laser vibrometers

[MarkUMSU]

2 questions:
Does anyone have positive or negative comments about using laser vibrometers like

http://www.bksv.com/1067.asp?

If i were to redirect the beam though a fiber optic cable, is the signal still usable? How would one correct for the change?

 

[jimkinney]

They are excellent for non contact displacement measurements, but the ones that I'm familiar with are not that fast. You can't measure transients, and the larger the measuring area the longer the dwell time, so you have to watch the input levels for fatigue sensitive components.

Jim
Kennedy Space Center, FL

 

[GregLocock]

I've used their older model. To be honest it wasn't that useful in our context (noise and vibration development on cars). I did use it to measure the axial vibration of a crankshaft, which would be fairly difficult with a traditional accelerometer.

The frequency response was only 400 Hz, and there are amplitude related limits as well.

The fibre optic idea is interesting, the blurb says you can measure through glass, ask B&K

One of the big problems is that it is not an inertial reference, so you measure the vibration of the instrument as well as the target.

Cheers

Greg Locock

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

 

[Rob45]

What a timely thread!

We're investigating the use of Polytec's scanning laser vibrometer, and I am curious as to whether anyone here has experience with it?
Our intended application would be in determining the vibrational modes of a cabin backwall as relate to a severe boom problem. The panel is roughly 8 feet by 8 feet, and we're interested in modes around 60 Hz, so we're predicting that a few hundred points will give us usable results.
The Polytec unit can scan this in broadband (white noise) mode in under 10 minutes, followed by a narrowband "fastscan" in about a minute at 10 x the original number of points.
The output will be an animation superimposed over a digital image of the panel, as well as FRF data on all 200 points.

 

[GregLocock]

I've been involved in a project using a similar instrument many years ago. It used doppler interferometry to make contours of displacement. What can I say? it worked as advertised.



Cheers

Greg Locock

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

 

[Rob45]

This system does a point-by-point measurement - much as I would do if I were to grid the surface and put an accelerometer on a couple hundred points.
Which, given our current engineering budget, looks like what I'll be doing, starting this morning...

 

[GregLocock]

Only 200? well, that shouldn't take long. Another technique I've seen, but never used, is SPATE.

One technique we did use that was fun was to use a mic as a roving pickup, as the local sound pressure is largely a function of the local surface velocity.

Cheers

Greg Locock

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

 

[Rob45]

Greg:
I've used that microphone thing too, when doing a study of door skin "ring" - door slam noise. Found it gave essentially the same results as an accelerometer in measuring the effectiveness of damping sheet materials.

The weather's been too nice for me to try this yet though.
(When it's clear out, I do on-road testing...)

Regards,
R

 

[rob768]

We use it very succesfully to measure torsional vibrations in ship's propulsion systems, and especially on crankshaft (measuring on flywheel or damper hub). The only disadvantage it has (in troubleshooting practice) is that is doesn't give any torque information, so, for power or torque, we still need straingaugges.

 

[Rob45]

rob:
Could you correlate torsional vibration level at an rpm to measured strains at the same conditions?
Then you'd have a guide to future torque levels based on torsional vibration?
Oh, wait: you're interested in overall torque, not torque variation, right?
Sorry.

 

[rob768]

depends, we are usually interested in both. After all you have to evaluate torsional vibration levels. And if there is a shaft you can put a strain gauge on, you can use the information to a part of interest by using a torsional vibration calculation.

 

[molossi74]

I used Laser just dor Panel Mobility with good results !

bye

 

[Rob45]

Molossi:
Did you use a single-point or a scanning laser?
What did you use to drive the door panels?
What was your point density?

Thanks,
- R

 

[xydrex]

Yeah, we've been using Polytec to scan car panels, it is really convenient to scan hundreds of points (excited by shaker obviously) & let it run for a few hrs. Your arm would drop off if you were to hit with a hammer in that many positions. Also used it for tennis racquets, baseball & cricket bats. In fact in response to the guy who said about transients, we had a pitching machine shoot a ball at a bat & used the laser signal to self-trigger the measurement of the transient response, so you can do whole range of measurements.
Cheers.

 

[xydrex]

...obviously for one point only (the bat).

 

[Rob45]

Yeah... one point!
Speaking of your arm falling off:
We gridded just a part of the cab panel yesterday, and studied 49 points (a 7x7 grid) and it took 4 hours just to get the data - I don't even want to think about the time required to analyze that much data.
Ah, no: we don't have any modal software. We're working on the old graphical modal analysis model.