Crazy displacement field for a simple model that should be giving reasonable results

[OverWorked453]

So, I'm a new guy with respect to Abaqus. I ran a model (for a very basic robotic arm) and I am getting reasonable stress values. However, I am getting an outrageous displacement values. I originally did not notice this since Abaqus automatically scales your results with the "Deformation Scale Factor". But once I changed the Deformation Scale Factor to 1, I got a displacement field that was complete nonsense (the arm expanded in size and then flattens like a pancake).

It's common to make simple mistakes and have outrageous errors in FEM software, but the reason this inaccurate displacement is so unexpected is because the stress distribution is reasonable and relatively small. Also, the model behaves basically as it should when the Deformation Scale Factor is at the automatic value that Abaqus calculates (about 4.5E-6 for my case).

The material properties are that of aluminum, there is no error in units, and the loads are of reasonable magnitude. The only items that are really unique to my otherwise basic model is that I use a kinematic coupling constrain to input a moment and there are no boundary conditions on the model, just forces.

Do you guys have any troubleshooting advice? Again the stress distribution is reasonable but the deformation is outrageous. It does not make sense to have a regular and expected stress value, then an outrageous deformation. Any help/ advice would be appreciated! Thanks!

 

[IceBreakerSours]

Uploading an image or, better, the INP will help us.

What kind of analysis are you performing? What type of element are you using? Do you have "sufficient" number of elements? Is the material property set correctly? Do you see any warnings in the .dat/.msg files?

http://www.eng-tips.com/faqs.cfm?fid=376

http://www.eng-tips.com/faqs.cfm?fid=1083

 

[akabarten]

Hi,

I got a displacement field that was complete nonsense (the arm expanded in size and then flattens like a pancake).

For me it looks like you ran "Scale Factor" or "Harmonic" animation instead "Time History" in Abaqus/Viewer.
Please try Animate -> Time History.

Regards,
Bartosz

 

[OverWorked453]

These comments are great. Thanks.

When I am back to my work computer I will check on the mesh, material properties, etc as well as whether or not I actually used the "Time History" animations in my viewer. If not I will try to upload the INP.

 

[OverWorked453]

Just checked, and the material properties and sectional assignments are fine. Also I checked and I have indeed been looking at the time history animation, as opposed to a harmonic animation. So, unfortunately I have not solved my problem.

I have stress values around 10 MPa on aluminum, so I should not be getting outrageous deformation. I have realized something interesting though that perhaps you guys could help me out with. As I stated in my original post, there are no boundary conditions in my model, just forces/moments. When I input boundary conditions on my model, I get reasonable deformation (I restrict all degrees of freedom of a random group of nodes - this has no physical meaning, I just added them in the troubleshooting process). The type of deformation I get is essentially unnoticeable - which makes sense to me for 10 MPa on Aluminum.

So with boundary conditions, the job works. However, without boundary conditions, the results are still ridiculous. I obviously know that if the forces don't sum to 0, there will be translation, acceleration, etc, but this should be essentially a rigid body translation. What I am bothered about is the outrageous deformation , not the translation. The model expands and flattens, which is clearly inaccurate.

So far you guys have been helpful in my troubleshooting. Again, I am new to Abaqus so the problems might be obvious. I would greatly appreciete any advice.

THanks

 

[IceBreakerSours]

It would be far more convenient for me to have a look at the INP than try and play Sherlock

http://www.eng-tips.com/faqs.cfm?fid=376

http://www.eng-tips.com/faqs.cfm?fid=1083