Professor José Torero gives expert evidence on testing standards and fire performance. Critiques UK's reliance on pass/fail testing, explains Queensland's first principles engineering approach, and exposes fundamental flaws in BS 8414 test including inadequate failure criteria and inability to detect encapsulation failures at different heat exposures.
00:00:26 good morning everyone welcome to today's hearing
00:00:29 hearing today we're going to continue hearing evidence from professor tarero so i'll ask professor torreira to come back into the room please [Music]
00:00:46 good morning professor good morning welcome back thank you all ready to carry on i hope yes good thank you very much yes mrs griffin yes thank you good morning professor torreira so when we broke off yesterday we were in the
00:00:58 broke off yesterday we were in the middle of discussing the system that the queensland government implemented after the grenfell tower fire and i now want to turn to testing and have a look at
00:01:10 have a look at their approach to testing and if we could go to the queensland submission to the inquiry again that's a qld 702 and i want to look at page 30
00:01:22 and i want to look at page 30 of this submission
00:01:26 so strategy five was the testing of materials
00:01:30 materials and uh underneath the text in bold under testing protocols it states this to reflect the first principles engineering approach presented within the cpd course
00:01:41 approach presented within the cpd course for fire engineers a testing protocol was developed in association with the university of queensland to allow fire performance to be tested at an individual material level this approach
00:01:52 individual material level this approach was preferred over existing techniques such as large-scale facade testing as it allows a fast and conservative analysis of the fire behavior of individual cladding components based on fundamental
00:02:04 cladding components based on fundamental principles and then i also want to look at the third paragraph there it says ultimately the purpose of this testing protocol was to provide the necessary material properties to quantify the external fire
00:02:16 properties to quantify the external fire spread hazard presented by the combustible cladding present on a building to inform a performance solution
00:02:23 solution and then they go on in this respect the testing regime complements the tests previously adopted by the queensland's regulatory system those are
00:02:33 those are as15301 15303 and 155113
00:02:40 while providing a cost-effective solution to the issue of product substitution which has resulted in significant loss of confidence in the as approved documentation of a building so that's what's said in the queensland
00:02:51 that's what's said in the queensland submission so does this mean that at least initially queensland decided to fundamentally change its approach to the testing of construction products that is correct
00:03:03 that is correct and can you just help with writing the first line that i read it says to reflect the first principles engineering approach presented within the cpd course can you just help what does that mean
00:03:14 does that mean yes so the there is a the queensland government clearly recognized that there were two problems at hand one of them was to identify the risk of existing buildings and the other one was
00:03:27 existing buildings and the other one was to provide solutions to these buildings now the identification of risks we got quite complicated because there was a very rapid realization that while
00:03:38 was a very rapid realization that while many materials were labeled in a certain way there were minor changes in composition and those minor changes in composition many times have a significant uh change in fire
00:03:49 have a significant uh change in fire performance and therefore it was uh first important to be able to identify what this materials that already existed in buildings uh where
00:04:00 already existed in buildings uh where and and that required quite an extensive protocol of material testing that used a number of state-of-the-art equipment that enabled to characterize this material so that we at least had a
00:04:13 this material so that we at least had a good understanding of what was on these buildings now the the information that was extracted from that material testing complemented uh with a number of ad-hoc
00:04:26 complemented uh with a number of ad-hoc flammability tests that were being conducted at a small scale at the university of queensland provided information very much in line with my presentation the values of fee thermal inertia ignition temperatures all these
00:04:39 inertia ignition temperatures all these different things as well as establishing as professor bisbee described mechanisms of failure for these different types of materials so it provided fundamental first principles information for uh
00:04:52 first principles information for uh practitioners who have been educated through the cpd process to be able to do some level of quantification uh of um
00:05:02 uh of um of the risk that this building's uh imposed
00:05:05 imposed now obviously in the australian system uh given that um we do not have a statement strategy but instead there was a possibility of different uh levels of
00:05:16 a possibility of different uh levels of of equation the focus was obviously on quantifying the potential for fire spread and uh so the fire spread velocity was the the
00:05:27 the the uh the key target uh of this first principles analysis enabling the engineers to actually establish you know the fire spread velocity and the extent
00:05:38 the fire spread velocity and the extent of the potential damage and involvement of the rest of the building that was possible but the intention was to deliver a holistic analysis so they could actually and analyze the full impact of having a fire and on the basis
00:05:52 impact of having a fire and on the basis of that be able to propose solutions that are not only were effective at delivering an adequate fire safety strategy but also they would potentially be much more cost effective than just
00:06:04 be much more cost effective than just completely removing for example everything from a building just because it didn't meet some simple criteria of combustibility yes thank you that's very helpful so just to be clear those tests
00:06:15 helpful so just to be clear those tests that are described in that last paragraph i read on the page the as1530 etcetera tests now they're similar aren't they in terms of some of our national small scale
00:06:27 of some of our national small scale tests and also the as1 5113 is similar to the bs test yes so as5114 was adopted after uh the the
00:06:39 the the docklands fire and uh and it's basically a
00:06:42 a a version of a414 is almost identical and uh one five three zero uh point one and one three five three zero point three are your non-combustibility tests
00:06:54 three are your non-combustibility tests very similar to uh to the test that we have in the uk yeah so similar to the 476 part four test that we have in the uk and i think what you've just described is that what they did is take those tests that they'd used for a while
00:07:05 those tests that they'd used for a while and then complement them supplement them with uh additional tests which provided data for the crucial fire performance metrics
00:07:15 metrics that you need to be able to assess an external wall system exactly yeah and how successful was that system while it was operating well i mean it's it's difficult to to describe a metric of
00:07:28 difficult to to describe a metric of success other than the fact that the system uh
00:07:32 system uh was implemented and and the cpd courses happened and and people were asked to provide the materials
00:07:43 to provide the materials in their buildings so that the university of queensland will make the assessment and publish all those materials openly in the materials library and um the government paid for the tests and uh and everybody
00:07:55 the tests and uh and everybody volunteered their materials and therefore the library is populated with i think several hundred uh different types of products right now and um and many people have used the data of the
00:08:06 many people have used the data of the library and not only in conducting assessments for performance but also particularly to try to identify the specific products so in that sense i think the approach was extremely successful now when it came to the
00:08:18 successful now when it came to the actual practice of conducting these risk assessments then it's much more difficult to establish success because obviously there are certain practitioners who embraced uh the the
00:08:30 practitioners who embraced uh the the cpd educated themselves to the inappropriate level and started delivering an appropriate service and uh and i believe many of them were very very successful at uh demonstrating
00:08:41 very successful at uh demonstrating appropriate schemes that enabled to have much safer buildings
00:08:46 buildings but
00:08:47 but in that sense that
00:08:50 that more or less never got fully resolved because it never became mandated that these individuals who were accredited according to this process will be the
00:09:02 according to this process will be the only ones who actually could do that and uh and and i think what ended up happening is that the system uh start being used by people who did not have
00:09:13 being used by people who did not have necessarily the competency and and it's not very clear exactly where we are right now yes i understand and and i'm going to ask you a little bit more about the cladding materials library which you began to explain just
00:09:25 library which you began to explain just then in more detail in a moment but just just pausing there what was the role of the large-scale testing within that first principles engineering approach was it
00:09:37 engineering approach was it just to validate check uh
00:09:41 uh conclusions that had already already been reached by applying uh the results from smaller scale testing or what was known about the materials that that was potentially the case i
00:09:52 that that was potentially the case i think in many cases it was just a validation so you have as part of the process people will actually predict the spread rates in the test now the test
00:10:03 spread rates in the test now the test has a very very coarse estimate of the spread rate in the sense that you only have two points to calculate how fast is spread but nevertheless you have videos and you have other information so many people actually use the tests
00:10:16 people actually use the tests as a mechanism to validate the analysis that they were making now other people use those tests also for additional information so it one of the good things that this approach did is that it uh
00:10:28 that this approach did is that it uh encouraged practitioners to actually interrogate the information that they had from the test in a much more thorough way so instead of just taking a
00:10:39 thorough way so instead of just taking a very simple uh pass fail criteria and when i say pass fail criteria i don't mean that there is a pass it's just a binary uh line that says you failed and
00:10:50 binary uh line that says you failed and um
00:10:51 um but take taking this sort of binary criteria which is the approach that we have taken uh in the uk instead what people did was interrogate the test so they they were looking at photographs of
00:11:02 they they were looking at photographs of the aftermath they were trying to in tests that were conducted later they were interrogating the debris because in many ways what they had to do is gain comfort that the calculations
00:11:13 is gain comfort that the calculations that we're making you know were in many ways um
00:11:18 ways um supplemented and validated by the test yes i understand does that mean that it had more instrumentation on the rig compared to what we would see with an 8414 test
00:11:29 with an 8414 test that is a very interesting point that you make because i have heard of numerous cases where the practitioners required much more instrumentation and also required modifications to the test
00:11:41 also required modifications to the test so for example uh something that was done um
00:11:44 done um several times was there were certain buildings that had very minor areas with these types of materials and and the practitioners requested
00:11:56 and the practitioners requested uh
00:11:58 uh a number of different tests where different options were tested of what to remove what not to remove in such a way that they would create fire breaks for example so so yes i mean
00:12:09 example so so yes i mean it really as i say it really encouraged practitioners to push the bounds of the test beyond what the standard required yes so to make those large-scale tests more bespoke depending on the phenomenon
00:12:21 more bespoke depending on the phenomenon you thought you might get and that you needed to validate that's right exactly yeah
00:12:26 yeah so we know that queensland established this cadding materials library it's available for anybody to log into you you register don't you and and you can get a log in easily and look at the
00:12:39 can get a log in easily and look at the library so i've done it anyone can do it yes yes um let's go over to page 31 of this um submission where we see that cladding materials library explained in more detail
00:12:51 more detail um
00:12:52 um and
00:12:55 we can we can see there that they say using the testing protocol developed by the university of queensland the queensland government supported investment in the necessary specialist equipment staff and
00:13:06 necessary specialist equipment staff and product samples to establish a database or clouding materials library the data will assist competent fire engineers in assessing the fire hazardous cladding materials on existing buildings
00:13:17 materials on existing buildings dhpw developed a methodology for sourcing and cataloging hundreds of product samples taken from government-owned buildings to overcome the challenge of unknown products and product substitution a sample collection
00:13:29 product substitution a sample collection strategy was developed with a range of samples required for testing is largely based on the building grouping assessment and then so they tested according to different categories
00:13:40 according to different categories um and
00:13:41 um and i think what they're saying there is with a high-rise building for validation reasons they require a minimum of 10 samples and up to 15 samples to check that
00:13:52 samples to check that what they were sampling was consistent is that correct yes it was an issue of consistency the bigger the building the more the surface area so you have to get more samples to make sure that certain parts of the building were not clouded
00:14:03 parts of the building were not clouded with different materials in other parts of the building yes
00:14:06 yes and if we go back to your report and what you say about this in your report this is at jtor 706 page 189.
00:14:18 and at paragraph a216 you say this you say the most extensive among all these studies is the university of queensland cladding materials library this work was part of a comprehensive programme funded
00:14:29 part of a comprehensive programme funded by the state of queensland that aim to address the assessment of risk placed by combustible clouding materials the library followed a rigorous methodology that first provided methods of identification of the different
00:14:40 identification of the different components of facade systems these methods used state-of-the-art diagnostics including many of the diagnostics used by mckenna attal the identification of the material was followed by a comprehensive set of small
00:14:52 followed by a comprehensive set of small scale flammability tests with the aim of extracting all relevant flammability properties the properties aim to enable engineers to use this information in models that could serve to predict the
00:15:03 models that could serve to predict the risk imposed by different materials used in facade systems a series of tests aimed at providing further data in support of these models was also proposed for this purpose a
00:15:14 was also proposed for this purpose a professional development program was designed to ensure that professionals conducting risk assessments had adequate competency to conduct these analyses the professional development program was assessed by means of examination and a
00:15:26 assessed by means of examination and a certificate provided by the university of queensland and i think you can confirm when you go to the online library and you put in a product such as pir insulation you get a
00:15:36 you get a a very significant quantity of information provided in graphs diagrams tables that goes for some way down the page giving you all sorts of information about the flammability properties that's
00:15:48 about the flammability properties that's right isn't it yes you will get a very extensive amount of information and also you will get a number of variants so the pir there's many versions of pir so you
00:15:59 pir there's many versions of pir so you will see the different versions of pir that were submitted and you can see the differences between the different products many of those actually will be uh sometimes labeled with the same
00:16:11 uh sometimes labeled with the same product name and nevertheless they're distinctively different yeah so presumably that that did require the cooperation did manufacturers buy into this process as well and submit their
00:16:22 as well and submit their materials or was it just come from government buildings uh it mostly came from government buildings i am not aware of any manufacturers volunteering uh their products but
00:16:34 their products but i'm not 100 sure um i didn't follow every single one of the products i got submitted yeah and we saw there that you said um it's about six lines up that a series of tests
00:16:47 six lines up that a series of tests aimed at providing further data in support of these models was also proposed do we take it from that that those series of tests didn't end up being carried out yes the the approach that the
00:16:57 that the the program intended to follow was that if you rely on the very very basic material data that enabled you you just to identify the product and use the data
00:17:09 to identify the product and use the data that was collected for that specific product then the decision-making process when it came to the way in which you addressed the performance of the fire safety strategy
00:17:21 performance of the fire safety strategy had to be enormously conservative so that was a principle now if you progressed to incorporate an analysis that included a lot of the flammability variables with the small-scale tests
00:17:32 variables with the small-scale tests that were more fire type of tests and you could quantify things in a slightly more precise manner that enable you to be slightly less conservative but remain within the bounds of appropriate safety
00:17:43 within the bounds of appropriate safety now the idea was that the next level was to move to more of a system performance where you actually had intermediate scale tests very similar to the ones that professor bisbee described and uh and and then you
00:17:55 bisbee described and uh and and then you will have understood better many of the failure modes and that will allow the engineer to make an even more informed decision that potentially will open the door for even a broader range of
00:18:06 door for even a broader range of materials and the ultimate goal was to conduct a number of validation large-scale tests to actually demonstrate all these forms of behavior now the latter two stages never happened
00:18:17 now the latter two stages never happened uh because that's where the program stopped
00:18:19 stopped right yeah and can you explain why that program stopped um i think i think in essence um uh it's it's the the the the very common conflict that
00:18:31 the the the very common conflict that happens when you get opposing uh opinions uh that somebody at the end has to anoint the expert in this particular case it was a minister and the ministry
00:18:42 case it was a minister and the ministry got a lot of pressure from a number of different spaces and they effectively because they recognized that the process proposed by the university of queensland was a long
00:18:55 of queensland was a long and significant process that got weighted in the decision of of making of listening to different parties and then at the end the decision was to stop
00:19:08 then at the end the decision was to stop where they were not go any further and allow the market uh to decide uh what was going to happen under the constraints of a very simple ban just like
00:19:18 like new south wales and victoria has done yeah
00:19:21 yeah if you were designing an idealized testing system would you say that that ought to involve taking the something like the queensland data and and taking it through to system
00:19:33 data and and taking it through to system level so that you then would have a database which covered individual material performance product performance and then system performance uh yes i mean it's no secret that i to
00:19:44 uh yes i mean it's no secret that i to some extent was involved in in this process and um and i had already left the university of queensland but i have a close relationship with the people that are there so this
00:19:55 that are there so this model uh is is very consistent you know with my way of thinking so if i i think uh if if you look at the uh sort of the the range of conformity
00:20:08 uh sort of the the range of conformity um
00:20:09 um levels that i put in my diagram in in in in the report and that that that we discussed during my presentation i think one of the things that we need to recognize is that if we are a
00:20:20 to recognize is that if we are a manufacturer and we don't provide any relevant information then the onus is on the later stages so we're passing passing the problem to the later stage and of course because we're
00:20:32 later stage and of course because we're throwing different products and those products are unqualified and quantified and characterized then the the complexity of the job of those who come next the designers and the people who
00:20:43 next the designers and the people who are constructing and implementing the building is obviously extremely high extremely complicated and requires a level of competency that is extraordinary but if i responsibly take my own product and i
00:20:56 responsibly take my own product and i developed it characterized do all this background work that is in the materials library and provide all the the information that allows the engineers to
00:21:07 engineers to confidently make an appropriate assessment i have by doing so reduce the onus on the later stages of the process and enable these engineers to actually
00:21:18 and enable these engineers to actually operate
00:21:19 operate in an efficient and safe manner without necessarily having to understand the whole chain of competency that goes from polymer chemistry you know all the way
00:21:31 polymer chemistry you know all the way into the thermomechanical behavior of an encapsulation system so
00:21:35 so every
00:21:38 every part of the chain has a responsibility to take part of the knowledge because it's extremely broad in nature it's a very multi-disciplinary problem so so if if i was a product manufacturer i would
00:21:50 if i was a product manufacturer i would make sure
00:21:51 make sure that i fully characterized to as much as i am capable the product so that i deliver to the next stage of the design process something that already has a great
00:22:03 something that already has a great certainty that i do understand the behavior the failure modes and and so forth yeah but i suppose that might depend on whether they're producing a single product
00:22:14 product um or some kind of system which which then is for use in the market if they're if they're producing a whole cladding system and selling that that's one thing but if they're just producing an individual product
00:22:27 producing an individual product doesn't any competent system or good systems still need to have people further down the chain that can make an assessment of system performance of course i do i mean i think i think it
00:22:39 of course i do i mean i think i think it is just where you uh hand over the baton i mean that's that's the whole thing no it's a if if you are just the material manufacturer you know then you're going to characterize the
00:22:50 you're going to characterize the material as as a material and uh and if you're going to pass it on to a product manufacturer that then is going to include an encapsulation then it is the product manufacturer that should be
