autodesk mechanical MES automatically cancelled

[grausch]

Hi Guys,

New poster here; I frequently read posts to help solve my issues, however, this one is stumping me. I'm trying to run a MES but am having a whole heck of a time with it. I've attached an image of the setup.

What I'm trying to do is run these two rollers together while applying a rotation to both of them, and a downward force on the top roller at the bearing surfaces to essentially squeeze (or what's called "****" in our field) the two rollers together. What I'm trying to find is the displacement of each roller in the center where the largest deflection would be. Think of the rollers as bowing outwards because of the forces applied at the bearings creating a gap in the center between the rollers.

Anyways, I think I have the setup correct, however, the simulation keeps canceling on me sometime during the process. It appears that the simulation runs, and once it finishes and before the results are gathered, it gets cancelled. I'm running the simulation on the cloud, which has been somewhat of a pain so far.

Does anyone have any ideas of why it is canceling before I receive the results? Perhaps I have setup the simulation incorrectly.

Thank you guys in advance for the help!

Regards,

Garrett Rausch

[zhuangs]

Did you try local running?  After check the model, is there any warning message or error in the ds.ldd file?

-Shoubing

[grausch]

I just got done using the check model feature and although it took a couple hours to complete, it did finish. No errors popped up, but the when animating the model, there is no rotation visible on the rollers when in reality, they should be rotating.

I think I must be making mistakes setting up the loads and constraints. Would anyone be able to comment on my setup?:

1. created surface pin constraints (to act as bearings) on the shafts of the lower roller so the roller can rotate only - fixed axially and radially

2. weld contact with beam element attached to roller so rotational displacement can be applied to the roller (I'm not sure I'm doing this correctly, do you apply the rotational displacement on the center node of the shaft? - the center node being the one that lies upon the axis of rotation)

3. apply a surface bearing load to both ends (shafts) on the top roller to model a downward force on what would be the bearings keeping the roller in place. The idea is that the bottom roller can only rotate, and the top roller can rotate and move in the y direction as would be dictated by the force on the upper bearings. The bearings on the top roller would be able to translate in the "Y" direction to engage/disengage the contact of the two rollers.

4. The same type of rotational displacement applied to the top roller - again, not sure I did it correctly.

5. I put a prescribed displacement on the center nodes on each end of the upper roller. The nodes are fixed in the "X" direction - I did this to keep the rollers perfectly aligned on top of one another. I'm not sure if this step is needed.

6. I created joint meshes on the "bearing surfaces" of the shafts on each end of each roller. I split the face of the shafts on each end in inventor so that there is a distinct surface for the bearings.

For whatever reason, I've had zero luck with getting this simulation to run. To reiterate, I'm trying to see how the system reacts to a downward force on the "bearings" of the upper roller while both rollers are rotating at a constant angular velocity.

Any suggestions/help would be greatly appreciated.

Regards,

Garrett

[AstroJohnPE]

Hi Garrett,

Your description sounds reasonable, but you did not mention setting up any type of contact between the top and bottom roll. If there is no surface contact, the two rolls can pass through each other.

Also, your first post indicated you were trying to calculate the displacement (vertical or Y direction?). In this case, what effect does rotating the rolls have, other than to increase the runtime by several orders of magnitude? If the rotation does not affect the vertical displacement, then you should remove that complication.

You mentioned that the animation did not show the rolls rotating, but you did not mention whether you had other results, such as displacements and stresses. The "Check Model" does not produce results. The only reason that I think you may have tried that is if you were hoping that the results were downloaded but you were prevented from entering the Results environment. That is, the Check Model gets you into the Results environment when the tab is grayed out.

[grausch]

John,

Thank you for your help thus far! I believe the solution to this particular problem lies with the idea that rotation is not of importance. I think I'm getting close to the solution.

Regards,

Garrett