Swimming pool modeling issue

[Michael Hurd]

I modeled a swimming pool building with a temp/RH of 82 degrees Fahrenheit and 60% RH inside, in the Washington D.C. area. A small portion of the exterior wall is 12" thick concrete, Air space, 6"fiberglass batts and 5/8" gyp board. I first modeled this wall without a vapor retarder to compare the results with a version with a high quality low perm vapor retarder directly behind the gyp. While the assembly with the vapor retarder performs much better than the wall without, the results are confusing. The fiberglass insulation has an ASHRAE 160 (the older version) violation for months and the Concrete wall and the fiberglass gain moisture over time. The concrete wall was modeled per the white paper on the WUFI website.

My problem here is that I really don't believe what WUFI is telling me - If I install a high quality vapor retarder and seal it up tight, how is the moisture from the pool environment getting into the insulation and concrete? Doesn't make sense to me. I tried a similar wall using 3.5" of spray foam direct to the back of the concrete and it behaved very similarly. I'm concerned, if this wall was built this way, that there will be mold growing on the dirt that gets in the insulation during construction.

[Manfred Kehrer]

Hello Michael,
nice to hear again from you.
How can you be sure the moisture from the case with low perm retarder comes from the inside? Could also be built in moisture fro the concrete.
Also I would not use the old ASHRAE 160 Standard evaluation criteria, as we know it is too strict.

[Michael Hurd]

I know it is coming from the pool area because both the concrete inner layer, mid layer and the insulation all are gaining moisture over time, there isn't enough moisture in one of those products to elevate the moisture in all 3. The concrete certainly cannot add moisture to itself. Additionally, since the mid layer of the concrete basically is locked down for moisture flow, it cannot be coming from the outer layer of the concrete or outdoors since the mid concrete layer is locked down. Those are my thoughts.

[Manfred Kehrer]

I presume you have checked the inputs for your low perm vapor retarder and for the case without vapor retarder many times, correct?

[Michael Hurd]

Not sure what you mean. I'm using a vapor retarder that is furnished with the program, it is not something that I have entered into the program on my own. I've used the same vapor retarder in multiple wall assemblies for years without encountering this situation.

[Manfred Kehrer]

Which one do you use?

[Michael Hurd]

The PE membrane in the "generic materials" catalog.

[Manfred Kehrer]

To be sure that the vapor retarder is the only difference between the two calculation, I would duplicate that case, delete the retarder and recalculate.

[Michael Hurd]

I believe that basically I have already done that, but I'll try it again. I've run this wall with multiple configurations, different concrete types, insulation types and basically it spits out the same results +/- some small variations.

[Michael Hurd]

Same results starting from scratch. Makes no sense to me. Ran it in WUFI version 4 also and it did the same thing.

[Manfred Kehrer]

Then I presume numerical issues. How are the balances in the "Status of last calculation"?

[Michael Hurd]

Dead on exactly the same and no convergence issues either.

[Manfred Kehrer]

Weird, I contacted Fraunhofer IBP

[Michael Hurd]

Thanks!

Also, Do you know why the results for the exact same walls can be very different from WUFI versions 4 to versions 5?

[Manfred Kehrer]

Hard to say,
We did put a lot of effort in to ensure that 4.x and 5.0 are the same. Of course, we cannot test this for millions of calculation but certainly did this for hundreds.