00:23:01 product manufacturer that should be characterizing the encapsulation taking into consideration the potential applications of the product they're going to market a product with a functionality and because they're marketing a product with a functionality for a specific type of application then
00:23:13 for a specific type of application then they should understand for example the failure modes of the encapsulation now obviously then the the organization that is going to assemble it into a cassette and now they're going to include several
00:23:24 and now they're going to include several materials and they're going to include an array geometrical arrangement fixations and so forth then they have it is the onus on them on characterizing how that system behavior operates and then that's going to be put in a
00:23:36 then that's going to be put in a building and then it is the designer of the building that should take all this information and make sure that it works with that building you know and taken into consideration height and and use of the building and so we go through the
00:23:47 the building and so we go through the chain now the the the question is at which stage each individual group hands over the responsibility and what information they
00:23:58 responsibility and what information they are required to provide if we don't provide anything from the beginning and we let the whole process go all the way till the end then those people who are actually designing and implementing the building have to take
00:24:09 implementing the building have to take the full responsibility over an extraordinary wide range of competencies and that is not possible because i mean you will you need a renaissance man to be able to do that so at some point the
00:24:20 be able to do that so at some point the baton needs to be handed yes and it has to be done in a responsible way yes i understand so i think what you're saying is if you were the manufacturer of an acm
00:24:28 acm composite material product two skins of aluminium with a pe core there's actually quite a lot the manufacturer could do to faithfully explain its performance in fire and the potential ways that the
00:24:40 fire and the potential ways that the encapsulation by the aluminium can fail with different fixing systems that then would really inform those at the next stage of the process even if that's then going to be used perhaps with an insulin
00:24:52 going to be used perhaps with an insulin in a system absolutely and i think that that was the principle behind the materials library it was that stage more or less that brings you to what we
00:25:03 what we assume is the responsibility of the manufacturer so it brings you to that level that's where it stopped yeah but would it be possible for somebody to pick up that data and keep moving forward with it and
00:25:15 data and keep moving forward with it and that's what the cpd was for so the cpd was to enable engineers to pick up that data and be able to actually do the fire engineering you know and deliver an adequate holistic fire safety strategy yeah thank
00:25:27 holistic fire safety strategy yeah thank you
00:25:29 you um
00:25:30 um and and just to be clear i think you touched on it earlier in terms of the the assumption in queensland um and indeed across australia more generally what's their approach to stay put versus evacuation or phased
00:25:43 put versus evacuation or phased evacuation strategies yeah as i indicated yesterday i think in general the conventional approach would be a phase two equation strategy and therefore the focus in this particular case was the rate of flame spread yeah
00:25:55 case was the rate of flame spread yeah so that's what they were trying to measure with all this data yes to check that that time was less than the time to evacuate exactly and i think as you've already touched
00:26:06 and i think as you've already touched upon queensland has recently abandoned that holistic system and has gone to more blunter tools in terms of dealing with the cladding system is that right yes i mean i think
00:26:17 system is that right yes i mean i think uh
00:26:18 uh it's not that they have abandoned it it's just they stop enforcing it and uh and by doing so it was equivalent you know to abandon it because the simpler approaches uh became more the
00:26:29 simpler approaches uh became more the norm i believe yes and you've given some reasons why different aspects of the model didn't end up finding favor and didn't get the enforcement behind it
00:26:41 and didn't get the enforcement behind it behind it that it was perhaps originally intended
00:26:44 intended can you just summarize for us the key reasons why this system didn't succeed i i think the the the fundamental reason was that there was a perception
00:26:57 was that there was a perception that um that one it was going to take too long
00:27:01 too long uh to be able to put all this machine in place and and there was a pressing need for much faster answers
00:27:13 for much faster answers and because keep in mind that if you follow this process you have to do a very thorough analysis of every building and that very thorough analysis brings a cost and it brings time so i think the
00:27:24 cost and it brings time so i think the cost and the time you know was clearly um a very significant uh variable on it and uh and i think um and you know i i i want to repeat that i
00:27:35 and you know i i i want to repeat that i think the the other one is is the uncertainty of who's right i i think that that in in in a space where professional practice is not clearly defined
00:27:47 clearly defined there will always be two parties and uh at least that are going to argue what if the solution is appropriate so what you saw was a lot of
00:27:58 you saw was a lot of conflict
00:27:59 conflict so you have an engineer that will apply all these first principles that will take all this and embrace this knowledge and propose a solution and obviously that solution carried a very significant
00:28:10 that solution carried a very significant cost now then you go into litigation there will be another engineer that will say that's not necessary can i just intervene um i'd like just to go back a little bit to what you were saying about manufacturers providing information
00:28:23 manufacturers providing information about their the behavior of their products
00:28:27 products and and i can see the force of the suggestion that if they do that then the designers and so on have something reliable to work on
00:28:37 does what you suggest envisage that there should be some legally binding requirement on manufacturers to test for certain types of properties and to publish the results
00:28:48 publish the results uh i i i
00:28:50 uh i i i it is a very interesting question i think um
00:28:54 think um from my perspective legally binding can be in numerous different places and and i think that when it comes to safety
00:29:05 safety you actually do need that legally binding requirement that is that is placed on whatever entity is putting forwards a
00:29:16 whatever entity is putting forwards a product now that legally binding requirement doesn't necessarily have to be met as a matter of compliance or as a matter of delivering safety but it could
00:29:27 matter of delivering safety but it could be a legally binding requirement to report
00:29:30 report a list of information that pertains that material
00:29:33 material so the the the issue of conformity from my perspective is what it is important if you require a minimum amount of information to be published on that material openly and
00:29:46 published on that material openly and for those manufacturers who are selling the products to actually be open to accepting
00:29:52 accepting people analyzing those for conformity you know the equivalent of a label and uh and i i think i think that that that to me is something that i believe is is fundamental because nobody knows
00:30:05 is is fundamental because nobody knows better than the manufacturer what they are producing and they safety is a functionality so they have to be accountable for that functionality that they're providing yeah
00:30:15 yeah that's helpful thank you very much thank you mr andre yes thank you just one last question um about the queensland system or indeed australia more generally has anyone grappled um in any detail with the problem of
00:30:27 um in any detail with the problem of those with with disabilities with mobility problems was that something that was ever addressed as part of the queensland system absolutely i mean i think the moment that you put forward a holistic fire
00:30:39 that you put forward a holistic fire safety strategy um and particularly if you're going to rely on an egress strategy there has to be a consideration and appropriate provisions with
00:30:50 with everybody that can actually enter the building and so so so clearly uh this type of approach enables you to assess carefully what would be the performance
00:31:01 carefully what would be the performance of the building in regards to the integrity of the population and so it doesn't be it doesn't become an assumption you know that that that people can stay and it will be fine it becomes an
00:31:12 and it will be fine it becomes an explicit analysis that shows that people can stay and they will be fine and and i think that had to be taken into consideration so the fact that most people will evacuate doesn't necessarily
00:31:23 people will evacuate doesn't necessarily prevent
00:31:24 prevent uh creating a safe space for a smaller number of people that actually will have to remain within the building so does that basically mean things like refuges
00:31:34 refuges that's an example for example a a way to do it yes yeah so because you've got an egress strategy you've got to think about whether these people are going to stay if they're going to stay where is that going to be
00:31:45 going to stay where is that going to be safe if not how long is it going to take them to evacuate or use a lift or whatever is that is that the point yeah so so so proactively you put all the necessary provisions and you make explicit how those probations are going
00:31:57 explicit how those probations are going to work you know it it it is about to be honest with you for me it is really more about having a plan and demonstrating that the plan works and uh and and because you're addressing
00:32:09 and uh and and because you're addressing a fraction of the of the population then that plan is going to require resources that are commensurate you know to the expectation that you have of of what fraction of the population of
00:32:22 of of what fraction of the population of the building is going to be required that that to support yeah thank you so i now want to turn back to some some quite detailed questions about the
00:32:33 quite detailed questions about the current uk regular regulatory approach to testing um now in your presentation and you've touched on it again this morning you identified these key performance metrics
00:32:44 identified these key performance metrics which are in your opinion key in terms of assessing external flame spread including ignition flame spread extinction and thermo mechanical performance and you highlighted the need to quantify
00:32:55 and you highlighted the need to quantify heating length and split flame spread velocity as key parameters which may be particularly important in order to understand whether particular fire safety strategies are going to work or not yes
00:33:07 are going to work or not yes now
00:33:08 now in your report what you've done as well is you've analyzed some reaction to fire tests that have been used in the uk um and you you've you've critiqued them
00:33:19 um and you you've you've critiqued them essentially and i just want to start with some just a few questions because you've covered it in detail in your report about the bs 476
00:33:27 476 testing regime and i want to ask you about bs 476 part 6 and part 7. so part 6 we know is the fire propagation test and if we can see a copy of this
00:33:39 and if we can see a copy of this standard at ctar 6016
00:33:47 there we have it it's a method of test for fire propagation for products and if we go to a diagram of the test which appears at page 13
00:33:57 this is a diagram i think of the rear of the test just to remind us what the apparatus looks like for it and just to recap it's that chimney at the top which records temperatures of
00:34:08 the top which records temperatures of the gases leaving the chamber is that right yes
00:34:11 right yes and the output of this test is expressed as a fire propagation index which is established by measuring the temperature on in the gases in the chimney of on the one hand the calibrated sample and then
00:34:22 one hand the calibrated sample and then on the other hand several specimen samples is that your understanding yes that's correct now does fire propagation index does that have any correlation to real
00:34:33 real fire performance um
00:34:37 um not in itself um i think that uh that obviously the measurement that you're making is a thermocouple measurement so you're measuring a temperature increase and the temperature increase is directly
00:34:48 and the temperature increase is directly related given the condition to the of the test to the combustion that has occurred on the inside of the chamber and in as such you can
00:34:59 and in as such you can take that information and extract the heat of combustion and the heat of combustion is one of the parameters that governs um flame spread so in principle you're getting one component you know of
00:35:11 you're getting one component you know of the whole picture and uh but still requires analyzing the information of the test to extract the variable the index in itself it is fundamentally an arbitrary index
00:35:23 it is fundamentally an arbitrary index that uh that sets certain thresholds and uh and those thresholds are are indicative but they are not related to the physics of the problem yeah and is it possible to establish the
00:35:35 and is it possible to establish the thermal exposure of the sample well
00:35:40 well not with the test as designed and the only thing that you can establish is that you are under conditions that normally will ignite most materials and uh but as as the test is is utilized uh
00:35:54 uh but as as the test is is utilized uh you will have to instrument it much more heavily to be able to get some sense and characterize the thermal exposure does that mean that
00:36:05 thermal exposure does that mean that it's quite difficult to relate the performance of this sample in this particular test with any particular design scenario uh it it
00:36:15 uh it it it limits the information you can take so you know that at those temperatures it will ignite so if the material is combustible it will ignite and uh and it will give you a sense of how much heat releases in the
00:36:28 a sense of how much heat releases in the process of ignition but so in many ways it limits what you can get purely to just that heat of combustion after an analysis you cannot get any further
00:36:39 analysis you cannot get any further information so for example the moment that is
00:36:42 that is defined as a as a propagation apparatus it really doesn't tell you anything about how fire propagates because since you don't know what the external heat flux is the exposure the
00:36:54 external heat flux is the exposure the propagation is just arbitrary and it's just purely dependent on the test itself so i wouldn't rely on anything associated to propagation all that i can get a sense of is
00:37:06 all that i can get a sense of is and for me more than the index is from the thermocouples is how much heat these materials are capable of releasing yes thank you that's very helpful and and then moving on to the bs 476
00:37:17 and and then moving on to the bs 476 part 7 test that's the method of test to determine the classification of the surface spread of flame of products we've got a picture of this test being performed in your report if we could go
00:37:28 performed in your report if we could go to that jtor 706 page 131
00:37:33 page 131 let's figure 20 at the top of the page so that is a picture of that test being performed so we've got the the radiant panel to
00:37:44 so we've got the the radiant panel to the left and then the sample held in its frame on the right yes yes and we can see there that a flame is progressing um horizontally across that panel and
00:37:55 um horizontally across that panel and there's various lines we can see which are various parameters that are being tested yes yes now
00:38:03 now so is it right that essentially this test measures whether and how fast a flame spreads laterally across a product is that correct that's correct
00:38:14 is that correct that's correct does it tell us anything value about ignition
00:38:19 you could treat the data and extract an ignition temperature out of it so the north american version of this which is astm e1321
00:38:30 astm e1321 basically is fundamentally the same test but it incorporates a methodology of data analysis and the methodology of data analysis is much more thorough than uh than the british test and it
00:38:41 than uh than the british test and it allows you to extract out of the results of the test and ignition temperature so it could potentially give you information about ignition if analyzed correctly but if used
00:38:53 correctly but if used the way in which the standard describes that information is never extracted right yes and does it tell you anything about the propensity for vertical flame spread
00:39:04 spread again um so there is uh i explained this parameter fee that uh that represents the amount of energy and the the heating length and that that value of v can be extracted
00:39:16 that that value of v can be extracted for the test and that value of e will correspond only to lateral spread now
00:39:24 now i could analyze this using a much more complex approach and try to change uh the interpretation of fees so that i
00:39:35 uh the interpretation of fees so that i can get relevant information that could potentially allow me to extrapolate these results to vertical fire spread but without that detail analysis and that is a complicated analysis because
00:39:46 that is a complicated analysis because you have to reanalyze the flow structure and transform completely the way in which the heat from the flame goes to the material and you can bring it back and extract
00:39:58 and you can bring it back and extract something but then uh it will be have to it will have to be caveated by a big error bar because that extrapolation is not trivial yes i understand
00:40:09 yes i understand um
00:40:11 um and i think in your report you refer to studies of european tests which correspond with bs 476 part 7 which in fact give very different results for the same materials is that right yes i mean
00:40:22 same materials is that right yes i mean what what ends up happening is a lot of these tests because they're run under different conditions and those conditions are not characterized appropriately and uh and in principle
00:40:33 appropriately and uh and in principle the
00:40:33 the the measurements and the results are extremely simplistic and the analysis method is extremely simplistic every test is going to change some variables and you're going to get a
00:40:44 variables and you're going to get a slightly different answer and therefore what you get is conflicting class what appears to be conflicting classifications but really what it is is that we're running tests in the blind and so we are we're really just running
00:40:55 and so we are we're really just running a test and getting getting some indicative values but those values are not directly linked to the fundamental parameters and therefore it is extremely difficult to compare
00:41:06 it is extremely difficult to compare this test you will get very contradictory results right so you could run
00:41:11 run five tests on the same product and get wildly varying results is that what you're saying yes i mean there there's a famous study by howard emins which is a professor at harvard you know from the 1970s when he takes all these tests and
00:41:23 1970s when he takes all these tests and he explains how you can actually run a fire spread test around the world on the different standards and basically get results in all directions and uh and and that but but but in essence i think the
00:41:35 that but but but in essence i think the point that i want to emphasize it is that it is not the test itself it is the way in which we use them that we are not really extracting the the fundamental information what we're doing
00:41:46 fundamental information what we're doing is trying to get a simplistic output without considering the complexity and all the variables that are associated to the test i could use this test and do a proper analysis and actually get some
00:41:58 proper analysis and actually get some information from the data you know make the right measurements and that information if i was to do a different test
00:42:07 test and
00:42:07 and get do the same level of high you know highly sophisticated analysis and i will probably get the same fundamental parameters yes can you comment then on the
00:42:19 yes can you comment then on the appropriateness of using those two tests we've seen that in our system or certainly in the the guidance in approved document b that the combination of these two tests was used to get for example national
00:42:31 was used to get for example national class nor can you comment on the appropriateness of using those tests in the way they were used so not the detailed analysis you're talking about to assess external fire performance uh
00:42:42 to assess external fire performance uh it's highly inappropriate i mean it it doesn't represent the the full set of parameters uh it's too simplistic of an analysis and particularly is is highly
00:42:54 analysis and particularly is is highly inappropriate when it comes to the assessment of products that are composed of more than one material because the composite form of behavior is one that for example this particular
00:43:07 is one that for example this particular test provides absolutely no information because it's a surface flame spread test so it is only the exposed material that becomes relevant and whatever is happening behind
00:43:18 and whatever is happening behind that really doesn't um materialize in in the test so in general i think
00:43:24 i think these tests are are used uh the way in which their use is too simplistic for the types of materials that we're talking about so it's highly inappropriate yeah
00:43:35 inappropriate yeah um and another point you make in your report is that that radiant panel that we see there imposes a heat flux of between 5 and 32 and a half kilowatts per meter squared but as you've said and as we've referred to many times
00:43:47 as we've referred to many times external fires in buildings can impose a heat flux of something like 120 kilowatts per meter squared exactly which fundamentally disqualifies this test as a scenario test you know because
00:43:59 test as a scenario test you know because it doesn't represent the scenario so even if you want to say well because it worked like this here it will work the same way there that that not even that is possible because from the perspective of a scenario test is completely
00:44:10 of a scenario test is completely disqualified by the fact that it it introduces a much lower uh heat flux yeah
00:44:16 yeah and and so in summary um is this fair what you say in your report is that these tests actually don't provide some certain
00:44:24 certain types of key information certainly in the simplistic way in which they're run and reported on in the uk about the key parameters that you need to assess external flame spread that is correct
00:44:35 correct um and in fact they weren't intended to do that that was not never their intention no they were never intended and and it's therefore always going to have been necessary although people
00:44:47 have been necessary although people might not have appreciated it for the competent professional to take the data from these tests and very carefully analyze it to see what it's actually telling you and what it's not telling you is that fair yeah that's absolutely
00:44:59 you is that fair yeah that's absolutely correct and there is one additional step to that i think it is to the competent professional like for example in the case of an acm to be able to recognize that this test doesn't provide any information at all
00:45:11 information at all uh on the performance of of of an aluminum composite panel so so so i think i think the responsibility from my perspective of a competent professional is not only knowing how to interpret the data appropriately but also it is a
00:45:24 data appropriately but also it is a responsibility to be able to disqualify the use of certain tests that are highly inappropriate for a specific product yes um so i now want to ask you some questions about the european reaction to
00:45:36 questions about the european reaction to fire classification and testing regime um and we know that there's a overarching classification system bs en 13501 part 1.
