The Demonstration of Energy Efficiency Potential (DEEP) – Implication of Project Findings for Historic Building Retrofit

Hannah: Hi to everybody. Thank you very much for joining us today, and taking this time out of your day. So, my name's Hannah Reynolds. I'm a conservation architect, and I work within the Historic Building Climate Change Adaptation Team at Historic England. We're a multidisciplinary team, and we focus on historic building climate change mitigation and adaptation within the wider historic environment. We undertake research, we write technical guidance, and we support a whole wide range of projects and workstreams on this vast topic. My work in particular focuses on understanding, providing best practice, advice, and guidance on implementation of robust and resilient mitigation measures in historic buildings. The DEEP Project was one of the largest research projects into solid wall retrofit undertaken to date, and it successfully demonstrated that the whole-house approach is a fundamental part of reducing risk.
This directly aligns with the work we do at HE to support good decision making in the historic built environment, and reduce the risks associated with retrofit whilst maximizing energy and carbon savings. So, we're very pleased to be following on from our previous webinar on this topic. And today we'll be taking a deep dive into the interesting findings that we weren't able to fully cover previously. So, the first webinar presented a summary of the project and its findings, and there is a recording on our website if you haven't seen that, but I hope you have. So, on your screen now, there's just a load of links to various bits about the DEEP project itself, from the DESNZ and the Leeds Beckett University website, and the previous recording is at the top. I believe you will have been emailed these before, and they'll also be put into the chat now for you.
So, joining me today, we've got Roger Littlewood and Professor David Glew. So, Roger is from the Department of Energy Security and Net-Zero, and is based in the Energy Research Team, where he provides technical advice to policy teams, as well as leading a range of research initiatives which included the DEEP Project. Professor David Blue is Director of Leeds Sustainability Institute and Head of Energy Efficiency Policy. He manages a large range of interdisciplinary retrofit evaluation projects, and undertakes building performance evaluations. He was lead for the DEEP Project Consortium for DESNZ. So, it's now my great pleasure to hand over to Roger for a short introduction before we jump into today's discussion. So over to you, Roger.
Roger: Thank you very much, Hanna, and afternoon, everybody. Yes. Now, The DEEP Project, as Hannah mentioned, was a multi-year research project. It's domestic energy efficiency, specifically in retrofit of solid walled homes. We commissioned this way back in 2019, But I'd like to say, we have now, at long last, published all the reports which were published on gov.uk a couple of months ago, or towards the end of October. It was delivered for us by a consortium led by Leeds Beckett University, but also involving a few other partners as well. The general purpose of the project was to compare the energy saving performance and risk that is from things like moisture, overheating, taking the whole-house approach, versus the piecemeal retrofit approach, i.e. piecing things together over time with no particular consideration for how different retrofit measures interact with one another or with occupants, and so on.
Dave and the team used direct measurement, physical measurements, looking at the change in heat transfer coefficients both across retrofit and across stages of retrofit. Where multiple measures were installed, they examined different models and how they can be improved, and observed and made recommendations on practical delivery in the field. And like any government research project should be, we were one of the projects to gather evidence to inform policy, change innovation standards- sorry, industry standards, our future innovation programs, and of course, future research itself, some of which we've actually already commissioned, and indeed there will be a couple of projects that directly follow on from DEEP, will be published in the next couple of months.
Cost wasn't the primary focus of DEEP, but there is some useful data that's been gathered by Dave and the team nonetheless, and presented in the reports, which you can browse at your leisure on gov.uk, as I mentioned. But just to summarize what the major findings were- there were a lot more than this, but these are the ones that we thought were the most important, that we brought to the attention of our ministers when we were seeking approval to publish the report a couple of months back. The whole-house approach does indeed have fewer unintended consequences than a piecemeal approach, as you'd hope. This is some real direct evidence in UK homes of that. The solid wall insulation was the most effective measure individually, with between 19% and 55% heat loss across the case study homes that were retrofitted.
There are, as you know, high pre-retrofit enabling costs, on average 20%. That is things like removing asbestos or doing structural repairs that are necessary before that have nothing to do with the retrofit measures themselves, and a long financial payback to a number of the measures. But there are, however, non-financial, or at least not immediately that you could state as financial advantages, such as warmer surface temperatures, better control of ventilation and heating, and reducing the risk of overheating. SAP, or RdSAP was used for retrofit currently, and therefore EPCs, which they used to produce, overestimated the amount of heat lost by 42% on average versus what Dave and the team measured in the case study homes. But they also recommended improvements, for how that can be improved in the future.
So, clearly things like using measured inputs and the use of dynamic modeling. My colleagues who were producing the home energy model, the successor to SAP, have been engaged throughout the project, and since the funding has been produced, to incorporate those into HEM as they develop. It's still in progress at the moment. So, with that, that's an overview of the project, why we did the project. and its general findings. But to delve further into the technicalities, I shall hand over to Dave.
Hannah: So, thanks, Roger. Sorry, I'm just going to jump in before you get going, David. So, today we're going to begin discussing what data was studied before moving on to the interesting findings on insulation, draft proofing, and air leakage, overheating, some of the most beneficial actions and some of the next steps that came out of this project. So, David, starting off on the dataset studied, we understand that the majority of solid-wall buildings studied were brick-built. There was one concrete, one stone, and one of nontraditional construction. But the report didn't speak about their wall thickness, whether they're half-brick or solid wall. So, the latter being solid walls of 300 or more. So, I was wondering if for our audience you could expand on this, and does this mean the findings are representative of mass solid walls buildings as well as thin solid buildings, or might this be an area of further research? And do you think the benefits and risks might be different in these cases?
