Episode 6

Air Tightness, Under-Slab Insulation and the Ongoing Performance Gap

9 April 2026 · 42 minutes · Hosted by Brian Haines, Matthew Graham and Andrew Hooper

Summary

Builders are already delivering air tightness well beyond what NatHERS assumes, and getting no credit for it. This episode asks what happens when the modelling catches up — and what it will mean on site when a house has to hit a tested number to get an occupancy permit.

The second half turns to under-slab insulation, where the team have opposing answers from two authorities on the same question. The ABCB told one of them it is not required; the Victorian regulator told the other that it is. The discussion traces how three other states quietly wrote the problem out of their own schedules, leaving Victoria alone with it, and what NCC 2025 does to fix the wording.

The episode closes with a run through recent questions from the NatHERS Assessor Network Facebook group, covering a difficult climate zone 3 project, determining true north when the title has no bearings, and a thousand dollar bounty that has gone unclaimed.

In this episode

• 0:16Air tightness in NatHERS, and what it opens the door to

• 1:13Does it need to be in the NCC first?

• 1:45The industry has already moved from twenty air changes to six

• 3:03Post-construction verification, and giving builders the credit

• 4:41How much a rating actually moves

• 5:04What happens if the finished house misses the number?

• 6:22Tolerance, and the learning curve for builders

• 8:41Under-slab insulation: 13.2.64, 65 and 66

• 10:43Two authorities, two opposite answers

• 13:02Who actually regulates, and who only writes the code

• 13:33How three states wrote the problem out, leaving Victoria alone

• 15:01The comma that decides it

• 17:27What NCC 2025 changes

• 19:10Waffle pods, custom builds and where it actually bites

• 21:03Weekend review of the assessor network

• 23:14A climate zone 3 project stuck at 6.2 stars

• 31:37Finding true north when the title has no bearings

• 33:04How much a single degree of orientation is worth

• 36:23Verification using reference buildings, and a thousand dollar bounty

Key points

The industry has already improved, quietly

Referencing a recent industry webinar, Andrew notes that where Australian housing was previously running at fifteen to twenty air changes per hour, it is now closer to eight to ten — and NatHERS is producing around six. The argument that follows is simple: if builders are already delivering that, why should they not get the credit for it?

NatHERS does not have to wait for the NCC

Brian's assumption was that air tightness could not be used for regulatory compliance until it appeared in the NCC. Andrew's position is that because NatHERS is performance based rather than prescriptive, it can bring in factors that are not in the code, as part of demonstrating the overall performance of the shell. You would still have to test it — a blower door and post-construction verification — but it becomes an optional path within an assessment, in the same way that choosing a custom window over a default is optional.

What it is worth

Matthew, who models passive house projects, runs them out of regulatory mode at six air changes per hour out of curiosity. The difference can be significant — as much as half a star — though he is careful to say it is highly design dependent. The corollary is that builders currently constructing at four to six air changes are being assessed as though they built at ten, and losing tenths of a star they have already earned.

The problem nobody has solved

If air tightness becomes a rated input, the house has to achieve the tested figure once it is sealed up. Miss it and you are potentially unable to obtain a certificate of occupancy on a finished building. The team's instinct is that testing would need to happen earlier, when the envelope is sealed but before internal linings go on, so there is a chance to correct — and that some tolerance would have to exist, in the way glazing carries a plus or minus five per cent substitution allowance. Nobody in the discussion is confident about how large that tolerance could safely be.

Two regulators, two answers, same clause

Brian was told by both the Victorian Building Authority and the ABCB, some years ago, that insulation under the slab was unambiguously required. Andrew was told by the ABCB more recently that it is definitively not, and that they do not consider slab edge insulation to wrap around to the underside — though they would not put it in writing. He then put the same question to the Victorian regulator and was told the opposite: that under-slab insulation is a requirement on the wording of Specification 42.

The ABCB is not the regulator

An important distinction the team draw: the ABCB can tell you the intent of the code because they wrote it, but they do not enforce it. Enforcement sits with the state — in Victoria the building regulator, elsewhere its equivalent. If the enforcing body says something must be done, then it must be done, regardless of what the code's authors intended.

Victoria is the last one holding this problem

New South Wales has BASIX, which overrides that part of Specification 42. South Australia rewrote its schedule specifically to exclude under-slab insulation. Western Australia instructed that climate zone 6 be treated as climate zone 5, which removes the question entirely. That leaves Victoria as the only jurisdiction still reading the clause as requiring it — a position that turns, as Andrew explains, on where a comma falls and whether the reference to slab edge insulation applies to both subclauses or only one.

