How do I model a two-variable sinusoidal load? 1

[kman1986]

I have an isotropic plate of shell elements that is simply-supported in the xy-plane. I'm trying to model a sinusoidal pressure in the +z-direction of the form:

P = sin(pi*x/3) * sin(pi*y/3)

where "x" and "y" are my variables.

I have a feeling that if I go to MODEL ==> LOADS ==> ON SURFACE ==> FORCE/AREA ==> FUNCTION (while keeping only the Fz box checked) that I could specify my load there. Unfortunately, it appears that this would only work for a one-variable case. Even in that case, I am having trouble understanding how to go about that. Please, can anyone help me solve this problem? I would really appreciate it.

 

[rb1957]

a dumba$$ approach (i specialise in them) would be to calculate the pressure at each node. that leads to the though that perhaps you need to map the pressure surface and make sure your mesh accurately reflects the varyng pressure (NASTRAN will assume a linearly varying pressure over a single element).

 

[kman1986]

yeah that is another approach, however, like you said it sucks because it involves heavily modifying the input file to mimic the sinusoidal load, drasically increasing computational time and the window for stupid errors. But there HAS TO be a way to model the load in FEMAP...arghhhh! thank you all the same though for responding so promptly.

 

[rb1957]

try the helpdesk ... though it can be hit-or-miss (sometimes good, sometimes not)

another thought, maybe a little easier ... integrate the function for each element, then at least you don't need to fuss the mesh, and apply the load as a uniform pressure on the element. mathcad or maybe excel should be able to do it; maybe exercise the grey matter ...

 

[iulianM]

i'm just a beginner in femap using nastran. i'm searching for a week the answer to my question and i didn't find it. please anyone can tell me how to put a hidrostatic pressure to a beam element? thank you in advance...