Where to find

[dicer]

Torsional vibration measuring equipment?
I googled and can find plenty of services but nothing to purchase.

 

[SomptingGuy]

It's been a few years for me, but I've never seen an off-the-shelf instantaneous angular velocity measurement device.

We tend to drill and tap into casings and then target inductive probes at gear teeth. Using (very) interesting signal proessing techniques, we can resolve angular motion higher than tooth-passing frequency. All you need is some (cheap) inductive probes, a suitable A/D converter and something like Matlab or its clones.

Other methods involve fitting shaft encoders, but (I think) that's over-kill.

- Steve

 

[TPL]

I don't know what your application is, but I have got useful information on centrifugal pumps using this piece of kit:

http://www.gepower.com/prod_serv/products/oc/en/downloads/141581.pdf

 

[GregLocock]

B&K sell a laser doppler TV device. If you want toothed wheels etc then you just roll your own as SG said.

TV analysis of toothed wheel data is a bit more tricky - the only commercial analyser I've seen is a steam powered device from the 50s. I'd guess Ono Sokki make something more modern.

What you are actually looking for is an FM demodulator with appropriate frequency ranges. It would be a fairly trivial Matlab script.






Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[MachineryWatch]

I have used all the methods discussed. The B&K Laser Doppler device is something they used to rent out from their facility in Atlanta and I rented it from there a few years ago, but I do not know if they still will rent them. They are quite expensive to purchase, but the setup doesn't get any easier.

Polytec has a similar device that I have never used. Here is a link to their website:

http://www.polytec.com/usa/158_423.asp

If you want to go the encoder route Phil Copp makes a little 9 volt battery powered demodulator. The encoder signal can be created as described by SomptingGuy, or you can use your imagination to get equally spaced pulses using an optical pickup. Phil's website is:

http://www.copp-tek.com/.

Phil also has a calibrator available for purchase for his device.

There is also a device from Ono-Sokki.

http://www.onosokki.co.jp/English/hp_e/products/keisoku/revo/fv1300.htm

I hope this helps!

http://www.machinerywatch.com

 

[jeyaselvan]

To my knowedge the best way is to use incremental encoders (say few hundered ppr). Yo may find plenty of them in the market.If your drive train speed is not so high, just do a high speed acquisition of pulses with any DAQ system. You can write very simple algorithm using Hilbert transform to get the phase modulated component out of it, which is nothing but the torsional vibrations. If drive train speed is pretty high, then you may need to go for counter/timer based measurements. Both of these can be done with a simple NI DAQ card with LABVIEW/MATLAB.

If you have access to the drive train elements, probably, you can also use strain gauge & telemetry based torque sensors, and do dynamic measurement of torque. This could also provide information on the torsional vibrations.

More details, look into

http://www.binsfeld.com

Goodluck
Jeyaselvan

 

[dicer]

What about a torsiograph? I thought it was something that could be attached to the end of the crankshaft or what ever the rotating shaft is. Old stuff?

And what about magnetic pickups and reluctor wheels, and output to a dual trace scope? Would it be accurate?
Similar to one of the links. And thanks a bunch for the links.