NCC 2025 fixes the wording

The good news is that the clause has been renumbered and clarified in NCC 2025 to make clear it applies to slab edge insulation, using essentially the same approach South Australia took. Andrew emailed the Victorian regulator to ask whether this changes their position on NCC 2022 projects, or whether the industry will run two different requirements side by side, and had received no response at the time of recording. The team also note a likely consequence: builders with small single storey projects may start choosing to move forward to NCC 2025 rather than stay on 2022.

Where it actually matters

Most volume projects use waffle pod slabs, which generally do not need it. Custom builds are the opposite — Brian puts it at close to ninety per cent — because they carry more glass, less optimised orientation and a coupled slab that is hard to get to seven stars without it. So the daily push-back comes from volume builders, while the projects that genuinely need it are the custom ones.

One degree of orientation

On the question of determining true north where a title carries no bearings, Andrew's practice is to order the title himself if it is not supplied and bill it back — noting that in Victoria the Building Act requires bearings on drawings, so it is rare not to have them, but that is not the case in every state. His advice is blunt: never trace a north arrow off a plan. On why it matters, Brian describes a file where a single degree of rotation moved the result by half a star, though he is clear that was an error in the file rather than normal behaviour. Matthew's real-world figure is that five degrees is worth around a tenth of a star — which can still be the difference between double glazed and double glazed thermally broken, and a significant cost to the client.

Documents referenced

• National Construction CodeAustralian Building Codes Board — the code itself, and the Housing Provisions referenced throughout.

• NatHERS publicationsTechnical Note, assessor handbook and scheme guidance material.

• Building and Plumbing Commission VictoriaPublisher of the Victorian practice notes, including EE-04, and of Regulation 233 guidance.

Transcript

Lightly edited for readability. Hesitations and repetition have been removed and technical terms corrected; the substance is unchanged. This episode opens mid-conversation, continuing from Episode 5.

0:05It depends a little on the NatHERS software tool, but I agree, Andrew — you have to start somewhere. I was quite excited when air tightness was introduced into NatHERS, because in the same way, would I have liked more? Sure. But we have more coming, and it paves the way for heat recovery to be introduced once we have air tightness in. So it does need to start somewhere, and hopefully it evolves and grows over time.

0:46But are we not just waiting for air tightness to be in the National Construction Code before we can use it within NatHERS software for regulatory compliance? There was a bit of that in NCC 2019 for one of the verification requirements, where you could use blower door testing, but we are waiting for it to move into Part 13 itself. How long before something like that would be available?

1:13I do not believe it is necessary. Again, we are talking performance based versus prescriptive based. NatHERS is performance based, so it can bring in other factors that are not in the NCC, because it becomes part of the big picture of the performance of the shell.

1:36I like that idea, but that is not the way it was explained to me. I certainly do like it though.

1:45This is my understanding. If NatHERS brings in the air tightness requirements — and there was a webinar recently talking about how the building industry has already shifted. We have improved air tightness over the past ten years. Where we were previously at fifteen to twenty air changes an hour, we are now down to around eight to ten, and in NatHERS we are producing around six. So if builders are already doing it, why should they not get the credit for it? That is where performance based comes into play, because we can demonstrate that the performance has improved within the NatHERS scheme.

2:40You are still going to have to test it. You will still have your blower door and all that sort of thing, but it becomes part of an assessment. If you want to do that, these are the steps you follow. If you do not, you go the other way and these are the steps you follow. Do you use a default window or a custom window? You are not required either way, you can choose.

3:03We can do post-construction verification to determine the air changes per hour of the actual house, and adjust that figure within the NatHERS software accordingly. Builders are being penalised at the moment because we are assuming ten, but building at around six. If you build at four, or at three, you are losing points of a star when it is calculated.

3:49I feel like the ten is a bit of an old wives' tale. It has just been a variable air change rate — it varies. It is that it has not been exposed in the NatHERS software, and now it is, so we can see what that value is. For the passive house projects I work on, where I am doing the NatHERS component mainly for regulatory compliance, out of curiosity I will take it out of regulatory mode, set it to six air changes per hour, run it and see what difference it makes. It can be quite significant — maybe half a star.

4:41But it is very house and design dependent, because you can get half a star out of two air changes per hour depending on the design.

5:04What becomes interesting is if you have definable levels of air tightness and the project does not achieve the number. Once it is built and sealed up and you have tested it, that is difficult. What are you going to do — keep going until you get there? You cannot get your certificate of occupancy until you manage to hit the magic number. That is always going to be a difficulty with something so prescriptive.