David: Hi, everybody. Thanks. Hopefully you can hear me. Yeah. Let's have a look, then. So these are our case study homes. And as you can see, as you just described, there's a lot of brick homes now. Now, they're nine inch bricks, most of those. there's really only one stone home in there, and there's a couple of concretes as well. Now, the objective of the project was to find homes that were similar to homes getting retrofit under ECO. So, that's what we've kind of set out to do. It wasn't to compare really thick walls versus really thin walls, or concrete versus brick, or anything like that. It's not to be representative of UK housing stock, but to inform energy policy. Having said that, it's not been... What we don't think though, therefore, is that there's a specific finding that's only going to be relevant for brick walls rather than stone walls, and we'll come onto it in a bit more detail a bit later on.
The first thing that we did, though, we took some core samples of these homes, and we took some core samples from some other homes, eight in total. One of the major findings that we had was actually that the variability in moisture content in those bricks is tremendous. So, you can have 10 times more moisture content in one of the bricks we sampled in London, and flattened brick that we sampled, compared to one of the homes that we sampled in South Yorkshire, for instance. That's really important to consider. You can't really archetype on brick type, because bricks themselves are so variable. So, that was one of the major things that we spotted. Regardless of, again, if it was a super thick wall or just an old nine-inch brick, all of these homes- and you'll notice not all of- some of them actually already have some external wall insulation, some have some very thin internal wall insulation, but where they were uninsulated walls, regardless of the original substrate, they all had cold surfaces as modeled by thermal bridging analysis.
They had surface condensation risks present, as described by the temperature being below 0.75, which is a kind of threshold analysis, a rigid 0.75 value, regardless of the brick type. And that's just a model. It doesn't consider whether or not there really is a risk, or whether or not the internal environment is being maintained and ventilated, etc.. But, you know, we have models, and this is what they tell us. Okay. One of the things we did is, we measured U-values. Some of the homes had pretty much similar U-values to what was considered in EPC. So, the home in the top left and the one below that, and the across of the middle, on the right hand side, they had a pretty similar U-value to what APC was assuming. Some had much better U-values, so much, much lower U-value. So the home, the top in the home, the stone home in the middle, they had a better U-value. And some of the homes we found had much worse U-value.
So, again, the performance is kind of all over the shop, depending on what brick type, or what the substrate is. It was interesting, but from a point of view of solid wall insulation, the substrate that you're starting with becomes really not important when you then install the insulation. The insulation just dominates the U-value. Whether you got a higher or lower starting U-value with your brick, whether it was the same, it kind of didn't really matter, because as soon as you put a little bit of insulation on there, it doesn't matter what you started with, the insulation is the thing that's driving the U-Value. We will talk a little bit about EWI versus the IWI. There is no difference physically whether you put the insulation on the inside or outside, but it will affect a dynamic analysis, the heat of the cool down.
But the methods that we use and the models that we use, they're all looking at steady state performance. But we will touch on the IWI stuff a little bit later. So, I think that's why I've got on this one. Did you want to- I hope that captured lots of things. I'm trying to capture it. I know there's lots of sub questions that we've been chatting about. There's so much to talk about.
Hannah: Yeah, Thank you.
David: I don't know when you want to pick up some of the questions, or whether we just jump straight through to the next.
Hannah: Yeah, thank you. That's great. Well, there's one that came up that was in my head as well, and it was asking if the U-value varies owing to the moisture content of the bricks. So, the more they were, the worse they were performing?
David: Yeah, I do get that. We did analyze the water content in the bricks that we sampled. To be honest, it's really not such a big deal compared to things like adding a little bit of insulation. It just makes tremendously more difference. So, I don't know if we need to get caught up On whether or not we measured. It wasn't particularly moisture contents or not... It wasn't the purpose of the experiments anyway.
Hannah: Okay, brilliant. Thank you. So moving on to insulation, I'm conscious of time. We've got lots of interesting things to talk about. So, I thought one of the really interesting overarching findings was relating to the longevity of retrofit. So, your report said that since the lifetime of a home exceeds that of the retrofit, insulation material should be installed so that they can be replaced at the end of their expected lifetime. And I thought that was just worth noting to the audience because it highlights an important consideration that people quite often don't account for when they're looking at these. And it kind of feeds into the circular economy, and all those sorts of things that we should be thinking about. So, I thought it was worth noting. But as for our discussion here, an interesting finding for me was that installing IWI was always found to increase moisture risk, even when only a thin layer was installed, and interstitial condensation risk was also introduced, which we are all aware of on the most part. Probably the audience too.
And the report found that increasing the IWI thickness, of course, increases this moisture risk, with the risk Increasing more rapidly below a U-value of 0.8. So, if it's at 0.8 or better in terms of its thermal performance. I thought this was really interesting. I wondered if you could expand on the implications of this finding and its relevance to traditional construction?
David: Yeah, this is great. This is a really exciting finding, isn't it? So, with something that's similar that we've had in our other project, the thin internal wall insulation project, which is also on the DESNZ website, you can look it up. But yeah, that comment that we've made that it always introduces more risk. Well, it's quite controversial, really, a quite boring state when you think about it, because actually all we're saying is, you're making the structure of the building colder. Anything that's cold has a higher risk than anything that's warmer. So, that's kind of just a general comment. But the specifics are the really interesting bits, I hope, and what we've got here is some really nice modeling that was done by Loughborough University as part of the project. And there's three graphs here.
