Resonance Sweeps 1

[KevinEgle]

We have a couple of different vibration test fixtures that both exhibit the same type of behavior, so I will make this a more general question for now.

The setup we have is a base fixture bolted directly to the shaker, and then part fixtures bolted onto the base fixture. Before we run vibration tests, we do a resonance sweep on the fixture, in the range that the test will see.
This is a relatively new process here, so we do some experimentation with accelerometer placement.

When we did these sweeps on two particular fixtures, we noticed some things that we didn't quite know how to handle. When running a 1 g sine sweep on the base fixture alone, we saw relatively low response levels. When the part fixtures were added, we saw different response levels in the fixture, depending on where the control accelerometer was placed. When the control was placed on the slip table, the response was very erratic. When the control was placed near the bottom of the fixture, with the response toward the top, we saw a magnification of approximately 4x the control acceleration. When the control was placed near the top of the fixture, near where the response was measured, this magnification was reduced to approximatly 1.5x the control acceleration.

My questions are as follows:
A control channel for a test is placed in such a way that the part being tested sees similar acceleration to what is being input. Is this correct?
If the way to reduce acceleration magnification is to place an accelerometer near the point of measurement, the point of interest, is this acceptable?

What level of magnification of a signal is acceptable before the fixture is no longer appropriate for the test? Is a 1.5X magnification acceptable?

 

[tomirvine]

Ideally, slip tables and fixtures would provide unity gain. The laws of practical physics, however, yield fixture resonance conditions.

I recommend that the control accelerometer be placed as close to the test item mounting point as possible. Furthermore, a good practice is to use at least two control accelerometers.

The vibration control computer can then notch the voltage input to compensate for fixture resonances, so that the test item is subjected to the desired base input level.

Tom Irvine

http://www.vibrationdata.com

 

[IRstuff]

One caveat is that slip tables can resonate with top heavy test items. The oil film will behave like a damped spring.

Anything that you test needs to low and close to the table.

Having that large a Q is both irritating and undesirable. Ideally, you could subtract out the Q from the table, but there is potential that the test item modifies that response, so it makes it very difficult to eliminate table behavior.

TTFN

 

[jeyaselvan]

Are you mounting your test items on to the part fixture?
Have you tried evaluate the natural frequencies of your part fixtures. The base fixture may be more rigid and may not be the case with the part fixtures.
The another way may be to mount your control acclerometers ( may be 2or3)on to the part fixture rather than on to the slip table or base fixture.

Jeyaselvan