00:45:48 13501 part 1. if we look at page 138 within this report
00:45:52 report at figure 22
00:45:56 we can just remind ourselves of the list of classifications that the european system uses and uh in the those columns we can see for example that
00:46:08 for example that um often a particular classification is achieved only by a combination of tests that's right isn't it yes so a1 is achieved by combination of the en iso
00:46:19 en iso 1182 test
00:46:21 1182 test and the en iso 1716 test
00:46:27 and it that's a thing this is a single classification system isn't it that's applied in in europe um but is it right and i think you make the point in your report that it's attempting to capture a very broad range
00:46:40 attempting to capture a very broad range of physical phenomena relevant to fire and that's they're trying to achieve it by the broad range and complexity of the resulting testing protocols is that a fair description yes that that
00:46:53 is that a fair description yes that that that is correct i think uh this test is a very good example that illustrates um what is behind the definition of a test and i uh
00:47:04 and i uh if you don't mind i would like to take a step back
00:47:07 step back and compare the physics of what this test is intending to to describe do you mean particularly the um the sbi the single burning item test uh
00:47:18 the sbi the single burning item test uh yes the single bending anti-i i can test which in many ways also describes the room corner test yes let's look at a diagram of that um so at page 149 of
00:47:29 diagram of that um so at page 149 of your report so we know that this sbi test is critical to the classifications for many of these classifications b to d and then a2
00:47:40 b to d and then a2 um
00:47:42 um and just by way of background you tell us in your report that all the other tests that go to the classification system already existed or adapted from existing tests
00:47:53 existing tests but
00:47:54 but en13823 this single burning item test was specifically created to fill a gap in the data is that correct uh yes the single burning item was was created uh to
00:48:08 re reduce the scale of a phenomena to a level where it was still appropriate yes so that the test could be run faster and and cheaper
00:48:20 and cheaper than than
00:48:22 than than a full compartment which is a room corner test yes so i think it's right to say isn't it that it's based on a test called the room corner test that was iso 9705
00:48:34 9705 is that right and was that a large-scale test yes
00:48:37 test yes um
00:48:38 um and the single burning item test uses the same reference scenario as that room corner test is that right yes and so here we have a diagram do you
00:48:49 and so here we have a diagram do you want to just explain to us the difference in terms of the physics of what this test is measuring compared to the physics of what you need to look at for external wall performance yes so so
00:49:01 for external wall performance yes so so this is this is a very very good example because i think in many ways illustrates uh how we sometimes miss the nuances so if you if you think about the purpose
00:49:12 so if you if you think about the purpose of this test the purpose of this test is to try to establish how much time people have available when they are in a compartment where a fire starts
00:49:23 compartment where a fire starts and uh and the the sort of reference event
00:49:27 event that is used is flashover so you know that when you reach flashover then conditions in the room will be completely untenable and therefore by
00:49:40 making sure that everybody has evacuated that room way before flashover has happened you can have a measure of safety that that that is that is the basic principle so so what this test
00:49:52 basic principle so so what this test does is it basically looks at flash over or the time to flash over as the mechanism that uh is going to be used to classify
00:50:02 classify wall linings and so it is designed originally for the purpose of assessing wall linings and um and and and it's looking at a phenomena that is specific which is the time to
00:50:15 that is specific which is the time to attain flushover so the process until you get to flashover goes from the origin of the fire until the moment you attain flashover by
00:50:26 until the moment you attain flashover by compromising the linings that's the idea so
00:50:30 so the initial fire the initiation fire is very small because you're at the very early stages of the fire and the compartment in itself is cool so what you want is to make sure first is
00:50:43 what you want is to make sure first is that that initiation event is consistent with a typical scenario so this is equivalent to a waste paper basket the the next step is you want to put that in a position in which it is in the
00:50:55 that in a position in which it is in the worst possible position which is you put it in the corner why do you put it in the corner because it compromises two walls
00:51:02 walls so very rapidly it's doubling the role of the lining so you're putting it in a worst case scenario and you're you have an ignition source that is consistent now that ignition source is extremely small
00:51:13 extremely small and uh in the room corner test is 100 kilowatts and 300 kilowatts a little bit later and in in the sbi is approximately 30 kilowatts so it's a very very very small ignition source very consistent
00:51:25 small ignition source very consistent with
00:51:26 with the experiments that professor bisbee did
00:51:30 did so it's a very very small ignition source now the entire compartment is cold so there is no no any extra energy coming from anywhere it's a compartment it's only
00:51:41 anywhere it's a compartment it's only the linings that are going to ignite and they're going to start spreading a flame now
00:51:46 now here you have a flame spread process and that flames per process as it grows what matters is how rapidly it grows because that's what creates a smoke layer that
00:51:58 that's what creates a smoke layer that is going to descend it also matters how much energy is releasing thus you need to measure the heat release rate because that energy is going to be deposited in the smoke layer and increases
00:52:10 in the smoke layer and increases temperature another thing that is going to matter is how much
00:52:15 how much soot or how dense your smoke is because that increases the emissivity of the smoke and increases the heat feedback to the linings and does accelerate the spread bringing you faster to flush over
00:52:27 spread bringing you faster to flush over this is why you have the figara and the smoker as the two indicators of performance because you want to know how dense the smoke is your smoker and you want to know how fast your fire spreads
00:52:38 want to know how fast your fire spreads that's your figure and your heat release rate is a third indicator and together all those enable you to assess how that smoke layer grows because it is the radiation from the smoke layer that will
00:52:49 radiation from the smoke layer that will bring you to flashover yes so what you're assessing in this test is purely how that smoke layer is evolving and it's affecting the spread of a fire so
00:53:00 so the material creates the heat that develops the smoke layer and the smoke layer accelerates the progression of the fire to flush over that is the mechanism yes okay so it is in in many ways
00:53:13 it is in in many ways all these things are then put in a box shaken and turned into classifications and that's the figure of the smoker and all these things lead to the abcde classification and is intimately linked
00:53:26 classification and is intimately linked to flashover now if you tell me let's take all the measurements of the spi let's break down all the information from the sbi i could extract the heat of combustion i could extract flame spread
00:53:37 combustion i could extract flame spread rates i could extract ignition temperatures i could extract all this different information that i could effectively use to try to assess external fire spread but if you tell me then instead can you
00:53:50 but if you tell me then instead can you use that classification the answer will be absolutely not there's no problem with the test and the information i can extract from the test the problem is me then using a classification that is
00:54:01 then using a classification that is intended to address a smoke layer into a problem that has nothing to do with the smoke layer because the smoke is going away so in the case of the external flame
00:54:12 so in the case of the external flame spread you're not going to have a small ignition source you're going to have a post flash over flame plume that is going to impinge which brings you to 100 kilowatts per meter square at least while here you
00:54:24 meter square at least while here you never get more than 30 or 35. so the scenario is completely different the corner configuration is meaningless because what you're going to have is a spill plume
00:54:35 spill plume and you have a whole series of other parameters that you need to consider like for example the pre-heating length and all these things that are to some extent is relevant here because you're focusing
00:54:45 focusing in this particular case on the development of the smoke layer so if i want to use the information and do a thorough analysis of this test and extract all the parameters i could probably get four five out of the ten
00:54:57 probably get four five out of the ten parameters that i need for vertical spread but if you ask me to transpose that classification uh that would not be possible
00:55:05 possible yes thank you that's that's very clear so it's providing useful data it's so it's not a not a useless test it's useful data that you can extract key parameters from exactly
00:55:16 key parameters from exactly but i think what you're saying is the problem comes when you try and take a classification that's fundamentally based on different parameters and apply that classification to an external wall situation
00:55:30 external wall situation that's unreliable is that what you're saying yeah it's beyond unreliable it's just incorrect and incomplete and and so so it just simply cannot be done the problem is extrapolating a classification without any further
00:55:42 classification without any further analysis uh it's not possible so does that mean that in the approved document where we saw in diagram 40 that it could be a class naught or a b that you would say it's entirely
00:55:54 that you would say it's entirely inappropriate to start putting those european classifications into that kind of guidance on external wall spread yeah is is wholly inappropriate in as much as class not is inappropriate
00:56:05 class not is inappropriate so the two of them are inappropriate yeah
00:56:08 yeah there's no equivalency between them and neither of the two represents the physics that you want to analyze yes
00:56:16 yes and if we look at your report at page 152 just where you you summarize some of this for us right paragraph 2.4 sorry 12.4.81
00:56:28 you say here so you say this is very important because this value is much smaller than that which an external fireplane will impose so you're talking about the 30 to 40
00:56:39 talking about the 30 to 40 uh kilowatts per meter squared that's imposed in the spi test so because this value is much smaller than that which an external fireplane will impose on a facade for the scenario relevance which
00:56:50 facade for the scenario relevance which is greater than 100 kilowatts per meter squared
00:56:53 squared therefore a poor performance in the sbi can equate to a poor performance in the event of external fire but not the opposite
00:57:03 opposite a good performance in the sbi could still result in an inadequate performance in an external fire so i think what you're saying is it can tell you what might not be appropriate
00:57:15 you what might not be appropriate quite clearly but just because something's got a b or even well i'll ask you about a1 and a2 in a moment but just something's got a b let's say i think what you're saying is that doesn't equate to good performance
00:57:28 that doesn't equate to good performance in an external fire spread situation yeah i mean this this is a quite complicated because i mean first of all the the 100 kilowatts per meter square as opposed to the 32 to 40 kilowatts per meter square this
00:57:40 kilowatts per meter square this qualifies the sbi as a scenario test so that's that's the first step now the the second step is is in in both cases now you are at a scale where there are going to be some some
00:57:53 where there are going to be some some form of system behavior because you're going to have now products you could actually put almost a cassette for example in in in the sbi so there are elements for example like mechanical behavior the
00:58:05 example like mechanical behavior the thermal mechanical behavior that uh that if you put 100 kilowatts you will induce certain failure modes of the you know encapsulation for example that uh that will fail
00:58:17 that uh that will fail uh at 100 kilowatts but will never fail at 30. so in many ways you can uh your encapsulation can work beautifully in the sbi but the moment that you have
00:58:28 in the sbi but the moment that you have this applied in an external facade and you get a post flash over fire that imposes 100 kilowatts your encapsulation fails
00:58:36 fails and the problem is you are completely blind
00:58:39 blind into what gets exposed so if you are relying fully on encapsulation what is happening is that what is exposed at that point um is
00:58:50 what is exposed at that point um is completely unnoticed by the test so if the encapsulation fails and you have polyethylene in the back or you have expanded polystyrene for example in the back this will immediately burn
00:59:02 the back this will immediately burn vigorously and become a very significant hazard you will not notice that in the sbi because the sbi does not allow the encapsulation to fail but when you transpose the scenario
00:59:15 transpose the scenario then you will have that failure and the result could be an enormous amount amount of heat being released instantaneously yes but i think what you're saying is as well is that but if if you did your sbi on your aluminium uh
00:59:29 if you did your sbi on your aluminium uh composite cassettes with pe and in that sbi test it performs badly with those lower heat fluxes and in that situation then then you certainly should be
00:59:41 then then you certainly should be thinking it's not going to perform well with an external fire no because the scenario is more onerous in in in nature so yes i mean so if it performs really badly on the sbi i will have to think
00:59:53 badly on the sbi i will have to think uh that it would perform even worse under under real conditions yes and uh but if you manage to uh create a system that performs
01:00:05 system that performs well because it is protected mechanically and so forth um i have no guarantee that once i change the scenario that is going to be still the case so just to be clear in
01:00:17 still the case so just to be clear in your opinion if you run the sbi on the cassettes
01:00:21 cassettes acm pe cassettes and you're getting an e or even a fail because you can't classify it um that's bad that's bad and and
01:00:32 that's bad that's bad and and it would be very difficult to then try and say that in an external wall it's going to be better it will perform better yeah it won't
01:00:41 um now if we could just look at page 152 um
01:00:48 um of your report uh
01:00:50 uh paragraph 12.4.86 yes so it's at the bottom of that page so this is where you're reinforcing the point you say the one aspect of the test that cannot be used are the
01:01:02 that cannot be used are the classifications emerging from the sbi test the classifications are based on specific measurements obtained during testing these are mostly physical measurements such as temperature rises in the duct which intend to represent
01:01:13 in the duct which intend to represent the contribution of the sample to the overall heat release rate and if you go on to page 153 12.4.90
01:01:22 you say this you say given the significant differences between the reference scenarios for external facade fires and the sbi the test specific nature of the parameters measured to define the classifications the lack of direct
01:01:34 classifications the lack of direct correspondence of the parameters measured to the flame velocity and heating length and the abstract nature of some of the parameters like the figure it's clear that the classifications resulting from the test
01:01:46 classifications resulting from the test have no relevance or meaning when addressing the performance of facade systems
01:01:52 systems and is that something that you consider should have been appreciated within this jurisdiction at a time when they were doing their transposition
01:02:02 absolutely i mean i think needless to say this knowledge has existed for 30 years
01:02:10 so this is not a this is this is not a difficult um conclusion to reach that it has no relevance to the scenario no and if we also look at page 155
01:02:27 um if we look at paragraphs 1 12 5 10 and 12 5 11.
