I recently watched this: https://www.youtube.com/watch?v=sqmDW1OpwKM , which sort of confirmed a suspicion that I already had - glass wool insulation is a lot less effective in some applications than the raw numbers would have you believe.

But this got me thinking, does the same apply to mineral wool products? These micro-convective losses are something that doesn't seem to get talked about much - people just seem to quote U/R-values and leave it at that. Obviously mineral wool boards tend to be denser than fibreglass batts, but I guess they're still considerably more gas permeable than e.g. EPS.

Does anyone have any concrete info on this? I'm looking at products for EWI and I don't want to make an expensive mistake - particularly given that EPS seems like it's by far the cheapest board-type insulation available to me. From an idealistic "green" perspective I'd rather not use a petrochemical product, but from a pragmatic point of view, I want something that is effective at a sensible price point, and savings will only be reinvested in various other efficiency/sustainability projects. If I can do more for less, it may well be that spending the money elsewhere has a net positive impact that will outweigh the downsides of using EPS (which at least has a *relatively* low carbon footprint by comparison to some of the other synthetic materials)

The purpose of mineral wool insulation is to reduce convection heat loss and this is already taken into account in the quoted thermal conductivity. However, mineral wool is more vulnerable than rigid insulation boards to so called "wind washing" where bulk air movements penetrate into the quilt and increase the heat loss. To get the advertised thermal conductivity you need to make sure the mineral wool is contained in a box with a wind barrier on the outside and the air barrier immediately on the inside. Both wind barrier and air barrier should be sealed at junctions with other materials to prevent thermal bypass.

The video compares mineral wool insulation site fitted to a timber frame with a SIP panel where the insulation is injected between two OSB sheets in a factory environment. If all SIP panel joints are sealed with polyurethane foam, a breather membrane is fitted directly to the outer face and an air barrer membrane to the inner face then there is no risk of wind washing, the risk of thermal bypass is greatly reduced and the airtightness is likely to be very good. So its not surprising that the SIP house performed better when built by a typical builder.

The same results can be achieved with mineral wool, but you need to make sure there is a sarking/sheathing board & membrane on the outside, an airtight membrane on the inside, all insulation is tightly fitted and all junctions are correctly detailed.

The same applies to EWI construction. Whether you use mineral wool or rigid insulation boards, you need to be particularly careful that there are no voids between the outer face of the masonry & the inner face of the EWI. This is typically done by parge coating the outer face of the masonry with a thin render or by covering the masonry with gap filling adhestive. In both cases, a render applied directly to the insulation is likely to perform better than a rain screen with an ventilated gap between it & the EWI.

David

Posted By: davidfreeboroughIn both cases, a render applied directly to the insulation is likely to perform better than a rain screen with an ventilated gap between it & the EWI.

Doesn't that depend on the insulant and the method of installation David?

Imagine a Larssen truss boxed out, filled with insulant (cellulose, EPS bead, whatever). Surely the boxing stops the wind washing.

Posted By: Ed DaviesThat Canadian experiment is interesting but without knowing how the houses were used it's impossible to say if that 50% energy difference was significant. E.g., maybe they should get the occupants to swap houses for the following year and see if the difference stays the same.

The other major thing that went unmentioned was the relative airtightness of the two houses. This is not related to the insulation and could easily result in a 50% energy consumption difference.

David

Posted By: Ed DaviesAny suggestions as to how you'd actually do that?

http://www.npl.co.uk/reference/faqs/how-can-i-measure-the-thermal-conductivity-of-my-insulation-material-and-what-sort-of-accuracy-can-i-expect-%28faq-thermal%29

This FAQ seems to answer at least the definition but not Ed's question:
http://www.npl.co.uk/reference/faqs/which-is-the-appropriate-heat-transfer-property-to-quote-for-my-construction-product-or-structure-(faq-thermal)

Note that the GBF software removes the end of the link so you'll need to cut and paste the two parts of the link together before trying to access the page.