# Boundary edges not recognized
**URL:** https://discourse.paraview.org/t/boundary-edges-not-recognized/11349
**Category:** ParaView Support
**Created:** [February 7, 2023, 9:43am UTC](https://discourse.paraview.org/t/boundary-edges-not-recognized/11349 "2023-02-07T09:43:19Z")
**Posts on this page:** 8
**Page:** 1
### Author:  [@Frankx9](https://discourse.paraview.org/u/Frankx9)
#### Post date: [February 7, 2023, 9:43am UTC](https://discourse.paraview.org/t/boundary-edges-not-recognized/11349/1 "2023-02-07T09:43:19Z")
Hi,
I’m trying to extract the airfoil surface from my unstructured data set to plot Cp over airfoil surface. I was following the procedure given in:
> [@Compute pressure coefficient on the cylinder surface](https://discourse.paraview.org/t/compute-pressure-coefficient-on-the-cylinder-surface/9185/6):
>
> I assume the “cylinder wall” you want is the circular boundary of the hole in the middle of your dataset. For your particular dataset, you can do that with the following sequence of filters. I’ll give an explanation of what each one does and why you need it for your data. (Don’t forget to hit Apply after each one.) Clean to Grid: Your dataset is not completely connected. The data seem to be broken into blocks and then stitched together. The rendering makes everything look contiguous because t…
but when I select boundary edges and then I try to extract cells by region none of the edges are recognized.
My case is unstructured and 3D whilst the case given in the topic is 2D
Any idea on how to solve?
Is there any other option which I have to use given that my case is 3D?
Sincerely
---
### Author:  [@Kenneth\_Moreland](https://discourse.paraview.org/u/Kenneth_Moreland)
#### Post date: [February 7, 2023, 2:16pm UTC](https://discourse.paraview.org/t/boundary-edges-not-recognized/11349/2 "2023-02-07T14:16:33Z")
If you have a 3D mesh and want a surface, then you probably want to skip the `Feature Edges` step.
You didn’t post your data nor even show a diagram or image, so I have to guess what your data looks like. I’m guessing the airfoil you want to see is a hole in the middle of your mesh. In that case, `Extract Surface` should generate this surface. But it will also generate the outer surface that envelops it, so you won’t see it. If you fly the camera inside the outer envelope, you should be able to see the airfoil in the middle. As described in the other page, you can use tools like `Extract Cells By Region` or `Clip` to remove the outer envelope.
---
### Author:  [@Frankx9](https://discourse.paraview.org/u/Frankx9)
#### Post date: [February 7, 2023, 2:20pm UTC](https://discourse.paraview.org/t/boundary-edges-not-recognized/11349/3 "2023-02-07T14:20:55Z")
Sorry for the few information I gave, you got everything right about my domain.
I was messing around and I’ve got to this point
I only have to remove spurious cells now, any help on that would be appreciated
Sincerely
---
### Author:  [@Kenneth\_Moreland](https://discourse.paraview.org/u/Kenneth_Moreland)
#### Post date: [February 7, 2023, 2:51pm UTC](https://discourse.paraview.org/t/boundary-edges-not-recognized/11349/4 "2023-02-07T14:51:13Z")
My guess is that those spurious cells come from discontinuities in the mesh. That is, there are 2 polyhedra that are supposed to share a face, but the topology connection is not correct. Usually this is from errors in the mesh although discontinuous methods might do this on purpose.
You can try using the `Clean to Grid` filter (_before_ `Extract Surface`) to merge coincident points and hopefully get shared faces to be shared topologically.
---
### Author:  [@Frankx9](https://discourse.paraview.org/u/Frankx9)
#### Post date: [February 7, 2023, 2:53pm UTC](https://discourse.paraview.org/t/boundary-edges-not-recognized/11349/5 "2023-02-07T14:53:25Z")
This is indeed for DG type solver but the `Clean to Grid` filter before `Extract Surface` has been already applied but was not successful (see on the left the pipeline browser)
---
### Author:  [@Kenneth\_Moreland](https://discourse.paraview.org/u/Kenneth_Moreland)
#### Post date: [February 7, 2023, 2:58pm UTC](https://discourse.paraview.org/t/boundary-edges-not-recognized/11349/6 "2023-02-07T14:58:15Z")
You could try using the `Connectivity` filter to identify and extract the largest connected surface, which should be the airfoil after clipping.
---
### Author:  [@Frankx9](https://discourse.paraview.org/u/Frankx9)
#### Post date: [February 7, 2023, 3:01pm UTC](https://discourse.paraview.org/t/boundary-edges-not-recognized/11349/7 "2023-02-07T15:01:32Z")
Getting closer but still issues on the Leading Edge zone
---
### Author:  [@Kenneth\_Moreland](https://discourse.paraview.org/u/Kenneth_Moreland)
#### Post date: [February 7, 2023, 4:02pm UTC](https://discourse.paraview.org/t/boundary-edges-not-recognized/11349/8 "2023-02-07T16:02:26Z")
At least one of those cells must be touching the airfoil. I can’t think of any automatic way of removing those. My only thought is to remove them with an `Extract Cells By Region` with a carefully placed plane.