01:02:35 12 5 11. so you make the same point about the scenarios being different at 12 5 10 and then at 12 5 11 you say it's therefore my opinion that any class b to f classification through the european
01:02:46 classification through the european reaction to fire protocols should not be stated in the approved documents or building regulations as providing a demonstration of adequate compliant performance of an external wall system or product used for this purpose
01:03:00 or product used for this purpose now what about the a1 or a2 classifications we know that they require slightly different tests um as well as sbi does it follow that you'd also have concerns about those well i
01:03:11 also have concerns about those well i mean direct extrapolation to some extent yes because there's an element of arbitrary threshold of what non-combustible is but we are landing in a space uh where you are discussing materials
01:03:23 uh where you are discussing materials that actually contribute very little from an energy perspective so so i think uh it is much easier to make a case uh that there is some element of
01:03:35 uh that there is some element of coincidence because the only real parameter that the a classification is providing
01:03:41 providing is really the heat of combustion and we have seen that most of our tests direct us straight into the heat of combustion so so i could provide um you
01:03:52 combustion so so i could provide um you know some element of of analysis that would link me that could link a you know with some of the existing approaches now uh if if you allow me the this this is i
01:04:05 if if you allow me the this this is i think i think something that is is really important and i think it has been emphasized that there was a pressing need you know by european directives to try to make this matches of classification
01:04:18 to make this matches of classification and and i do recognize that those pressures are obviously uh there and uh and there had to be something that needed to be done uh and i think this is
01:04:29 needed to be done uh and i think this is something that i find extremely difficult to to digest in many ways because you know we do recognize that the sbi and the european classification
01:04:40 the sbi and the european classification is not appropriate for this particular type of configuration which is the external building envelope and so we do re
01:04:49 re recognize that and uh and the approaches that we're using in the uk we recognize that they are not appropriate either you know nevertheless the sbi has a wealth of information and that wealth of
01:05:01 wealth of information and that wealth of information can be exploited the part that i really don't understand is instead of taking what is on the shelf and just going and saying well we just pick class 0 and we match it without
01:05:12 pick class 0 and we match it without absolutely no consideration to the physics and in a completely arbitrary way
01:05:18 way why we didn't really do a proper analysis of the sbi extracted from the sbi all the parameters that we could possibly extract them construct a test
01:05:29 possibly extract them construct a test locally that supplements it in a way such that we created that coincidence in a fundamental way and that actually represents the potential need for safety
01:05:40 represents the potential need for safety for the system so it's almost like the option is oh we're being pushed to do that so we do the simplest thing which is just take something off the shelf and we don't think twice about what that means
01:05:50 means instead of actually putting the effort to try because as i say the the sbi has a wealth of information and that information could potentially be useful
01:06:01 useful in in a classification is if it's treated just as a scenario test and you're just copying the classifications you're going to enter a space of being uh of misusing
01:06:12 enter a space of being uh of misusing the classifications but if you take a step back and you treat it as a test and then you analyze the information that you get and you supplement it appropriately we could have come up with an equivalent system
01:06:24 an equivalent system that would have not infringed the european directives and at the same time deliver us a testing regime that was highly appropriate yes thank you
01:06:36 and and is it right that um none of the the european tests uh give any information relevant to thermo mechanical performance at the moment no i mean i i think none of the tests
01:06:48 no i mean i i think none of the tests that we have addresses the thermal mechanical behavior
01:06:52 behavior at this point yeah and i think we'll come on to that in the context of 8414 as well um in a moment i just want to look at some of the evidence we heard in module 6 um
01:07:03 evidence we heard in module 6 um relevant to this and i want to look at the evidence we heard from dr debbie smith the former managing director of bre global if we could go to the transcript this is day 234
01:07:14 transcript this is day 234 of the transcript and i want to look at page 115
01:07:22 and
01:07:25 so the question begins at line four from mr millet and he says do you agree that the sbi test was in basic terms introduced in order to replace the room corner test
01:07:36 in order to replace the room corner test by providing a method by which the results of that test could be inferred without incurring the same cost in time and money and she says no now just pausing there do you agree with that
01:07:50 technically speaking i do in the sense that it was never intended to replace nevertheless um the objectives of both tests are fundamentally the same
01:08:01 fundamentally the same and
01:08:02 and and they're just simply at a different scale and
01:08:06 scale and so i i do agree that technically it was never intended to replace yeah and uh but as i say it's they're they're they they have similar objectives and i think she actually goes on to
01:08:17 and i think she actually goes on to clarify so then it says no you don't agree
01:08:21 agree no i don't the room corner test was the reference scenario which was used for the development of the single burning item test it was never intended as far as i'm aware that the route to classification would be using the iso
01:08:33 classification would be using the iso 9705 test method that's the room caller test it was a reference scenario like would you agree with that way of describing it yes i mean i think the two of them have exactly the right the the same sort of reference scenario and uh
01:08:47 same sort of reference scenario and uh and and and the it was it was always considered the iso 9705 was
01:08:53 was too complicated and uncertainties are big because of the size the scale you know to use it as a route to classification so in a way by reducing the scale making it more confined putting better measurements
01:09:05 putting better measurements you could actually have a classification that was a bit more robust yes let's look at the next page page 116. um
01:09:19 sorry actually go back to page 115 i want to pick it up there at line 22 right at the bottom so then another question did you ever yourself consider whether that test that's the sbi test
01:09:30 whether that test that's the sbi test was useful for assessing fire behavior in circumstances far removed from the test scenario for example on the external face of a high-rise building and then over the page
01:09:41 and then over the page she says yes i did consider that and it clearly is not suitable for that and much of the work in europe has focused very much on the identification of relevant reference scenarios which then form the basis for the development of a
01:09:53 form the basis for the development of a european test standard now so she's saying she considered that and it was not suitable for that do you agree with that statement i i i i think
01:10:04 agree with that statement i i i i think uh there's i i feel that there's a misuse of words i think uh what she is referring to is the
01:10:11 the the standard and the classification that it provides that is not suitable but she is not really thinking carefully about the fact that as a test that is
01:10:22 about the fact that as a test that is producing data the data could be useful so so i think uh you know i'm not disagreeing with what dr smith says i think it just requires a
01:10:33 dr smith says i think it just requires a bit of clarification because i really think that she's thinking as a standardized test the product is the classification and as such is highly inappropriate yes whereas i think what you're saying is
01:10:44 whereas i think what you're saying is the position is more nuanced because actually there is some data in that test that a competent professional could use provided they know what they're doing and then they understand the differences between the scenario in that test and
01:10:56 between the scenario in that test and the one they're investigating exactly and then if we go to page 118 and look at lines 1 to 12.
01:11:07 the question comes at the top of the page during that process harmonization or later did you ever say in very very simple terms almost late terms if i may look why have we got a test based on flashover being applied to the external
01:11:18 flashover being applied to the external build-up of a high-rise building and she says not in those terms no but i mean and then the question why is that and then she said it was well understood through the fire sector in the uk and in
01:11:30 through the fire sector in the uk and in europe that the whole premise of the classification systems were based on reference scenarios and there's a european guidance document that exists that sort of explains all of that
01:11:41 that sort of explains all of that now do you understand her to be saying there that it's the classifications that are the problem not the test data obtained itself exactly so now she comes back and and the terminology is used in a more precise manner so she is talking
01:11:53 a more precise manner so she is talking about the classifications and the european guidance document that she's talking about actually does discuss the scenario and and what is what the room corner test and the sbi intent to
01:12:05 corner test and the sbi intent to represent so so i would agree with uh with the statement yes i just want to pick up on another aspect of dr smith's evidence if we could go to her witness statement now at bre
01:12:17 her witness statement now at bre 405624 and i want to look at page 8.
01:12:24 and i want to read with you paragraph 24.
01:12:28 24. so she says this following a fire in a refurbished block of flats in liverpool in 1991 it was accepted that a large-scale fire test method was needed to provide a better means for assessing
01:12:39 to provide a better means for assessing and controlling the potential fire hazards associated with external cladding systems that test would involve the complete system as installed as closely as possible to what would actually be installed onto a building
01:12:51 actually be installed onto a building this was clear recognition that it was not possible to rely on small scale fire test data on individual elements materials or components in a cladding system as a means to try to predict and
01:13:03 system as a means to try to predict and thereby control the fire performance reliance on small-scale fire tests had been tried and had been shown to be inadequate now
01:13:15 do you agree with what she says there that
01:13:22 it's not possible to rely on small scale fire test data and individual element material data or components as a means to try and predict and thereby control fire performance yeah this is again i i
01:13:35 fire performance yeah this is again i i think there's a little bit of a confusion of words here because i think i will fully agree that you cannot rely uh
01:13:44 uh fully on small-scale test data because the reliance implies that you have everything that you need to be able to predict performance and clearly as you're moving into a system behavior and
01:13:55 you're moving into a system behavior and and you have multiple components that are assembled then those components are going to come at a certain scale and you have to start moving into that scale to try to understand uh the the pieces that
01:14:07 try to understand uh the the pieces that are missing from the small scale test so going back to the queensland model that is sort of the process that the queensland model wanted to follow which is get as much as you can from the small
01:14:18 is get as much as you can from the small scale then move to the next level get whatever is next and then finally you know get that and then validate so i think that that um i i would not disagree
01:14:28 disagree at all
01:14:29 at all with the concept that you cannot fully rely for this particular type of problem in small-scale tests but the inference that therefore
01:14:40 that therefore you have to go straight to a large-scale test
01:14:43 test it's the part that i i think is a bit problematic in the sense that that inference is not um direct you get to a large-scale test when you realize that you cannot get the
01:14:55 realize that you cannot get the information otherwise and because we all recognize the problems the uncertainties the difficulties of the large-scale tests and that's the basic reasoning behind moving from uh the room corner
01:15:06 behind moving from uh the room corner test to the sbi so so in many ways the large-scale test is the last resort where you can actually not get the information that you need so so i think the the two terms in the same sentence
01:15:18 the the two terms in the same sentence um you know are are a bit complicated for me to to understand and uh but i do recognize the process that eventually can lead you into a large scale test yes
01:15:29 can lead you into a large scale test yes so i think what you're saying is one of the problems here is that it's treating it as
01:15:32 it as if it's a binary problem exactly either you small scale test or you large scale test and in fact that's ignoring the complexity of the problem yes and it's the mindset that i was describing
01:15:43 it's the mindset that i was describing during my presentation that if you don't understand anything then you go and make the big burn and that almost becomes the substitute you know to understanding and to knowledge and to
01:15:54 understanding and to knowledge and to competency yeah just one final passage of dr smith's evidence um if we go back to day two three four i want to look at page one four seven
01:16:06 and i want to look at lines 5 to 15 here
01:16:11 here and the question comes but generally speaking your point about the unreliability of the small-scale tests she says yes i mean the key thing with the small scale tests and in the case of class naught and the part six and part
01:16:23 class naught and the part six and part seven tests is that you're unable to install them with a representative representative fixing system you can go as far as to install them with a different substrate behind them
01:16:34 with a different substrate behind them or with an air gap but that's about as far as you can go so you can't replicate some of the you know important design variables that are potentially important when it comes to
01:16:45 potentially important when it comes to you know large scale systems now again
01:16:50 now again if
01:16:51 if the implication of this evidence is that in contrast to small-scale tests large-scale tests can replicate all of the representative details which will give you the answer to whether that
01:17:02 give you the answer to whether that facade can be used do you agree with that
01:17:05 that unfortunately not because i think again it goes back to to the words i mean the the generally small scale tests are input tests where you run the test to harness information
01:17:16 harness information that's mostly the should be the objective of them and and
01:17:21 and it a large scale test can become a scenario test in which case you're trying to replicate something but you cannot mix the the the the two i
01:17:32 you cannot mix the the the the two i think these things need to be broken down in a much more refined way because i think what we are doing is is mixing two different things that cannot be
01:17:43 two different things that cannot be mixed
01:17:44 mixed and uh and the ultimate consequence of doing so it it once again reinforces the mindset
01:17:51 mindset that if you cannot replicate if you cannot understand if you don't cannot get the information then you have to go and burn big and uh and i don't think that that is uh an appropriate way of proceeding thank you mr chairman is that
01:18:03 proceeding thank you mr chairman is that a good moment for the morning break yes it is thank you very much well we'll have a break now professor gerrera we'll come back at 25 to 12 please and again please don't talk about your evidence when they're out of the room thank you
01:18:14 when they're out of the room thank you all right thank you
01:18:20 thank you ms grange 25 to 12 please
01:35:09 would you ask professor torreira to come back in please
01:35:21 right professor ready to carry on thank you very much yes mrs grinch yes thank you can i just go back to one passage in this morning's transcript um just to ask you to clarify one thing before we move on
01:35:32 on um when we were discussing the sbi test um there was a passage where you said so in the case of the external flame spread so in contrast to sbi you're not going to have a small ignition source you're
01:35:44 to have a small ignition source you're going to have a post flashover flame plume that's going to be impinging now um is this right that you don't necessarily have to have a post flashover fireplume to get an external
01:35:57 flashover fireplume to get an external fire going and i think that's we just explored that at phase one of your evidence oh yeah yes of course you don't need to yeah but eventually you will yes so
01:36:08 eventually you will yes so the the fire in the room is going to continue and uh it is it is very likely i mean obviously depending on on the ventilation conditions and the fuel available the room might never flush
01:36:20 available the room might never flush over like in the case of of grenfell so you don't need to have that but the reality is that if you're a designer and you are designing a safe system you have to consider that your fire will progress to flash over that you will have a plume
01:36:32 to flash over that you will have a plume post flash over plume venting and that becomes your design scenario basically yes thank you that's very helpful so um i now want to turn to bs 8414 and br135
01:36:43 br135 um
01:36:44 um we've heard about the history of the development of these two regimes from professor bisbee and you cover it in your report as well and is it right that in your report you're principally considering the 8414 test and the br135
01:36:57 considering the 8414 test and the br135 classification system in terms of what they can usefully tell you about the performance of external walls is that your focus yes and then you explain in your report that the 8414 test was
01:37:09 your report that the 8414 test was conceived as a scenario test i.e a test that's designed to assess something which consider is considered realistic
01:37:16 realistic however like professor bisbee you highlight that despite being conceived as a scenario test it was it made expressly clear certainly in the third edition of br135 in 2013 that it does not directly correspond to
01:37:29 that it does not directly correspond to the end use scenario that's the key point you make yes it count
01:37:34 count um and if we go to your report um again jtor
01:37:39 jtor 706 at page 102.
