Computer Simulation for Sound Analysis / Reduction 1

[MDGroup]

There are lots of different computer programs that can perform simulations on 3D CAD models to calculate real-world results.

FEAs can calculate stresses and displacement based on loads, you can simulate drop tests to see how an enclosure reacts, CFD can calculate flow through a system, you can calculate Heat Transfer through different components, Mold Flows, etc...

Are there any programs that can calculate noise level in a system?
Example: You have an assembly that has a motor mounted on the inside. Are there any programs where you can use your CAD files, generate a sound level at a specific point (motor) and then be able to measure the sound level outside of the assembly at different locations.
Then you could change part geometry or materials, or add baffles, or add foam, or move openings, … and it would tell you how those changes affect the sound level outside the unit.
The goal is to reduce the noise level as much as possible. Currently we are doing this through many iterations of physical prototypes, which takes time and becomes more of a guess and check to see what makes improvements.

Anyone know of any computer programs that can simulate this?

Thanks,
Lou

 

[FEA way]

Of course, numerical methods (including FEM) can be used for acoustic simulations as well, not to mention optics or electromagnetism. You just have to find a software that offers such capabilities. Possible options include COMSOL, ANSYS and Abaqus.

 

[IRstuff]

Seems to me that's only half the problem; you'd still need to know what the amplitude and spectrum of the noise is, and whether there's any interaction between the source and its enclosure.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[MDGroup]

Thanks for the quick responses.

Agreed, I would need to be able to fully define the sound in frequency and amplitude. And it will have lots of interactions with all of the components and materials surrounding it in the assembly.

To simplify my product, consider a household vacuum. It has a vacuum motor on the inside creating noise, along with all of the noise of the airflow through the unit and out the exhaust.
Is it possible to simulate this complete system, to analytically determine the optimal design of all the components and make it as quiet as it can be? Or is that really not practical, and is it actually faster and more accurate to do iterations with different physical prototypes?

 

[IRstuff]

You might want to read the section on acoustic engineering here:

https://www.jamesdysonfoundation.com/content/dam/pdf/Standalone_DesignProcess.pdf

It's basically not the tool, or tools, but the brains behind the tools.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm

 

[GregLocock]

Let's take the simple example of an exhaust system. We know the pressure waves inside the exhaust, so we should be able to predict the noise radiated by the exhaust's skin, and the vibration at its mounting points. We could then use these to drive the structural/acoustic model of the cabin to get noise at the drivers ear. Does everybody agree that sounds reasonable? Good. It can be done. But wow are you in a world of hurt.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[rb1957]

"Is it possible to simulate this complete system" … a tape recorder ?

We (in the aircraft business) do exactly what Greg is talking about, as I'm sure many others do as well. Yes, we work with a full size fuselage, with microphones at passenger ear level, and as Greg says "a world of hurt (well spent dollars)" later we have an anti-noise system.

How do noise cancelling headphones work ?

another day in paradise, or is paradise one day closer ?

 

[CWB1]

Sounds like you need to speak with a NVH engineer, those quiet-domed devils. Not sure your goals or complexity but quite a lot can be done with simple with FEA and CFD to dampen vibration and control airflow. As with any simulation tool, your results are only as good as your inputs and a ton of testing is necessary to understand and build accurate models.

 

[IRstuff]

There were(are?) people who can do bits of FEA in their heads and come up with solutions that FEA will back up.

The fact that you're doing guess and check suggests that you need to hire someone, even temporarily, who's experienced in doing sound suppression. Otherwise, you're simply going to be doing guess and check in FEA, with no intuition as to whether what you are simulating will even work correctly. I've done the same myself, and since the cycle time is shorter, all you wind up doing is way more guesses and checks, without any clear path to a real and viable solution.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm