FEA acoustic analysis

[pjnit]

I have a sound pressure level spectrum and I have a part that I need to determine stresses, forces, and accelerations of when subjected to this acoustic spectrum. I have Nei Nastran. Can I simply convert the acoustic SPL from dB to psi^2/Hz and apply that pressure as a random vibration load?

 

[GregLocock]

yes, as described

Cheers

Greg Locock


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

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

 

[amanuensis]

However you need to know the FRF's of the part in view to assess the input force even if it's a PSD.
So you should first carry out a modal analysis of the part in vacuum, then you should determine the frf of the part in numerous points. Next you calculate the coupling forces between your part and the air.
Finally you run a calculation using the STRESS function in nastran.

 

[GregLocock]

No.

Cheers

Greg Locock


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

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

 

[GregLocock]

In more detail

However you need to know the FRF's of the part in view to assess the input force even if it's a PSD.
OK, but I'm not sure assess is the right word

So you should first carry out a modal analysis of the part in vacuum,
No FEA program I am aware of defaults to allow air mass laoding on structures for a modal.

then you should determine the frf of the part in numerous points.
OK

Next you calculate the coupling forces between your part and the air.
Not required. A pressure is a pressure is a pressure, and if we are ignoring phase (as we must with a psd approach) then coincident frequencies etc will have to be ignored.



Cheers

Greg Locock


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

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