Jump to content

Thermal bridge calculations?


Recommended Posts

Therm is not very intuitive and great care is needed in building the model. The results need further manipulation to get the psi-value and the added bit of maths varies with junction type if I remember correctly. Also refer to BR497 from BRE on the calculation conventions. I use Psi-Therm 2D and 3D - expensive but easy to use.

Link to comment
Share on other sites

3 hours ago, SteamyTea said:

Do you want an exact value (certification) or an approximate value, I think the building regs have some charts to calculate it.

I did write up a method on here a while back, can't remember what the title was though.

It basically took a topological approach to turn a 3D shape into a 2D shape, but keeping the dimensions and thermal properties the same.  Trouble is I have no way to verify it.

Link to comment
Share on other sites

Sounds neat - I can see that working as heat doesn't know or care whether it's going up or down or side-to-side :).

 

A couple of related threads here:

 

https://aecb.net/forums/topic/psi-therm-software/

https://aecb.net/forums/topic/psi-therm-software/

 

The Therm video is fun - reminds me that some in the world still use BTU / ft^2 / degree F !

Link to comment
Share on other sites

There a few commerical products that make them easier to calculate, but none are exactly cheap.  If what you need to do is farily limited though, you could potential do them within a trial period.  The only low cost (free) approach I know is Therm though, but it takes some time to learn and you need to ensure you use the correct boundary materials and temperatures. Then on top of that you need to caculate thermal bridge psi value seperatly.  There is some guideance here:  https://peterwarm.co.uk/resources#therm

 

The altnerative approach is to get hold of a book of standard details and use the psi values for these if you are not doing anything really complex/different.  Using values taken from standard details is AFAIU acceptable for PH certification if you go down that route, and it's not a requirement to calulate every single bridge.

Link to comment
Share on other sites

Thanks @Duncan62

 

I feel I'm making some progress but it seems a hard nut to crack. It appears that most paid-for FEA software is actually a nice user interface sitting on a public domain back end e.g. Nastran. All quite expensive considering it's 'required' for SAP, PHPP etc. As @Dan F says, standard details are available if that's what you have.

Link to comment
Share on other sites

Junctions picture showing all the TBs that need consideration: 05-Junctions-for-Psi-Values.thumb.png.557b461b90553eba9c37cf0cdc2e4d2c.png

 

 

Appendix B had a list of all the TBs required: S10TP-05 - Treatment of thermal bridges - V1_2.pdf

 

Corners are quite straight forward: 

Screenshot_20240801-0658372.thumb.png.8481c4c798ecfb05406c3361ef0c4d7b.png

 

Windows are more involved. The frame conductivity needs to be calculated and input as a custom material for the frame in the materials library. If the head/jamb/sill have different UValues then you'll need to drop it for each.

Screenshot_20240801-0656572.thumb.png.31aeab7b115ef4441b8a375e483a3dfc.png

 

Ground slab calculations are a two step process. I think they are explained in the previous ThermInstructions.pdf I uploaded above.

Edited by Duncan62
Link to comment
Share on other sites

Thanks for that, and yes, those instructions are well written. Fusion makes some pretty temperature plots (in 3D) (see below) but doesn't do the last step of calculating total heat flow.

 

Do you have an idea of how well the standard details match real cases? Presumably if they work for standard modern estate builds they're a 'bit high' for better insulated designs. I see from the doc that the BRE have doubled some of the numbers between SAP 2012 & 10. They say part of the rational is to encourage people to do their own calcs. Well great if that was easy.

 

Re: windows I found that e.g. Rationel publishes U-Values and psi-values. I assume other suppliers do the same although I have not looked. How reliable are they - anyone's guess.

 

image.thumb.png.c7f4605ed235efdf57b248a1f2730f85.png

image.png.5eb836466b8a717c20807c4bff1601f3.png

 

image.thumb.png.b6f84ac7c7483a73497810543b96df9f.png

Link to comment
Share on other sites

@Alan Ambrose Remeber that the SAP and PHPP calculations are different.  One uses internal measurements and one uses external, also the boundary conditions and/or external temps might be different (can't remember).

 

I've seen defaults of 0.040 used as slightly conservative a number for(presumably good spec and well installed) windows and 0.1 used for thresholds.  Great installation and this could be less though, and if you have a blind system or poor insulation this could be more, but I think those are good starting points.  If rational publish psi values then that will be based on assumptions about the installation situation and fabric make-up, so may not be accurate or valid. Also need to check if rational numbers use SAP or PHPP methodology. 

 

Window/doors installation of course aren't the only thermal bridges, you have all the building juntions too. 🙂

 

Edited by Dan F
Link to comment
Share on other sites

18 hours ago, Dan F said:

One uses internal measurements and one uses external, also the boundary conditions and/or external temps might be different (can't remember).

 

 

Yes this is correct.

 

For example the corner detail: PHPP uses the external wall length. SAP uses the shorter internal dimension.

 

Boundary conditions and Rse values differ. From memory the external conditions are 0degrees and 0.04Rse. i.e. there is no allowance given for ventilated facades.

 

This is from memory so if anyone has a reference, please share.

 

 

Link to comment
Share on other sites

Yeah, I was sort of aware of that (I don't know much about SAP). Seems to me that PHPP is a bit more sensible in that, if you don't put in any bridges, it probably overestimates the straight U-Value heat loss and partially compensates for the missing bridges.

 

Have installed Therm, so will give it a go.

Link to comment
Share on other sites

You can create a DXF for your structure that you'd like to model in Therm, then trace it in Therm with the auto snap feature in Therm 7.8.

 

That makes it easy.

 

Look for "underlay" or similar in the menu 

Edited by Duncan62
Link to comment
Share on other sites

OK some progress. Now I need to check all my inputs and back out the psi thermal bridge value...

 

Also had to trace my dxf file for now (rather than getting a direct import).

 

Couple of hours work so far.

 

image.png.440ad2ef4dcc6e5c78f53a62c8462186.png

 

image.thumb.png.8f04d45f86b6f61b87254bd57128b247.png

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...