Showing results for 
Search instead for 
Do you mean 
Reply

Modal frequency response with NX nastran and FEMAP


Hello,
I'm performing a modal freq response analysis applying a unit acceleration load
along x on a node attached to other 4 nodes by a RBE2. Amplitude is constant between
20/2000 Hz. When I plot the output in format amplitude/phase on output nodes, amplitude
is always negative for some nodes and always positive for other nodes (curves symmetric
wrt y=0 axis). Shouldn't the amplitude be always positive? Or am I wrong?
Thanks for help.
Marco
2 REPLIES

Re: Modal frequency response with NX nastran and FEMAP


Marco,
I would expect to see that the phase of the response changes with frequency. At
some frequencies it will be in-phase with the input (postive) and other frequencies
it will be out-of-phase (negative). It will change phase at the frequencies corresponding
to modes.
Other than the input point and points close by, I would not expect the response
to be constant phase for all frequencies.
Mark





"Marco" wrote:
>
>Hello,
>I'm performing a modal freq response analysis applying a unit acceleration
>load
>along x on a node attached to other 4 nodes by a RBE2. Amplitude is constant
>between
>20/2000 Hz. When I plot the output in format amplitude/phase on output nodes,
>amplitude
>is always negative for some nodes and always positive for other nodes (curves
>symmetric
>wrt y=0 axis). Shouldn't the amplitude be always positive? Or am I wrong?
>
>Thanks for help.
>
>Marco

Re: Modal frequency response with NX nastran and FEMAP


Mark,
yes phase is correctly not constant with frequency. The problema is that the amplitude
(which is not constant either) is always < 0 for some nodes...why? isn't it positive
by definition? Examples in FEMAP help always give positive amplitudes in freq response
analyses.
Marco

"Mark Donley" wrote:
>
>Marco,
>
>I would expect to see that the phase of the response changes with frequency.
>At
>some frequencies it will be in-phase with the input (postive) and other frequencies
>it will be out-of-phase (negative). It will change phase at the frequencies
>corresponding
>to modes.
>
>Other than the input point and points close by, I would not expect the response
>to be constant phase for all frequencies.
>
>Mark
>
>
>
>
>
>
>
>
>
>
>"Marco" wrote:
>>
>>Hello,
>>I'm performing a modal freq response analysis applying a unit acceleration
>>load
>>along x on a node attached to other 4 nodes by a RBE2. Amplitude is constant
>>between
>>20/2000 Hz. When I plot the output in format amplitude/phase on output nodes,
>>amplitude
>>is always negative for some nodes and always positive for other nodes (curves
>>symmetric
>>wrt y=0 axis). Shouldn't the amplitude be always positive? Or am I wrong?
>>
>>Thanks for help.
>>
>>Marco

>