And essentially, these three graphs are showing a solid wall. And you can't really see my cursor, I don't think, but if you look on the left-hand side, and at the very bottom, that U-value, essentially right on the left-hand side is the solid wall uninsulated. Okay? And then, as you get to 1.6, and 1.4, or 1.2, you're adding more and more thickness of insulation. So, you get to 1, and then down to 0.8, and then all the way on the right-hand side, you've got 0.2, a really- really insulated wall, loads of insulation on that wall. Okay? So that's kind of, if you like, the journey of that wall depending on how much insulation you put on it. Okay. Now, what are we actually looking at here? so, we've got an orange line, or red line if you like, on each of those graphs, and that represents the heat loss. So, on the uninsulated side, where there's no insulation on the wall, it's really high, high heat losses.
And as you add more and more and more heat loss, as you would expect, that orange line drops. So, right at the other end, really insulated. You've got much lower heat losses. But the interesting one here is the blue line. So, if you look in the middle model there, on the left-hand side, uninsulated, obviously there's no interstitial condensation because there's no insulation on the uninsulated wall. But as you start adding thicknesses of wall insulation, internal insulation on there, that risk increases, increases, increases, as you would expect it to do. But that's the middle graph, and that's an EPS solution. So, the three graphs are comparing different insulation products on the same wall. So, the top graph, model A, is wood fiber. And if you look at its journey, it starts off on the left hand side low.
And actually, as you add more and more insulation of wood fiber, it doesn't really increase much until about 0.8, and then it starts to go up. The EPS with an air and vapor control layer on it increases almost linearly. And then, actually, what we've tried to do in Model C at the very bottom, is do that same foam solution, but without that control layer. And what we can see there is that now you're adding more and more insulation, actually there's a really small increase, similar to wood fiber, really. You're not really increasing the risk at all. And then, at 0.8, again, there's a much deeper increase in risk. And then, at 0.8, again, there's a much deeper increase in risk. And what that's telling us is actually because of the wood fiber and the no AVC layer on the piece is allowed to dry out the outside and dry out to the inside.
That's meaning that there is less moisture accumulation in that strip, in that structure. And until you get to a U-value of about 0.8 for this particular model, actually the risks are not that great compared to what they originally were. But, you're still getting 80% of the energy saving. So that's kind of where really- that's what that means. It is just one wall, it is using models, it is using defaults, and what we've just told you in the last slide is, of course, the existing substrate will really influence what the graph looks like. So, we need more testing on more different types of wall types. So, the existing substrate doesn't really matter from a thermal point of view, but it really matters from a moisture point of view. I think that's the key. Same sort of thing that we're finding out for the insulation as well. It's dominated by the U-value drying into the outside we've already talked about.
Matt: Could you just explain the acronyms for us, please?
Hannah: And there's a few questions about EPS, and I guess...
David: It's expanded polystyrene, or... Yeah, I think that's EPS. Or, extruded- Sorry, extruded polystyrene. And then, AVCL is the air and vapor control layer. Yeah. And are there any others on there? No. Okay, there we go. Yeah.
Hannah: I don't believe so.
David: Right. That was the query about the moisture risks. And then, I know one of the queries that, Hannah, you had, was about the thermal performance of the different types of insulations that we added. So, this is a graph which shows the heat loss reductions of every different type of retrofit that we did across our 14 case studies, where we had a specific test for it. The zero line effectively is, if you're touching that zero line, we couldn't really measure statistical significance before and after because the test methodology wasn't sensitive enough to pick up a change. The change was kind of too small to pick up properly. And really, the take-home is to show that actually solid wall insulation was the only insulation type, retrofit type that always gave us a big improvement, a big reduction in our heat losses, even where you only had quite small wall area to window area ratio.
So, you might have these really big windows and actually a terraced house, so it's not got a lot of external wall, even in those scenarios, solid wall insulation was still more effective than the other types of insulation. One thing I was going to mention here is that you can get more savings from an EWI than an IWI, not because it's inherently any better, but just because, in the last slide shown, we tried to reduce the risk by reducing the thickness of the IWI. We don't have that problem when you're doing EWI. You can just do it as thick as you like. So, the product is the same, the physics is the same, but from a risk point of view, you just choose to put less on with IWI. That's why you might get smaller savings, but still significant savings. Of course, it's still worth doing all of those other things, and especially where you can combine them, you get a cumulative effect.
So, this is a graph to just show that. This is what we did. We actually did lots of things together as well as separate, but still, the best thing to do from a thermal point of view was to do those things with the solid wall insulation, where you might get something like 60% reduction in heat loss. So that was, just hopefully covering some of those queries. I don't know if you want to go to other questions or take them, at the end or whatnot. What's next?
Hannah: Yeah, that was really interesting. Lots of information there, and I think there are lots of questions coming through, some of which we are going to cover as we move forward. And hopefully people are catching up. But of course, it's very detailed, and kind of in-depth, which is what we were hoping for for this webinar. But I hope people are keeping up. So, I think... Moving on to the draft proofing in air leakage. So, the report found that one of the simplest and most cost-effective measures to reduce heat loss due to air leakage was fitting carpets on timber floors. So, this is interesting because that guidance already highlights that. So, that's nice. But I wondered if you might expand on other draft proofing solutions that were tested, and just answer whether the research identified other solutions like this, that would be of use to our audience.
David: Yeah, this is good. And we were very conscious draft proofing... This is- you know, remember, this is a long time ago. Draft proofing has moved on a lot. And it means a lot different things to a lot of people. We were specifically looking at measures that you could do where people are in there, not stripping kitchens and bathrooms, going back to brick membranes, types, you know, all that good stuff that you might see in this type of retrofit. So, this is what you can just do potentially under ECO. Yeah. So, that's what we were specifically focusing on. I'm going to run through a couple of nice images, because I wanna show something pretty. The types of things we did. So, what can you do if you're not going to go back to brick? Well, you can seal up against penetrations. You can create new boxed-in areas around where the plumbing comes in and out of the wall.
