Inertia Relief Parameters
Altair Employee
Dear Mr. Grassmansdorf,
Thank you for your reply last time. We are making progress in our simulation. Now I have a specific question which is not about the software itself but more about a model problem which I don't know excactly what to do with.
As you know I am making a model about a carbon fiber monocoque. After doing some research I found that I should make use of Inertia Relief Analysis. So I made my constraints with SUPORT1 and got myself a working model. After doing some more research about Inertia Relief Analysis with Altair Hyperworks I stumbled across the controll card PARAM in which you can choose your settings for the Inertia Relief Method.
I noticed quite a big difference in my simulation between these settings. I am sending you this mail because I wanted to know the opinion of an expert in these kind of simulations about what I should do. Can you give me your opinion or maybe tell me if my reasoning is correct?
In the picture below you can see the constraints (suport1) I created for my model.
--
Met vriendelijke groeten,
Kind regards,
Jens Sneyers
Answers
-
Dear Jens,
as you have already noticed, the Param INREL for the Inertia Relief does have an impact on your simulation. Here is an extract from the online help on this option:
INREL controls the calculation of inertia relief.
-2 requests inertia relief analysis without the need for a SUPORT/SUPORT1 entry.
-1 requests that inertia relief be performed. SUPORT or SUPORT1 cards are required in the bulk data section to restrain rigid body motion. The total number of degrees-of-freedom specified on SUPORT and SUPORT1 cards must be less than or equal to six.
0 requests that constrained analysis be performed. SUPORT and SUPORT1 cards are treated like SPC in this case.
That means, with -2 and 0, your set constraints are ignored (-2: determined automatically in the mesh center of your model, 0: changed to SPC instead of Inertia Relief).
PARAM, INREL -1 would be probably the correct option, if the inertia (= accelerations at determined points to constraing your static loads) should act at specifically defined points.
Jan
0 -
Dear Jan,
Thank you for your reply. Indeed INREL set to zero would not be correct in my case since I want to simulate the suspension forces acting on the chassis. So I set the parameter to -1 and changed my constraints (because only 6DOF should be constrained). After some lookup I found this explanation:<?xml version="1.0" encoding="UTF-8"?>
And I have made my constraints according to this. But what do these constraints mean? And how can I explain that the displacements on this symmetric model are not the same on both sides while the composite stresses are?
What does it mean if I set the param at -2 (so ignore these parameters)? Wouldn't it be better to set the constraint in the center of the mesh? I am not very experienced, but I think this information will be helpfull for me to understand it better.
Thank you for the help
0 -
Hello,
basically Inertia Relief is meant for situations where you don't have a reaction force/ Constraint like with a bolt or a a screw, at which the force will be transferred to ground or another part. Instead your object will remain 'stable' and statically determined, without a fixed point or 'bolt'. Think of an airplane in flight or in your case the monocoque without wheels, where there is no defined point at which the structure is fixed.
So Inertia Relief will apply accelerations on your body to hold it stable and statically determined in your simulation. You can either prescribe the points at which the accelerations are applied --> option -1
Or you can let OptiStruct apply the accelerations at the center of your model if you do not know your exact locations (like in the airplane example).
Hope this helps a bit.
About the unsymmetric results: These could be probably only be judged by us, if you upload the model and /or result file.
0 -
@SneyersJens how did you solved your problem? I'm in the same position as you were: trying to optimize a composite monocoque, and I'm thinking that the best option is using inertia relief.
In your case, what was the best option: -1 or -2? Why the displacements in the monocoque weren't symmetric?
0 -
Please make sure to have symmetric mesh if you are expecting displacement/Stress output to be symmetric.
0 -
Ok thanks @Rahul R, I was just wondering if the problem was given by the suports that SneyersJens used in Inertia Relief type -1. Those suports seem right to you? Or is it better to put them in another way?
0 -
The choice of points doesn’t affect stress results. However, it does affect how displacements are displayed – the point with 3 constraints defines the “zero-displacement” point in the model.
Param>inrel>-2 Requests inertia relief analysis without the need for a SUPORT/SUPORT1 entry (user do not have to pick support points).
0 -
@Ivan I'm trying to use the type -2, but my elements are not beeing constrained, is there a major reason why this could be hapening?
0 -
This is unusual. Can you share the model?
0 -
Yes of course. I've attached the fem file because the model has too much MB.
If you have any doubt about the model, don't exitate in replying in this post!
Thanks!
0 -
In the model you shared there were no supports, spcs or inertia relief defined. So I applied automatic inertia relief by analysis>control cards>param>inrel>-2.
0 -
This is weird... Did you tried to run the analysis to see if it showed something normal? Because I think that I had defined that parameter in my model...
0 -
Maybe you shared the wrong model, because it was not constrained at all so it displayed rigid body mode (large displacements). Anyway, the problem is solved, right?
0 -
No, the model was right, but I guess I found the problem: I was changing the parameters I wanted directly in the cards in the model interface (where there are the components, load collectors, etc), instead of going to Analysis>Control Cards>Parameters. Now that I went there I found that my parameters were disabled. Maybe that was my problem. I'll post here if the problem is solved or not. Thanks @Ivan
0 -
It worked, I only had the Parameters disable. Dumb fail
/emoticons/default_wacko.png' title=':wacko:' />
Thanks for the help!
0
