20:17.526
Like five, ten centimeters.
00:20:17.526 --> 00:20:18.047
How big was?
00:20:18.086 --> 00:20:18.147
it.
00:20:18.147 --> 00:20:28.921
It was, uh, six centimeters, six, okay, from from the gypsum board to the, to the back of the pv module and the backing of the pv modules?
00:20:29.301 --> 00:20:33.048
uh, there was backsheet that was non-combustible.
00:20:33.048 --> 00:20:35.102
Or was it the plastic exposed to the cavity?
00:20:36.436 --> 00:20:42.157
It was a plastic backsheet, so some type of Because I wondered.
00:20:42.298 --> 00:20:58.494
In some of the aluminum composite panels we often observe this phenomena when the glue is releasing and the aluminum backsheet is, you know, falling off the panel and and creates a like a second cavity which exposes the plastic interior.
00:20:58.494 --> 00:21:05.728
And this is usually connected with, uh, quite bad fire outcomes of such a test.
00:21:05.728 --> 00:21:08.478
So I wondered if the peeling is also something you could observe here.
00:21:08.478 --> 00:21:14.307
But if you had a plastic backing, then there's no peeling needed to to expose the plastic.
00:21:14.307 --> 00:21:17.176
So, uh, so perhaps this is irrelevant.
00:21:17.176 --> 00:21:21.163
Um, the general conclusions of the test.
00:21:21.163 --> 00:21:23.467
Was it good, bad, uh?
00:21:23.467 --> 00:21:27.157
How did it look against a typical facade you would install?
00:21:27.157 --> 00:21:28.118
Would it pass the?
00:21:30.021 --> 00:21:34.347
No, this did not pass the test on basically two reasons.
00:21:34.347 --> 00:21:45.199
One was that large heavy objects well, basically entire PV modules were falling down from the facade and that the fire was able to propagate all the way up to the top of the facade.
00:21:45.199 --> 00:21:53.050
So this solution was modified and changed until the building was built.
00:21:54.336 --> 00:21:54.919
Is it a secret?
00:21:54.919 --> 00:21:57.385
Can you tell us how it was modified?
00:21:57.385 --> 00:22:01.285
I'm not sure what can you reveal, but how can we make it?
00:22:01.285 --> 00:22:03.501
Let's speak on general terms.
00:22:03.501 --> 00:22:05.903
If you had such a facade, how would you improve?
00:22:05.963 --> 00:22:06.023
it.
00:22:06.636 --> 00:22:23.125
I would definitely look into the difference of using glass glass modules compared to glass polymer modules so in this case, you basically remove all the combustible material out of the pv panels or at least expose most of it.
00:22:23.807 --> 00:22:30.326
Well, not, not really, because the glass glass modules they still have the encapsulant plastic material.
00:22:30.326 --> 00:22:39.109
So most of the plastic material would still be in the PV module, but it would be protected by the back layer of glass.
00:22:39.109 --> 00:22:48.605
So I would assume that the back sheet of glass would protect the plastic encapsulant better than the thin plastic sheet.
00:22:48.605 --> 00:23:03.303
I think it would be a matter of how how large fire you need in the cavity to keep exposing new and fresh plastic from the, from the encapsulant in the pv and it perhaps will be less self-sustaining.
00:23:03.723 --> 00:23:14.299
Uh, seeing that most of the damage you had in your previous experiment happened around the heptanes, like on the lowermost level, you did not destroy every single panel upwards.
00:23:14.339 --> 00:23:32.538
So, yeah, it kind of makes sense to, to, to layer up yes, and of course in in our facade we had very limited additional combustible material, so that would also be very careful in including other combustible materials in the cavity.
00:23:32.538 --> 00:24:21.459
The amount of heat that you need to propagate it further into the cavity is very limited and it would definitely try to reduce the amount of combustible material as much as possible and also try to see if we can generate some escape route for that heat out of the cavity instead of just leading all of the heat generated further up into the cavity above and in the real building is is the entire facade energy generating or the?