5:42One would think you would test when the envelope was sealed, before you put internal linings on, just to see where you were, and then you would know roughly what was going to happen after that. It opens a Pandora's box: if the house is complete and does not meet regulatory compliance, you cannot get an occupancy permit. There would have to be some leeway in those calculations.

6:22How much leeway could you have? With windows, for argument's sake, we have plus or minus five per cent when substituting something in. If you applied a similar allowance to an air change and you were targeting six, five per cent is not a big difference — you are not talking one air change an hour. If you are targeting six you are going to have to be pretty close to six. If it looks like you will be off, you might have to start talking to your assessor about options.

7:14It would be a learning curve for builders. There will be builders all over this who know exactly what to target and how to do it, and others who say we have not done this before, we are going to learn from it and we might struggle, but the next house we will be better. It is not going to change overnight, and we need to see what clauses they put in about how it will work.

8:03These are easy wins we are not currently getting. A little easier to achieve seven stars if the air tightness is more secure. If we are already building at four to six, and that is worth two or three tenths of a star, why would we throw that away? Though we would have to go through that testing phase, which is time, money and effort.

8:41Let us move on to under-slab insulation in NCC 2022, specifically 13.2.64, 65 and 66. Many years ago — it was the day you started on LinkedIn, Matthew, because you responded to my article telling me how wrong I was. I had to join LinkedIn. I think that is my only post. You have made two, and they were both directly relating to my article.

9:51So, 13.2.65 specifically, where it talks about the requirement for a minimum under-slab insulation level of R0.6 in NCC climate zones 6 and 7 — slightly different for 8, but we can talk about that separately. And 13.2.64, floor edge insulation, talking about slab edge perimeter insulation, again at a minimum R0.6 at the time.

10:30At the time I had spoken to the Victorian regulator and to the Australian Building Codes Board, and both gave me the same answer: it was one hundred per cent required underneath the slab. I believe you went through the same process, Andrew, and got similar answers.

10:49The ABCB rang me just a few months ago, in December, and said no, absolutely, it does not need to be applied under the slab, and that they do not consider the edge slab as wrapping around to the underside. So they were saying the edge is one thing and the under is separate, and the under is not required. Of course they will never put anything in writing. I asked whether we could get that in writing and they said no, but that if anyone wants to email them they will get back to them. It took them two months to get back to me.

11:34That was the ABCB's comment. I posted the exact same question to the Victorian regulator and they said yes, absolutely, the under-slab insulation is a requirement according to the wording of Specification 42. So two different responses.

12:22Regardless of what the ABCB have said to either of us, I do not want to be the one who leaves the under-slab insulation out of the NatHERS report and then has the regulator or the building surveyor come back and say, did you forget this clause in the NCC, because it needed under-slab insulation and you did not do it — and now the slab is poured, and now we have a big problem.

13:02I will point out that the ABCB is not a regulator. They can say what the intent of the building code is, as they wrote it, but it is the state regulator who enforces it according to their standards. From that point of view, if they say something has to be done, then it should be done, and that is the same in any jurisdiction.

13:33We talked last week about the differences in the jurisdictions. If we look at climate zones 6 and 7, New South Wales has BASIX, so it overrules that Specification 42 component. South Australia reworded theirs in their schedule to specifically exclude under-slab insulation for their jurisdiction. Western Australia changed theirs to say that if you see climate zone 6, treat it as climate zone 5, so they do not need to worry about it. That just leaves Victoria as the sole one reading it this way. So we are the only ones with this problem at the moment.

14:26So do we or do we not? NatHERS is performance based software, so by rights we should be able to give and take accordingly. However, Specification 42 does say 13.2.65, and it also carries over with the wording for slab edge. There is a comma between the two, and this is what the regulator said: because there is a comma here and not a comma there, if you refer it to this paragraph there is no comma, so they are mutually exclusive. This one is mutually inclusive. It is confusing.

15:21And 13.2.66 only talks about the requirements for slab edge insulation, so it would make sense that the comma is in that position where it says for slab edge. I do not think it does, because as you said, it only refers to that. So why say refer to this for slab insulation, for slab edge insulation? You do not double up on what you are saying. It says for slab edge insulation, so I look at that and say only that part is relevant.

16:11We can debate the way it should have been written, but it is also written more clearly in the verification method. It is actually written the way I think it should have been written there, believe it or not.

