Get therm and run some numbers. Remember your neighbour should also be heating their house to a similar temperature, and so not all of the heat that enters the wall goes over to their side and into their house. The main difference will be the outside of their masonry wall will be cold. Yours will be warm (behind the insulation). So at the interface with your wall and their wall there will be a temperature gradent and that is where the heat loss will be. (I think!)

Hi ikimiki

I'm not a Therm user but am wondering why you want to quantify the losses when there's not much you can do about them?

You could get a reasonable idea by just doing a steady state calculation. For the bridge concerned use the cross sectional area (by far the worst case)

Q = UA delta T

Heat Loss = U-value x surface area x temperature difference between inside and out.

Posted By: Mike GeorgeHeat Loss = U-value x surface area x temperature difference between inside and out.

U and surface area of what, given that it's a complex junction with oblique/curved heat flow?

Posted By: Mike George
''Heat Loss = U-value x surface area x temperature difference between inside and out.''

FT wrote:
''U and surface area of what, given that it's a complex junction with oblique/curved heat flow?''

To some extent I am with Mike, in that if the bridge simply cannot be avoided, don't spend too much time worrying about it, but I also sympathise with those who wish to know exactly what they are losing and why.

Fostertom's point - what is the surface area you are measuring? - is apt, bcause who is to say where the thermal bridge starts? At what point back down the party wall can a thermal bridge be said to be, or not to be, in existence?

If you were doing IWI you would return the insulation at internal and party walls, by roughly the length of a piece of string, to ''cloak'' the thermal bridge. Strictly there iss no reason why you should not do this (IWI to cloak the bridge) as an 'improvement' to the (uncloaked) thermal bridge with EWI. To me the only issue (apart from cost and disruption) is whether it can be done without looking stupid.

As a general point (not specifically related to Ikimiki's Q), in terms of other uncloaked thermal bridges, check carefully the situation where capping details are used (where, for example, there is no oversail at verge or eaves, and the insulation is capped with a plastic or aluminium moulding). Has this left any uninsulated areas within heated rooms? If so, what do you do? IWI above the picture rail? As above, is it feasible? Will it look stupid?!

Thanks for weighing in on this gents.

I am somewhat reassured by the general tenor of the comments here -- namely that it's not worth worrying about in the greater scheme of things, not least of all because there is very little that can be done about it.

...Except... ...if it is not a minor efficiency loss, then lack of co-ordination between semi units might turn out to be a factor holding back the rollout of EWI across the solid-walled semi housing stock.

The effect would be similar in nature to the "slumlord's dilemma". Two neighbouring slum highrises are owned by different slumlords. Upgrading of one will substantially increase the value not only of the upgraded highrise, but also of the neighbouring highrise. The one paying for the upgrade of his own building cannot appropriate the full value of the investment, because it accrues to the owner of the neighbouring highrise. So each slumlord simply waits for the other to implement an upgrade, with the end result that neither undertakes the upgrade.

In the case of EWI for semis, one unit's installation of EWI won't directly raise the value of the other, but it *will* raise the efficiency of the second-mover's EWI investment.

But, if people generally aren't aware of this efficiency loss -- or if it is felt to be negligible, as is clearly the case based on what has been posted here -- then it shouldn't affect willingness to invest in EWI after all.

Posted By: ikimikiIn the case of EWI for semis, one unit's installation of EWI won't directly raise the value of the other, but it *will* raise the efficiency of the second-mover's EWI investment.

And similarly, the second-mover's insulation will benefit the first-mover. So if the first-mover's investment makes it more attractive to the second-mover to actually get off their behind and move, so much the better.

I do think though that the best strategy for flats, terraces, semis, in fact anything other than fully detached, is to encourage street-at-a-time upgrades. It's tricky though, both in terms of costs and what people will accept aesthetically and culturally.

Well I made a model, but didn't get round to running numbers.

It doesn't look terrible (I will post a picture up later)

*Verrry* curious to learn the results of your model Timber!

Ok, well here are the numbers -

I assumed a solid 215mm brick wall with a conductivity of 0.77 W/mK giving a U-Value of 2.2 W/m2K.

I assumed a 150 mm thick layer of EPS for the EWI giving an overall U-Value of 0.2274 W/m2K.

The model resulted in a PSI value of 0.311 W/mK which is appalingly bad, really.

Here are the Isotherms and heat flux

heat flux

That is the external/party wall junction in plan. One house having the EWI, the other not.

As for fixings, renders, plasters etc that has all been omited for simplicity.

Well backstop worst case PSI values for new build are 0.15 W/mK.

A decent, well designed junction can have PSI values as low as 0.01 W/mk.

IF both houses had an EWI system, the PSI value would be almost (if not 0 (although the gemoetric effects of internal and external surface area come into play)).

Very 'pretty' images -- thank you Timber! :-)

But as for actual informational content, it seems, to my untrained eye, from the isotherm representation that the right-ward shift of the gradient zone is fairly modest.

Am I right in supposing that one may compute the total energy (& CO2) loss, relative to a fully EWI-clad semi, by introducing the height of the party wall into the calculation?

Is it a simple formula? Or is it more complex than that?

I'd say, "Almost no chance whatsoever."
The probability is greater than zero, but not by much! :-/

The overlap idea sounds good, though without central govt/LAs stepping in to enforce this possibility, I can't see a lot of semi EWI-ers succeeding in getting the neighbours to agree....

minimal except at the outside corner of the room(s)