ADAMS Request output steps 1

[josefl9]

Hello,

I 've tried a lot of different formulations and settings when trying to output a request file from and ADAMS model, but none of them managed to keep a completely steady output step. What I mean is that even thought I request an output about every 9e-5 seconds (10240 Hz sampling frequency), at some steps, the output time in the request file is larger, about 1e-4 seconds. Is there a way to guarantee an exact output step?

Thanks in advance!

 

[GregLocock]

Have you carefully read the Fixed Step integrator Option in solver integrator help? One other option is to import the data in matlab and interpolate it over a fixed timebase.


Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[josefl9]

Hello,

Yes I did try using the FIXIT formulation for the integrator, even using a larger HRATIO so that the integration runs smoother, but still, when the request file is ouput the time values vary a lot.

The only other solution that comes to mind is to use an output step even smaller than the one I actaually need, so that the results are more accurate, but this whould require extra computations time.

Your idea of interpolating the time series in matlab is really good, but since the problem is not linear, won't the interpolation might result in false results?

Thanks in advance,

JK

 

[GregLocock]

Not if you start with a time step 1/10 of what you want!

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[josefl9]

I found a solution which didn't require decreasing the time step, one that fits my problem right now anyways.

I compared the ADAMS FFT plot results with the ones I compute using matlab and calculated a correction factor. This actually got the results pretty close and is quite consistent as a method so far.

Regardless, thanks for your advice. You 've been helpful in every question I posted here so far!

JK