gravity load using concetrated mass

[hyperen]

hi all

I want to simulate the weight of a structure using concetrated mass at certain points of the mesh. However I am not so sure whether by applying inertial load over the entire model, by this way nastran takes under consideration the weight at those points. The procedure I used was teh following:
1)created an concetrated mass at a certain point of the structure 2) Applied an inertial load over the entire structure (with g=-1 in the y direction).
The rest of the structure is modelled with materials that have no density and shell elements (I thought that by this way the model would be computationally more efficient)

Was this the correct way?

Cheers

 

[bjpil]

The difference is appling a single force to a single point and a distributed force across an area. Does the mass lie outside the area of interest? Do you really need gravity?

BJP

 

[hyperen]

cheers BJP,

true , may be I do not need gravity. But applying only a concetrated mass does this result in application of a force at a single point? I inserted the mass property for the conm2 element from the property set card. But when I view the results and the reaction forces, it seems that the solution doesn't consider the masses. Obviously I am doing something wrong!

Any help, anyone?

 

[bjpil]

See if this will help.

http://www.mscsoftware.com/support/online_ex/Nastran/Nas105/appendix_C2.pdf

also, is the mass attached to the rest of the model?

BJP

 

[hyperen]

cheers bjpil,

the answers seem to be different therefore by applying an inertial load of g=-1, the concetrated masses must be taken under consideration in the solution

 

[shlr1]

You have two option: using gravity or force that act on some nodes.Howevere, you must add the density. I suggest for more accuracy to multyply density by E-6.
shlr

 

[saski]

Method is good. but no particular saving. The loads were based on the mass points but in F=K*X solution solving for the displacements is a simple matrix multiplcation ( not expensive). Computer time is saved if you run a large group of load cases in a single run or even better is to learn restart methods and save the inverted stiffness matrix. This means the structure can not change, only analysis or loads or output requests..