As a SAP assessor I am finding the issue of thermal bridging a constant problem. Surely we should have a catalogue of details where the Psi value has been calculated for designers to refer to. Is there such a thing? The Accredited and Enhanced Construction Details are very limited.

How is everyone else managing. Surely it is not practical to calculate all the bridging details for each dwelling or development.

The thermal bridging has such a large effect on the calculation.

Posted By: tonyDetails For Passive Houses is pretty good

Indeed so. Worth noting that PHPP computes psi values in a different (and better IMHO) way than SAP does though. I think there are some conversion methods, but I don't know if they are accurate.

Timber -Thank you for the constructive details website. That is useful for masonry construction.
Why do you say -
'The current ACDs and EST details are not a valid solution in ADL2010, so should not be used for new construction built to this version of Approved Document L.'
My particular problem is for larger blocks of flats where a steel frame is needed and where there are balconies.

To add, it shouldn't be your problem as an assessor. Detail present or not? If not, backstop of 0.15 should be applied. It is problem of the architect and main contractor to sort out the details, not yours.

I agree use 0.15 or pass problem back to designer.

If you are not building to the ACDs, you can't use the values. Different details = different psi values!

You can calculate your own details and use those in the SAP/PHPP calculations. This is the best bet, especially if you are building better than building regs.

Be warned, SAP and PHPP use different PSI values (i.e. SAP uses 'internal' calculated details and PHPP uses 'external' calculated details. IIRC)

No, that looks like Y acute as indicated by:

http://en.wikipedia.org/wiki/Code_page_850
http://www.fileformat.info/info/unicode/char/00dd/index.htm

No idea which language uses it.

Upper and lower case psi are:

http://www.fileformat.info/info/unicode/char/3a8/index.htm Ψ
http://www.fileformat.info/info/unicode/char/3c8/index.htm ψ

However, I think it would be better, or at least less obscure, if people just wrote the name of variable rather than the name of the symbol, e.g., “linear conductivity” instead of “psi” and “conductance” rather than lambda.

Posted By: Ed DaviesHowever, I think it would be better, or at least less obscure, if people just wrote the name of variable rather than the name of the symbol, e.g., “linear conductivity” instead of “psi” and “conductance” rather than lambda.

And even better if they used the right words!

lambda is conductivity - an inherent material property. Conductance is an extensive property of a particular sized and shaped block of some material. And usually it's the transmittance or U-value rather than the conductance that people are interested in.

psi is linear thermal transmittance, not linear conductivity. i.e. it is an extensive property of a particular sized and shaped detail, not an inherent material property.

About 2 years back the Irish govt commissioned BRE Scotland to calculate Psi values for our ACDs
The ACD's can be found here , very similar to UK details

http://www.environ.ie/en/TGD/#Part L Supplementary Documents

The current Irish Part L 2011 includes the results of the BRE commission

https://docs.google.com/viewer?url=http://www.environ.ie/en/Publications/DevelopmentandHousing/BuildingStandards/FileDownLoad,27316,en.pdf&pli=1

Tables D1-6 towards the back indicate Psi values for the ACD's with latitudes indicated for varying U Values.

And finally - the Y Value

SEAI recently published this tool

http://www.seai.ie/Your_Building/BER/BER_FAQ/FAQ_DEAP/Building_Elements/How_should_thermal_bridging_be_accounted_for_in_DEAP_.html

( By the way DEAP is the Irish SAP )

This stuff is labour intensive ....

Posted By: Timber'Why do you say'...

I say it because it is true!

Read the relevant section of ADL2010 and you will see for yourself. There are only 3 ways of demonstrating compliance

1) Use backstop worst case of 0.15
2) Calculate details
3) Use ACDs that are part of a quality controlled scheme to be approved by DCLG.

Option 3 never happened (as DCLG cocked it up), so only options 1 and 2 are available. As you can see, however, the old ACDs and EST details don't fall into any of those 3 options and are generally considered defunkt!

However most people don't realise this/don't read ADL2010/get bad advice from their local BCO and use them anyway (or just tick the box and build whatever the hell they like anyway).

Would be interesting to see how a UK BCO would treat the use of the Irish details ( for SAP calcs ) which have been processed akin to option 3 above ....

Timber are you saying that "Option 3" is requiring details signed off as-built before their Psi values may be entered into SAP ? Was / is that the idea ?

If so you are correct to state that the Irish details don't provide for that. They do form part of a reference set of design guides that I think Bft was seeking out.

Actually, you can either use approved designs (option 1) which is accredited construction details for England, Wales and Northern ireland, the Psi values for which are in SAP appendix K or BR IP 1/06, or Accredited Construction Details (Scotland).

Or Option 2 - Psi values calculated by a person with suitable expertise and experience following guidance set out in BR 497 and BRE IP 1/06 - No confidence factors or adjustment applied to those either.

Or a mixture of 1 and 2 above.

Or if neither applies, option 3 is use SAP default of y=0.15.

If some individual Psi values aren't known, then table K1 in SAP and in SAP conventions also gives individual defaultPsi values for each junction, which are set at double the 'Approved detail' level - so you could mix and match between 1, 2 and individual defaults.

If nothing else, the Irish details may come under the second option as far as SAP is concerned, if not the first (either way, no confidence factor applied) <- assuming they've been produced in accordance with BR 497 and IP 1/06).


The ACD's were published in 2007, based on generalised values taken across construction types and with constructions based on much poorer U-values than are now built to - really they desperately need re-doing to cover U-value ranges. The Psi values for constructions with better U-values may be somewhat worse than these levels (or better if designed well).

The Constructive Details thermal bridging details (BBA/Robust details) seem pretty well done, covering a number of options for masonry using aircrete blockwork internally at least.

Above is paraphrased from what it says in consultation SAP2012, which is what we are moving towards and follows pretty much what Appendix K of SAP2009 and the SAP conventions v3.0 (section 5.06) says (this does differ a little from what it says in L1A).

These are available at http://www.bre.co.uk/accreditation/page.jsp?id=2296

Unfortunately the CLG workshops to try and develop Accredited Thermal Modelling schemes came to naught due to lack of overall agreement as to how to make it all work (particularly as regards to checking that details actually followed and that details actually achieve in reality what the modelling suggests). In the current era of red tape challenge etc, it was deemed too costly to implement for Govt and Industry compared to any benefit.

One of the proposals in L1A 2013 / SAP2012 is for implementation of an overall confidence factor to be applied to Dwelling Emission Rate of 3% unless the developer can provide proof that they have followed a Quality System or Publicly Available Specification to show that they have adequately followed specification, which presumably would include confirmation that details produced by someone competent and that junctions constructed in accordance with those details - but it's a little wooly still IMO as to whether this will actually be checked by anyone.

I also agree with you, in that I suspect that some builders / developers just state 'in accordance with ACD's' or whatever and then don't actually draw it that way (I've seen that myself on drawings many times).