01:37:45 i just want to look at what you say paragraph 12.2.19
01:37:50 in the middle of the page and you say despite being clearly conceived as a scenario test br135 states the classification applies only to the system as tested and detailed in
01:38:01 to the system as tested and detailed in the classification report the classification report can only cover the details of the system as tested it cannot state what is not covered when specifying or checking a system it's important to check that the classification documents cover the end
01:38:13 classification documents cover the end use application and you go on and you say thus while the test was inherently designed as a scenario test it is not assumed to directly correspond to the new scenario
01:38:25 directly correspond to the new scenario this emphasizes the need for competent interpretation of the results of the test
01:38:30 test and has implications for how applicable the test output actually is so um
01:38:37 so um now just aside from that warning which only came in in the third edition of bl 135 would it have been evident to any reasonably competent fire engineering professional that this test would not
01:38:49 professional that this test would not correspond to an end-use scenario it should have yes
01:38:54 yes and what you do is you you analyze the development of br135 in your report and you explain that the very first edition of br135
01:39:03 of br135 was devised as a scoping study and if we look at what you say at page 103
01:39:10 103 over the page at 12.2.28
01:39:17 so you say upon inspection of this report that's the the 1988 br135 it becomes clear that this experimental series was originally devised as a scoping study and was not in any way
01:39:28 scoping study and was not in any way intended to be the development of a test of these systems performance now can you just explain for us exactly what you mean by the fact that it was you say it was devised as a scoping study yes i
01:39:40 was devised as a scoping study yes i think that uh that for me the difference between a scoping study and a study that is designed to develop something that is of general application is the breadth of
01:39:51 of general application is the breadth of the different systems conditions and uh and and characteristics that you are going to use to support the development of the test so if uh if the conditions
01:40:03 of the test so if uh if the conditions are very very limited and um then
01:40:06 then what happens is the the
01:40:09 the the information that you are collecting might pertain only to a very narrow set of systems now if you want that to become a standardized test and you want that to be used broadly then you would
01:40:21 that to be used broadly then you would have had to do a much much much broader set of scenarios characteristics and so forth until you narrow down exactly what is the condition that gives you the most information or the information that you
01:40:33 information or the information that you actually need yes and and you explained that that first edition looked at certain specific systems including one system which was an aluminium-faced cladding system incorporating expanded polystyrene with
01:40:46 incorporating expanded polystyrene with the extent of the cavity being limited by fire barriers so a very particular type of system exactly and then at 12.2.29 just going back to that page
01:40:57 12.2.29 just going back to that page you say this you say furthermore the focus is on risk of flame spread surface and cavities but it bears no relationship to the objective of supporting a fire safety strategy so
01:41:09 supporting a fire safety strategy so when the objectives indicate that the purpose is to identify the design principles they relate to an undefined concept of risk of flame spread so i think you're critical of the fact
01:41:20 so i think you're critical of the fact that there's no defined objectives which tell you
01:41:24 tell you what concept you're seeking to classify in this standard is that right exactly
01:41:30 exactly and then you say in the following paragraph at the bottom of that page as a result and from the outset br135 and the research supporting it leaves it to the interpretation of the competent professional to determine what the
01:41:42 professional to determine what the objectives of the test are and how to embed the concept of risk of flame spread surfacing cavities in the assessment and definition of the fire safety strategy so i think what you're saying is because
01:41:53 so i think what you're saying is because it only ever is addressing this generic risk ill-defined risk of flame spread the competent professional has got to take that but then
01:42:04 take that but then work out what what type of fire safety strategy they're designing and whether that risk of flame spread is compatible with that fire safety strategy is that right yes and and beyond that the the competent professional has to be able to
01:42:16 competent professional has to be able to extract the information and determine whether that information um
01:42:21 um enables them to define
01:42:24 define uh appropriately the adequacy of the fire safety strategy but also they should be able to establish when that information is either insufficient or inappropriate yes so for example if you were to alter the
01:42:35 so for example if you were to alter the system significantly and or you have a system for example that has a balcony and that balcony has certain characteristics that might completely invalidate the test the competent
01:42:46 invalidate the test the competent professionals should be able to make that assessment yes and the one of the points you make is about the extent of focus in br135 all through
01:42:58 the extent of focus in br135 all through all three editions on cavity barriers and i just want to look at some of what you say about that if we go to page 105 now
01:43:08 and we pick it up at paragraph 12.2.37
01:43:13 you say this you say the second recommendation indicates that flame spread will be inhibited if access to further fuel is prevented by an effective barrier this is a specific way to attain the
01:43:24 this is a specific way to attain the heating length equals zero in a manner that is not intrinsic to the combustibility of the materials involved but by means of a mechanical barrier and you go on thus both recommendations
01:43:35 and you go on thus both recommendations are specific to systems to the system studied and are not influenced by assessments of combustibility but by assessments of thermo-mechanical behavior
01:43:46 behavior when you go on you say the second recommendation is important in the context of subsequent historical perception of the risk posed by rain screen facade systems as stated above the external panels considered at the
01:43:57 the external panels considered at the time were uniformly non-combustible likely representative of the products on the market at the time judging by the scope of the document therefore spread of flame over the external surface was
01:44:08 of flame over the external surface was not a possibility would not have been observed and thus was never addressed in this document and then you go on at 12.2.40 the principal hazard identified was the elongation of flames contained
01:44:21 was the elongation of flames contained within a cavity and seemingly as a result of which ever since the publication of this document has historically been perceived as the most critical threat in regard to these systems
01:44:32 systems thus the provision of cavity barriers has remained the primary focus of industry and regulation and was typically the sole consideration when assessing the risk posed by rain stream
01:44:43 assessing the risk posed by rain stream facades containing no combustible materials
01:44:47 materials so
01:44:48 so is it your opinion that we can trace the focus of industry on cavity barriers and cavity barriers effectively solving your problem of external fire spread at
01:45:00 your problem of external fire spread at least in part to this first edition of br135 which had this acute focus on flame elongation within a cavity yes and i mean i at least that's that's sort of the impression that that i get
01:45:12 sort of the impression that that i get by reading the the document and and again that is caveated by the fact that these were very specific systems in which the cavities have certain dimensions that seem to promote the elongation of
01:45:25 that seem to promote the elongation of flames which is not always the case uh with with cavities so so yeah i mean i think there seems to be a coincidence between the two of them but i say it's an impression that i get by reading the documentation yes
01:45:38 reading the documentation yes do you think professor bisbee's right that cavity barriers are something which has received particular focus in this jurisdiction but which hasn't received the same attention internationally in many other jurisdictions for the for the
01:45:50 many other jurisdictions for the for the transcript the reference to that is at day 290 page 162 lines 4-14 yep um the cavity barriers are um
01:46:01 cavity barriers are um are something that will appear in in most jurisdictions and uh and and obviously there is a recognition that under certain conditions a cavity in itself represents a chimney and uh and
01:46:14 itself represents a chimney and uh and under those conditions it has a potential to accelerate uh fire spread and uh but um the the enormous emphasis that is given in the uk is is not
01:46:25 that is given in the uk is is not parallel elsewhere where it is just simply seen as yet another element that needs to be complied with as many others and um and i think i think here we do get
01:46:35 get um a very significant emphasis on cavity barriers without really paying too much attention to exactly what they're doing and whether it is as beneficial as it is
01:46:46 and whether it is as beneficial as it is perceived as they are perceived to be yes so would you say that the focus on the cavity barriers in this jurisdiction has potentially been disproportionate to the benefits that they can potentially achieve
01:46:57 achieve yes the the focus has to be disproportionate to the effort that has been put in understanding how beneficial they can be it's been an assumption that if we put cavity bars according to certain characteristics those cavity
01:47:09 certain characteristics those cavity bars will be immensely beneficial and uh but the that requires an element of effort to understand if effectively that is the case and i think that effort has never really been
01:47:20 think that effort has never really been there
01:47:21 there do you consider that their role in the approved document and their promotion in the uk market has given a false sense of security that provided cavity barriers are present the cladding system will be
01:47:33 system will be safe and will resist the spread of fire i mean this is this is the case with multiple um you know fire safety systems and the
01:47:41 and the cavity barriers is one of them sprinkler systems are another that uh that in many ways there is this over simplistic extrapolation that if i put this then i am safe and that false sense of security
01:47:52 am safe and that false sense of security is i find something very problematic
01:47:57 and you make the point in your report that later editions of br135 continue to focus almost exclusively on the threat of fire propagation through cavities with very little reference to the risks
01:48:09 with very little reference to the risks posed by flame spread through the external panels themselves is that correct yes and if we can look at what you say um on page 106 now the next page in this
01:48:21 page 106 now the next page in this section paragraph 12.2.44
01:48:26 you say this you say while this perception of the risk posed by modern facade systems may have changed in light of the external fire behavior at grenfell tower there is still a significant industry focus on the type
01:48:38 significant industry focus on the type location and orientation of cavity barriers this is despite the fact that external fires of the nature of that of grenfell tower clearly spread over the external facade a clear technical assessment of the role of cavity
01:48:51 assessment of the role of cavity barriers in reducing or preventing external fire spread is currently not available this again is demonstrative of the endemic perception in industry of where the risk lies when addressing rain
01:49:02 where the risk lies when addressing rain screen facade systems i want to pick up on that penultimate sentence where you say
01:49:07 say a clear technical assessment of the role of cavity barriers in reducing or preventing external fire spread is currently not available what do you mean by that so if if you look at the entire
01:49:18 by that so if if you look at the entire technical literature in in in fire safety you will find almost
01:49:25 almost a complete absence of any studies that really address cavity virus in a systematic way i i think i'm i'm probably one of the few people that has actually studied cavities
01:49:38 actually studied cavities globally this is there's no emphasis at all in trying to really understand how they work
01:49:45 they work and and is that a complex piece of work that really needs to be done in order to understand how they work better yes of course because uh it it is it is a complicated problem in the sense that
01:49:57 complicated problem in the sense that that the size of the cavity uh is critical in the way in which the cavity barrier either promotes or in many cases hamper uh the the progression
01:50:09 many cases hamper uh the the progression of of a fire and i think uh in many ways it bears a significant correlation with the actual uh with other functionalities of a building envelope because the reason why we have cavities
01:50:21 reason why we have cavities um you know is is is not because of fire the reason we have cavities because of other functionalities and we introduce all these components that have a detrimental effect on these
01:50:34 that have a detrimental effect on these other functionalities of the building envelope without really having a true understanding of what benefit we're gaining by introducing these elements and obviously from a constructability perspective so many other areas that get
01:50:46 perspective so many other areas that get affected by the fact that we keep introducing these cavity barriers so we really need to understand it and it's an area that has received almost no attention okay thank you now let's just look at what the later
01:50:58 now let's just look at what the later versions of br135 do say about external fire spread through the panels themselves if we look at the third edition from 2013 this is at bre three zeroes
01:51:08 zeroes one six six six five
01:51:14 there we have the third edition authored by sarah colwell and tony bacon if we could go to page 22 in the right hand column so under the word combustible
01:51:26 so under the word combustible so in the second half of the page there's a bold so it says this combustible panels are typically based on vinyl or glass to reinforce plastic although various new products are being
01:51:38 although various new products are being developed in this area some of which also contain insulation materials these products generally have good surface spread of flame characteristics to prevent rapid fire spread across the surface of the system but once the
01:51:50 surface of the system but once the panels become involved in the fire they have the potential to generate falling debris add to the overall fire load and provide a route for fire to propagate up the outside of the building
01:52:01 the outside of the building now you make the the point in your report that there's nothing more than that in this document about how to assess the risks that are being described there
01:52:12 the risks that are being described there is that is that fair yeah that's correct um now there are performance criteria in the appendix let's just look at the appendix uh of this third edition on
01:52:24 appendix uh of this third edition on page 29.
01:52:28 this is appendix b it's the performance criteria and classification for the 8414 part 2 test on steel frame systems and if we turn to page 31
01:52:42 we can see there's a heading performance criteria and classification method there on the left hand side
01:52:50 and then if we go to the top of the next column top of the right hand column in the second line it says the performance of the system under investigation is evaluated against the following three
01:53:01 evaluated against the following three criteria external fire spread internal fire spread and mechanical performance and then over the page if we go to page 32
01:53:12 32 we can see that there are some criteria then set
01:53:16 then set for uh
01:53:17 for uh performance um but we can see that well actually if we go back a page we see the criteria for external fire spread
01:53:27 spread at the bottom of the page it's right isn't it that this is expressed purely in terms of failure is that right failure due to external fire spread is deemed to have occurred if the temperature rise and then it goes
01:53:39 if the temperature rise and then it goes on to explain what the specific requirements are for failure yes absolutely i mean i think i think that the the title of section b2 is misleading because it is not a
01:53:51 misleading because it is not a performance criteria and it is not a classification all it is is a statement of failure and uh and so so it doesn't provide you any further
01:54:02 so it doesn't provide you any further information other than a certain set of systems that will actually fail the criteria yes
01:54:10 yes um
01:54:11 um and do you understand the rationale behind these failure criteria we've asked a number of witnesses at module six to explain it do you do you understand what the rationale is for those failure criteria
01:54:23 those failure criteria there is no rationale behind them other than just simply typical values that you will use uh to characterize the presence of a flame
01:54:35 flame so if you if you look at the data that is available on on flames what you will find is that um there is an area of the flame that is called the continuous flame which is an area that is always burning and normally
01:54:47 area that is always burning and normally the temperatures in that air will be about 800 900 degrees and uh and then you get what is called the intermittent flame which is a period of pulsation and uh and in the middle of that you will be
01:54:58 uh and in the middle of that you will be around six seven hundred and uh and then by the time you get out of the area that is actually burning and just get hot gases you're down to 400 and uh and and i think it's just simply
01:55:10 and uh and and i think it's just simply a
01:55:11 a they've used a characteristic value that is
01:55:14 is commonly recognized as being somewhere in the range of things that burn right but so but you can't it's arbitrary
01:55:23 arbitrary i think it's arbitrary yes yes um
01:55:26 um and you also make the point in your report for the transcript that's at page 97 paragraph 12.2.11 that having this simple temperature system is not capable of distinguishing between
01:55:38 is not capable of distinguishing between a small flame close by or the smoke plume from a larger flame at a distance exactly because i mean as i described you have these different sectors of the flame so as the flame increases in size
01:55:49 flame so as the flame increases in size everything increases in size so if i have a big flame i will have to be very far away to be at 600 but if i have a small flame i will have to be very close to be at 600. so
01:56:00 have to be very close to be at 600. so so yes it doesn't allow you to distinguish what is really happening yes and it doesn't enable you to tell whether it's the smoke that's heated that or the flame itself exactly so you
01:56:11 that or the flame itself exactly so you could have for example a membrane where you have a little flame that is creeping up and if it gets to the thermocouple it will get to 600 degrees and you know as opposed to if you have a very big fire
01:56:22 opposed to if you have a very big fire underneath and uh and then the smoke could be a 600. and is this one of the reasons why you say that the failure criteria simply don't relate to the objectives of the test
01:56:32 the test yes because at the end uh we are talking about fire spread so the the objective of the test should be to characterize the capability of the system to sustain
01:56:45 the capability of the system to sustain or adequately resist fire spread and uh and it's not necessarily linked purely to the characteristics of potentially the smoke for example so there's not a one-to-one correlation between the two
01:56:56 one-to-one correlation between the two yes
01:56:56 yes do we get any guidance within this document as to what the significance of a pass is
01:57:02 a pass is uh to br 135 oh i my understanding is that there's no such a thing as a pass
01:57:12 it is just a failure criteria and uh and and the past is defined
01:57:18 defined by the competent professional in the use they make of that information they take the information they apply it you know to their building and they decide
01:57:29 decide if it actually meets uh the requirements of the fire safety strategy so that their holistic fire safety strategy meets the regulatory requirements of the five
01:57:40 regulatory requirements of the five functionality pillars yes so you just failed the test but i think what you're saying is that doesn't tell you anything about whether then you're satisfying a particular fire
01:57:51 then you're satisfying a particular fire safety strategy and meeting a functional requirement yes absolutely do you agree with professor bisbee his characterization of it was that the only thing this test really tells you is what
01:58:02 thing this test really tells you is what would be a really bad idea to use do you agree with that yes as i i described it in a different way i said basically your failure criteria cuts the tail
01:58:14 criteria cuts the tail you know of the really really bad systems
01:58:17 systems yes
01:58:17 yes and for the transcript that that was day 290 page one four three lines seven to eight
01:58:23 eight um so again does that leave it to the designer to decide what the significance is of the fact that you've not failed this test
01:58:31 this test but i think what you're saying is it certainly doesn't tell you that it's okay just to crack on and put it on an external wall it's clearly not a pass so it it all that it tells you if it
01:58:42 so it it all that it tells you if it doesn't fail is that you have to take that information and then make good sense of that information to guarantee that it fulfills the needs of your fire safety strategy
01:58:53 strategy yes
01:58:54 yes and if we look at page 109 of your report
01:58:57 report um
01:58:58 um at
01:58:59 at jtor706 page 109 and i pick it up at 12.2.62
01:59:07 so you say as has already been discussed the process of interpreting the performance requirements defined by and big ambiguous statements in the building regulations must be inherently linked to the fire safety strategy of a building
01:59:18 the fire safety strategy of a building whereas a fire safety strategy with a controlled evacuation can allow for a conservatively defined degree of external flame fire spread a fire strategy with a stay put policy in
01:59:30 fire strategy with a stay put policy in place cannot allow any and then you go on if i pick it up at 12 264 you say it logically follows that a facade system can meet the br135 performance criteria and yet still be
01:59:42 performance criteria and yet still be entirely incompatible incompatible with the fire safety strategy of a high-rise residential building with a stay put policy in place
01:59:52 and then you go on in the next paragraph to explain that the information what the information provided by the thermocouple arrays do not establish including the ignition temperatures of
02:00:04 including the ignition temperatures of the products and whether they're sufficiently high the length scales of any flames and whether or not the flame will tend to extinction and with regard to a controlled
02:00:15 and with regard to a controlled evacuation strategy it also doesn't tell you the actual rate of external internal flame spread the length scales of the flames established or certain material properties
02:00:29 yes that's correct would it be possible to run these large scale tests but in a way which collected a larger pool of useful data absolutely i mean i think i can even
02:00:40 absolutely i mean i think i can even think of the most sort of simplistic approach you already take videos of this test and if you could actually do a very careful analysis of the video we'll give you uh
02:00:52 you uh some estimate of the rates of flame spread for example yes
02:00:57 yes if we go on to page 110 the next page at 12.2.72
02:01:04 you say right at the bottom of the page the br135 bs 8414 approach is presented by approved document b 2019 as an alternative to following the more simplistic alternative that restricts
02:01:15 simplistic alternative that restricts material combustibility in spite of this the bl-135 assessment does not translate data from the test to a classification of combustibility that could be used to assess the adequacy of a system
02:01:28 assess the adequacy of a system so just to be clear would it be possible to translate data from a good large-scale fire test into a classification of combustibility that could be used to assess the adequacy of a system uh yes i think again you know
02:01:41 a system uh yes i think again you know it all goes back to the less you measure the less you observe the more difficult it becomes and the more competent the person has to be but but clearly there is information in the test that
02:01:53 there is information in the test that can be used and uh and as it stands with the simplistic nature of what we measure
02:02:02 measure uh probably you will have to use that information to infer something extremely conservative but but yes of course you can always extract information from a test yeah
02:02:13 extract information from a test yeah just going back to the third edition of br135 again so bre three zeros one uh three sixes five at page 32.