00:24:21.499 --> 00:24:35.015
the panels are alternating between the generating ones and, let's say, a fake ones, because in poland we would have to, uh, provide a 1.2 meter part of the facade which is fire resistant.
00:24:35.015 --> 00:24:43.349
So it would be probably challenging to create a continuous combustible facade, at least on the larger building.
00:24:43.349 --> 00:24:47.766
What was your experience on this building that you were participating in the design?
00:24:47.766 --> 00:24:50.564
It was a continuous facade all over the building.
00:24:51.694 --> 00:25:06.263
No, as far as I understand, parts of the facade was made from wood, wooden cladding and parts were supposed to be made with integrated PV modules as the facade cladding and from the other facades with PV modules.
00:25:06.263 --> 00:25:21.987
That I know it varies a bit because if you have a large, open, rectangular facade without any windows or openings or other difficulties, then you could easily cover the entire facade with PV modules.
00:25:21.987 --> 00:25:33.875
But where you need to integrate doors or openings or windows or different angles, then you could also custom make all of these shapes into PV modules.
00:25:33.875 --> 00:25:41.184
But then the cost would be much higher than if you can use more standardized sizes of PV modules.
00:25:41.806 --> 00:25:56.734
Okay, cool, let's leave the facade alone for a while now and I would like to ask you some questions about your previous work where you were doing assessment of fire safety of other PV installations.
00:25:56.734 --> 00:26:06.365
In this case, the paper is treating mostly about roofs and sloped roofs and real buildings from Norway.
00:26:06.365 --> 00:26:15.684
So can you tell me what was the reason of that study and what were the main outcomes of that survey?
00:26:15.684 --> 00:26:19.423
What did you find out about the PV installations in Norway at that point?
00:26:19.934 --> 00:26:22.263
Yeah, we looked at the sloped roofs.
00:26:22.263 --> 00:26:33.539
We were inspired by the work that Jens had done on flat roofs and wondered what about all of these single-family homes with sloped roofs and PV installations?
00:26:33.539 --> 00:26:41.625
How does building attached PV modules on top of those roofs influence the fire properties of the roof surface?
00:26:41.625 --> 00:26:47.458
And in Scandinavia we use the roof fire test method number two.
00:26:47.458 --> 00:26:58.063
That's number two of total four different ones across Europe with a relatively small wood crib fire placed on top of the roof surface.
00:26:58.144 --> 00:26:58.786
Okay, this one.
00:26:58.786 --> 00:27:00.500
I think we also have that one in Poland.
00:27:00.500 --> 00:27:01.723
It's a funny test.
00:27:02.836 --> 00:27:19.750
Yeah, it's a Huge crib, yeah, it's a small crib, meant to resemble some flying brand that flies with the wind from a neighboring building on fire and lands on the roof and should not make a larger damage on the roof than a certain length.
00:27:19.750 --> 00:27:41.164
And then we studied how this standard roof testing setup was influenced by adding a PV module, or in that case we used a simulated PV module made of a steel plate, basically so just seeing how the physical barrier above the roof would influence the fire dynamics on the roof.
00:27:41.654 --> 00:27:44.500
So the crib was underneath the panel, but on the roof layers.
00:27:46.017 --> 00:28:15.624
Yeah, so we tested the roof, started to test it in small scale, basically similar to the standardized fire test, and just added a steel plate representing PV module and at different distances from the roof, and we didn't really see much change before we approached the roof, I think the distance from the roof to the PV module was the lowest we tested was six centimeters, so that's quite a narrow narrow gap.
00:28:16.255 --> 00:28:20.942
That's pretty much like your ventilated facade from the first experiment.
00:28:21.855 --> 00:28:29.726
So that's probably lower than most of the building attached PV modules that are installed on roofs.
00:28:29.726 --> 00:28:41.099
But for the higher gap heights we didn't really see much change in the damaged length of the roof surface when installing the PV module.
00:28:41.099 --> 00:29:03.503
At 9 or 12 or 15 centimeters above, the damaged length was approximately the same, but at 6 centimeters we saw an increased length of damage, indicating that the flame was forced closer to the roof surface and that more of the heat was contributing to further fire spread upwards in the roof membrane.
