Applying Gravity constant to model

[viking012]

Using FEMAP 9.3, I know to apply a acceleration constant you input your valuel within Model-Load-Body and you add your density to the material card. My first question is what do you put for a value to similate gravity. Looking at the help section, it says that the inputted value isn't G's. However comparing hand analysis versus the model of a cantilevered plate element using a value of -1 in the z-direction(normal to my plate), this is the only way to get similar numbers. However checking the freebody display, the loads are half of what they should be.

I also tried a value of -2. Checking the freebody display, the loads correlate, but the displacement is off. I tried three different assumptions: a cantilevered beam with a distributed load, a mid span point load, and an end point load. The end point load had only 5% error but the other two had much larger errors. I was assuming a gravity load would act like 1 or 2 not 3. This probably relates to my first question.

 

[ThomasH]

Hi

It's hard to be certain since you don't mention what solver you use. However, since you use Femap it might be one of the Nastran's. Then typically, as you say, you can use acceleration to simulate gravity.

That means that you should use -9.81 m/s2 (SI units).

Then you can check the models mass and the resulting freebody reaction should be mass * 9.81 = xx (N).

As always, make sure you are consistent with the units.

Good Luck

Thomas

 

[rb1957]

well it depends (doesn't it) ! ...
if you're using a model in inches, then its 386 in/sec^2
if your load is in terms of weight (ie lbf), then apply the acceleration as "g"s

whatever, you have to be consisent and produce force units at the end of the day