02:02:24 uh three sixes five at page 32. i want to ask you now about mechanical performance we can see there what it says under the heading on the left hand column is no failure criteria have been set for mechanical performance however ongoing
02:02:36 mechanical performance however ongoing system combustion following extinguishing of the ignition source shall be included in the test and classification reports together with details of any system collapse spalling delamination flaming debris or pool
02:02:48 delamination flaming debris or pool fires
02:02:49 fires the nature of the mechanical performance should be considered as part of the overall risk assessment when specifying the system now is that another example of the requirement for a designer to assess
02:03:01 requirement for a designer to assess this information and then apply it to their particular situation
02:03:05 situation yes and i think in this particular space it is it is described in a manner that completely underestimates the complexity of that problem because you have so many
02:03:16 of that problem because you have so many forms of mechanical behavior that need to be considered you know from the failure of the encapsulation for example
02:03:24 example you know two components falling off to the formations that will allow smoke to migrate to the upper floors for example there's so many aspects of mechanical behavior
02:03:34 behavior that are that are all summarized uh in in those um you know eight lines that uh that that i think um it misrepresents
02:03:46 um it misrepresents you know the importance of that particular problem yes so i think you're saying that's simply not a satisfactory way of dealing with mechanical performance not at all and um
02:03:57 and um is it your evidence that it would be possible to provide some form of classification system or criteria for mechanical performance if one wanted to do so
02:04:06 do so yes i think i think you know if you really understand what you're aiming at uh you should be able to uh you know create a a a criteria just to give you a very
02:04:18 a a a criteria just to give you a very very simple example um you know we introduce this fire stops between the the facade system and the concrete slab and we filled them in okay so i could introduce for example a
02:04:31 okay so i could introduce for example a criteria of mechanical behavior that does not allow for a deformation that is bigger
02:04:37 bigger than a certain gap with that that stop can fill in you know by its own natural expansion so so you can actually look at the systems and say what i want is this
02:04:49 the systems and say what i want is this gap to never open and because i don't want that gap to ever open then i have to restrain expansion to some level so that all the components will fill its
02:05:00 all the components will fill its function so so there are ways by which you can establish some performance criteria and and you can make them more complicated or simpler i mean you can simply say that uh that
02:05:13 mean you can simply say that uh that you know you don't want to allow any smoke to be above the barrier and put a measurement of of a small concentration right above and the moment
02:05:24 concentration right above and the moment that that smoke gets uh beyond a certain threshold you can say well then the smoke has creeped in and therefore the mechanical deformations are allowing smoke to fill uh the the compartment above so so there
02:05:37 uh the the compartment above so so there are many many ways in which you can do it it is a complicated problem and needs to be carefully assessed but uh
02:05:44 but uh but
02:05:45 but but it can be done yes
02:05:48 yes um then if we can look at back at your report jtor706 page 111
02:05:58 i want to see what you say at 12.2.76 you say this here in in respective range screen systems deformation of such systems as described in the main body of the br-135 report second edition can
02:06:11 the br-135 report second edition can lead to the creation of gaps that allow flame penetration behind and through the facade system and directly into the unit above
02:06:18 above the absence of another opening above the fire compartment in the bs 8414 test makes even a visual assessment of the propensity for such a compartmentation failure to occur almost impossible
02:06:30 almost impossible thus failure by deformation of the system is not an aspect which is assessed by bs8414br135 so is this right i think what you're saying there is that for rain screen
02:06:42 saying there is that for rain screen systems given specific ways that they can deform and gaps can be created it might be a good idea to have some large-scale fire tests which actually have additional penetrations
02:06:54 have additional penetrations above the original that the original fire compartment so that you could at least observe how those uh how the fire reacts with those additional penetrations is that right yes that's correct and and uh even
02:07:07 right yes that's correct and and uh even beyond that uh sp 105 actually does have that
02:07:11 that so so the swedish uh test does include a penetration in the in the floor above so that you can actually assess the conditions behind now they
02:07:22 assess the conditions behind now they don't look at mechanical behavior either you know but nevertheless they do make an assessment of what happens in the compartment above yes and and having windows and a compartment above um
02:07:35 above um is that always going to be necessary would it vary depending on the system that you were assessing well i mean i think it depends mostly on your fire safety strategy so if you are relying on a staple strategy
02:07:48 relying on a staple strategy and you're putting no other provisions uh to compensate uh for the fact that your compartmentation might fail then you do need to make that assessment
02:08:00 then you do need to make that assessment because you're fully relying on the integrity of the compartmentation and the fact that there will be no smoke or flames penetrating into the compartment above yeah but is there a role for i mean i
02:08:12 but is there a role for i mean i appreciate everything you've said about what the role of large-scale testing is is there a role for a better large-scale fire test which
02:08:22 which has for example as a standard assumption an additional penetration in it yeah i'm i'm actually convinced that there in this particular case because you're talking about complex systems
02:08:34 you're talking about complex systems that are at a scale that is the scale of the compartment that there is a role for a large-scale test now
02:08:43 now i think the objection that i have to the current uh large-scale test and the approach that is followed towards large-scale tests is that they are not based on needs so it's not a demonstrable need that has come from a
02:08:55 demonstrable need that has come from a very systematic and rigorous assessment of what is what the information that i need to collect you know they're just a substitute you know to the fact that we don't understand how the systems behave
02:09:06 don't understand how the systems behave so i at no point i am saying that uh that there is no need for large-scale tests i'm actually quite convinced that for a system that you will need some form of large-scale tests it just simply has to be designed
02:09:18 has to be designed you know and instrumented you know and analyzed in an appropriate way yes so i think what you're saying is you could have better large-scale tests but standardized ones
02:09:29 standardized ones as part of any comprehensive system yeah i mean i think i think the last word that you use when you mentioned when you said standardized i mean i think that um that
02:09:40 think that um that to to me
02:09:42 to to me that's you know where we run a little bit into trouble because for me standardization is the product of understanding so once you've you've understood something fully you can come up with a
02:09:55 something fully you can come up with a standard simpler version or a standard version that enables you to obtain the information in a systematic way and uh and i think the problem is that when you're talking about this type of products the way in which they evolve
02:10:07 of products the way in which they evolve and the fact that they're being used with a building and the buildings are going to be generally always different then there is an element of bespokeness
02:10:18 then there is an element of bespokeness that is very difficult to break so the concept of a standardized test
02:10:25 test might be much more valuable at the smaller scales while the large-scale test will always be an ultimate need you know when you have a bespoke system that needs to be
02:10:37 have a bespoke system that needs to be assessed because you're now talking about the full system so i think we have to be a bit careful in the way in which we look at this because uh standardizing large-scale tests can only be done if
02:10:49 large-scale tests can only be done if you truly understand the problem that you're working with and if the problem is going to vary for every building then that is actually not possible so so we have to be quite careful with the use of standardized but
02:11:00 standardized but yes i think i understand so i think what you're saying is that in any ideal system
02:11:04 system you could have um large-scale testing to validate results and to provide data and information but not necessarily the same test for all systems exactly and you might need
02:11:17 all systems exactly and you might need to have some flexibility in the way you arrange your test depending on what it is you're seeking to exactly understand exactly yeah
02:11:30 now um link to that point if we go to page 191 of your report
02:11:39 uh paragraph a223 so this is in the section of your report it's in the appendix or one of the appendices where you're analyzing um whether there's good literature out there on the way in which
02:11:50 literature out there on the way in which external fire performance should be assessed
02:11:54 assessed and you explain at paragraph a 223 you say an attempt to analyze the shortcomings of the different test approaches used in the uk mainly bs8414 was conducted by
02:12:05 the uk mainly bs8414 was conducted by schultz at al and the footnote takes us to a number of individuals we can see there contributing to a paper called a critical appraisal of the uk's regulatory regime for combustible
02:12:17 regulatory regime for combustible facades that was published in 2021 when you go on you say this study aims to establish a critical assessment of the current testing regimes of relevance to facade systems the authors correctly
02:12:28 to facade systems the authors correctly conclude that the process by which combustible facades are permitted on tall
02:12:34 tall buildings in england does not fully does not ensure fire safety and insist that a new and more robust approach is required and then you go on you say the scope of the paper is limited to the bs 8414
02:12:45 the paper is limited to the bs 8414 testing protocol the performance criteria in nxa of br135 and the mhclg sponsored tests those are the ones carried out after grenfell i think yes the analysis discusses details such as
02:12:57 the analysis discusses details such as the ambiguities within the bs8414 protocol reproducibility strong sensitivity of the outcomes to many variables and the lack of alignment of the criteria in bl-135
02:13:08 of the criteria in bl-135 to any specific objectives of the fire safety strategy it is clear that reproducing realistic facade systems is very difficult in a test so some degree of extrapolation will always be necessary the authors use
02:13:20 will always be necessary the authors use the mhcld test to illustrate this problem and how small differences can significantly alter the results finally a series of recommendations are provided that correctly stipulate the need for
02:13:31 that correctly stipulate the need for more reproducibility better measurements and more clear reporting and transparency in regards to the objectives and motivations for the test and you go on and you provide an opinion
02:13:42 and you go on and you provide an opinion in the next paragraph a2.25 you say while it is impossible to disagree with any of these statements this study fails to establish the fundamental weaknesses of the test or to
02:13:54 fundamental weaknesses of the test or to establish the key variables that will affect the quantification of external fire spread in a manner that supports the fire safety strategy furthermore the study provides no critical analysis of
02:14:05 study provides no critical analysis of what is required for a correct extrapolation a test delivers data and that data needs to be interpreted data that has more imperfections does not mean that the data is useless it only means that a
02:14:17 data is useless it only means that a more complex interpretation is necessary without this critical analysis this study does not enable any advancement towards a better compliance regime and just to orientate ourselves if we
02:14:28 and just to orientate ourselves if we bring up that study it's inq3014943
02:14:38 that's the study um in question there now what exactly do you mean when you say at the beginning of the paragraph i just read that this study fails to establish the fundamental weaknesses of the test is that it fails
02:14:51 weaknesses of the test is that it fails to recognize the fundamental weaknesses of eight four one four exactly so uh it is
02:14:56 is at the core is the things that we have been discussing uh that uh that fundamentally make this test limited and uh you know from
02:15:07 you know from issues like not being able to assess mechanical behavior to other issues of very poor instrumentation uh there's so many of these fundamental weaknesses that uh that are not touched upon in
02:15:19 that uh that are not touched upon in detail in that study
02:15:24 and when you say that it fails to establish the key variables that will affect the quantification of external fire spread does that go back to your presentation where you identify exactly variables
02:15:35 but do you think it would be possible to make the 84 test 8414 test much better in terms of the data it produces so that it was more useful to fire safety professionals if it was chosen as a test
02:15:46 professionals if it was chosen as a test to be used yes but but you have to always um you know consider the fact that it is a test and as such is not there to fully
02:15:57 and as such is not there to fully reproduce reality but to provide information for a competent professional but by improving the test you reduce the onus on the competent professional and um and and and you create a much more
02:16:09 um and and and you create a much more effective and robust uh set of data that can be used uh in support of your fire safety strategy yes but i think what you're saying is that you could approve the test but it still
02:16:21 you could approve the test but it still wouldn't matter how good the test is because there would always be a need for some extrapolation of the test results and judgment by the competent professional exactly and i think i think for example the issue of
02:16:33 for example the issue of very small changes can make an enormous difference uh i mean i think this is very common with problems of encapsulation and we saw it in professor bisbee's tests that uh very very small changes can have
02:16:45 that uh very very small changes can have a very drastic difference in in in the outcome and um and that is something that um that is not a flaw of the test
02:16:54 the test it is actually a weakness of the strategy that we're following to protect the combustible material and i think by by directing the attack on the test because of those changes what we're
02:17:06 because of those changes what we're doing is failing to recognize the intriguing weaknesses of the strategies that we're using to protect combustible materials
02:17:14 materials yes thank you now that takes us nicely to desktop assessments and some questions about those um and i'm going to ask you some questions about those in about 94 bs
02:17:25 questions about those in about 94 bs 9414 and the guidance that that now gives
02:17:28 gives to those carrying out such assessments but before we do just stepping back and thinking about this generally are desktop assessments or extrapolations from test data a
02:17:39 extrapolations from test data a scientifically valid basis upon which to make assessments about external wall performance yes i mean a desktop study is on another expression that describes an engineering assessment
02:17:51 an engineering assessment and uh and as a competent engineer i will
02:17:55 will most likely always have to supplement the information that i have with calculations with analysis with other additional knowledge and and once we document that into a report
02:18:08 and once we document that into a report that is what becomes a desktop yes
02:18:11 yes now let's look at the section of your report in which you deal with this if we could go
02:18:16 could go back to your report jtor706 at page 114
02:18:23 and if we look at 12.2.93 this is where you've got the section on desktop studies you say as described earlier the use of testing to establish the performance of systems to be used in buildings
02:18:35 systems to be used in buildings potentially requires multiple levels of analysis parameters of verified magnitudes can be introduced into models and potentially combined with different approaches to performance testing so as to determine if the system
02:18:47 to determine if the system in the context of its application fulfills the requirements of the fire safety strategy this information has to be analyzed by a confident professional and delivered through a desktop study thus a desktop
02:18:58 through a desktop study thus a desktop study is an inherent component to the practice of fire engineering now just pausing there have you yourself ever carried out desktop assessments yes many times yeah and then carrying on with 12.2.93
02:19:11 and then carrying on with 12.2.93 you say this any parameters that are deemed critical to attaining the required outcomes of the fire safety strategy should be independently verified for conformity the parameters are combined through a desktop study to
02:19:22 are combined through a desktop study to define either the performance of a subsystem or that of a building envelope the fire performance of the subsystem or the building envelope also needs to be verified for conformity the parameters
02:19:33 verified for conformity the parameters describing the fire performance of the building envelope will serve as an input to the performance assessments in the fire safety strategy thus the fire performance of the building envelope is not in itself compliant but it can be
02:19:44 not in itself compliant but it can be deemed by the competent professional to conform with the fire performance expected of it by the fire safety strategy
02:19:52 strategy the vehicle by which these assessments are presented and recorded are the desktop studies the assessment of a fire safety strategy might incorporate several desktop studies now in terms of that last um
02:20:04 that last um comment that the assessment of a fire safety strategy might incorporate several desktop studies can you just give us some examples of what you mean by that yes so for example um if
02:20:16 by that yes so for example um if if if i'm looking into the performance of a facade system that actually has uh combustible elements in it and there will be an analysis that will be
02:20:28 there will be an analysis that will be primarily led by a fire safety engineer to look into how those combustible materials are going to enable the spread of
02:20:36 of of of a fire now
02:20:39 now that can be complemented for example by an analysis that is done jointly but this time led by a facade engineer where a fire engineering is providing the input temperatures to look at the
02:20:50 input temperatures to look at the formations and and those deformations will produce gaps and uh and the definite and and the quantification of those gaps then is going to be fed into the fire safety engineer report and
02:21:04 into the fire safety engineer report and showing that those gaps do not create conditions that challenge the fire safety strategy so just with a simple example i have talked about three desktop studies that could be led by
02:21:15 desktop studies that could be led by three different individuals and um nevertheless they all form part at the end of the fire safety engineers um fire safety strategy demonstration of
02:21:26 fire safety strategy demonstration of performance yes now
02:21:29 now just some very general questions and these might be hard to answer but i want to get some sense of this from you how difficult are these assessments to do proper properly and competently
02:21:41 do proper properly and competently uh they they they are difficult they they do require a very competent professional and it also requires a professional that has some level of understanding and is interacting