00:29:04.976 --> 00:29:06.742
In this case because it was a mock-up.
00:29:06.742 --> 00:29:10.037
You had no interaction between the combustibles in the PV panel and the roof itself.
00:29:10.037 --> 00:29:11.603
So it's difficult to tell probably what would happen if it was a mock-up.
00:29:11.603 --> 00:29:13.569
But you had no interaction between the combustibles in the pv panel and the roof itself.
00:29:13.569 --> 00:29:15.936
So it's difficult to tell probably what would happen if it was um a real people.
00:29:15.936 --> 00:29:17.421
What was?
00:29:17.421 --> 00:29:25.300
The slope of the roof was like 45 degrees, 30 degrees we follow the the standard at 30 degree slope.
00:29:25.622 --> 00:29:53.635
So we started with basically the same test setup as the B-roof T2 test and then from that we used the same roof slope and roof materials and configurations and did medium and large scale tests as well to see what if we use a larger fire source and use larger wood creeps, larger wood cribs.
00:29:53.635 --> 00:29:58.089
And in those medium and large scale tests we used a gap height of 12 centimeters, in which that we didn't really see any influence in the small scale tests.
00:29:58.089 --> 00:30:13.567
But when we started adding a larger wood crib we were able to see increased damage length and with the largest wood crib that we tested the fire was propagating all the way up to the top of the roof.
00:30:13.567 --> 00:30:27.267
So this shows that it's not only the gap distance but also the amount of heat that you generate in the initial fire that also is determining whether or not the fire will propagate or not.
00:30:28.055 --> 00:30:30.744
In this setting, the fire is coming from the roof membrane.
00:30:30.744 --> 00:30:35.105
What about the fire that initiates inside the panel?
00:30:35.105 --> 00:30:46.548
Have you looked into like how can a panel ignite or what causes the fire of the panel itself, and then perhaps how it transitions into the roof membrane?
00:30:46.567 --> 00:30:57.365
No, yeah, of course, one of the interesting aspects here is what kind of fire would we expect from an electric fault in a PV installation?
00:30:57.365 --> 00:31:07.071
The standard roof test is basically a small flying brand that is carried with the wind, so it's a very small and light ignition source.
00:31:07.071 --> 00:31:18.459
But the most probable ignition source in the PV installations would be some electric arc igniting some of the plastic materials and starting a fire from that.
00:31:18.459 --> 00:31:24.742
So how much heat release would we expect from the electric arc?
00:31:24.742 --> 00:31:32.248
That's one source of heat, and then how much plastic materials would the electric arc ignite?
00:31:40.650 --> 00:31:43.836
That is a slightly different ignition source than something small and light that is carried with the wind and burns for a short time.
00:31:43.836 --> 00:31:46.923
In your paper I also saw an interesting summary of mechanisms to prevent this.
00:31:46.923 --> 00:31:51.050
Ignition in the PV panels, so you've mentioned electrical arcs.
00:31:51.050 --> 00:31:52.832
Ignition in the PV panels so you've mentioned the electrical arcs.
00:31:52.832 --> 00:32:03.259
Are there any other, like typical for PV modes of ignition and are there any means that engineers can implement to actually protect against them?
00:32:04.040 --> 00:32:22.724
Yeah, the DC arcing in PV installations is probably the most common cause of fire, as the electric arcs are more stable in a direct current system like a PV installation than in a normal alternating current system.
00:32:23.730 --> 00:32:25.778
Sorry, where is this arcing happening?
00:32:25.778 --> 00:32:28.259
Is it between the connectors between panels?
00:32:28.259 --> 00:32:33.001
Is it on the connections to the cables to take the electricity?
00:32:33.041 --> 00:32:33.241
away.
00:32:33.241 --> 00:32:44.398
The arcing can occur basically anywhere where you have a broken circuit, and one common place for these broken circuits are the connectors between the modules.