16:29We all know what it was meant to be, and we all know how it is, and that is the difficulty. The intent probably was not how it has been interpreted by Victoria. However, who wants to be the person who does not do it and then gets caught out, because it will happen? I cannot say I have — certainly before I spoke to the regulator there were jobs where we got away without the slab insulation and did not put it under there, particularly when the builders and clients did not want it, because it causes problems on site.

17:13The simple fact in Victoria is, how many jobs need it? Most, if you are not doing a waffle pod. It helps a lot.

17:27So what is happening with NCC 2025? The good news for Victoria is they changed the wording. They have clarified with the extensions of the letters and subclauses that it is just the edge, removing the under-slab — basically exactly the same wording South Australia changed theirs to. So it has taken out the confusion and said it is for slab edge insulation. Consequently I emailed the Victorian regulator again and asked whether this changes their policy — are we going to have 2022 projects and 2025 projects as two different requirements? Unfortunately in the past two months there has been no response.

18:24But we can assume it is probably going to be cleared up for 2025. And then, barring any other significant change for Class 1 in NCC 2025, I could see people moving to NCC 2025 rather than 2022 — for example if they have a small single storey house that does not necessarily need under-slab insulation to achieve seven stars. Where everyone was trying to drag it out as long as possible before, now you might find they want to jump forward.

19:10I personally feel projects with concrete slabs on ground are likely to have rigid insulation both under and at the slab edge. Projects with waffle pod slabs are unlikely to require them or have them. They are just different types of project.

19:40We both get push-back every single day about under-slab insulation, purely because most volume build projects are waffle pod. But any custom build project we do, it is almost ninety per cent. Custom builds have greater glass, the arrangements are greater, and they are generally not optimised to the site, so it is harder to get a custom job through without the under-slab. You are going to struggle to get to seven stars with a coupled raft slab.

20:36Do we want to talk about what constitutes a waffle pod, or is that another show? I reckon that is for another episode.

21:03So I thought we would do a brief weekend review of the NatHERS Assessor Network Facebook group, which as of this episode has more than 1,600 members. For those unfamiliar, it was created by Michael Young from Transformed, a well known registered training organisation, and it is moderated by these two random guys who try to keep the group on the straight and narrow.

21:40This week saw the launch of the first Rating Room podcast. A big shout out to everybody who has watched it. We are just starting out, and did we have a learning journey — so many technical difficulties. The microphones did not work properly because the gain was too low. The main camera was only recording in 1080p instead of 4K. The next attempt, the camera overheated and shut down, and we did not realise until an hour and fifty in. But we are determined, and the positive feedback over the last week has been motivating. We will need a little time to iron out the kinks, but hopefully in the next couple of episodes we will feel confident enough to get some guests along.

23:14I have cherry-picked a few posts from the group over the last week. The first is from Calvin Groen, who asked whether anybody has suggestions on what design changes will significantly improve a rating in climate zone 3. He has one sitting at 6.2 stars, pretty much maxed out on insulation and glazing, and the only way he can get it across is to change from a suspended floor to a concrete slab, which then complies easily. A friendly reminder to everybody: if you are genuinely seeking feedback, include as much information as possible — floor plans, elevations, whatever describes the problem — so the group can give a detailed response.

24:13Andrew, you followed up on that one. I did. I got hold of the floor plan and quickly rated it up. It was a very simple house, so it was not done accurately on my behalf, just to see how it stacked up, and sure enough I found exactly the same thing. In climate zone 3 it was very difficult. The advantage for him was that it was in Queensland, so as we talked about with the jurisdictional differences, he could apply the ceiling fans and outdoor living area and get to the equivalent of seven stars. The big question is what you would do if it was in Western Australia, which has climate zone 3 without those credits.

25:13There were a few good solutions that came up, such as changing the floor to something with an amount of thermal mass, which would help. What sort of increase were you getting from that? On other projects I have had two, three or four tenths of a star. That is about right — like everything, climate specific. I have not seen up to four tenths, but to be fair a lot of my testing was in a temperate climate in New South Wales. It is design specific, and it is not a standalone solution — it is part of the jigsaw.

27:21Thanks to Calvin. I contacted him and asked whether he was okay with us sharing the post, and he said that was fine. That is the idea — that we can share some knowledge. We are definitely stronger working together than apart. Questions were asked, answers were found, and there was a result. I would encourage others to do the same, and I would love to see what the final solution was — if you ask the question, come back and say we ended up doing this, and that is how we got it over the line, so the next person benefits.

