# Normal extrusion
**URL:** https://discourse.paraview.org/t/normal-extrusion/8704
**Category:** Feature Requests
**Created:** [December 27, 2021, 3:19pm UTC](https://discourse.paraview.org/t/normal-extrusion/8704 "2021-12-27T15:19:29Z")
**Posts on this page:** 2
**Page:** 1
### Author:  [@FEngineer](https://discourse.paraview.org/u/FEngineer)
#### Post date: [December 27, 2021, 3:19pm UTC](https://discourse.paraview.org/t/normal-extrusion/8704/1 "2021-12-27T15:19:29Z")
Hi,
whenever shell models are analyzed with structural FEA, it’s very convenient to display the results on them using solid representation instead of the regular surface one. ParaView already has a Linear Extrusion filter but for non-planar shells (cylindrical, spherical or otherwise curved) it would be necessary to extrude in normal direction, not just in one of the global axes. In this thread a solution was already proposed:
> [@Adding thickness to a surface mesh](https://discourse.paraview.org/t/adding-thickness-to-a-surface-mesh/6568/3):
>
> Hello, A simple work around is to use the Delaunay 3D filter if adding thickness to a sphere. See the attached state file: [delaunay\_3d.zip](https://discourse.paraview.org/uploads/short-url/9q9IvUlkzaDse1H30ck9Mg0APie.zip) (393.0 KB) But, the cutting surface by the Clip filter with box might not be very sharp as shown in the upper left figure.
It seems that it wouldn’t be difficult to implement a filter like that in ParaView. Thus, I decided to ask for such feature to be included in upcoming versions of the software. I wonder what you think about it.
---
### Author:  [@Kenichiro-Yoshimi](https://discourse.paraview.org/u/Kenichiro-Yoshimi)
#### Post date: [December 30, 2021, 2:25am UTC](https://discourse.paraview.org/t/normal-extrusion/8704/2 "2021-12-30T02:25:17Z")
Hi,
The **Linear Cell Extrusion** filter (vtkLinearCellExtrusionFilter) is already available as a filter for creating three-dimensional finite elements extruded in the normal direction. However, this filter extrudes in the cell normal direction and not in the nodal normal direction as is now desired. If we are going to implement it, it would be a good idea to add an option to **vtkLinearCellExtrusionFilter** to extrude in the normal direction at nodes. For example, an implementation like the one attached below.
[test\_LinearCellExtrude.zip](https://discourse.paraview.org/uploads/short-url/cMCdsVa9Z0LR8wgogjMytpugJs7.zip) (5.9 KB)
This is an inefficient implementation due to the mixed branching of cell normals and nodal normals, but if this can be improved, it may be adopted by the development team.