CORRELATION OF ACOUSTIC MEASUREMENTS AND VIBRATION ANALYSIS RESULTS.

[COBWARD]

Hi
Does anyone know of a good source of information correlating acoustic measurements (sound power/intensity) with vibration analysis of small (1 or 2 cubic meters) products during production?
Thanks

 

[electricpete]

what kind of products?

 

[tomirvine]

Sound can cause structural vibration and vice-versa.

Most of the available research is for large panels and cylinders.

A scholarly reference is:

L. Cremer and M. Heckl, Structure-Borne Sound, Springer-Verlag, New York, 1988.

The following reference has a section that gives a practical calculation method:

D. Steinberg, Vibration Analysis for Electronic Equipment, Wiley-Interscience, New York, 1988.

Also, refer to the Franken method, as shown, for example, in:

NASA-HDBK-7005 Dynamic Environmental Criteria

The quick answer, however, is that you will probably need to take your own measurements.

Tom Irvine

http://www.vibrationdata.com

 

[COBWARD]

The products are small pumps approximately 1.5m cubed.
I can't find the Franken method in the web version of the NASA handbook. Can anybody give another reference to the method.

Thanks

 

[tomirvine]

The Franken method is given in NASA-HDBK-7005, Figure 5.5.

Tom Irvine

http://www.vibrationdata.com

 

[vanstoja]

I doubt that anyone designing and building pumps would do a vibration analysis for the purpose of correlation with radiated airborne sound measurements. If you have to meet airborne noise criteria, the best approach is to test and not waste time and money on complicated external vibration radiation analysis.When serious vibration analysis is done, which is quite a rare event in the pump world,it is done primarily to determine reliability of internal critical components like impellers, bearings, seals, etc. or for reasons of resonance avoidance to prevent damaging vibration levels or excessive stuctureborne noise transmissions. If measured airborne noise and analyzed vibration data exists, it is probabably unconnected in purpose. For example, a university researcher doing a radiated airborne noise test on some vendors pump that just happens to have been analyzed for vibration by the vendor.

 

[kplim1178]

I don't think there is any reference on this. I have used the acoustic/mechanical vibration to locate and diagnose a noisy vacuum pump due to the roots rubbing. I use a 4 channel spectrum analyzer with microphone and accelerometer simultaneously to correlate the waterfall daigram in real-time.

 

[COBWARD]

Thanks for the advice on this.
Unfortunately I still can't work out the Franken Method. The reason for asking the original question is that people within our organisation are keen to reduce the amount of acoustic testing of products in production and rely purely on vibration analysis as it is easier to do. I was trying to see if we could totaly replace acoustic testing with vibration analysis - it seems we can't because the two are not necessarily related.

 

[GregLocock]

That isn't strictly true - if you measure enough vibration data and know enough about the system then ultimately you will be able to accurately predict the radiated acoustic energy from the vibration measurements.

But that sentence has many weasel words in.

The topic of predicting noise from vibration for engines is covered well in an SAE or IMechE paper from the 80s, one of the authors was Yorke, working at Perkins.

Typically you need to classify each radiating surface of the machine as one of several types, then analyse it to get a radiation efficiency, vs frequency, and then measure the average panel vibration spectrum, using say 5 accelerometers.

My friends at

http://www.prosig.co.uk

once developed a system to do just that.

Cheers

Greg Locock

 

[COBWARD]

Hi there
I have just found a reference to ISO/TR 7849 in ISO 9614 (determination of sound power levels using sound intensity). The title of the ISO/TR document suggests that it is possible to determine sound power from vibration measurements.
Does anybody have any practical experience with the ISO/TR 7849 document or the procedures described therein?