That was great where it was a direct air leakage. What we found was, though, if that was then an air pathway which was going into an intermediate floor, or the ground, or into another riser, essentially all you were doing was removing the draft here, but it would still occur somewhere else. You're just kind of redirecting the air, not getting rid of it. We did take some sash windows out, repaired them. That was fine. And again, where it was a direct air leakage, there was an improvement, but it was quite small relative to the rest of the air leakage that was going on in the house. So, it made a difference locally to that specific area for comfort maybe, but you're not going to see any kind of major energy savings from it. Other things we looked at- So, this is post-retrofit, and because the insulation was so thick, we had to have a bespoke bit of joinery for the loft hatch, and where there's bespoke joinery, you're not getting kind of a certified product. You're relying on the carpentry skills. And actually, we found that that was an area of weakness. So, be careful where you're having to do bespoke carpentry and stuff like that. Floor sealing.
So, not just carpets, but coverings as well. This is actually plywood that was put down. That was really effective, same as carpets. But, what wasn't effective at all was, they were doing things like mastic around the sides to join where the wall and the floor meet, and they deteriorated really quickly, especially under our testing conditions, where we were superheat charging the home and stuff. So yeah, they're really not going to be very effective at all. Adding a new servicer. This is an NVHR that we put in, and it was really hard to then actually make those new penetrations air-tight too, especially in this instance, because we had no fine construction. And that is just a really crumbly, horrible material to have to work with. It just made it really tricky. So, we were trying to do some good, and we actually ended up creating challenges for ourselves.
And then, the last thing we did we were observing a lot of- we measured a lot of air exchange between neighbors. So, across the parting wall. So, we revealed where the timber joists are going through into the parting wall, and we could see the gaps in the mortar, cracks, lots and lots of potential areas of infiltration through into the neighboring properties. We exposed those edges of the walls, tried to then reseal those, mortar in the cracks and gaps. It's a very disruptive and quite a difficult thing to do, very costly, because it was labor intensive, and really very ineffective, because you are trying to do a local job, local solution. And really, when you're trying to do that without properly stripping everything out, you're really up against it. So, that was not very effective to do.
So, what did we do in total? So, here's seven homes. We did several things in some of those homes, and this is the graphs to show what the air tightness did at each step. The first dot is always where it started from, where we started. In this home, 17BG, you can see those general ceilings were really effective. So, it went from 50 meters cubed per meter squared, 50 Pascals down to 10. And that's because there's loads of really obvious penetrations that we could see that were going direct to the outside. But then, in the other homes where we did something similar, we didn't get such a benefit. Here's where we looked at the floors and tried to do sealing. Really, nothing improves the air tightness until we put the carpets on in this particular home. Same for this house as well.
So, the take-homes are, it can actually cost you a lot, it's quite labor intensive and you really have unpredictable results. You don't know when you start whether or not this is going to be beneficial. Floor coverings seem to be the best thing that you could do. So yeah, really, really interesting stuff. And of course, there's lots more that you can read in there, the reports. I think you're on mute again, sorry.
Hannah: Thanks, David. IT assistance. So, there were a couple of comments, just to clarify that we're not talking about doing draft proofing without any adequate ventilation alongside it. That's always a prerequisite for us. So, there was a question about that. We'll get on to air tightness and ventilation and things later. So, just a note that it sort of generally seems a less nuanced set of draft proofing measures that you've studied than HE describes. So, there was another question about passive measures in the questions. So I mean, for us, we think about including wall finishes, window joinery, and doors, chimneys, and floors altogether, which kind of adds to the package of tools that people might have. It's just that this project didn't particularly study every single one of those. You've got to have some limit.
David: Yeah. And I think there's really good evidence that if you don't look at it like that, you're potentially going to not have any benefit. So, you kind of have to go to the extent of maybe going back to brick, and re-plastering and, you know, doing all that sort of stuff, because it's so unpredictable.
Hannah: And I was going to say, for clarity for the audience, and when we're particularly looking at historic buildings. So, the findings are suggesting that where fabric heat loss is being greatly reduced by insulation, so, air leakage therefore becomes more impactful, the best thing technically we could do is strip the building back to achieve a continuous air tightness line with membranes and tapes, which then it's quoted as may have more potential to achieve savings. And of course, for us, this level of solution could be extremely disruptive and expensive for the owners, but also, in a heritage setting, might result in the loss of lots of historic fabric and internal detail. So, it's just noting that for those reasons, the level of intervention needs to be really carefully considered, resoundingly benefit, and provide longevity. So, I didn't know what your thoughts were on that sort of potential benefit versus disruption. Whether it's... Just something for our audience to consider.
David: Yeah, I think because we weren't designing the experiment, it wasn't designed to evaluate different approaches to the disruption, it's really hard to tell. We know there are case studies that have been really successful in reducing air leaks by having that really invasive, really disruptive approach, which, essentially, you're just almost building from scratch again, building a new air tightness layer on the site. That is really not appropriate for the vast majority of situations. We were specifically looking at easy wins and quick gains here. That would be a real boon for policy, if this had shown that it's really reliable and successful, and sadly, it's not. It's much more complicated than that.
Hannah: Yeah. So I think it goes back to sort of... People love when we say horses for courses. We've really got to look at what we're dealing with, and what might be appropriate or not in that particular situation. On that subject, I wondered if in terms of maintenance and repairs of buildings, did the research consider that micro cavities in external walls might be less when they are in good condition, and that if we've got renewed line plaster render layers, it can have positive impact and help ensure the building remains in equilibrium with its environment? Presumably that wasn't necessarily covered.
David: No. Well, I mean, it was in part. I don't know if the next slide might help inform that. I don't know if I can click forward to...
Hannah: I think we both clicked it.