02:21:52 understanding and is interacting uh in a in an effective manner with professionals in other disciplines as i described for example in the case of a facade you're going to have an architect you can have a facade engineer and you're going to have a fire engineer and
02:22:03 you're going to have a fire engineer and to be able to do a proper assessment of mechanical behavior you probably need the input of all three at least and uh so so they are difficult you know to to be done i mean i think
02:22:16 you know to to be done i mean i think i think there is only one caveat that i have to put to that is that um you can make them simple and you can make them simple by being very conservative it still requires you
02:22:28 very conservative it still requires you to be very competent but nevertheless you can simplify the analysis to a point where um
02:22:37 where um you know you you don't need a one-year study you know to resolve that you could actually do it within the time scales of a typical engineering analysis and it will still require a competent individual and many times simplifying uh
02:22:50 individual and many times simplifying uh requires more competency than than than doing things brute force but um but i think
02:22:57 think the concept of conservatism and factors of safety is a mechanism that we have you know to make more mainstream our calculations and but do that do requires
02:23:09 calculations and but do that do requires a level of conservatism that many times might be unacceptable to the construction industry yes and that was my next question which is how cautious should those carrying out such assessments be and if it sounds
02:23:20 such assessments be and if it sounds like you're saying it it depends and if you're doing a simplistic assessment you might need to have lots of factors of safety built in to ensure that you're the right side of
02:23:31 that you're the right side of any line
02:23:32 any line exactly i mean your analysis has to be robust
02:23:35 robust it has to be extremely explicit and if you're going to make those assumptions to make your analysis simple doesn't mean that the analysis is simplistic is that your calculations are
02:23:46 simplistic is that your calculations are simple and therefore all the assumptions have to be listed have to be very carefully articulated and they have to be made available so that they can be verified for conformity yes and we've seen i think you've already
02:23:57 and we've seen i think you've already made the point that tiny differences we saw it with the edinburgh experiments for example just perforations in a a tiny foil facer or difference in method of fixings can
02:24:09 or difference in method of fixings can make a huge difference in fire performance absolutely and this is where you become conservative no so if you have a a protection system or an encapsulation system that you know it's fragile then you might simplify your
02:24:21 fragile then you might simplify your problem by just assuming it doesn't exist
02:24:23 exist yeah and the moment it doesn't exist you have immediately the ignition and uh and then you assess your system uh as a function of an unprotected uh
02:24:33 uh system yes was it an integral part of the queensland system
02:24:39 system to teach people how to do proper extrapolations yes that was that was at the core of the of the cpd you know it's basically walking uh
02:24:50 cpd you know it's basically walking uh the professionals that were in the course uh through the process of of building all the blocks until you finally deliver the performance of the system according to your fire safety
02:25:01 system according to your fire safety strategy yes we've heard from professor bisbee about some historic desktop studies that he was asked to specifically look at are you aware of his report and his evidence on that yes
02:25:12 on that yes and if you considered his analysis of the six specific desktop assessments which he was asked to review uh yes and do you agree with his criticisms of these desktop assessments absolutely what does that tell you about
02:25:26 absolutely what does that tell you about the state of competence of those operating within the industry prior to the grenfell tower fire in terms of their ability to carry out such desktop assessments well i think that if you take those
02:25:37 well i think that if you take those cases as examples of the practice and you assume that uh that they represent uh in a faithful manner the the the the practice i think that uh that those
02:25:48 practice i think that uh that those studies show that uh the people who were conducting them did not appreciate one the complexity of the problem and to the potential magnitude of the consequences of the mistakes and uh
02:26:01 of the mistakes and uh the the way in which they were done the very significant lack of justification provided for many of the assessments um shows you know very uh clearly
02:26:14 shows you know very uh clearly that um that the necessary attention to detail was not placed in in those studies were you surprised by what you saw in those assessments no
02:26:25 by what you saw in those assessments no no
02:26:26 no because why are we not surprised well i've done many third party reviews and i've i've done assessments of other people's studies many times and
02:26:37 of other people's studies many times and observing more or less the same approach to practice and how would you describe that practice very poor
02:26:46 very poor now i want to turn to bs 9414 briefly now um which is a new standard which was published after the grenfell fire in 2019
02:26:54 2019 which you deal with in this section of your report if we can go to it it's a bsi four zeros one nine three four
02:27:07 so here we have it it's called fire performance of external cladding systems the application of results from bs8414 part one and for part two tests and if we go to page nine of it and look at
02:27:20 we go to page nine of it and look at paragraph one under scope it says uh there at the top this british standard specifies procedures and rules used to evaluate variations and changes
02:27:31 used to evaluate variations and changes to products and systems which have been tested in accordance with bs 8414 parts one or part two and where appropriate defines options and limits for preparing reports based
02:27:42 and limits for preparing reports based on the direct and extended application criteria provided so is it fair to say that what this is is is some guidance that's intended to be followed and to help people
02:27:54 to be followed and to help people carry out desktop assessments by extrapolating from 8414 data well that's what the paragraph says but that's not what the document is
02:28:05 but that's not what the document is how would you describe the document an absolute oversimplified simplification of a complex problem that is full of mistakes from beginning to end
02:28:15 now just taking sticking with the document a moment if we go to page seven and look at the introduction
02:28:23 it says there it says at the top there are a number of practical limitations on the size and design of construction elements that can be evaluated by a standard method of test for example if these elements are larger or of a
02:28:35 these elements are larger or of a modified design or type it's necessary to be able to confirm that the integrity of the test result would be maintained if such changes to the system were implemented this can be achieved through the use of standardized rules now do you
02:28:48 the use of standardized rules now do you agree with what's said there no as i explained to you before what the objective of doing the assessment of a facade system is to establish if it actually meets
02:29:01 is to establish if it actually meets what we need from that facade system for the purposes of the fire safety strategy it is the fire safety strategy that is compliant
02:29:09 compliant now the moment that this is going to be up this information is going to be applied to a specific building with a specific fire safety strategy then the
02:29:20 specific fire safety strategy then the use of the word standardized becomes very limited there's very few things that you can actually standardize when what you're aiming at is a bespoke assessment
02:29:32 aiming at is a bespoke assessment yes and if we go to the third paragraph here it says this british standard is concerned with evaluating the scope of the results obtained from the 8414 tests defining the parameters which can or cannot be varied and to what extent
02:29:45 cannot be varied and to what extent it provides direct application rules for a system which has been subjected to a single bs eight four one four part one or part two test and an interpolation for systems which have been subjected to
02:29:56 for systems which have been subjected to multiple tests and then if we go to table one on page 14
02:30:02 14 we can just get look at some example some examples here of the type of guidance that's given so and you pick this one out in your report so you've got something that's
02:30:14 report so you've got something that's tested on a masonry substrate as the tested substrate and the variation is proposed that you're going from masonry to a board frame steel frame system
02:30:26 frame steel frame system and that's not permissible we can see that under option for change then underneath you're going from the tested substrate is a face seal steel frame system
02:30:35 system and the variation proposed is from a board-faced test
02:30:40 test sorry board-faced steel frame system test to a masonry test and that's said to be permissible uh provided you've got some additional evidence of temperature in the
02:30:51 evidence of temperature in the insulation layer fire resistance of substrate thickness of the brick block or concrete now you single this out as just one i mean this is actually the first piece of guidance that's given specific guidance
02:31:02 guidance that's given specific guidance in the report you single this out in your report as a piece of guidance that you simply don't agree with can you just explain why yes i mean if if if you think of
02:31:13 why yes i mean if if if you think of professor bisbee's analysis his his tests um
02:31:18 um one of the the the very significant conclusions of his test is that the insulating capabilities uh of uh whether it is the mineral wool
02:31:29 uh of uh whether it is the mineral wool or cellular plastic insulation have an extraordinary impact on the capacity of the aluminum composite panel to enter this runaway phase in the absence of those insulating
02:31:40 in the absence of those insulating properties this doesn't happen the moment that you have the insulating properties you get to this position now the moment that we say for example that if you're going from masonry you know to still frame systems and you say
02:31:52 know to still frame systems and you say it's not permissible that already in itself implies an assumption of what all those thermal properties are with absolutely no mention of what those
02:32:03 with absolutely no mention of what those thermal properties are you might assume that um that under standard conditions of masonry is is a much higher conductor than what
02:32:16 is is a much higher conductor than what a steel frame system is but the reality is that i can come up with a product where that gets reversed in which case i land on the other case so effectively what you're doing is
02:32:27 so effectively what you're doing is omitting in this table the entire complexity of the problem you're making lists and lists of assumptions you're omitting physical processes you're making all these
02:32:39 you're making all these jumps from one place to another one and landing into um into a decision not only that
02:32:48 that on top of that you are requiring additional evidence that obviously needs to be analyzed by a competent professional so on top of that you're putting this
02:32:59 so on top of that you're putting this qualification so on the one hand you're discounting the need for a competent professional and you're delivering rules because that's what it says at the beginning and uh and on the other hand you're demanding that a competent
02:33:11 you're demanding that a competent professional does additional analysis so in a way i mean just those two lines show clearly the inherent contradiction the sort of lack of care to the assumptions that are embedded in those
02:33:23 assumptions that are embedded in those decisions and complete disregard for many physical processes that um that can affect the performance of the system an sfs system for example can deform
02:33:34 for example can deform much much more significantly than a masonry system and we have taken absolutely no regard of that workmanship the the implementation of an sfs system requires a very high level of
02:33:47 sfs system requires a very high level of workmanship while masonry doesn't require that and we have eliminated all the things and then land you know with a conclusion that clearly does not necessarily cover anything beyond maybe
02:33:59 necessarily cover anything beyond maybe what it is conventional and uh and trivial to the naked eye yes
02:34:04 yes in your report you say that what this piece of guidance does is substitute a desktop study which should be conducted by a competent professional with simple tables
02:34:13 tables and you're very critical of this piece of guidance is that yeah yes i mean i think i think that you will find that there is a recurring theme in in all my reports that i'm extremely critical of
02:34:24 reports that i'm extremely critical of any document that substitutes competency and that's what this document does yeah because what you're saying is these systems these processes they're far far too complicated
02:34:36 too complicated to be able to be simply understood or interpreted by reference to simple tick box tables essentially exactly
02:34:46 and the the point you also make in your report is that because the 8414 test doesn't provide important relevant information for assessing relevant fire
02:34:57 information for assessing relevant fire performance you're building error upon error that's your view is that right exactly so you're starting from from a difficult place
02:35:05 place and you're just making it worse
02:35:08 so um does this 9414 document address any of the gaps that you've identified in the bs8414 testing and classification regime not in my opinion
02:35:21 and is it right that your view is that it compounds and amplifies the gaps and weaknesses in bs8414 exactly because it is presented as a substitute to competency
02:35:36 if we could go to page 118 of your report jtor 706 page 118. we just get your summary paragraph on
02:35:47 we just get your summary paragraph on this
02:35:48 this so you you've said in more detail in your report while you've got concerns and then at the paragraph at the top of that page you say in summary bs849 bs 9414 is a
02:36:00 you say in summary bs849 bs 9414 is a document that follows a long tradition of attempting to ignore the need for protection professional competency by substituting the need for professional assessment by means of over-simplistic
02:36:11 assessment by means of over-simplistic non-comprehensive and flawed prescriptive rules when tables one to six of bs 9414 establish whether a proposed variation is permissible or not this assessment is many times incorrect furthermore in most
02:36:24 many times incorrect furthermore in most cases the tables will also indicate the need for additional and competent professional assessment therefore this new standard is not only flawed in its technical content but it attempts to hide the need for the professional
02:36:36 hide the need for the professional competency that the standard itself invokes
02:36:42 now is there anything equivalent to the bs
02:36:46 bs 9414 guidance internationally has any other state gone down the route of providing this kind of guidance for desktop assessments uh well the the nfpa document the the nfpa um developed this software
02:37:00 nfpa um developed this software and uh and but they're quite explicit in what they're doing so they're not developing a table what they're doing is a collected expert opinion
02:37:10 opinion and uh and on the basis of that expert opinion they created a software that allows you to make a preliminary assessment of
02:37:20 of of a facade system now
02:37:23 now i mean i'm completely against doing that too
02:37:26 too nevertheless as opposed to uh bs
02:37:30 bs 9414 first of all they never qualified it as a standard or something that people should be following it is just a software that they offer the community and second the expert opinion is so
02:37:41 and second the expert opinion is so conservative in nature that uh that unless you have a brick wall it will reject everything basically so so so you know in in a way it's quite cautious
02:37:50 cautious of how far they they they go and it's because of the nature of of the fact that it's just a compilation of expert opinion so that's the only i would say similar
02:38:01 i would say similar um
02:38:02 um system that i know uh
02:38:04 uh that um that that has been proposed uh in in the past yes
02:38:10 yes and looking at this guidance and taking it as a whole in terms of all the guidance it's giving would you describe it as a cautious approach to the assessment of external wall behavior no
02:38:22 wall behavior no okay
02:38:24 okay now um just to pick up on a few more topics dealt with in your report in your report you discussed some historic
02:38:33 historic large-scale reconstructions that were carried out for forensic fire analysis and as part of that you've provided an analysis of the bre's reconstruction of the initial fire which occurred at
02:38:45 the initial fire which occurred at grenfell tower in flat 16. and if we could just go to that that's at met 3040237
02:38:59 there we have it it's dated the 24th of may 2019
02:39:03 may 2019 we can see it's called the grenfell tower fire investigation reconstruction and um is that the large-scale reconstruction you're referring to in your report yes
02:39:14 your report yes now i need to ask you this because when the phase one report was prepared um we didn't have the benefit of your views on this reconstruction you'd given evidence already for example about
02:39:27 already for example about the extent to which all the methods by which flame exited flat 16 um and and entered the cladding but um can you confirm this that in your opinion the fire which was delivered in
02:39:39 opinion the fire which was delivered in this reconstruction was a very different fire event from the one seen in flak 16 including the fact that this fire attained flashover unlike the flat 16 fire
02:39:51 unlike the flat 16 fire that's correct and and is it your view that hence this reconstruction and that's just one reason you give but it's a big reason you give this reconstruction does not provide any information which would contradict the
02:40:02 information which would contradict the views you expressed in your phase one report or indeed the views expressed by the chairman in his phase one report that is correct and just to confirm do any of the observations or measurements
02:40:13 observations or measurements uh that the bre took during this reconstruction change any of the views you expressed in your phase one report no they don't change anything okay thank you
02:40:23 you now you've also discussed other research activities in the appendices uh to your report um i just want to go to to that briefly if we could look at page 185 of your
02:40:34 if we could look at page 185 of your report jtor 706 page 185
02:40:44 so this is where you you analyze what research was available on um aspects of facade fires prior to the grenfell tower fire
02:40:53 fire and at the bottom paragraph a23 you say if we pick it up in the bid in bold you say it is therefore clear that prior to the grenfell tower fire the
02:41:04 prior to the grenfell tower fire the scientific community was not delivering significant and useful information that was capable of supporting a fire safe design of facade systems such as that which was implemented at grenfell tower
02:41:16 which was implemented at grenfell tower and then over the page you say this you say a different issue is the one of awareness
02:41:24 awareness and you say the fire safety engineering scientific community globally was aware of the external facade fire typology but nevertheless was incapable of securing
02:41:35 nevertheless was incapable of securing the support necessary to conduct the research required to quantify the fire performance of facade systems so can you just help us with this um
02:41:46 so can you just help us with this um why was it if there was knowledge of these external wall fire typologies why was it that the fire safety engineering community in your view was incapable of securing the support necessary to
02:41:57 securing the support necessary to conduct the research required to quantify
02:42:00 quantify the fire performance of these systems i mean the the the unfortunate thing is that uh for researchers to be able to conduct uh research and provide information
02:42:12 information you always need funding
