Differences in water content results between WUFI Pro and WUFI 2D with identical setups

[Cyprien]

Hi,

I am currently using both WUFI Pro and WUFI 2D. To validate my model in WUFI 2D, I have created the exact same building envelope assembly as in my WUFI Pro model.

Although the interior and exterior climates, material properties, and initial conditions are set up identically in both software programs, I am getting different water content levels, especially i dont understand why in the exterior LP SmartSide siding material the result are not the same despite the exterior climate are the same. While the temperature profiles match closely across the envelope, the water content shows noticeable discrepancies.

Could you please help me understand why the water content is not the same and how to resolve this?

Additionally, is it theoretically possible to get the exact same results in WUFI Pro and WUFI 2D if all parameters are identical? Does this discrepancy typically depend on the numerical mesh generation, or are there underlying differences in the calculation engines for 1D vs. 2D transport?

Thanks in advance for your help!

[Cyprien]

Update,

My issue regarding the exterior material has been resolved. However, I am still encountering difficulties in obtaining consistent relative humidity and water content values within my materials.
For example, within my envelope model, the studied surface is in contact with the OSB layer. Despite replicating the same mesh (by transferring values from WUFI Pro) and using identical calculation parameters (Maxit = 600, convergence criterion = 1e-6, URF matching WUFI Pro, and NSW = 1000) as well as identical materials and settings, I am unable to achieve the same relative humidity or overall water content in WUFI Pro and WUFI 2D.
When placing an observation point in WUFI Pro within the OSB layer and comparing it to the same location in WUFI 2D, the results differ. I am unsure of the cause of this discrepancy. Please see the attached figures for reference:

OSB WUFI 1D Relative Humidity.png (25.65 KiB) Viewed 90 times

OSB WUFI 2D Relative Humidity.png (19.71 KiB) Viewed 90 times

WUFI envelop

Capture d’écran 2026-06-18 142251.png (15.76 KiB) Viewed 90 times

[Christian Bludau]

HI Cyprien,
could you share a direct comparison showing how large the deviations between the curves actually are? WUFI Pro and WUFI 2D results can be compared in one diagram using WUFI Graph.

It would also be helpful to check the numerical quality of both simulations:

• In WUFI Pro, how do the balances look?

• In WUFI 2D, are there any convergence errors reported? How do the residuals deviate from the specified convergence criterion?

If convergence issues occur, the results are usually not directly comparable.

Another important aspect is the discretization. In WUFI 2D, you currently have many elements in the y-direction. Unlike in 1D, this can introduce lateral redistribution effects. To achieve a more direct comparison between 1D and 2D, it is advisable to reduce the model to only one element in y-direction.
In addition, exterior air layers are often numerically challenging to converge. If not already done, you could try using the standard air layer material (short name), instead of “air layer without additional moisture storage”, as this sometimes improves convergence behaviour.

If you include rain in the simulation, it would be advisable to first perform a comparison without rain. The rain treatment can differ between the two versions, especially due to the larger extent of the construction in the y-direction.

I assume that all other settings (materials, boundary conditions, surface transfer coefficients, etc.) are identical in both models.

Christian

[Cyprien]

Hi Christian,

Thank you very much for your help and insights.

To isolate the issue and follow your recommendations, I have performed a new comparison with rain treatment completely disabled in both models.

Regarding the specific points you mentioned:

Air Layer: I am already using a standard, classic air layer material, as I previously experienced convergence issues with the non-storage option.

Y-Discretization: The envelope is 406 mm high, and I have reduced the grid to 2 elements in the y-direction.


Here are the numerical quality stats from the simulation runs:

WUFI Pro:
No. of Convergence Failures 0
Sum of diffusive fluxes (left/right) [kg/m²] 0.165 -0.101
Sum of capillary fluxes (left/right) [kg/m²] 1.0E-3 0.000
Balance (mass vs. sum of surface-fluxes) [kg/m²] -1.312 -1.312

WUFI 2D:
Convergence criterion not reached 9 times
iteration interrupted 56 times

Please find below the direct comparisons from WUFI Graph for the water content in 3 different layers. Despite identical material properties and identical horizontal grid layouts, the deviations remain noticeable (especially for the OSB layer, which later causes a significant difference in the Mould Index when analyzed with WUFI Bio).

Water Content results WUFI Pro/ WUFI 2D:

Extruded Polystyrene.png (30.6 KiB) Viewed 12 times

LP SmartSide Siding.png (27.29 KiB) Viewed 12 times

OSB.png (23.39 KiB) Viewed 12 times

[Cyprien]

I couldn`t add more attachments, so i complete here:

Capture d’écran 2026-06-22 112710.png (3.15 KiB) Viewed 12 times

Capture d’écran 2026-06-22 111338.png (24.65 KiB) Viewed 12 times