28:24The next one: the industry summit returns this year, bigger and more global, with speakers from around the world. Did either of you attend last year? Yes, very good. They ended up with something like fifteen thousand people online over the course of it, and a significant number from overseas universities. The quality of the speakers was great, international, multiple countries represented. I walked away with a lot, so I have signed up again. This year is a little different, with two streams running at the same time — I have gone the performance stream, the thermal side, but you can sign up for both and watch the recording of the other.

31:37The next one is from Gemma Shakespeare, asking for advice on accurately determining true north on projects where bearings are not shown on the certificate of title or deposited plan and no site survey has been done. She has a current project in Western Australia and is realising that many title plans only show internal angles between boundaries relative to adjoining boundaries, with no expressed starting meridian or datum reference and only a vague north arrow.

32:16I responded to that one somewhat jokingly at first and said if you want to get the project started, put it in Google Earth and Google Maps and at least you can start moving. But I would throw it straight back to the building designer to work out where north exactly is, because you can get a significant difference if you are a few degrees out.

33:04A couple of months ago I got it straight from CSIRO. We had a project at a specific orientation, and if it moved one degree in either direction there was a half star difference. Now, they had a specific term for what the issue was with the file — it was an error — but it shows that if you are one degree out you can have significant problems. That was an error, but the point is you need to be right. The most I have seen is maybe five degrees being a tenth of a star. And if we are talking a tenth of a star and you are struggling to get seven, it can be the difference between double glazed low E and double glazed low E thermally broken. That is a significant price difference for your client, so you really need to get this right.

34:00I consider myself fairly fortunate because pretty much all the work I have has some sort of title re-establishment, and I have all the title boundary bearings in degrees, minutes and seconds. Andrew, you have mentioned that on some of your projects you do not have that information, and I think you have a good solution.

34:33I tend to be a bit particular about a lot of things. I have it in my fee structure that if you do not provide me with the title with bearings, or the title information preferably on the drawings, I might order it from the titles office myself and include it as part of my fee, which they pay for later. That was not too bad when it was about thirty dollars — it is now maybe a hundred and six.

35:07In Victoria it is in the Building Act that you are required to have bearings on your drawings, so because I work mostly in Victoria we end up with them anyway. It is pretty rare not to, and most of the time if it is not there and I ask for it, I get it. That is not the same in every jurisdiction — South Australia you do not have to have them, and I do not think Western Australia does either. So clearly there is a problem. The technical note allows for internet sources to determine north, which then raises the question of how accurate you actually need to be if internet sources are good enough. It does not spell that out. I have only ever actually had to buy a title once, but the threat of it works — you generally explain to someone that if you give me this information I can give you the most accurate result. And I never trace a north arrow off plans. I just do not ever trust it.

36:23Then some fellow called Andrew Hooper asked whether anybody is successfully producing verification using reference building assessments, because he has a client wanting to consider them and wants samples modelled — and the person producing them would need to defend them when questions are asked. And then some other fellow called Brian Haines posted a thousand dollar bounty.

37:16I will point out that the request was legitimate. I have a client I have worked with for a number of years who wants to explore alternatives, and one that keeps coming up is verification using a reference building. I do not do it and I do not support it, but they want to explore it, so I am legitimately looking for someone who can produce one successfully and prove it.

37:49Brian would like to see the results of that, because most of those reports do not give data. They just give an outcome. So nobody can really prove whether it was done correctly. A lot of the responses were not positive — except the thousand dollar bounty, and I stand by that. If someone can do them, I would be very interested in talking to you.

38:35Is this another episode? It sounds like we have to dedicate one to it. I will get some people in who can talk about it in a lot more detail than we can, but from my understanding and research over a number of years, it is not real. And it is even worse in NCC 2022. If you are doing them, shame on you, and if you are accepting them, shame on you as well. All the liability sits with the certifier or the building surveyor. Try getting your professional indemnity insurance updated to cover them — no chance. This is a class action waiting to happen, coming closer every day with mandatory disclosure, and I think it will be a huge issue for the industry at some point.

39:52And lastly, someone launched the Forge program. It is really just my attempt to see whether I can craft some more elite subject matter experts. We have some and they are pretty good, but I feel there are not enough of us. We are much stronger working together than apart, and we need to connect our expertise and share our insights — which is what we are trying to do here in the rating room as well. So it is forging us in a way, to push forward and drive continuous growth. It does not matter whether you are a proven presenter, an experienced assessor, or just somebody who is really driven and knows where they want to be but does not know how to get there. Hit me up on the NatHERS Assessor Network Facebook group.