02:42:15 funding and funding has to come from somewhere government was much more interested in supporting things that were more targeted in the direction of the
02:42:26 targeted in the direction of the reduction of energy consumption and and so forth and the goals of sustainability were very very strong in in in the period
02:42:35 period and anything that was perceived as being a decelerator to that direction uh was considered as unwelcomed uh research and uh and
02:42:46 unwelcomed uh research and uh and clearly this falls very much in that category so governments were not funding research in this area at all and
02:42:54 and in the industries that were actually expanding their market on the basis of uh the reduction of energy consumption again they will look upon this type of
02:43:08 again they will look upon this type of research
02:43:09 research in a very negative way and in many cases actively opposing it so people will not fund this work and whenever this work was
02:43:20 this work and whenever this work was done even if it was done independently you will get very rapidly reactions from multiple stakeholders trying to quench any information that could come out and has anything changed in that regard
02:43:32 and has anything changed in that regard since the grenfell tower fire no that's still the space the actual exactly exactly the same right that's still the space that academics are operating in yes
02:43:43 academics are operating in yes yes
02:43:46 and in your view is there a real need for research to
02:43:52 to to be carried out into the quantification of the fire performance of these facade systems i mean i i do hope that it is if you know with my presentation i really do hope that if there's one thing i have convinced you
02:44:04 there's one thing i have convinced you is that this is an incredibly complicated problem i think i did manage to confuse you completely with my equations and so forth and uh and it is a very very complicated problem that we actually
02:44:16 complicated problem that we actually know very little and that it has massive implications on numerous sectors and in the relationships between different areas of competency and overall in the construction industry i i
02:44:27 construction industry i i i think it is a fundamental area that requires an enormous amount of work if we really are going to come up with properly optimized systems that are going to meet all the functionalities
02:44:38 going to meet all the functionalities including the functionality of fire safety yeah
02:44:43 i've just got one more topic before we look at your conclusions from your report and this topic is about postgrenfor modeling related studies
02:44:54 modeling related studies now it's right isn't it that there's been certain modelling studies conducted after the grenfell tower fire and one such program of tests and modelling was conducted by effectus and published
02:45:05 was conducted by effectus and published in a series of papers between 2019 and 2020 is that right yes and you've provided some analysis of these in an appendix to your report appendix a2 yes um let's just look at the first
02:45:17 yes um let's just look at the first paper that was published this is inq three zeros one four nine zero eight
02:45:30 that was the first paper it was a study of fire behavior facade mock-ups equipped with aluminium composite material-based claddings using intermediate scale test method and that was published by the fire
02:45:42 and that was published by the fire materials journal in 2018 and the work was a series of tests and modeling which you say followed a building block approach and which aimed to give
02:45:53 to give some validation at each stage of the process can you just explain what you mean by that yes
02:45:59 yes the the model that they use is a complex computational fluid dynamics model called the fire dynamics simulator which is a state of the art this particular model uses a um
02:46:11 uses a um a numerical scheme that is called large eddy simulation that breaks the domain in blocks that are significantly large so you don't resolve the equation for
02:46:22 so you don't resolve the equation for every point you basically take a block and then you resolve everything for that big block and uh and then you have to move to the next one and the next one the next one and those blocks can be in
02:46:33 the next one and those blocks can be in in a smaller scale space can be five centimeters by five centimeters but if you want to mock up or create a model of grenfell tower those blocks become one meter by one
02:46:44 those blocks become one meter by one meter so what you need to do is try to find ways
02:46:48 find ways of resolving what is happening in the smaller blocks and for that purpose then you have to start with a smaller experiment a simpler experiment try to characterize what is happening in this
02:46:59 characterize what is happening in this small block so then you can use that and put it in the bigger block and uh and that calibration or that information then gets transferred blocked by block by block so they started with something that was quite
02:47:10 started with something that was quite simple quite small they moved to the next level the next level the next level until they finally tried to model grenfell
02:47:17 grenfell and i'll come to some more detailed questions about this modeling but in your opinion how useful is this kind of modeling in terms of an accurate prediction of external fire spread
02:47:28 prediction of external fire spread well it you know as as the authors recognize in in multiple places it depends what you're looking at so if i want to get a sense of how the fire propagated around the building after
02:47:39 propagated around the building after doing all these calibrations then i have more or less all the velocities right and and therefore i can see visually how the flame is moving around
02:47:51 the flame is moving around now if
02:47:52 now if for example i want to try to understand the failure modes and uh and look at the experiments that professor bisbee did then i would have to go
02:48:04 then i would have to go many scales below to be able to refine and see when the aluminum plate splits the their starting point is already an order of magnitude bigger of a scale so
02:48:16 order of magnitude bigger of a scale so they have to make a lot of assumptions about the
02:48:19 about the how the aluminum composite panel works so for example one of the assumptions that they make is that they don't look at the failure mode and the split but
02:48:30 at the failure mode and the split but what they assume is when certain conditions are attained the aluminum and the polyethylene disappear so they basically the protection of the aluminum is gone and the polyethylene burns
02:48:43 and the polyethylene burns and so for example phenomena such as dripping and melting and the burning that happens underneath cannot be predicted and the whole process of splitting
02:48:52 splitting of the aluminum plate cannot be predicted so they obviously once everything has split and the fire has gone into their fast growth growth phase then they can
02:49:03 growth growth phase then they can calibrate and then they're at the scale in which their model is actually predicting quite correctly what is happening in the rest of the building so it the the adequacy of the model is more
02:49:14 it the the adequacy of the model is more about
02:49:15 about what you want to get out of it and uh and the model is perfectly adequate for very large scale microscopic type of observations but it doesn't have the resolution to go into
02:49:27 doesn't have the resolution to go into the into the real details of for example all the early stages you know of the failure modes yeah now there are as we've discussed there are a series of papers that built up
02:49:40 are a series of papers that built up and then there were four final papers directly about grenfell tower the first paper was about an analysis of the images and some hypotheses the second paper uses modeling software to
02:49:53 second paper uses modeling software to explore different possible scenarios that you described in your phase one report
02:49:57 report the third paper attempted to model the propagation of the fire on grenfell tower and the fourth and final paper looked at
02:50:05 looked at grenfell
02:50:06 grenfell lateral flame spread so different aspects of the grenfell uh situation it is the reason you looked at these in such detail to understand what use could be made of this work and what the limitations are of it
02:50:18 limitations are of it well i mean i think it was it was a former i think in principle it was very important given that we were doing this experimental study that i was trying to support professor bisby it was very
02:50:30 support professor bisby it was very important for me you know to try to understand to what extent we could use this information in support of what we were doing and um and and
02:50:41 and i mean i came to a conclusion that while
02:50:45 while the results that pertain most of the later papers are quite quite well supported
02:50:52 supported by the earlier papers in this building block approach they
02:50:57 they basically coincide very well with all my observations and uh and all the the comments that we have made in the phase one uh report so i gained confidence that their
02:51:08 so i gained confidence that their modeling in many ways did not disagree at all with anything that we were saying and uh but at the same time i came to the realization that for the purposes of what we were doing with professor bisbee
02:51:19 what we were doing with professor bisbee the model did not have the capacity of being that granularity that was necessary to be able to understand the earlier stages and the contributions of the different materials yes so is this a
02:51:30 the different materials yes so is this a fair summary that you concluded that this work was broadly consistent with your conclusions in phase one and indeed with the chairman's conclusions in his phase one report um
02:51:40 um but that you concluded that the tests and models couldn't answer the detailed questions which you and professor bisbee were seeking to understand as part of the edinburgh experiments that is
02:51:52 the edinburgh experiments that is correct yeah and was it your view that this work couldn't ex properly explore the role of encapsulation of the products and how failure of encapsulation might occur yes that that is uh that is one of the
02:52:05 yes that that is uh that is one of the key uh limitations of this model and uh that it doesn't really have the capacity uh to
02:52:13 uh to understand the failure of encapsulation it can it it basically mimics encapsulation by trying to model it in a simpler way and uh but it doesn't look at all into
02:52:24 and uh but it doesn't look at all into the failure of it because it just disappears the encapsulation when it's necessary yes now i think since you've produced your report you've been provided with some further work and data
02:52:35 provided with some further work and data from effectis and i just want to ask you some brief questions about that if we go back to your report at jtor7 page 195
02:52:45 page 195 i want to look at paragraph a2.37
02:52:50 so here you say it's right at the bottom of the page the test utilizes a gas line burner as an ignition source this particular burner provides a heat flux to the facade assembly the facade
02:53:01 to the facade assembly the facade assembly is placed 250 millimeters and if we go over the page above the surface of the burner the heat flux imposed by the burner on the facade system is not characterized and therefore is not it is not possible to
02:53:12 therefore is not it is not possible to establish the heat flux distribution imposed by the burner on the facade assembly
02:53:18 assembly now
02:53:20 now would you accept that the heat flux from the burner in the intermediate scale test iso 13785 139 has been characterized by a test on plasterboard
02:53:31 plasterboard um in in subsequent data yes i mean i think that that is uh that is something that the authors actually touch in in one of their own papers so the the this is a standard test and as
02:53:43 the the this is a standard test and as such the the ignition source has been characterized in the past uh using a plasterboard as a substrate and extrapolating out the the the heat
02:53:55 extrapolating out the the the heat fluxes so i'm perfectly aware of that but as the author state um there are a series of variations that were done in this test that changes slightly the geometry of the burner which has it has a tendency to change the heat flux
02:54:08 a tendency to change the heat flux distribution in a certain way and also the presence of the facadin itself does have a completely different thermal boundary conditions so it actually changes um the
02:54:20 changes um the potentially the the the values of the uh of of of the heat flux imposed by the burner to the surface now to be able to for me to be able to
02:54:32 to be able to for me to be able to understand again i go back to what are you using the model for for me to be able to understand the failure modes which is what we were looking for this information is critical
02:54:43 this information is critical while for the authors of the paper because we're looking more at the spread later on and they're not looking at these failure modes then this becomes almost insignificant and i actually
02:54:54 almost insignificant and i actually touch upon that a little bit later when when i say that once you are past a certain distance then it just doesn't matter and for that purpose then you know my comment in in a237 uh then
02:55:08 know my comment in in a237 uh then becomes meaningless so so i think i think uh
02:55:12 think uh man many times when we get to this level of detail
02:55:16 of detail things are actually not disagreements but they are just simply coming from different angles might be perceived as disagreements yes and uh and i think this is exactly one of those cases where
02:55:27 this is exactly one of those cases where i think we know exactly what we're talking about it's just that we're focusing on different things yes now in terms of the two papers on computer modeling from 2019 there were two papers one was called the numerical
02:55:39 two papers one was called the numerical simulation of fire behavior a facade facade equipped with aluminium composite material based claddings model validation at intermediate scale that was june 19
02:55:50 was june 19 and then another with the same title but model validation at large scale was dated august 2019 now the papers themselves do not provide the
02:55:59 the numerical
02:56:00 numerical modeling that was used for that computer modeling
02:56:03 modeling but um i think you've now been provided with further details of that model is that correct yes um do you agree that the effect numerical model does take account of the properties of both the aluminium layers
02:56:15 properties of both the aluminium layers encapsulating the polyethylene core of the acm for the heat conduction convection and radiation yeah that's that's again you know we go back to exactly the the same thing if i
02:56:26 back to exactly the the same thing if i am looking into the failure mode which is the way i presented this in my report um i need to be able to resolve the heat transfer between the aluminum and the
02:56:37 transfer between the aluminum and the polyethylene and for that i need to refine the calculation to the dimension of the thickness of the plate now they take into account the thermal properties as as they say
02:56:49 properties as as they say but they take them into account as a bulk because what they're doing is they're heating the bulk of the aluminum and then disappearing it so for the purposes of what they're doing of course they're taken into
02:57:00 doing of course they're taken into account they have to take into account the properties of the aluminum in in in in their modeling but for the purposes of what we're looking for it is not sufficient resolution and it's not
02:57:11 not sufficient resolution and it's not the way in which we will have to take into account this to be able to look into the failure mode so again i think we completely agree and uh in this and the information that they provided just confirmed uh what i
02:57:22 they provided just confirmed uh what i already knew it's just that we're coming from different angles so it doesn't change any of the views you've expressed in your report no not at all no um if we look at page 208 of your report
02:57:33 um if we look at page 208 of your report um at paragraph a287
02:57:39 this is where you're still dealing with these papers and you say this you say nevertheless this information even if treated in a qualitative manner is interesting and useful and shows that these two specific features can serve to
02:57:50 these two specific features can serve to enhance the rate of vertical fast spread and then you say this the observations that cavity barriers will have little impact on the fire spread rate is also in agreement with the descriptions
02:58:01 in agreement with the descriptions presented in my phase one report the overall vertical fire spread rates are consistent with those reported in the phase one report now
02:58:10 now based on the numerical data that you've now seen do you agree that that numerical modeling shows that the cavity barriers delayed the initial propagation of the fire by about seven minutes
02:58:21 of the fire by about seven minutes uh well that that is that that is difficult to agree with because i think uh in in um in point a two to 87
02:58:33 uh in in um in point a two to 87 basically what we're seeing is that the modeling at that scale very rapidly shows that um
02:58:42 that um that the flame basically is bypassing the cavity barrier so it really has very little impact and therefore there is consistency between uh what i discussed
02:58:53 consistency between uh what i discussed in in phase in my phase one report and the modeling now
02:58:58 now um
02:58:59 um in their modeling results effectively yes the cavity barriers um delay the
02:59:06 the the the propagation by seven minutes or uh the the number that they specify but that is under the assumptions and simplifications that they have made
02:59:17 simplifications that they have made because if you again go back to this idea of the building blocks they cannot resolve the cavity so they cannot resolve the flow field in the cavity they cannot resolve the gaps in the cavity virus none of these things are
02:59:29 cavity virus none of these things are resolved so if if you caveat what they're saying with uh based on the coarseness of the model and you know there is an estimate off from a qualitative perspective
02:59:41 qualitative perspective you know i cannot disagree with it but the model really is not telling you that in the sense that the model is too coarse to be able to attain that level of precision okay thank you mr chairman
02:59:53 of precision okay thank you mr chairman just one more question on this numerical modeling if i may and then i finished the topic
03:00:01 that if we could go to inq three zeros one four nine zero nine this is the numerical simulation model validation at uh intermediate scale and
03:00:12 validation at uh intermediate scale and it's about the performance of clouding systems incorporating different types of insulation behind the pe called acm do you agree that the nature and rate of flame spread on cladding systems
03:00:23 flame spread on cladding systems incorporating foil face pir and foil phase phenolic and unfaced mineral fiber insulation in all cases with pe called acm rain screens have found to be comparable in modeling
03:00:35 comparable in modeling uh yeah at that scale you know once everything has taken off and i think that's actually consistent with professor bisbee's observation so so once it's taken off uh yeah i mean
03:00:46 once it's taken off uh yeah i mean there's no reason for me to disagree with that yeah and just briefly is there a role for these kind of cfd models as part of assessing facade fire performance
03:00:58 assessing facade fire performance um
03:00:59 um yeah of course i mean the cfd
03:01:02 the cfd model provides an enormous amount of information but we have to recognize that cfd models are also quite limited because we do have to do a lot of uh
03:01:14 because we do have to do a lot of uh playing around with them to get them to work appropriately so i think in the hands of a very competent individual they can actually help understand certain things and so so i think that
03:01:25 so so i think that that
03:01:26 that cfd obviously has a very significant value and but it has to be used in a very cautious way okay thank you thank you i'm sorry for going slightly over but it's right good to get that topic finished good thank you well i think
03:01:37 finished good thank you well i think it's time we break for lunch thank you professor herrera say we'll stop there we'll resume please at five past two and again please don't talk about your evidence when you're out of the room all right thank you
03:01:55 thank you very much five past two
03:02:02 you