David: Maybe we'll get onto that with this next slide, which is where we did- we didn't just look at these 14 homes for air tightness. We actually did a survey of 150 homes, which did include some cavities as well, but we tried to focus as much as possible on solid walls. This is just a histogram- So, the age ranges. So, you can see we did look at really old homes and some really new homes. And then, collectively, we threw all that data together and produced this graph, which is just to show... Have I gone forward again? If I do the next graph... That's then all of these homes' air tightness results all lined up against each other. And what you can see is, there are some really airtight homes, down to five meters cubed, and some really, really poorly performing homes, really struggling with air tightness issues up here.
These were those poorly maintained homes that were, you know, where they were missing bricks, and there were cavity problems, and holes, and stuff like that. But what we tried to do is a little bit more nuanced, see if there's any trends that we could spot and that we could archetype, you know, to say, okay, it's this age, therefore it's going to have this air tightness. And there really wasn't anything like that that we could pick up from an age band. I should say, this red line as well. That's the new build standard of eight meters cubed. And for interest, most of the homes that we assessed in SAP, the EPC assumed those homes would have about 15 meters cubed per meter squared. So, what can we see? Well, actually, most homes have got better air tightness than we think in EPCs.
So, that's interesting. But most are worse than new builds. We kind of would have assumed that. The only analysis that showed any kind of trends - age didn't, EPC band didn't. What we did find was a small correlation between whether or not the wall had wet plaster on the solid wall versus dry lined wall, for instance. So, the dry lined walls were much worse performing. They tended to be. It wasn't such a strong result. The most important driver, if you like, if you're trying to predict where your home was, was whether it had a suspended floor or solid floor. So, suspended floors were much leakier. Not much leakier, they tended to sometimes be leaky. It wasn't a very, very strong driver. And the solid floors tended to be down towards the more air-tight thing. But that was just an indication that, yes, actually, homes aren't archetypal. And that condition, which is to say how well-maintained they are, which you were getting at, and how much they've been messed around with, is almost a bigger determinant than any of its physical characteristics. So, it was really interesting.
Hannah: Thank you. That was really great.
David: That's for air tightness. Sorry, there we go.
Hannah: Yeah, no, I think we're moving on. I was going to say thank you. That's really interesting. I think that kind of answers the questions that our audience might have. There's some coming up, but we'll get to those later if we have time at the end. So, moving on to overheating, then, which kind of feeds into what we're talking about with air tightness as well. Just wondered if you could explain a little bit about the approach that was taken in the project to overheating risk assessment, and the findings of these? It might answer some of the questions that we've had as well.
David: Oh, great. Okay. Yeah, I'm very happy to... I'll go offline as well. I'll try and answer as many of these as possible, because they're all really interesting. Right. So, for the overheating, we assessed two different approaches. Leeds Beckett did a lot of evaluations of every single of those case study homes that we did against every single retrofit that we did, and then Loughborough took two of the case study homes and applied 10-year weather files to them and looked at some additional scenarios. And I'll just quickly run through what we found in two slides. So, this is the first one that we did. So, we took... This is a graph that shows every retrofit that we did. It's in on all of our case study homes, what its impact on the likelihood of that home overheating was. So, the zero line that you can see is there's no change when you apply the retrofit.
So, if you're above the zero, that means that that retrofit increased the chance of overheating. And if you're below the line, it means it decreased the chance of it overheating. So, what we can see here is there was a more- and I should say from the very beginning, all of the homes pretty much overheated. So, they all almost failed, all of them overheated, because that's just the way that it was. The retrofit then just- this is the relative change that that overheating risk had. So, draft proofing made it slightly more likely to overheat. You can imagine why. Just less air changes taking place over summer evenings. Loft insulation. When you've got better loft insulation, it slightly increased the chance of overheating. Again, that's because on a summer's evening there's less heat- the heat is less able to be lost to the sky.
The room and roofs had the opposite, slight improvement in overheating, and that's because essentially uninsulated rooms and roof, as we all know, act like greenhouses sat on top of the house. They get really, really hot, and so insulating that out is beneficial. New glazing was very beneficial, not because of the kind of posh glass, but because old fashioned PVC windows had really small openable areas, and more modern windows are able to open big and wide. And that's really positive for heritage buildings, which have got large sash windows. You can really get loads of ventilation happening in those types of buildings. So, that's a positive. Suspended timber floor, slight overheating. That's because on a summer's day, the air underneath the floor void is sometimes cooler than ambient. And so, that's a cooling benefit.
And if you insulate your floor, you stop the home from being able to benefit from that cooling in the summer. The walls in the whole house are the same thing, really, because they both include a solid wall. And really that's the same as the room and roof. You're just essentially stopping the amount of solar energy getting into the home to begin with during the summer, so it's less likely to overheat. So, those are our findings. Loughborough took two of those homes and they evaluated them in a little bit more detail. So, we just did our normal assessment according to our criteria for one year. Loughborough decided they could do it over 10 years, which is really much more interesting. So, you don't get one hit. You can see how many years out of 10 something will overheat, And that's what these numbers represent.
So they picked up our homes and they put them in London, Manchester, and Glasgow, to see what would happen. Scotland didn't overheat at all. Manchester overheated a little bit, and London did seem to overheat. The base case is uninsulated in the whole house. HWR is the whole house retrofit version. So, a little bit of overheating when you've retrofitted, you're not in London, but what Loughborough did was, they introduced shading and increased the amount of ventilation capability in those homes, and that basically solved the overheating problems that those homes had to begin with. So really interesting. We know a lot of heritage buildings were always built with shading, even in the UK.
So, we should be really encouraging policy, especially in London, to start reintroducing shading as part of whole house retrofits. The next thing they did is, they looked into the future. It's just the same story, but it just gets a lot worse. It's a real problem. And, you know, even in London in the future, even with shading and ventilation, it's not looking great. Probably going to be some overheating happening in those homes. I think that's all for that one.
