Hi all,
I have been trying to record the deflection of a diaphragm vs frequency in the side of a Helmholtz Resonator. The scan seems to work but I keep getting zero output to the SaveScalar solver. I have included a .png of the device and zip file of the whole project. Could someone look over the code and tell me what is wrong.
If you wish to run the project you will have to re-mesh using the .brep file to Gmsh and run the .msh file through ElmerGrid 14 2 -autoclean -merge 1. The .mesh files were too big to fit the 1 meg file limit.
Gary R.
Diaphram displacement on Helmholtz Resonator
Diaphram displacement on Helmholtz Resonator
- Attachments
-
- CabinetProject.tar.gz
- (40.12 KiB) Downloaded 61 times
-
- Posts: 2237
- Joined: 25 Jan 2019, 01:28
- Antispam: Yes
Re: Diaphram displacement on Helmholtz Resonator
Bodies 2 and 3 have no material
Body 3 is solid elements, the SMITC solver is for elastic plate elements, stresssolver is for solid elements
Body 3 is not attached to the box and has no displacement constraints, which means if any load is applied it will move to infinity
Pressure Wave 1 = 1000 is not a recognized load for the linear elastic solvers
https://github.com/ElmerCSC/elmerfem/tr ... zStructure
Body 3 is solid elements, the SMITC solver is for elastic plate elements, stresssolver is for solid elements
Body 3 is not attached to the box and has no displacement constraints, which means if any load is applied it will move to infinity
Pressure Wave 1 = 1000 is not a recognized load for the linear elastic solvers
https://github.com/ElmerCSC/elmerfem/tr ... zStructure
-
- Posts: 2237
- Joined: 25 Jan 2019, 01:28
- Antispam: Yes
Re: Diaphram displacement on Helmholtz Resonator
If you can do the Helmholtz solution on its own, I can add the structural displacement part to it.
Re: Diaphram displacement on Helmholtz Resonator
Hello,
Attached is a gmsh model, based on your dimensions, of what looks like a bass reflex speaker. The attached sif file can be run after generating a mesh with gmsh, and converting the mesh using elmergrid.
Maybe someone with more experience with acoustics simulation can evaluate whether the results are sensible.
Rich.
Attached is a gmsh model, based on your dimensions, of what looks like a bass reflex speaker. The attached sif file can be run after generating a mesh with gmsh, and converting the mesh using elmergrid.
Maybe someone with more experience with acoustics simulation can evaluate whether the results are sensible.
Rich.
- Attachments
-
- Speaker.geo
- (2.03 KiB) Downloaded 51 times
-
- case.sif
- (2.4 KiB) Downloaded 61 times
-
- Posts: 2237
- Joined: 25 Jan 2019, 01:28
- Antispam: Yes
Re: Diaphram displacement on Helmholtz Resonator
I added the linear elastic part, assuming the box is rigid, no displacements. The diaphragm can deflect. Using the
Structural Interface = Logical True
results in a solution trivially zero since no pressure load is applied.
If you apply a pressure load using Normal Force then a solution is obtained, but this is not the pressure load from the acoustic model.
So the question is how is the pressure from the acoustic model applied to the diaphragm as a load?
Structural Interface = Logical True
results in a solution trivially zero since no pressure load is applied.
If you apply a pressure load using Normal Force then a solution is obtained, but this is not the pressure load from the acoustic model.
So the question is how is the pressure from the acoustic model applied to the diaphragm as a load?
-
- Posts: 2237
- Joined: 25 Jan 2019, 01:28
- Antispam: Yes
Re: Diaphram displacement on Helmholtz Resonator
https://github.com/mrkearden/cabinet
This version models the box around the air as well as the diaphragm. The box is restrained from motion at it's bottom.
Has the same issue, doesn't work with
Structural Interface = Logical True
This version models the box around the air as well as the diaphragm. The box is restrained from motion at it's bottom.
Has the same issue, doesn't work with
Structural Interface = Logical True
Re: Diaphram displacement on Helmholtz Resonator
Interesting, thanks for putting together another geometry model. From reading more about speaker design on the internet, I think the air in the room outside of the speaker should be part of the model. Think of two boxes connected by a moving diaphragm, where one box is small (the speaker) and the other is larger (the room). Adding the reflex port as a second connection between the two boxes would be close to a model of a bass reflex port speaker sitting in a room.
Maybe some of the shoebox FSI tests will be helpful for the coupling between sound waves and the moving diaphragm.
Rich.
Maybe some of the shoebox FSI tests will be helpful for the coupling between sound waves and the moving diaphragm.
Rich.
Re: Diaphram displacement on Helmholtz Resonator
Hello Kevin,
Looking at your cabinet model in ElmerGUI, I'd like to report something. If you hide boundary 5, and then select either boundary 3 or 6, the red highlight is mottled. This usually means there are two boundaries right on top of each other, and non-conforming.
Rich.
Looking at your cabinet model in ElmerGUI, I'd like to report something. If you hide boundary 5, and then select either boundary 3 or 6, the red highlight is mottled. This usually means there are two boundaries right on top of each other, and non-conforming.
Rich.
-
- Posts: 2237
- Joined: 25 Jan 2019, 01:28
- Antispam: Yes
Re: Diaphram displacement on Helmholtz Resonator
Thanks Rich.
I did a merge nodes in salome and it fixed the mesh, but no change in the results. I updated the files in the github.
I did a merge nodes in salome and it fixed the mesh, but no change in the results. I updated the files in the github.
-
- Posts: 2237
- Joined: 25 Jan 2019, 01:28
- Antispam: Yes
Re: Diaphram displacement on Helmholtz Resonator
There are 3 HelmholtzStructure test cases in the elmer fem test case directory. They all have the structure moving creating sound waves in the air. This case is looking for the opposite. Sound waves loading the structure.