CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   FLUENT (https://www.cfd-online.com/Forums/fluent/)
-   -   Two microphone impedance tube (https://www.cfd-online.com/Forums/fluent/180677-two-microphone-impedance-tube.html)

hannesjk November 28, 2016 06:46

Two microphone impedance tube
 
Hello!

I am trying to simulate a impedance tube in Fluent with an inlet, then 2 measuring points, an absorbing sample and lastly a rigid wall. The arrangement can be seen here:
http://www.scielo.br/img/revistas/jb...1/a11fig01.jpg

The idea is to measure the difference between amplitude and phase of the incident and reflected wave. There are a few things I am not sure about though.

1. Will there be errors if I dont have an outlet, just a velocity inlet?

2. If I specify a wall with zero shear at the end (termination) wall, will the incident wave be 100% reflected or do I need other conditions?

Can there be any other issues?

Thanks in advance!

LuckyTran November 28, 2016 15:05

You are good to go.

1. You should only have an inlet. That's the real physical geometry in an impedance tube. You should be aware though that for unsteady simulations, there are acoustic boundary conditions at the inlet that may not be so obvious.
2. Slip versus no-slip at the wall really doesn't do anything. What's important is the kinematic blocking of the wall (the wall normal boundary condition, which is simply a wall). The acoustic boundary layer has some cute results but this really isn't a viscous flow problem and even a laminar or inviscid simulation will get you most of the information you need. Unless...you have really exotic conditions with acoustically generated turbulence.

However, do not be surprised if the wave is not 100% reflected. In order to have 100% reflection, you would need to generate a perfect plane wave (which is already tricky) and it would need to be reflected completely in-phase. Then there's damping/attenuation, unless you are running an inviscid simulation.

Gefest May 15, 2020 08:06

Quote:

Originally Posted by LuckyTran (Post 627272)
You are good to go.

1. You should only have an inlet. That's the real physical geometry in an impedance tube. You should be aware though that for unsteady simulations, there are acoustic boundary conditions at the inlet that may not be so obvious.
2. Slip versus no-slip at the wall really doesn't do anything. What's important is the kinematic blocking of the wall (the wall normal boundary condition, which is simply a wall). The acoustic boundary layer has some cute results but this really isn't a viscous flow problem and even a laminar or inviscid simulation will get you most of the information you need. Unless...you have really exotic conditions with acoustically generated turbulence.

However, do not be surprised if the wave is not 100% reflected. In order to have 100% reflection, you would need to generate a perfect plane wave (which is already tricky) and it would need to be reflected completely in-phase. Then there's damping/attenuation, unless you are running an inviscid simulation.

Can you give me advice on how to generate a perfect plane wave? Got some issues with what.
Also, I noticed what wave depends on the quality of speakers that you using for generating sound. Personally, I using now 6x9 speakers for this experience. I found a good website with a detailed description of them named Audio direct. I found this site occasionally when I was searching to replace for my old microphone, I found their best budget microphone top.
Check this out, maybe you will find something useful for yourself.

LordOFImmortal May 15, 2020 14:05

Quote:

Originally Posted by Gefest (Post 770598)
Can you give me advice how to generate a perfect plane wave? Got some issues with what.

I think this topic is too old and dead.

LuckyTran May 15, 2020 16:14

The plane wave is generated by using a velocity inlet or pressure inlet. That part is straightforward. The issues come after...


All times are GMT -4. The time now is 08:17.