# "Failed with results" but results look normal **URL:** \ **Category:** Bug Reports\ **Created:** [September 22, 2026, 6:58am UTC](https://prepomax.discourse.group/t/failed-with-results-but-results-look-normal/3681 "2026-09-22T06:58:18Z")\ **Posts on this page:** 7\ **Page:** 1
**Author:** ![CosmoKramer](https://yyz2.discourse-cdn.com/free1/user_avatar/prepomax.discourse.group/cosmokramer/32/6636_2.png) [@CosmoKramer](https://prepomax.discourse.group/u/CosmoKramer)\ **Post date:** [September 22, 2026, 6:58am UTC](https://prepomax.discourse.group/t/failed-with-results-but-results-look-normal/3681/1 "2026-09-22T06:58:18Z")
I’m modelling a wall-slab intersection (v 2.5.0) and getting `Failed with results` but results look normal (graphically and numerically). First time facing such ‘error’. Anyone has any idea why this is happening? Screenshots below. Links: [.pmx](https://1drv.ms/u/c/eec7215cdc16258e/IQC6N_xlDAekR7cAgSC2whisAWfoBiY-U4ZVnU76p2VK5RE?e=KOgO9P) [output file](https://1drv.ms/t/c/eec7215cdc16258e/IQCfayfjY3zdSJP_IuqV94CvAfSOsxiKLkIu-nRyKvjWuoM?e=1BfQVU) [FreeCAD](https://1drv.ms/u/c/eec7215cdc16258e/IQACTL6eSAqrQqolxnEzFNd3ASQ7SX7dvd-W88Y0uYPx34c?e=PrDaul). ![image](https://global.discourse-cdn.com/free1/uploads/prepomax/original/2X/c/c4e5587580b7c4961b46e71d7f45aafe9434e514.jpeg) ![image](https://global.discourse-cdn.com/free1/uploads/prepomax/original/2X/d/d6deea6880ac65300b9dfcfb631cf1d44f69174f.png) ---
**Author:** ![FEAnalyst](https://yyz2.discourse-cdn.com/free1/user_avatar/prepomax.discourse.group/feanalyst/32/688_2.png) [@FEAnalyst](https://prepomax.discourse.group/u/FEAnalyst)\ **Post date:** [September 22, 2026, 7:10am UTC](https://prepomax.discourse.group/t/failed-with-results-but-results-look-normal/3681/2 "2026-09-22T07:10:33Z")
CalculiX’s log indicates that the analysis was completed successfully: ```auto Job finished ________________________________________ Total CalculiX Time: 19.707712 ________________________________________ ``` It converged in 1 increment (2 iterations): ![image](https://global.discourse-cdn.com/free1/uploads/prepomax/original/2X/9/9fb0d5e09095f080806a1fe2dca71be59ca7af3c.png) You can try submitting the analysis outside of PrePoMax: [Using standalone CalculiX](https://prepomax.discourse.group/t/using-standalone-calculix/846) And the results can be normally opened. I wonder why PrePoMax shows this incomplete status here. ---
**Author:** ![FEAnalyst](https://yyz2.discourse-cdn.com/free1/user_avatar/prepomax.discourse.group/feanalyst/32/688_2.png) [@FEAnalyst](https://prepomax.discourse.group/u/FEAnalyst)\ **Post date:** [September 22, 2026, 7:16am UTC](https://prepomax.discourse.group/t/failed-with-results-but-results-look-normal/3681/3 "2026-09-22T07:16:02Z")
Regarding the model itself, you should try meshing it with hex elements as discussed in your previous threads. And possibly use a compound with proper volume partitions (to use Transfinite Gmsh for hex meshing) instead of tie constraints. Only this rebar will be problematic. You may need some cutouts for it. ---
**Author:** ![CosmoKramer](https://yyz2.discourse-cdn.com/free1/user_avatar/prepomax.discourse.group/cosmokramer/32/6636_2.png) [@CosmoKramer](https://prepomax.discourse.group/u/CosmoKramer)\ **Post date:** [September 22, 2026, 7:53am UTC](https://prepomax.discourse.group/t/failed-with-results-but-results-look-normal/3681/4 "2026-09-22T07:53:28Z")
Thanks @FEAnalyst. I recall discussing that – I think meshing method / compound didn’t solve the issue I was having at the time and it was merely a case of simplification / remodelling in FreeCAD. I prefer tie constraints as it’s easier when assigning materials (I recall compounding the entire model also gave an error / weird results) ---
**Author:** ![MisiaKu](https://avatars.discourse-cdn.com/v4/letter/m/e5b9ba/32.png) [@MisiaKu](https://prepomax.discourse.group/u/MisiaKu)\ **Post date:** [September 22, 2026, 5:25pm UTC](https://prepomax.discourse.group/t/failed-with-results-but-results-look-normal/3681/5 "2026-09-22T17:25:18Z")
I don’t mean to be rude, but you’re making your calculations more difficult. You have a very simple model made of cuboids, which you can easily and quickly create from cubic elements. Extrusion or extrusion from flat surfaces is enough. The minor problem is probably the small elements on the vertical wall. Maybe some brackets every 0.5 meters? The mesh where the ring meets the wall is terrible. Nothing fits together. Even if you get a result, it will be unreliable, despite its nice appearance. More work means a faster result. You want help, and you won’t show the model. ---
**Author:** ![FEAnalyst](https://yyz2.discourse-cdn.com/free1/user_avatar/prepomax.discourse.group/feanalyst/32/688_2.png) [@FEAnalyst](https://prepomax.discourse.group/u/FEAnalyst)\ **Post date:** [September 22, 2026, 5:44pm UTC](https://prepomax.discourse.group/t/failed-with-results-but-results-look-normal/3681/7 "2026-09-22T17:44:47Z")
Honestly, I think that there’s no point in pursuing this anymore. We all know (and have discussed multiple times for this kind of thermal bridge models) what a good mesh should look like and that hex elements should be used whenever possible (but the geometry needs to be simplified and partitioned first), followed by a mesh convergence study. Maybe let’s just focus on the bug for now, unless the OP wants additional help with geometry preparation and meshing. ---
**Author:** ![CosmoKramer](https://yyz2.discourse-cdn.com/free1/user_avatar/prepomax.discourse.group/cosmokramer/32/6636_2.png) [@CosmoKramer](https://prepomax.discourse.group/u/CosmoKramer)\ **Post date:** [September 24, 2026, 11:06am UTC](https://prepomax.discourse.group/t/failed-with-results-but-results-look-normal/3681/8 "2026-09-24T11:06:51Z")
(I’ve stopped reading Misia comments) Thanks @FEAnalyst. Hopefully this will be resolved in the `2.7` version. I’ve just downloaded the `2.6` version and I’m excited to try out the reduced time required for tie constraints.