# Shell element surface loads position **URL:** \ **Category:** General Questions\ **Created:** [September 27, 2024, 1:08pm UTC](https://prepomax.discourse.group/t/shell-element-surface-loads-position/2038 "2024-09-27T13:08:39Z")\ **Posts on this page:** 7\ **Page:** 1
**Author:** ![synt](https://yyz2.discourse-cdn.com/free1/user_avatar/prepomax.discourse.group/synt/32/4658_2.png) [@synt](https://prepomax.discourse.group/u/synt)\ **Post date:** [September 27, 2024, 1:08pm UTC](https://prepomax.discourse.group/t/shell-element-surface-loads-position/2038/1 "2024-09-27T13:08:39Z")
hi, recently i have unexpected result of shell element due to surface loads. No mater i defined in position and sign, seems face load of shell element always applied to bottom faces. I’m not sure is this problem related to PrePoMax or CalculiX implementation. Maybe anyone have similar experiences or differently without a problem, thank you. ![2024-09-27 19_51_48-](https://global.discourse-cdn.com/free1/uploads/prepomax/original/2X/b/bcc3034124fbb82310088139c11973398508bfaa.png) ![2024-09-27 19_51_57-](https://global.discourse-cdn.com/free1/uploads/prepomax/original/2X/a/ab655f6064e298b4062204d4ad885ccf4b881ba7.png) ![2024-09-27 19_53_40-](https://global.discourse-cdn.com/free1/uploads/prepomax/original/2X/a/a42dda8a34c34e48fbd13dd83cc1402f44884124.png) ---
**Author:** ![synt](https://yyz2.discourse-cdn.com/free1/user_avatar/prepomax.discourse.group/synt/32/4658_2.png) [@synt](https://prepomax.discourse.group/u/synt)\ **Post date:** [September 27, 2024, 1:40pm UTC](https://prepomax.discourse.group/t/shell-element-surface-loads-position/2038/2 "2024-09-27T13:40:17Z")
probably, the problem is in front and back face definitions, flip in geometry and remeshing given expected results. ![2024-09-27 20_31_10-](https://global.discourse-cdn.com/free1/uploads/prepomax/original/2X/6/67580cac1187d8bb5727c3d2c3a3e2aef35a0cb3.png) ![2024-09-27 20_31_30-](https://global.discourse-cdn.com/free1/uploads/prepomax/original/2X/a/af1bd955872f6951ae140a2bf30e8c7c2614fc0d.png) ![2024-09-27 20_39_02-](https://global.discourse-cdn.com/free1/uploads/prepomax/original/2X/6/640d4cff2195b0ad0de2b5ca1f0cae5da228a040.png) \*_edited_ it’s a CalculiX implementation when looking few of example and test, default surface load position is back face or bottom, stack sequences in layering shell by composite also. ---
**Author:** ![Matej](https://avatars.discourse-cdn.com/v4/letter/m/a698b9/32.png) [@Matej](https://prepomax.discourse.group/u/Matej)\ **Post date:** [September 29, 2024, 5:44pm UTC](https://prepomax.discourse.group/t/shell-element-surface-loads-position/2038/3 "2024-09-29T17:44:23Z")
So the composite shell has a different force direction by definition in CalculiX? I tested normal shell loads, and the deflection is as expected. ---
**Author:** ![synt](https://yyz2.discourse-cdn.com/free1/user_avatar/prepomax.discourse.group/synt/32/4658_2.png) [@synt](https://prepomax.discourse.group/u/synt)\ **Post date:** [September 29, 2024, 6:07pm UTC](https://prepomax.discourse.group/t/shell-element-surface-loads-position/2038/4 "2024-09-29T18:07:30Z")
even without composite options, still shown face load applied to bottom face by default. In case of shell with normal ratio as thin part, discrepancy almost unseen probably. ![2024-09-30 01_13_37-](https://global.discourse-cdn.com/free1/uploads/prepomax/original/2X/2/2c184bb1a2e77834c434f4d03f306162f799857a.png) \*attached PMX files [shell\_face\_load.pmx](https://prepomax.discourse.group/uploads/short-url/8KbOmO2nJ8ArhwAGT9oV78cxZtD.pmx) (453.6 KB) ---
**Author:** ![Matej](https://avatars.discourse-cdn.com/v4/letter/m/a698b9/32.png) [@Matej](https://prepomax.discourse.group/u/Matej)\ **Post date:** [October 3, 2024, 7:28am UTC](https://prepomax.discourse.group/t/shell-element-surface-loads-position/2038/5 "2024-10-03T07:28:49Z")
You are right. The \*dload always acts on the same surface. Did you try using the surface traction? Surface traction is coded as a \*cload and as far as I understand, it will be applied to the generated knot. ---
**Author:** ![synt](https://yyz2.discourse-cdn.com/free1/user_avatar/prepomax.discourse.group/synt/32/4658_2.png) [@synt](https://prepomax.discourse.group/u/synt)\ **Post date:** [October 3, 2024, 8:53am UTC](https://prepomax.discourse.group/t/shell-element-surface-loads-position/2038/6 "2024-10-03T08:53:46Z")
indeed i was tried in Surface traction, it will have similar to assumption of classical 2D shell in mid-surface positions. Concentrate nodal loads distributed evenly in top and bottom nodes of new solid from shell expansion. commonly, surface load is in top face position instead of bottom. This should be in CalculiX implementation but did not why the reason to placed at bottom. Maybe an improvement in flip operation of PrePoMax in multiple or large selection to become fast and smooth, another in display artifact of model but just cosmetic. ---
**Author:** ![Matej](https://avatars.discourse-cdn.com/v4/letter/m/a698b9/32.png) [@Matej](https://prepomax.discourse.group/u/Matej)\ **Post date:** [October 3, 2024, 12:09pm UTC](https://prepomax.discourse.group/t/shell-element-surface-loads-position/2038/7 "2024-10-03T12:09:39Z")
> [@synt](#): > > another in display artifact of model but just cosmetic I notice it, but I am not 100 % sure why it appears. I think it has to do with drawing from nad back faces of the shell faces.