Joint force error - Bad 3D Contact 3

[joedunai]

Hi Experts,

I am trying to run a trivially simple simulation, a ball moving within a cone.

If I don't add a 3D Contact - Force Joint between the ball and cone wall then the ball will fly through the wall.

If I add the Force Joint then I get the following error message:
"Joint force error: No face on `Ball:1` object, contact 3D with penetration impossible !"

Another weird problem is that if I remove the Force Joint and try to move the ball in "Dynamic Motion" mode no force is applied to the ball when moving cursor as it supposed to do.

Can somebody help find the solution to these problems?
The Assembly file is attached to this post.
Thanks in advance.

Joe

 

[rollupswx]

Of course an iam file without parts is useless. This should have been covered in class long before a venture into Dynamic Simulation.
Search Amazon.com on book by Wasim Younis.
and/or attach ALL of your assembly here (including the parts).

 

[joedunai]

Dear Rollupswx,

Thanks for taking a look at this problem and for your advice.
The parts and assembly files are now attached to this post.

I have solved the same simulation in SolidWorks without any problems, but Inventor appears to be less intuitive and user friendly. Anyway, I want to compare the two solutions and find out the advantages/disadvantages of both in such simple simulations.

Joe

 

[joedunai]

Now after a month of presenting the problem there is still no suggested solution how to get Inventor to perform this simple simulation.

Therefore I will conclude that this is a bug in the software, or it is was not designed for simulations, and just ditch the idea to use it. We will have to use some other software packages which have successfully solved this trivial test.

 

[GregLocock]

So have you contacted Autodesk support? Strangely enough support desks are often interested in problems.




Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[rollupswx]

Now after a month of presenting the problem there is still no suggested solution

Sure there was - a month ago.
Did you read the suggested book?
Did you go through the built in Help Tutorials?
You have a surface body for your ball - no solid volume.
You need a Spatial Joint and a 3D Contact joint. This should have been covered in your training or the book.

 

[rollupswx]

I made the problem a bit more interesting (see attached).

 

[rollupswx]

I have solved the same simulation in SolidWorks without any problems...

Attach your sldasm here (including of course all sldprt files) - did you use a surface body for the ball in SolidWorks?).

 

[rollupswx]

On this one I changed the size of your ball to make it easier to see and changed the starting position.
I also changed your direction for gravity as it didn't make sense to me how you had it set.

Changed the cone to a transparent color.

 

[rollupswx]

Therefore I will conclude that this is a bug in the software, or it is was not designed for simulations...

...or that the most basic of training is lacking
Inventor or SolidWorks - assemblies must include parts
Dynamic Simulation should probably be of solid bodies
Direction of gravity?

 

[joedunai]

Thanks GregLocock for the tip; though I am convinced the Autodesk support monitors the related important fora (it would be a mistake not to). But now with your contribution we have got the solution advice from rollupswx.

Thank you rollupswx for finally revealing the trick why the Inventor was unwilling to perform the simulation. To answer your questions:

The purpose of this project is to evaluate and compare the usability, ease of use, user friendliness etc. of at least 3 software packages that can simulate dynamics. Inventor, Solidworks, and Ansys. I have mentioned this in my second post. Instead of asking the advice of others which is the best, we wanted to taste ourselves what is it like to work with each of them. After the decision is made it will be time to start learning in earnest and become experts in the chosen software. But we don’t need all of them.

From this purpose follows that it was out of question to thoroughly learn the usage of all software packages. We are not trying to become experts before deciding which software to invest in. This answers your question whether I have read the suggested book: I didn’t. However, I have gone through some of the help and tutorials in all 3 packages, which made it possible to get to the stage where I am.

The attempt to simulate the motion of a ball on the surface of a cone was successful without any difficulty with Solidworks and Ansys Mechanical – Rigid Dynamics. (Although I couldn’t get Ansys to plot the trajectory on the cone yet, but the coordinates are accessible in a table and on a separate diagram).

I have spent more time with figuring out Inventor than the other two packages, still could not get it work. This is why I needed help. I knew that a standard joint has to connect the ball to the cone. Therefore based on commonsense logic I added a Point-plane Joint because that is what connects the two objects in reality. Also knew that a 3D Contact force joint needs to be applied. If you check my original .iam file you will find both joints in there.

The difference between your setup and mine was that you have used a Spatial Joint instead of Point-plane Joint, and a solid body instead of a surface body.

OK, if it is so logical that we use most of the time solid bodies then how did I get a surface body? To find this out, I restarted the Inventor and in a new part repeated the drawing of a semicircle, closing the arc with a line, and applying a full sweep feature. With a bit of intelligence built into the software it supposed to offer the creation of a solid body by default since that is used more often than the abstract surface body. It did the opposite; it made a surface body by default, since I didn’t know that this will not work in simulation and did not specifically change it.

But again, if the Simulation detects that there is a surface body which it can’t handle, then it supposed to give an error message that it needs a solid body. Instead of a confusing "Joint force error: No face on `Ball:1` object, contact 3D with penetration impossible !" error message which didn’t help at all to find out what is wrong. It is also a mystery why a Point-plane Joint could not be accepted when in reality the contact of the ball and the wall is exactly that…

Some more answers to your further questions:
“...or that the most basic of training is lacking” – Sure I am not an Inventor expert; it was not my intention to become one before even deciding whether we would use it or rather another software.
“Dynamic Simulation should probably be of solid bodies” – probably…Interestingly there was no need to bother about this issue of in the other two software packages. They have some more commonsense built into the program to sort this out for me. This is why I couldn’t guess what the heck is wrong with the Inventor setup.
“Direction of gravity?” – Irrelevant. The purpose of the project was to get the program simulate a setup as I wanted it to be. Whether the ball moves to the left, right, up, or down is unimportant. It is also useless to create a fancy design for this purpose.

“…did you use a surface body for the ball in SolidWorks?” – This question never popped up with Solidworks. It created a body to which I could assign a material, and it was automatically accepted by the simulator.

For those who are interested, there are 3 zip files attached containing the files of the simulation projects for the Inventor (a new, functional version), SolidWorks, and Ansys and some screen shots.

Finally I thank you both again for helping me find the problems and finally make an objective, experience based decision. You get the stars you deserve.

Cheers,
Joe Dunai