Let's Learn about... Insulation (Part 2)
A deep dive involving maths. Because cosiest doesn't necessarily mean thickest
Part of my Cleanest Greenest Home Series: demystifying eco upgrades, healthier materials, and smarter ways to live well inside the homes we already have.
When you think of doing up your home you probably don’t imagine yourself sat at your laptop with a calculator trying to work out the insulating-ness of various combinations and thicknesses of materials. The thing is, having done extensive research now into the best aka cleanest greenest materials to use for phase one of my retrofit, now I need to work out exactly how much cosier I want to aim for.
But first let’s go back a step. Upstairs, my house has little more between me and the elements than a sheet of standard Gypsum plasterboard and some derisory fluff of decades old insulation, most of which appears to have slumped behind the boards to the floor ergo combined thermal/acoustic resistance negligible. Then it’s just the tiles hung on the outside with bitumen sheeting on the inner face.
Downstairs, it’s Gypsum plaster straight onto a single skin of brick. Net effect, although the house heats up a treat, when the boiler stops pumping, all that lovely warmth disappears pretty pronto.
Phase one then is to strip mine and small’s bedrooms back to the bones, as well as a secondary main living space downstairs. I’ve picked these rooms because the latter was easy to empty, and the rooms in which we sleep will always be my priority. But also, I want to protect the building. The whole point of this cleanest greenest project is to retrofit using breathable assemblies and sustainable materials.
So back to the calculator. Simply put, the cosier I want the house to become, the thicker the new insulation has to be, surely?
Well that’s what I thought.
Quick explainer: U value is the standard measurement of insulating-ness defined as watts (heat energy) per m² per degree of temperature difference, written as Wm².K. Ie how many watts of heat escape through one square metre of wall for every degree difference between inside and outside. Think of a high U-value wall as being like wearing a T-shirt in winter, the heat rushes out. Whereas a low U-value wall (what we’re aiming for) is like wearing a big fluffy coat ergo heat stays in. You’ll sometimes hear materials described in terms of their R-value too. This is the inverse of U-value, it denotes a materials’ thermal resistance, ergo how hard it works to keep heat in. If U-value is “how quickly your house loses heat,
R-value is “how hard your wall works to keep heat in.
Current UK building regulations stipulate certain requirements for new builds when it comes to heat loss (Approved Document L is the primary advisory document for understanding this). These regs state that the insulation factor for a standard new build wall (ie assuming cavity, sheathing, plasterboard, air layers, etc.) should provide a U value ≤ 0.18 W/m²·K. (For roofs it’s maximum U-value of 0.11 W/m²·K, and for floors maximum U-value of 0.13 W/m²·K). To get there with conventional insulation (mineral wool, PIR boards), this might take 10–15 cm of material — but at the cost of breathability and potential chemical exposure.
For a renovated solid wall, the target U-Value is cited as 0.30 W/m²K if you’re renovating >50% of an individual thermal element, though there is a threshold value of 0.70 W/m²K for internal insulation on solid walls when it is not technically or functionally feasible to achieve the target U-Value. Ie because you’d need such ridiculously deep linings that your rooms would become a lot smaller. It’s generally reasonable to accept up to 5% reduction in a room when insulating internally.
So… having narrowed down my material choices, and having laboriously read through all the technical specs, in order to get a starting point on the thickness of materials required, I chucked the whole lot (data sheets, Approved Doc L, the works) into Chat GPT to see what it said. It was most illuminating…
In short, you work the thickness out backwards from your U-value target (and any practical constraints like lost floor area), using a basic thermal resistance sum.
A) Set target U (in my case 0.30 W/m²K because it’s a retrofit).
B) Convert to target R value = 1/0.30 = 3.33 m²K/W (see explainer above)
C) Estimate R of what’s already there (brick, any air layers, sheathing, etc.)
D) Whatever resistance you’re short becomes the insulation requirement:
Factoring in the data I had from the materials I’m thinking of using (separate post to come) and remembering that plaster doesn’t move the thermal needle much this is more or less the result…
Option A (compliance-leaning): plan on roughly 140–160 mm of insulation to get near 0.30 on any single-skin/solid walls. The material I was thinking of using tops out at 100 mm per board, so this would mean doubling up (e.g., 60+80, 80+80, etc.) and detailing fixings accordingly.
