Question about combine acceleration excitation and displacement excitation in one FRF case

Mark Lin

Hi all

New I have a question that how to combine the different excitation in one FRF analysis.

For instance, I want to follow the test that in the lower picture

 

I set an acceleration excitation from 7Hz to 200 Hz, then set a displacement excitation from 18Hz to 50Hz that is fixed on 0.8mm.

Both excitations are RLOAD and combine in DLOAD. 

But, When I check the acceleration of excitation in result, it doesn't follow my input.

Whether are there any misunderstanding? and how to fix it?  

 

Gildas GUILLY_21156

Hello Mark,
Are accelerations and imposed displacements applied to the same node ?
If so, you cannot work like this, because applying an acceleration on a node is exactly the same as applying a displacement.
Any analysis you launch while applying an acceleration at a node is actually run by OptiStruct as an imposed displacements calculated by integrating the provided acceleration twice.
So your setup cannot work because you impose 2 different displacement the the same node.

In your case, where you have acceleration data on some frequencies, and displacements on other frequencies, you must translate everything in terms of acceleration (by derivating twice the displacement) OR displacements (by intergrating twice the accelerations).
As those data are of the form Mag(f)*sin(2.Pi.f.t), derivation/integrals are pretty simple.

Gildas

Mark Lin

Gildas GUILLY_21156 said:

Hello Mark,
Are accelerations and imposed displacements applied to the same node ?
If so, you cannot work like this, because applying an acceleration on a node is exactly the same as applying a displacement.
Any analysis you launch while applying an acceleration at a node is actually run by OptiStruct as an imposed displacements calculated by integrating the provided acceleration twice.
So your setup cannot work because you impose 2 different displacement the the same node.

In your case, where you have acceleration data on some frequencies, and displacements on other frequencies, you must translate everything in terms of acceleration (by derivating twice the displacement) OR displacements (by intergrating twice the accelerations).
As those data are of the form Mag(f)*sin(2.Pi.f.t), derivation/integrals are pretty simple.

Gildas

YES! After I derivate twice the imposed displacement condition, the value of acceleration is perfectly reasonable !

 

Really appreciate your assistance!