Hi,

I want to perform a sine vibration test for my model. Here I attached the required specification and corresponding input that I gave in the test. I tried the test only in the lateral direction. And I input the base excitation as two curves one with displacement (5-8 Hz) and one with acceleration (8-100 Hz).

The test needs to be conducted

The given input in Solidworks

1 - Is the way I gave input in the software correct for the given test level specification?

a. The way I split the frequency ranges into two base excitation curves

b. input the 34.5 mm DA as 17.25 mm displacement

c. any other mistakes in the given values?

2- How to mention the sweep rate in the software?

3- How to resolve the error message?

4- Most importantly since the natural frequencies of the model is too far from the sine test level ranges do I have to perform this sine sweep?

If there are any mistakes please help me to correct it.

**Details of model**

- The model is a satellite structure. It is having some electronic boards in each compartment and some panels at each outer face, I considered each of the boards and the panels as point masses acting at respective mounting holes
- The structure is having 8 mounting location 4 at each end, I gave the fixed boundary condition at these locations
- the structure is mainly made of two long frames and 12 short ribs connected by screws, To model this, I removed the default global contact option and gave bonded contact to the surfaces where the ribs touch the long frame
- I have done static and modal analysis on the same model for different mesh sizes and choose a mesh size which was giving a convergence, so used the same mesh size for this analysis also ( attached the mesh details and corresponding results obtained from the other simulations )

I am not sure is the approach of modelling the screw joints as bonded contact and the electronic boards as remote masses, correct or not. But i thought it may give a preliminary result

Regards

Suhail

1. Is the way I gave input in the software correct for the given test level specification?a. The way I split the frequency ranges into two base excitation curvesI don't think the curves can be split up this way in SOLIDWORKS Simulation. I tried it on a simple file and got the same error as you

One suggestion would be to break up into two harmonic studies: one for the 5-8 Hz range with corresponding base excitation and lower/upper limit and another one for the 8-100 Hz range.

EDIT: Didn't have enough morning coffee yet. My previous suggestion would work, but you should be able to do this - just need to extend both curves so they range the course of the analysis and input 0 values for when they are not supposed to be active.

So for instance the first curve might look something like this:

And then the other curve would start at 5 with value of 0, stay 0 until 8, then go to 1, and stay at 1 through 100 Hz.

Note: In other study conditions (such as Nonlinear analysis) you will want to be careful about setting a curve for something with a Displacement to 0 as this can act like a fixture rather than actually releasing the fixture, which is achieved by setting the value to Off. But it appears to work as intended on the base excitation.

Both curves need to span the lower and upper limit range

2- How to mention the sweep rate in the software?The sweep rate will be neglected in the Harmonic analysis. Harmonic analysis solves in frequency domain and provides the "steady state" solution for each point in calculation. This could be thought of as the worst-case peak result possible under sustained resonance. So essentially it is assumed that the sweep rate is very slow.

4- Most importantly since the natural frequencies of the model is too far from the sine test level ranges do I have to perform this sine sweep?I wouldn't expect there to be a significant difference from applying the equivalent acceleration in a static study until you get sufficiently near a resonant point. What qualifies as "sufficiently near" would vary based on geometry.

Another important question is, are you confident the natural frequencies being calculated are plausible?

I would double check to make sure that your model is not unrealistically stiff due to conditions such as:

Sharing screenshots or your model would be helpful if you want feedback from others.