Option B (space-leaning): cap at 80–100 mm and accept I may not reach 0.30, but I would still show a very substantial improvement. However upstairs, I was thinking to bias thicker regardless because the timber frame will create repeating thermal bridges (repeating cold spots as not covered by the insulation), so the “between-stud” insulation can’t fully deliver its headline R in the finished wall.
But here’s what I didn’t fully understand when I started running the numbers. Insulation doesn’t just change heat flow. It changes moisture behaviour. And moisture is where things get complicated.
In technical terms, this is called hygrothermal behaviour — the way heat and moisture move together through a building over time. And the key fact? When you add insulation internally, you don’t just keep the room warmer. You also make the original wall behind it colder. The colder it becomes, the more likely it is to sit below dew point meaning that moisture that once harmlessly evaporated could now linger. In the worst cases, it accumulates. And that’s when you get mould, decay, freeze–thaw damage in masonry, or timber rot where you least want it.
Let's learn about... Lime Plaster
Part of my Cleanest Greenest Home Series: demystifying eco upgrades, healthier materials, and smarter ways to live well inside the homes we already have.
This is why, when I spoke to Peter, a technical manager at Mike Wye, a leading supplier of eco building materials, and where I’m buying mine from, he gently suggested that once you move beyond slim internal insulation — say 40–60mm — you’re no longer just chasing energy savings, you’re altering the building’s moisture regime. What you’re after, he explained, is the “sweet spot” of enough insulation to increase comfort and lower bills, while protecting the building fabric and not losing too much floor space. In other words, heat loss is only one part of the story.
This was sobering. I had been so fixated on 0.30. In reality, retrofit is a negotiation.
He then sent me an excellent introduction to natural fibre insulation (which I’ve attached below so you can download it too) which reframed the entire exercise.
Most striking was the graph (copied below) showing the diminishing returns for internal wall insulation on a solid wall. It shows that the majority of energy savings are achieved within the first 60mm of insulation depth. Beyond that, the curve flattens. Yes, you can keep shaving off incremental heat loss. But each additional centimetre increases condensation risk while delivering proportionally less benefit.
This is the “sweet spot” Peter referred to. Not the thickest wall you can physically build. Not the lowest possible U-value. But the point at which thermal improvement, moisture safety, embodied carbon, space loss and practicality are most effectively balanced.
Because what I was doing, initially, was optimising one variable in pursuit of the cosiest possible house thinking this was about hitting 0.30. But buildings, specifically old ones, are systems. You cannot aggressively improve one metric without affecting others. Add too little insulation and you miss out on comfort and energy savings. Add too much internally and you risk trapping moisture in a structure that was originally designed to breathe.
This is why vapour-open, moisture-buffering materials matter. Natural fibre insulation allows moisture to move safely through the fabric rather than trapping it. Lime plaster helps walls dry in both directions. Ventilation becomes more critical as airtightness improves. Everything interacts. And so the question shifts. It’s no longer: how thick can I go? It becomes: what thickness allows meaningful thermal upgrade without destabilising the wall?
Upstairs, where the construction is timber-framed behind tile hanging, that might mean 100mm of insulation plus a continuous internal woodfibre board to reduce thermal bridging — but only after modelling confirms the moisture profile remains safe.
Downstairs, on single skin brick, it may mean being more conservative. Perhaps accepting a U-value that doesn’t quite scrape 0.30, but sits comfortably within the allowable zone, while protecting the building fabric for the next 100 years.
This is the part of retrofit that doesn’t make for glamorous before-and-after shots. It’s slower. It’s more nuanced. It involves spreadsheets, dew points and phone calls to technical experts. But it’s also where the real work happens. Because a healthier home is achieved by understanding how the whole system behaves — in winter and summer, in damp and in drought, in theory and in practice — and finding the point where warmth, dryness and durability coexist.
That, it turns out, is the real definition of cosy.
And if you enjoyed this post then please do hit the ❤️ button - it shares the love by making it more visible to non-subscribers. Thank you!
BTW I have a new website that brings together all the varied facets of my world in one place — the books, the Club, my collaborations, my Amazon Shop — all now easily accessible via a single URL. Check it out! michelleogundehin.com
Thank you for your generosity in sharing your extensive research and downloadable data.