Hannah: Yeah. Before we move on, I was gonna say this: it's really interesting and it's something that, you know, the team, and particularly my colleague Joanne, who works on risk and resilience, is really looking into this aspect of overheating. So, it's really interesting to see that the findings suggest that buildings in London need to be treated differently from those in the north, particularly when you're talking about wall insulation and suspended floor insulation, and how that impacts overheating. And it's also really interesting to see that this kind of demonstrates the importance of considering those adaptation and resilience measures alongside mitigation, and how much those will potentially help mitigate that risk. So, I think it's really interesting in terms of positive actions we could take to reduce this risk. So, we're mostly talking about external shading and increasing ventilation, as I understand it. Is there anything else that was highlighted?
David: Those are the only measures that were modeled. But of course, yeah, you could introduce lots more into those models to see what the effects were.
Hannah: Okay. So, I think the main thing from this for our audience is that we really need to better understand the risk of overheating from climate change depending on the location of the properties that are proposed for retrofit, and where the climate change hazards are not being considered. We're probably at risk of inappropriate works and overheating, which is something to really think about when we're specifying.
David: Yeah, definitely. And also the use of the current models which only use one weather file. Maybe that's not enough. We might need 10, because it's so much more useful to get that spread across time. So, that's a really good point, an important contribution that Loughborough have made.
Hannah: Yeah, perfect. Thank you. So moving on now, we've got about just under minutes perhaps left, and we hopefully will get on to some questions at the end. So, in terms of the most beneficial actions, I wondered from the project and its findings, could you expand a bit more on the lessons learned and the cost benefit of retrofitting solid walled homes, the most beneficial actions.
David: Well, the most beneficial thing to do is to do really good surveys. We want more good surveys, of course. But let's have a look. We didn't monitor homes. That wasn't part of the scope. But we have modeled the improvement of the homes, and we've calibrated those models, so hopefully they're more useful than you might think. So, this is a graphic which shows the fuel bill savings of each of the different retrofits installed in each of the different homes. Those code names there, 17BG, etcetera, equals a home. And again, it's a similar story really. What the take-home from... I won't talk about the colors of the graph bars yet, but really the story there is, look at the right-hand side, the solid wall insulation measures, where you do multiple measures. Those are the things that are going to get you kind of major fuel bill savings.
All of the other measures, regardless of what they were, are kind of still beneficial, but sort of down to about 5% reductions. The green bars, they're EPCs. And what you can see specifically for the room and roof, but some of the others too, they're really bad at predicting what the benefit is going to be. So, hopefully the new SAP update has actually integrated some... Well, this kind of validates some of the changes that are happening. That won't happen. There's different room and roof assumptions in EPCs in future versions. and the other colors are the DSM, The dynamic model is the orange, or the BREDEM, and the more advanced EPC is the blue. And they're just a little bit more accurate, that's all, really. So yeah, a little bit of overprediction in EPCs, but most beneficial is solid wall insulation for money.
Money isn't everything. People care about Their EPC bands. Solid walls, again, were the only measure that would get a band C, which people- I know is a very important thing for a lot of landlords too. Couple of other things to show. Another way of looking at cost benefits is, how much do you spend versus how much it saved. And the really big caveat of, our costs were not representative, because we were a strange university project one-off. But comparing across the board, really the story of this graphic is just that there weren't any really big winners. If you just looked at EPCs, so just the predictions for EPCs, the green, you might think room and roof is above that average line. That means it's saving more bang for buck. Actually, when you put the real numbers in, it isn't. The only thing you could possibly say which has the opposite is that loft insulation, where these homes are already insulated in the lofts, actually going back and topping it up, or replacing it with, you know, taking out replacing it, is actually probably still one of the most cost effective measures that you can do.
So, really, the story is, there's no real clear winner. We really need to be not selling retrofits on energy saving. We want to be looking at the benefits it has and sense of reducing condensation risk, improve air quality, improving maybe the appearance, the condition, the maintenance, the longevity of the buildings, reducing overheating risk. These are the reasons that we think people should be pushing people to have retrofits installed. Yeah, that was about it.
Hannah: Yeah. I was going to say, there was another question about the metric for the percentage of savings somebody asked earlier, and it feeds into this. So, it's just the percentage difference. Is it based on EUI, which is energy use intensity? And then there was also questions about per year, so maybe it's just...
David: It was just literally the total predicted annual costs. So, not relative to meter squared or anything like that. And we use a percentage because it was at a time of high energy price instability, should we say. So, we kept consistent costs and just did a percentage. As you can see. Again, Roger mentioned it, cost wasn't the big driver, but it's obviously a really important question for people.
Hannah: So, let's go on to next steps, and then we'll have time for some questions at the end. I think there are some good ones to have a look at. So lastly, just a word on the highlights and the key gaps in knowledge in this area, a need for future research and tools. Anything that you have to say on that?
David: Yeah. So some of this is probably quite obvious, as it has come through. The defaults that we're using for U-values, we need more information. we need greater range of defaults to be used to give us better... That's been really useful to confirm that, and we hope some will come of it. Similarly, the default that we use in moisture modeling is also something that we would really like to see more information of. This gets less attention. U-Values is kind of the thing that people know and hear about. The moisture property is less well known, and much more difficult to actually do the testing for. We've actually got to take core samples. But we think it's really, really important for risk assessments. When we're doing risk assessments, a lot of our risk assessments are based on thresholds.
