I've run into problems with the in-Revit thermal calculations in the way that it handles cavities. I outlined them in an earlier post: Here.
I am still having problems - and haven't been able to come up with a workable work-around.
Essentially it boils down to the fact that Revit treats a cavity as a layer of air - and calculates its thermal resistance as a product of the thickness and conductivity of the air material - this means, by default that an air gap of 50mm provides the same insulation as 50mm of insulation rather than the (in the UK at least) recommended value of R=0.18 m2K/W. This skews the results considerably.
50mm unventilated cavity : R should be 0.18 m2K/W (0.44 m2K/W for low-e foil to insulation to cavity face)
50mm Revit air material : R = approx 2.0 m2K/W
As an example - the correct U-value for a given wall construction with a 50mm non-ventilated cavity was 0.268 W/m2K, Revit calculated it as 0.1935 W/m2K - nearly 30% lower than it should be (a significant deviation).
I had tried to get round it by assigning a 'fake' insulation value to the cavity's air material to ensure that the resistance for that layer comes out at the desired figure - the trouble is - as soon as someone changes the thickness of that layer, the result is incorrect again. Also, I would rather not assign an inaccurate conductivity to a material, you never know how/if it will be mis-used further down the line.
Is it possible to have a material provide a fixed Resistance - regardless of it's thermal conductivity or thickness ? (ideally with an allowance for the emissivity rating of the external face of insulation)
Apparently Revit includes the correct internal / external face heat transfer resistance (Rsi and Rs) - but not for a cavity (Ra.)
If I could create a material that has a pre-defined Resistance - I could create the appropriate fixed R rating as needed for walls, floors etc.
At the moment - it means that the U-Values generated by Revit only work accurately if there is no cavity present.
Anyone else encounter this problem ? If so, how did you get round it ?
In case anyone else is searching, or has a similar query - Autodesk responded that they have passed the issue on to the development team, where it has been documented for possible resolution in a future version.
For the moment - I have created new materials - setup for specific cavity widths and named them as such: "Wall Cavity 50mm", " Wall Cavity 100mm" etc. and assigned the thermal conductivity as required to generate the specific R value (ie 0.18) at that specific width. It's not ideal - and is still subject to someone changing the cavity width and not changing the material - or vice-versa - but it is a work-around for the moment.