WUFI

Hello,

I'm testing a range of cavity wall constructions under both typical conditions and also high moisture conditions (walls facing towards driving rain etc).

All the calculations have 0 convergence errors.

When I compare the "Total thermal performance" steady state U-value in the component section, with the transient U-value readings via the "Thermal Transmission" post processing tool, I have some unexplained readings.

I assumed that for constructions where high levels of moisture are observed, that transient U-values would be higher than the steady state U-value for the same wall construction. As more moisture within insulation typically means a higher thermal transmittance.

In some cases (but not all) where moisture risk is observed, the transient U-values during winter months are in fact slightly lower than the steady state U-value for that same wall construction.

The wall constructions are insulated timber frame with masonry cladding and a slightly ventilated exterior cavity.

I have simulated the exterior wall cavity with an air change rate of 1ACH, could this air change rate have an impact on the unexplained transient U-value readings? Or is it something else I haven't considered?

Thank you.

Just to clarify if anyone is able to help me with this (which I really hope someone is able to!):

When comparing transient U-values between the same wall constructions but different moisture conditions (i.e. wall orientation, initial moisture content of the plywood sheathing etc), the transient U-value is lower in simulations where there is a significant identified moisture risk. Where the plywood sheathing is above 18% moisture content and at risk of decay, the transient U-value is lower than the same wall construction where simulated conditions show plywood remains dry.

This is the case of all but one wall construction simulated so there are discrepancies between wall types. I've tried turning off and on latent heat of evaporation/condensation but the results still remain unexplained. All other aspects of the simulations are as expected with 0 convergence errors.