And if we know that we've got huge uncertainty in the model, the use of an arbitrary or binary threshold pass-fail, maybe that's not appropriate, and we need to introduce other things into our decision making than just what the model says, the condition of the House, how well it's being maintained, etc.. Again, for risk assessment, why do we use one year weather files when we get more information from ten? We think that was a really good contribution, and we hope that will start to become standardized. We didn't think it was a massive contribution, but confirming solid wall is actually really beneficial. Seems to be a message that's been incredibly well received, which we thought was just a confirmation of something, but people really didn't seem to know. And then some other things that are quite small, that are not dismissed, but just not really giving very much weight, like the loft tough-ups.
And we didn't talk about actually, but secondary glazing is in the report as well. These are actually pretty good, even if the models say they're not, or people think that they've already been covered. And the theme running across every single retrofit was, oh my gosh, homes have been messed around with, and really getting a handle on that maintenance and the complications caused by the specifics that are going on in your specific home is so important. And again, we know that there's a lot more awareness of that, which is good. And then from an LBU point of view, we're doing lots of these to try and get the results out. But what I haven't mentioned here, we're actually really doing a lot of work on making sure that the methods that we're using to collect data from homes are as good as they can be.
That occupies a lot of our time, as well as making sure the models are as good as they can be. We do a lot of work on that now. This is a really scientific investigation. Case studies are fantastic. They're really good at giving us information that we can apply, and raise questions that we can kind of discuss. But what's really a big gap is that the majority, 90% of retrofits taking place probably have almost no evaluation whatsoever. And so, we think there's a really big gap there. We need to understand how we can get good data from every retrofit taking place, not just loads of data from a few retrofits. And then, most of you I hope will have noticed there's no people mentioned. People live in homes. The project of DEEP wasn't focusing on people and their lives, but we have a lot of research taking place around why people maintain their buildings, don't maintain their buildings, will retrofit, or why they build, why they will install low carbon heating or not as well. So, that's the sort of stuff that we are moving on to from DEEP now.
Hannah: Thank you. And then just handing over to Roger in terms of next steps from DESNZ point of view, or research that you spoke earlier about, a couple of future research projects that I understand you've already actually undertaken. So, it would be good to hear about those.
Roger: Yeah. One thing will be to maximize DEEP through doing things like this today. So, thanks again to Hannah and Co. for inviting us. Yeah, there's a wealth of information there, so I encourage everyone to take a look at the report. Start with the synthesis reports, and then more detailed reports into each individual work package that Dave and team carried out. And then yeah, as far as future, we have actually already commissioned and mostly completed two bits of follow-on research, one of which I mentioned in the chat already, which is on the air tightness of the UK housing stock, Which is carried our by Loughborough University, who also had a role in DEEP. The other is on deterioration of retrofit measures and their longevity, which has also been carried out by Leeds Beckett Uni, Dave and his colleagues. And both of those will be published... Well, fairly early -- we don't want to commit to anything -- in the new year. And we'll also be doing a follow-on research program for the next spending review period, which means from April next year, over the coming months as well. So, we'll engage with experts in the sector as we put that program together.
Hannah: Perfect. Thank you. That's really interesting. I think there is another format of screen that Matt has for us to then move on to answering some of the questions. So, I don't know if Matt like to do that. Here we are. And then we can go through some of the questions I've picked out some that I think are interesting, just a couple on the modeling, first of all. So, there's a couple of questions on, is there any particular reason why the dynamic model was used over PHPP, and then also why in terms of future modeling, you'll be looking at improving HEM non PHPP. So I think that's a good one.
David: Yeah, really good. Thanks for that. So, we need a dynamic because it's a different type of model. PHPP is still a steady state model. So, it's equivalencies there we wanted to compare against. However we have built PHPP models. We are in the process of building them. And the PHPP...
Hannah: David, can I just jump in? Before we've done it again with acronyms, it's Passive House Planning Package.
David: And Home Energy Model. There we go. Home Energy Model, let's start from the beginning, is a new version of SAP. The DESNZ, or the government is going to be producing what is actually a dynamic model as well. So that will be replacing what we currently know as SAP. Maybe. And that will be introduced in retrofits as well as new builds. PHPP, Passive House Planning Package is what is required for compliance under Passive House, and is... I'll get shot if I say this. It's the same as SAP, but has more inputs and is more detailed. And actually, the results that we're getting when we're comparing when we upgrade the SAP model in BREDEM, it's actually really similar to PHPP. Which you kind of hope it should, because it's physics, and they should have the same answer. They kind of do. So, we'll end up with the real measured values, the EPC value, the updated BREDEM value, the PHPP value, and maybe the HEM value. We'll be able to compare how all of these different models compare against the actual measured values. So, I hope that's... That's to come. It wasn't part of DEEP, but hopefully that will be a paper that we put out quite soon.
Hannah: Brilliant. Thank you. And SAP is Standard Assessment Procedure, for those of you that don't know. And RdSAP is Reduced Standard Assessment Procedure. One is applicable to new build houses and commercial, and one is applicable to existing buildings, and they're just different levels of detail or not detail analysis of your buildings to begin with. So, just answering some questions on insulation, just to clear up a few misunderstandings and questions that have come up. So, in terms of the room and roof insulation, I wondered if you saw any trends on the type of insulation that we used, because of course, you said this reduces overheating risk, and I think we are seeing that there is a difference in that level of reduction depending on whether you use the man-made polyurethane type insulations versus the natural fiber insulations that tend to have a better what's called decrement delay factor. So, ability to stop the transfer of solar radiation on the outside surface to the inside room, and therefore delay that heat gain. Was there anything in your modeling that kind of matched with that?