00:32:44.398 --> 00:33:13.964
If they're not well-made or well-fitted, of similar brands, they may start to have an increasing contact resistance over time and then this increasing contact resistance generates a little bit of heat, so you increase the temperature and then the contact can further deteriorate and start to corrode or to melt plastic and eventually it can generate enough heat so that the electric circuit is broken.
00:33:13.964 --> 00:33:31.537
And once the electric circuit is broken, then the current will continue to flow through the air in what we call an electric arc, and in these arcs you can have several thousand degrees temperature and it can easily ignite almost any combustible material around that arc, and that can easily ignite almost any combustible material around that arc.
00:33:32.356 --> 00:33:38.801
And I understand that the challenge is here, that there's no on-off button for the energy generation.
00:33:38.801 --> 00:33:46.667
As long as there's sun up there and the photons are hitting your silicon chip, the DC current is generated and it's not that easy to stop it right.
00:00:00.600 --> 00:00:01.804
00:00:01.804 --> 00:00:05.652
00:00:05.652 --> 00:00:21.553
00:00:21.553 --> 00:00:27.062
00:00:27.062 --> 00:00:30.782
00:00:30.782 --> 00:00:38.427
00:00:38.427 --> 00:00:46.814
00:00:46.814 --> 00:00:54.509
00:00:54.509 --> 00:01:02.551
00:01:02.551 --> 00:01:18.603
00:01:18.603 --> 00:01:30.751
00:01:30.751 --> 00:01:37.703
00:01:37.703 --> 00:01:43.765
00:01:43.765 --> 00:01:50.346
00:01:50.346 --> 00:01:58.751
00:01:58.751 --> 00:02:03.748
00:02:03.748 --> 00:02:06.614
00:02:06.614 --> 00:02:11.451
00:02:11.451 --> 00:02:19.367
00:02:19.367 --> 00:02:20.947
00:02:20.947 --> 00:02:22.048
00:02:22.048 --> 00:02:28.612
00:02:28.612 --> 00:02:30.062
00:02:30.062 --> 00:02:35.294
00:02:35.294 --> 00:02:41.965
00:02:41.965 --> 00:02:45.393
00:03:00.300 --> 00:03:03.450
00:03:03.450 --> 00:03:06.299
00:03:06.299 --> 00:03:17.233
00:03:17.233 --> 00:03:33.055
00:03:33.055 --> 00:03:44.715
00:03:44.715 --> 00:03:48.319
00:03:48.319 --> 00:03:55.753
00:03:55.753 --> 00:03:58.628
00:03:59.901 --> 00:04:06.991
00:04:06.991 --> 00:04:14.447
00:04:14.447 --> 00:04:23.814
00:04:23.814 --> 00:04:29.004
00:04:29.084 --> 00:04:29.225
00:04:29.225 --> 00:04:29.726
00:04:29.726 --> 00:04:32.375
00:04:32.375 --> 00:04:34.901
00:04:34.901 --> 00:04:53.413
00:04:53.413 --> 00:05:03.367
00:05:03.367 --> 00:05:05.632
00:05:06.680 --> 00:05:29.175
00:05:29.175 --> 00:05:33.471
00:05:33.471 --> 00:05:49.434
00:05:49.434 --> 00:05:53.091
00:05:53.091 --> 00:05:54.846
00:05:55.399 --> 00:06:04.800
00:06:06.608 --> 00:06:08.879
00:06:09.319 --> 00:06:18.232
00:06:18.720 --> 00:06:34.709
