Eliminating a component from vibration testing

[Engrni]

Hello All,

I have three components attached to each other in series and I am interested in the response of a third component when vibration input is given to first one. For testing purpose, we want to eliminate the intermediate second component (considerably stiff) and want to modify the vibration input spectrum so that the third component will have a same or bounding response compared to the case when all three components are in series.

I will greatly appreciate any help including suggestions of some references where I can possibly find more information about this approach.

Thank you.

 

[IRstuff]

This is not always possible and sometimes unwise. Unless you know for sure there is no interaction between the three components, decoupling them may result in incorret results.

IF this is not the case, then you could simply test the 1st component with mass simulators for the remaining two, and collect the resultant profile between the 2nd and 3rd.

TTFN

FAQ731-376
Chinese prisoner wins Nobel Peace Prize

 

[Engrni]

Thanks IRstuff for your reply.

Yes, the overall system behavior would change when the intermediate component is removed. However, we are interested only in the response transmitted to component 3. That's why we want to modify the input spectrum for the new assembly such that the response transmitted to component 3 in the when the intermediate component is removed will be same or bounding that the case when all three components are installed together.

Using simulators has been thought off, but the idea was discarded because of some other concerns.

 

[FeX32]

Decoupling them reduces the DOF number. Thus changing the response modes entirely.

"third component will have a same or bounding response compared to the case when all three components are in series"
Why?


Fe

 

[FeX32]

I meant to say . Why do you want to achieve this?


Fe

 

[IRstuff]

Well, then, it seems like you're asking to test an LR circuit by simulating the L with a voltage source. That's just not going to get you meaningful answers, it seems to me.

TTFN

FAQ731-376
Chinese prisoner wins Nobel Peace Prize

 

[BrianE22]

Have you heard of MTS's RPC software? It figures out how to drive an actuator(s) to get a desired response at some point on the specimen.

For instance, strain gage your specimen, record its response with all components connected together then remove the two components you want to remove, connect an actuator to the third component and have the computer drive the actuator to get the same response on the component.

http://mts.com/en/products/producttype/test-components/software/erpcpro/index.htm

 

[Engrni]

Thanks all for your responses.

@ FeX32, I understand that changing the structure will change the modes, however mode is not the point of interest. We are looking for the response spectrum transmitted to component 3.

@BrianE22, I will check out this software and see if it will help us achieve what we are planning to do. Thanks.

 

[FeX32]

I understand that changing the structure will change the modes, however mode is not the point of interest. We are looking for the response spectrum transmitted to component 3.

Yes, I can see what you are trying to do. But, I asked why you would want to do this?


Fe