David: That's really interesting. We didn't... We only installed what the landlord's wanted us to install, and interestingly, none of them wanted any foam products. I think we were close to Grenfell at that point. Maybe there's other reasons for breathability that people on this call might favor, but I don't think that was necessarily the thing that was driving it. So, we haven't got the value for overheating using the two comparisons in the same model. It could be done. However, the major overheating mitigation benefit of the insulation is its U-value, not the other things which are beneficial in terms of those dynamic effects. So, I would still expect a very similar trend regardless of insulation type, but we need to do more looking.
Hannah: Okay, perfect. I mean, on the same sort of ilk, there was a question about the project showing if there was much better benefit in thermal mass or inertia. Let me read that again. As the project... Has the project shown how much benefit in terms of thermal mass or inertia gives to energy efficiency and thermal comfort or danger of overheating?
David: Yeah, It's really interesting. Yeah. And this gets back to whether you Should you put your insulation on the inside or outside, and exposure thermal mass, and all that sort of good stuff. So, we mentioned at the very beginning we didn't have a monitoring program, so we had to explore that question of whether or not you get the inertia, and when you have multiple heat waves, multiple days during a heat wave, are you able to kind of get rid of the heat? You would need to do longitudinal monitoring, which we couldn't do. The models that we're using and the measurements that we were using were steady state. So, we don't have dynamic effects in those that we could investigate.
Hannah: Okay. So that's maybe an area of further research. It's one that we in our team are constantly talking about, And are very interested in.
David: Okay. The DSM overheating model checks that we did do included that, but because we didn't design the experiment to specifically look at high versus low thermal mass buildings and their propensity to overheat, we did just didn't have enough of a big sample to explain. Even if one of them happened to overheat less, we would need to do more work on that.
Hannah: Yeah. Okay, brilliant. And there was another question about IWI with or without a gap behind the insulation and any findings. I presume it probably wasn't something that you necessarily looked into.
David: Yeah. By this point, the gaps behind insulation was already a no-no. Best practice guides and all that.
Hannah: So I was going to expand on that for the audience and the person that put the question. It is best practice to fully bond your internal wall insulation to your substrate to avoid that air gap between the two layers to avoid interstitial condensation, because that's what we're finding is a problem a lot, that you get hidden moisture accumulation and mold growth that people don't know about, and that's what, with internal wall insulation, is one of the bigger risks. There is one exception to that rule, but that is if you're in a very, very exposed location, and your external wall is wet, and the only option you have is to put a gap behind it. But then that gap should be ventilated to the exterior and absolutely sealed from the interior of your building. So, you're essentially looking at a box in a box solution then, which is not one that, you know, often is...
David: It's really hard to integrate into multiple stories and adjoining rooms and all that sort of stuff. So very, very tricky to achieve in practice. What we did observe, though, was dot and dab. This is introducing a gap. People might know what that is, a very thin polystyrene back plasterboard, kind of 20 mil thick randomly in lots of the bedrooms, or one of the rooms in the homes that the landlords didn't even know about. This is where the survey is really very important. If we hadn't checked, we'd have just gone and installed our IWI over the top of that, and of course, compounded some kind of a problem that might have already existed. So, we had to remove that first, and then apply our solution.
Hannah: Excellent. I'm conscious of time, so I think we've got some other questions coming, but we've managed it in time, so that's really good. There's one more about how HE feels about external wall insulation. I just wanted to kind of cover that quickly, that of course, technically, external wall insulation is your most robust way to insulate your building, because you can wrap it in a nice little coat on the outside, and your details are much easier to deal with. Obviously, in a heritage setting that isn't always possible or appropriate. And that's when you start looking at internal wall insulation being an option, if your risk assessment suggests that it's appropriate and you're using the right kind of materials that are vapor open and... So, it's not just vapor open but liquid open too, so that they allow both types of moisture to move through the materials to stop any problems that you might see. So yeah, technically, EWI is the most robust, less risky option, but we appreciate that it is not always possible.
David: There was one question which I think I could cover really, really quickly. It was, what was the IWI value that we went for? And I think this is really important. there is no single IWI value. That's really important to say. It is bespoke to the specific building. One of them, for instance, I think it was about 0.5 or something, or 0.6, so it didn't go down so far. It could have gone down further, but this time with back to back, No through ventilation, and likely to be multi occupancy space. So, the designer raised the threshold and said, actually, no, we're going to have less energy savings to have a more safe approach. So, there is not a single IWI value to go for. You have to check out and take into consideration the things that the models don't take into consideration.
Hannah: Yeah, absolutely. And there's the things about just the HEAN opposing IWI wall insulation. Just to say that it's not an outright opposition. It never is. There's a difference between what in policy terms you could do and what technically you should do in your building. And that is the distinction between those two things. And if you could do it in policy terms, you then need to look if you should do it technically, and whether it's got low enough risk to your occupants and your building fabric, and whether it's got enough benefit and payback to then be appropriate for your building. So, you have to look at it in that holistic situation again. So, I think we're almost on time. We're 3 minutes over, but I think we've done very well to get through all of that. So, are there any last closing thoughts or comments from either of you, David or Roger? Otherwise, we will wrap it up there.
David: Lovely comment that we haven't got enough results. Yeah, we need more. We need more research funding into this, for sure. It's expensive business, sadly, and there's lots more to learn. So, thank you so much for having us.
Hannah: Brilliant. Thank you both for your time, and thank you to the audience very much for your participation as well. Just a last note from us in terms of Historic England's future work. So, we're doing a lot of literature reviews and reviewing the evidence base on insulating solid wall buildings at the minute, considering climate risks. And we're in the process of revising and upgrading our guidance accordingly. We've got various different research streams going on onto lots of things that we've spoken about today, and lots of gaps that people have asked questions about. So, do keep an eye out. Do keep looking at our nicely refreshed energy efficiency and retrofit pages, because we will be putting more up as and when we can. So, yeah, thank you all very much.

[END]