2D Axisymmetric Analysis with LISA

[BBnJ]

I am new here. I hope this isn't too simple a question for you guys.

I am evaluating the LISA FEA package. I am trying to get the "textbook" results to verify I am using the software correctly. I have come across a strange problem. I am doing a 2D axisymmetric analysis. It is just a solid ring with an ID and OD. I am setting a "weight" in top of it with the force evenly distributed across the face of the ring. So I am doing this by just drawing a x-section (a rectangle) in the X-Y axis that represents the ID, OD, and height of the ring. I adjusted the ID and OD to make the area 1 sq.in. I constrain the bottom row of nodes to have no displacement in the Y direction. When I put a pressure on the top face of 500psi, I get the expected 500psi of von-Misses stress in the ring. But when I put a 500lb force on the top row of nodes, strangely I get 125psi of von-MIsses stress everywhere in the ring x-section. Does anyone know why this is happening. I have the feeling it has something to do with my lack of understand of how to apply forces in this type of analysis. Thanks in advance for any help you can give.

 

[btrueblood]

Having trouble envisioning both the real world load case, and your description of your model. Can you sketch them both and post it, it may help us give you some useful suggestions.

 

[BBnJ]

Here is a picture of what I am doing. (attachment)

It is an axisymmetric analysis - with Y axis symmetry. The bottom nodes are constrained to 0 displacement in Y. The top nodes have a 500lb force on them.

So, since it is axisymmetric, a rectangle drawn like I have it, should represent a solid ring with rectangular x-section.

 

[corus]

The stress is force divided by the area. Divide your 500lb force by the area of the cylindrical wall.

 

[BBnJ]

Yes, the area is 1in.sq. When I divide 500lb by 1insq, I should get 500psi. But somehow I am getting 125psi.

 

[BBnJ]

One more bit of information - I was reading the LISA reference manual on applying forces. It says when you apply a force to a group of nodes it will evenly distribute that load across the nodes. I have tried now with 2 different mesh densities, having different number of nodes across the top face of the rectangular x-section (where I apply the 500lb load). Both times the result was 125psi.

 

[corus]

Are you sure the area is correct. You are exactly one quarter of the answer you expect, which makes me think you've used Pi.D^2/4 incorrectly.

 

[BBnJ]

corus - I believe you are exactly right. I do not have my FEA computer with me this weekend, but I have been thinking about what I did. And I am almost 100% sure you are right. When I calculated the dimensions I needed in Excel, I assumed diameters. When I put the coordinates in LISA, I was using radii. Stupid me. Sorry for all the confusion. And thanks to everyone for the help.

 

[JAE]

Check out Eng-Tips Forum's Policies here:
faq731-376

 

[rothers]

Not familiar with LISA but forces in a 2D axisymmetric analysis are usually normalised over 1 radian of revolution.

 

[corus]

Another thing to consider when using FEA is that when applying forces you need to apply half the load at the end nodes. That'll be why Lisa warns you that the total force is evenly distributed across all nodes so the end nodes will have the incorrect load applied. Think of it as applying the force to each element so that end nodes will have half the load, but nodes between two elements will have half + half the load.
In addition, as rothers says, some packages ask for the load per circumference (N/mm), or radian, or the total load (N). It's worth checking the manual and results to see what the package does when applying loads and calculating reaction forces.