topic Re: Problem with Event Simulation in Fusion Support Forum

Problem with Event Simulation

adam99a — Tue, 09 Jun 2020 15:42:13 GMT

I can't understand what is wrong...

Log file is in attachment.

Thanks in advance!

Re: Problem with Event Simulation

lee.taylor — Wed, 10 Jun 2020 12:50:52 GMT

The error message that you have received says that the RBE2 (which is a Nastran rigid body constraint) has repeated grid points in the list of grid points that defines the constraint. I cannot see your model but I'm guessing this RBE2 was created by a rigid connector and that, when you created it, you inadvertently selected the geometry for the constraint multiple times.

Re: Problem with Event Simulation

adam99a — Wed, 10 Jun 2020 13:55:25 GMT

I can't locate what RBE2 is... I've already cheched all of constraints and they look normally but there is an option that I do not understand my mistake. It's my first simulation 😞

Can I give you my project in PM?

Re: Problem with Event Simulation

John_Holtz — Wed, 10 Jun 2020 16:56:05 GMT

Hi @adam99a

Does your model have rigid body connectors that are attached to the same geometry that has constraints? For example, the rigid connector could be on the surface of a hole, and the constraint could be on the face at the end of the hole. The edge at the end of the hole is in common with the face, and that is a problem. The same nodes along the edge that the rigid connector is trying to control are fixed by the constraints. That is a discrepancy that the solver cannot handle.

(Technically, the conflict only occurs if the "Fixed Translations" set for the rigid connector, Ux Uy Uz, are the same as the fixed Axis of the constraint, Ux Uy Uz.)

After you check your connectors and constraints, if your model does not have any nodes in common between the connectors and constraints, feel free to send me a private message (hover over my username and choose "Send a Private Message") with a link to your model.

Re: Problem with Event Simulation

adam99a — Wed, 10 Jun 2020 17:32:21 GMT

Thanks a lot for answer! Now I understand what is wrong but I don't know how to fix this. I mean, I need this type of connection 😞

I sent you my project. Maybe you will have an idea.

Re: Problem with Event Simulation

John_Holtz — Wed, 10 Jun 2020 20:04:15 GMT

Hi @adam99a

Did your original model include rigid body connectors? The model that you sent does not include any rigid body connectors.

The model you sent has bolt connectors. If those were in the model that gave the error messages, then maybe a bolt uses a rigid connector for the "head" and "nut" (and a beam element for the shank of the bolt). Since the head and nut are touching surfaces that have a constraint, that would be the reason for the error.

I see many problems in your model, such as:

Event Simulation is limited to a 12 hour runtime. Of course I cannot predict how long your analysis will take, but 5.46 million nodes and 3.4 million elements will never solve in 12 hours. You need to drastically reduce the level of details in the model. (The time step size for the analysis is related to the smallest elements in the model. You have elements that are on the order of 1.05E-6 mm! You should strive to have elements no smaller than 1 mm.) Simplify, simplify, and then do more simplification if you want to perform an analysis.
1. Replace the motors with a block. No threads on the shaft, no nuts.
2. Remove the circuit boards (especially the ones "floating in the air"). If you want the boards for structural strength or to see if they break, then remove the chips and clips/connectors. The boards need to be as simple as possible (like 6 sided volumes).
3. Replace the camera with a lump. If someone is flying this into a wall at 15 m/s, they should expect the camera to break.
The time it takes to simulate the motion between the current position and the block could be significant. You should move the two components closer together so that the impact starts about 1E-6 to 1E-5 seconds after time 0.
With a gap of approximately 150 mm between the drone and the block and a speed of 15 m/s, it will take 0.01 seconds for the drop to occur. The event duration is set to 0.001 seconds. If the analysis would run, it would only get 1/10 of the way to the impact.
I think all of your constraints are all wrong. Constraints "glue" the model to the ground and prevent motion. Therefore, these are almost certainly wrong in your model:
1. The flat block (the ground? a projectile?) has constraints on all 6 sides and an initial velocity. The velocity is not going to make the block move if it is fully constrained to prevent movement.
2. There are 7 other constraints on the drone scattered throughout. I do not know what you are simulating, but I am quite confident that the real drop-test or test rig does not weld the drone to a test fixture.
There are parts that magically pass through other parts. For example, Component125:1 is missing holes that Emax_2306 v2:4 fits into.

If you have not gone through the Event Simulation samples, you might want to do that first. Alternatively, do a very simple model with the basic outline of the drone to find out how long it takes to run and how to set it up. If you can get that to work, then you can begin to simplify the real model and try to get it to run.

Re: Problem with Event Simulation

adam99a — Thu, 11 Jun 2020 21:57:11 GMT

Thanks a lot! So I'm an idiot... I've already started simplyfying but I'm wondering if changing this is worth results of simulation... And I will have generally different part 😕 But ok, I will try! Thanks again!