Modal analysis of small, light structures w/o accelerometers? 2

[McLeod]

What are some of the most common, most effective, etc. methods of gathering modal data from structural components which are too small to be instrumented with accelerometers?

The part in question is a flat surgical saw blade that is oscillated by a powered handpiece. Its low mass and high aspect ratios lead me to believe that anything larger than a small strain gage will introduce testing artifact.

Given that the time domain data can be displacement, velocity, OR acceleration, would it be feasible to use a laser micrometer or other alternate measurement system? Has anyone attempted experimental modal analysis using anything other than accelerometers? What was the outcome?

 

[EnglishMuffin]

Laser vibrometers are a hot item.
Expensive though !
See

http://www.mainewebs.com/polytec-cec03-3336/laser11.htm#hsv

 

[elogesh]

Hai,

I faced a similar situation long back.

Laser micrometer is not feasible at my end,because of cost.

Is there any cost effective solution.

I did it using FEM techniques.But experimental techniques to correlate will be fantastic!.

Regards,
Logesh.E

 

[GregLocock]

I agree with elogesh that a correlated FEA is probably the best way to go for a cost effective solution.

I have done a modal on an engine block using lasers (doppler) to get fringe patterns, and I have used the B&K laser vibrometer, as a single point measurement device. Both work, within their limitations.

Other non contacting sensor methods are a close coupled microphone, from which you can back-derive the surface velocity, a capacitative or inductive pickup, and SPATE. SPATE is stress pattern anlysis using thermal emissions, and is a camera based system, from SDRC.

Or you could sprinkle sand, salt or black pepper onto the blade to look at the mode shape.

Cheers

Greg Locock

 

[McLeod]

FEA is definitely part of the plan, and we just have to find an experimental method for correlation. Ideally, the empirical data would allow correlation of vectors and degrees of freedom as described here:

http://macl.caeds.eng.uml.edu/macl-pa/a-walk/a-walk.htm

However, a more qualitative comparison of the resonant frequencies and mode shapes may suffice.

Using particulate to visualize the mode shape is elegantly simple, but unless the blade was constrained all along the outside edge, I think the sand would fall off. Attached to the handpiece, it has a cantilever configuration. In use, the opposite end is also constrained somewhat in the Z-direction by the bone that is being cut. (I plan to model both constraint scenarios.)

Thanks for the equipment suggestions - I'll look into them and get cost information.

 

[MikeyP]

Why not include the mass of the transducers in the FE model? You could then validate the instrumented FE model with the instrumented test article. It's not ideal, but it is better than nothing.

M