Find sound pressure at the source

[Cheetos]

Hello, I have a motor that is connected to a dyno with big blowers. Most likely those blowers will dominate the noise for the entire system. I want to measure the dyno with blowers first (background) and then measure the total system (tot). To get the motor sound pressure level (source), can I subtract the background level from the total (logarithmically speaking)? See attachment.

Normally, the background noise is much less than the source, but, in my case, it's actually much louder. I would think the equation should still be valid.

 

[GregLocock]

Mathematically, yes, practically, no. If you have to do an in-situ measurement of the sound power of the motor in a noisy environment then your best bet is an acoustic intensity survey, but they are difficult to do well in noisy environments (to the extent that I wouldn't bother).

Cheers

Greg Locock


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

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

 

[IRstuff]

Hypothetically, there are other measurement approaches that can identify source locations and intensities, but I've never tried them in real life.

These approaches involve using an array of calibrated or matched microphones -- the simplest example would be 2, which would allow you to direction find a single source, much like the way your ears work and how stereo recordings work. If the sources are at different frequencies or characteristics, you could still use only 2 mikes to extract directional information, much like listening to stereo recordings of orchestras and distinguishing the reed section from violine.

Additional mikes would allow you to mathematically distinguish additional sources. There's the MUSIC algorithm which ostensibly processes multiple mikes to mathematically extract such information. Arrangement of the microphones could allow you to pinpoint locations in 2D space, or possibly even 3D space, much like computational axial tomography (CAT) scans.

I got to the point of identifying a requirement for a truckload of FFTs, and our group decided the computational requirements were too large for our munition and resorted to a simpler direction finding approach, but that was in 1995, so computational load is less of an issue today.

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]

I haven't seen MUSIC used for spatial field description, but as wiki says "MUSIC assumes coexistent sources to be uncorrelated, which limits its practical applications." which pretty much knocks it out of contention for the problems I worked on.

Cheers

Greg Locock


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

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

 

[IRstuff]

Possibly; we were looking for armored vehicles, an while they're ostensibly similar, each engine has its own idiosyncrasies.

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]

Specifically on engines almost everything is phase locked to half engine speed.

Cheers

Greg Locock


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

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