00:06:34.709 --> 00:06:41.545
00:06:41.545 --> 00:07:15.786
00:07:15.966 --> 00:07:30.026
00:07:30.026 --> 00:07:35.985
00:07:35.985 --> 00:07:40.353
00:07:40.353 --> 00:07:43.605
00:07:43.605 --> 00:07:47.754
00:07:47.754 --> 00:07:49.485
00:07:49.485 --> 00:07:52.687
00:07:53.519 --> 00:08:03.964
00:08:03.964 --> 00:08:24.274
00:08:24.274 --> 00:08:28.211
00:08:28.211 --> 00:08:40.769
00:08:40.828 --> 00:08:52.428
00:08:52.428 --> 00:08:54.466
00:08:54.466 --> 00:08:59.667
00:08:59.667 --> 00:09:02.649
00:09:02.649 --> 00:09:05.903
00:09:05.903 --> 00:09:15.294
00:09:29.748 --> 00:09:37.533
00:09:37.533 --> 00:10:02.469
00:10:02.469 --> 00:10:08.404
00:10:08.404 --> 00:10:21.160
00:10:22.222 --> 00:10:24.389
00:10:24.389 --> 00:10:31.428
00:10:31.428 --> 00:10:34.208
00:10:34.208 --> 00:10:40.587
00:10:40.587 --> 00:11:06.629
00:11:06.629 --> 00:11:08.707
00:11:09.441 --> 00:11:18.192
00:11:18.192 --> 00:11:19.995
00:11:19.995 --> 00:11:25.452
00:11:25.452 --> 00:12:05.852
00:12:06.813 --> 00:12:27.500
00:12:27.500 --> 00:12:39.927
00:12:40.830 --> 00:12:52.416
00:12:52.416 --> 00:13:03.533
00:13:03.572 --> 00:13:18.961
00:13:19.562 --> 00:14:10.451
00:14:11.480 --> 00:14:15.011
00:14:15.011 --> 00:14:25.033
00:14:25.033 --> 00:14:26.325
00:14:26.325 --> 00:14:29.330
00:14:29.330 --> 00:14:39.032
00:14:39.441 --> 00:14:45.714
00:14:45.714 --> 00:14:58.591
00:14:58.919 --> 00:15:00.025
00:15:00.046 --> 00:15:10.745
00:15:10.745 --> 00:15:18.677
00:15:19.320 --> 00:15:24.857
00:15:25.178 --> 00:15:26.883
00:15:26.883 --> 00:15:39.342
00:15:39.342 --> 00:15:50.975
00:15:51.380 --> 00:15:56.472
00:15:57.240 --> 00:16:03.591
00:16:03.591 --> 00:16:08.412
00:16:08.412 --> 00:16:15.653
00:16:15.653 --> 00:16:34.030
00:16:34.721 --> 00:16:35.840
00:16:35.840 --> 00:16:38.990
00:16:40.601 --> 00:16:42.187
00:16:42.187 --> 00:16:43.442
00:16:43.442 --> 00:16:58.812
00:16:58.812 --> 00:17:03.533
00:17:04.519 --> 00:17:14.046
00:17:14.046 --> 00:17:17.946
00:17:17.946 --> 00:17:19.866
00:17:19.866 --> 00:17:24.784
00:17:24.784 --> 00:17:30.518
00:17:30.518 --> 00:17:44.589
00:17:44.589 --> 00:17:44.691
00:17:45.595 --> 00:18:07.704
00:18:07.704 --> 00:18:32.767
00:18:33.596 --> 00:18:36.065
00:18:36.065 --> 00:18:38.963
00:18:39.476 --> 00:18:52.263
00:18:52.914 --> 00:18:56.345
00:18:56.345 --> 00:19:00.665
00:19:00.665 --> 00:19:03.182
00:19:03.182 --> 00:19:05.795
00:19:06.316 --> 00:19:06.876
00:19:06.876 --> 00:19:27.682
00:19:27.682 --> 00:19:30.731
00:19:30.731 --> 00:19:34.345
00:19:34.345 --> 00:19:46.625
00:19:46.625 --> 00:19:56.741
00:19:56.741 --> 00:20:12.684
00:20:12.704 --> 00:20:14.157
00:20:14.157 --> 00:20:15.821
00:20:15.821 --> 00:20:17.526
00:20:17.526 --> 00:20:18.047
00:20:18.086 --> 00:20:18.147
00:20:18.147 --> 00:20:28.921
00:20:29.301 --> 00:20:33.048
00:20:33.048 --> 00:20:35.102
00:20:36.436 --> 00:20:42.157
00:20:42.298 --> 00:20:58.494
00:20:58.494 --> 00:21:05.728
00:21:05.728 --> 00:21:08.478
00:21:08.478 --> 00:21:14.307
00:21:14.307 --> 00:21:17.176
00:21:17.176 --> 00:21:21.163
00:21:21.163 --> 00:21:23.467
00:21:23.467 --> 00:21:27.157
00:21:27.157 --> 00:21:28.118
00:21:30.021 --> 00:21:34.347
00:21:34.347 --> 00:21:45.199
00:21:45.199 --> 00:21:53.050
00:21:54.336 --> 00:21:54.919
00:21:54.919 --> 00:21:57.385
00:21:57.385 --> 00:22:01.285
00:22:01.285 --> 00:22:03.501
00:22:03.501 --> 00:22:05.903
00:22:05.963 --> 00:22:06.023
00:22:06.636 --> 00:22:23.125
00:22:23.807 --> 00:22:30.326
00:22:30.326 --> 00:22:39.109
00:22:39.109 --> 00:22:48.605
00:22:48.605 --> 00:23:03.303
00:23:03.723 --> 00:23:14.299
00:23:14.339 --> 00:23:32.538
00:23:32.538 --> 00:24:21.459
00:24:21.499 --> 00:24:35.015
00:24:35.015 --> 00:24:43.349
00:24:43.349 --> 00:24:47.766
00:24:47.766 --> 00:24:50.564
00:24:51.694 --> 00:25:06.263
00:25:06.263 --> 00:25:21.987
00:25:21.987 --> 00:25:33.875
00:25:33.875 --> 00:25:41.184
00:25:41.806 --> 00:25:56.734
00:25:56.734 --> 00:26:06.365
00:26:06.365 --> 00:26:15.684
00:26:15.684 --> 00:26:19.423
00:26:19.934 --> 00:26:22.263
00:26:22.263 --> 00:26:33.539
00:26:33.539 --> 00:26:41.625
00:26:41.625 --> 00:26:47.458
00:26:47.458 --> 00:26:58.063
00:26:58.144 --> 00:26:58.786
00:26:58.786 --> 00:27:00.500
00:27:00.500 --> 00:27:01.723
00:27:02.836 --> 00:27:19.750
00:27:19.750 --> 00:27:41.164
00:27:41.654 --> 00:27:44.500
00:27:46.017 --> 00:28:15.624
00:28:16.255 --> 00:28:20.942
00:28:21.855 --> 00:28:29.726
00:28:29.726 --> 00:28:41.099
00:28:41.099 --> 00:29:03.503
00:29:04.976 --> 00:29:06.742
00:29:06.742 --> 00:29:10.037
00:29:10.037 --> 00:29:11.603
00:29:11.603 --> 00:29:13.569
00:29:13.569 --> 00:29:15.936
00:29:15.936 --> 00:29:17.421
00:29:17.421 --> 00:29:25.300
00:29:25.622 --> 00:29:53.635
00:29:53.635 --> 00:29:58.089
00:29:58.089 --> 00:30:13.567
00:30:13.567 --> 00:30:27.267
00:30:28.055 --> 00:30:30.744
00:30:30.744 --> 00:30:35.105
00:30:35.105 --> 00:30:46.548
00:30:46.567 --> 00:30:57.365
00:30:57.365 --> 00:31:07.071
00:31:07.071 --> 00:31:18.459
00:31:18.459 --> 00:31:24.742
00:31:24.742 --> 00:31:32.248
00:31:40.650 --> 00:31:43.836
00:31:43.836 --> 00:31:46.923
00:31:46.923 --> 00:31:51.050
00:31:51.050 --> 00:31:52.832
00:31:52.832 --> 00:32:03.259
00:32:04.040 --> 00:32:22.724
00:32:23.730 --> 00:32:25.778
00:32:25.778 --> 00:32:28.259
00:32:28.259 --> 00:32:33.001
00:32:33.041 --> 00:32:33.241
00:32:33.241 --> 00:32:44.398
00:32:44.398 --> 00:33:13.964
00:33:13.964 --> 00:33:31.537
00:33:32.356 --> 00:33:38.801
00:33:38.801 --> 00:33:46.667