Vibration measurement

[geoffthehammer]

I have a 24" x 18" cast iron surface plate mounted on a bench using soft rubber pads. Bolted to the surface plate is a piece of equipment, part of which vibrates horizontally at amplitudes up to +/- 1.25mm (adjustable)and frequencies between 20 and 120Hz (adjustable). The vibrating mass is around 150 grams which has a small vertically offset out-of-balance.
We are trying to measure the amount of vertical movement of the vibrating mass induced by out-of-plane forces within the drive mechanism and the out-of-balance. We are expecting movements of up to 10 micrometres. We are reading the vertical motion using a capacitance sensor (sensing part of the vibrating mass) mounted to the surface plate via a magnetic base.
It seems reasonable to assume that we are also measuring any vibrations within the surface plate and the sensor mounting arrangement.


We have been told that if we place a very soft, thick rubber pad on the surface plate, then place a heavy cast iron block on the rubber and mount our capacitance sensor on the block it will eliminate the vibrations of the plate and the sensor mounting arrangement. The rubber pad acting as a filter.

The question is this: Is this argument correct and exactly what will we be measuring?

 

[CESSNA1]

GEOFFTHEHAMMER: At what unbalance (frequency and amplitude) do you want to measure the verical motion of the plate? Basically what you have is a spring - mass - damper system, but unfortunately the spring you have (the rubber) is non linear. I would suggest start off by going to a vibrations text and get a thorough understanding of the basics. You did not mention what the capacitance device is, I assume it is a variable capacitance accelerometer. MAximum amplitude will occur at the resonant frequency. Which is sqrt(k/m) radians/sec. A large mass and a soft spring will then have a low resonant frequency. Look at the transmissibility curve and you will see that the data should be taken at no more than 1/3 the resoant frequency. At higher than resonance the motion deminishes with increasing frequency.

Regards
Dave

 

[GregLocock]

The idea is good, but you'll find it very hard to build a sufficiently soft suspension. Why not build a gantry over the surface plate, screwed to the bench and mount the pickup to that?

Alternatively mount an accelerometer on the capacative pickup and subtract that reading from the measured capacitor acceleration. This will then give you the true absolute acceleration of the test part, if you do it perfectly.



Cheers

Greg Locock

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