WUFI

Dear colleagues

I have been doing a simulation campaign with different internal insulations using Wufi 1d and Wufi 2d and I have a problem in the cases of most unfavorable scenarios (when using an internal vapor barrier and not protecting the outer face by a rain repellent application).

I obtain a big difference between the results obtained of relative humidity after the insulation face with the two software (in the condensation point), where the relative humidity is very high and I want to know the limit of application of the constructive solution (taking into consideration 95% according to the recommendation of the WTA).

I have taken into account the considerations that have been emphasizing through different posts of the forum to obtain better results in the 2D simulations (Convergence criterion e-5, use a low value of URF (0.1-0.3), a sufficiently fine mesh x: 183 y: 73, I checked that I used the same boundary conditions).

As can be seen in the attached images with one example (glass wool insulation), after the insulation layer (in the adhesion plaster) in the 1D simulation it reaches 98%, while in the 2D simulation in that same point it has the same humidity profile but the reached value is about 94%. I don´t understand why there is so much difference between the two results.

I would greatly appreciate if someone could help me with this problem as it would be of great help to me.
Thank you very much
Best regards

Dear AMG,

for comparing WUFI Pro with 2D results, you have to use the same grid in both programs and the same convergence criterion, WUFI Pro uses 1e-6.

Christian

Christian Bludau wrote:Dear AMG,

for comparing WUFI Pro with 2D results, you have to use the same grid in both programs and the same convergence criterion, WUFI Pro uses 1e-6.

Christian

Dear Christian,

Thanks for your reply. I modified the convergence criterion to 1e-6 and on the other hand, the mesh I had was the same in the 2D simulation and the 1D simulation. In addition, for this case and to make the simulation faster, I simplified the model not including the wooden beam detail (I attach the detail of the simplified model 2d).
Despite the modifications, the results in the simulation are still the same as those obtained previously. I can´t get 2D values close to those obtained in 1D simulation. What can be the reason for this difference?

Thank you very much
Best regards

Dear AMG,

from the submitted images, it is not clear why the results do not match. Have you checked all materials, transfer coefficients and boundary conditions to see if they are identical? further the balances in WUFI Pro should match as well as the timesteps in in WUFI 2D should converge. THe evaluated data has to be in the same unit.
Can you post a picture with the comparison of the results from 1d and 2d?

Christian

Christian Bludau wrote:Dear AMG,

from the submitted images, it is not clear why the results do not match. Have you checked all materials, transfer coefficients and boundary conditions to see if they are identical? further the balances in WUFI Pro should match as well as the timesteps in in WUFI 2D should converge. THe evaluated data has to be in the same unit.
Can you post a picture with the comparison of the results from 1d and 2d?

Christian

Dear Christian,

Yes, I have checked all materials, transfer coefficients and boundary conditions. The timestep is 1h (3600s in Wufi 2D). I have made a last simulation with URF 0.8 (for temperature) and 0.3 (for humidity), but the obtained results have worsened a bit (which are the ones I attach in the images) compared with those obtained with 0.7 (for temperature) and 0.1 (for humidity).
I attach in the following link https://drive.google.com/open?id=1ts1Q3 ... Zlc3r4Tk2p both the 1D simulation, and the 2D (without the results to reduce the weight), as well as the mentioned screenshots of the relative humidity behind the face of the insulation that is the point that I want to evaluate. Also, the used climate data is in the folder.

Thank you very much for the help
Regards

Dear AMG,
the grid is still quite different between both your cases: Furtehr please go to Computational Parameters -> Advanced Settings and change the values for NSW from 4 to 1000. Thats the iterations for the inner solver loop. Sometimes 4 is to less. 1000 allows to use more iterations, if needed.

Christian

Thank you very much Christian.

The modification of the NSW parameter was what allowed the results obtained in 2D simulation to be adjusted with those obtained with the simulation 1D. Next I will try to perform the 2D simulation of the complete detail and carry out the simulations campaign with different materials.

Thanks for everything
Best regards

Christian Bludau wrote:Dear AMG,
the grid is still quite different between both your cases:

Zwischenablage01.png

Furtehr please go to Computational Parameters -> Advanced Settings and change the values for NSW from 4 to 1000. Thats the iterations for the inner solver loop. Sometimes 4 is to less. 1000 allows to use more iterations, if needed.

Christian