# Sweep mesh of thin part **URL:** \ **Category:** General Questions\ **Created:** [November 10, 2024, 2:22pm UTC](https://prepomax.discourse.group/t/sweep-mesh-of-thin-part/2168 "2024-11-10T14:22:25Z")\ **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:** [November 10, 2024, 2:22pm UTC](https://prepomax.discourse.group/t/sweep-mesh-of-thin-part/2168/1 "2024-11-10T14:22:25Z")
i tried different setting value in mesh smoothing of newly implementation by sweep. However, it seems mesh generated did not change the number of element trough thickness for thin part. Is there something i miss? or there’s current limitation needed to improve, thank you. ![2024-11-10 21_12_34-](https://global.discourse-cdn.com/free1/uploads/prepomax/original/2X/9/9cb8c92ac4c6ec8b6565e77c480e2c9b7b79fe36.png) ![2024-11-10 21_12_51-PropertyGrid](https://global.discourse-cdn.com/free1/uploads/prepomax/original/2X/b/ba1d272c2b57b2928893c66ecd9f1ae13145f867.png) ![2024-11-10 21_13_05-PropertyGrid](https://global.discourse-cdn.com/free1/uploads/prepomax/original/2X/9/93bccee6f1e146ce79ff6fb0652fce57fefa50d9.png) ---
**Author:** ![Matej](https://avatars.discourse-cdn.com/v4/letter/m/a698b9/32.png) [@Matej](https://prepomax.discourse.group/u/Matej)\ **Post date:** [November 11, 2024, 7:35am UTC](https://prepomax.discourse.group/t/sweep-mesh-of-thin-part/2168/2 "2024-11-11T07:35:01Z")
To change the number of elements through thickness, you can use Mesh refinement. Currently, you can only specify the size of the elements (not their number), but they are easily convertible. It suffices to prescribe the mesh refinement to a single sweep line. ---
**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:** [November 11, 2024, 9:57am UTC](https://prepomax.discourse.group/t/sweep-mesh-of-thin-part/2168/3 "2024-11-11T09:57:27Z")
actually i miss since its different methods compared to extrusion mesh setup, thanks for the hint. Increase number of element through thickness successfully generated by mesh refinement, but reduced or coarsening still not possible to be less than six layers (default maybe). Two and four layers of quadratic and linear hexahedral element is commonly used for thin parts, hopefully there’s some work around to coarsening and have possibility to set. ![2024-11-11 16_46_03-](https://global.discourse-cdn.com/free1/uploads/prepomax/original/2X/a/a7b4052887d826965501393520801f499bfc3a78.png) \*edited sorry i miss again, actually mesh coarsening or reduction of number element trough thickness also possible by modifying layer smooth steps instead of default value. ![2024-11-11 22_04_54-](https://global.discourse-cdn.com/free1/uploads/prepomax/original/2X/d/dd0857ada78e2385dd20a3e909037bef32d1b5c4.png) ---
**Author:** ![Matej](https://avatars.discourse-cdn.com/v4/letter/m/a698b9/32.png) [@Matej](https://prepomax.discourse.group/u/Matej)\ **Post date:** [November 12, 2024, 8:45am UTC](https://prepomax.discourse.group/t/sweep-mesh-of-thin-part/2168/4 "2024-11-12T08:45:37Z")
> [@synt](#): > > \*edited > sorry i miss again, actually mesh coarsening or reduction of number element trough thickness also possible by modifying layer smooth steps instead of default value. The number of smooth steps does not affect the number of elements in any direction. It only changes the positions of the mesh nodes. ---
**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:** [November 12, 2024, 11:20am UTC](https://prepomax.discourse.group/t/sweep-mesh-of-thin-part/2168/5 "2024-11-12T11:20:25Z")
well, i need to learn more about sweep mesh since not truly understand. I tested and i can confirm, coarsening not have relation with smooth steps. Also, number of edge division by mesh refinement is not always followed, another governed values exist e.g set to 4 & 8 element trough thickness of my model is hard to generates, not yet succeed here. ---
**Author:** ![Matej](https://avatars.discourse-cdn.com/v4/letter/m/a698b9/32.png) [@Matej](https://prepomax.discourse.group/u/Matej)\ **Post date:** [November 12, 2024, 2:28pm UTC](https://prepomax.discourse.group/t/sweep-mesh-of-thin-part/2168/6 "2024-11-12T14:28:37Z")
First, try to play with the mesh refinement with algorithms that are not Quasi-structured quad. There the mesh sizing should work. For Quasi-structured quad, Gmesh does not follow the edge seeds specified. So PrePoMax adjusts them as good as possible. ---
**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:** [November 12, 2024, 3:26pm UTC](https://prepomax.discourse.group/t/sweep-mesh-of-thin-part/2168/7 "2024-11-12T15:26:58Z")
further test as suggested, sweep mesh using Frontal-Delaunay(Blossom) and Quasi-structured quad algorithm. Refinement applied to limit of four element trough thickness for all part, quasi-structured quad still generates six edge division, Maybe it could be more common to set to be four whenever possible. Thank you. ![2024-11-12 22_21_17-](https://global.discourse-cdn.com/free1/uploads/prepomax/original/2X/3/320fd8670bea5acaf32963f2fdf5215eb59e98db.png) ![2024-11-12 22_15_46-](https://global.discourse-cdn.com/free1/uploads/prepomax/original/2X/a/aab3428ba94a0fb963601eecc05cda0a559d1944.png)