Simulation Mechanical Forums (Read-Only)
Welcome to Autodesk’s Simulation Mechanical Forums. Share your knowledge, ask questions, and explore popular Simulation Mechanical topics.
cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

Problem in contact pair

36 REPLIES 36
Reply
Message 1 of 37
pugAA
1249 Views, 36 Replies

Problem in contact pair

Hello,

I can't obtain realistic friction contact.

I have a top cylinder which has the spin move (0.632RPM).

A metal sheet is supported by two other cylinder that are free to rotate on their axis.

The top cylinder should drag the metal sheet by friction contact.

The problems are:

 

- The advancement of the metal sheet is not constant and doesn't match with the real value;

- The two bottom cylinder almost can't rotate at all.

 

Thanks for all suggestions.

I can't attached the file because is too big (8MB)

 

Angelo

 

 

 

36 REPLIES 36
Message 21 of 37
pugAA
in reply to: pugAA

Hi,

there were many mistakes in the load curves of this file.

The simulation time is 1s. The load curve 1 is (t=0;Multiplier=0)-->(t=1s;Multiplier=10).

The numer of revolution for step is 0,001055rev (for 10steps for second)

The load curve 2 for the Y-movement is different: after 1mm the contact occurs.

After that, the two bottom rolls move up till 5,245mm.

Maybe the load curve 2 could be: (t=0s;multiplier=0)-->(t=1s;Multiplier=5,245).

 

I'll wait your answer if you need further information.

 

Thank you.

 

Angelo

Message 22 of 37
zhuangs
in reply to: pugAA

Are you sure it is 5,245, not 5.245?  The radius of the two lower rolls is 645mm.

Message 23 of 37
pugAA
in reply to: pugAA

I'm sorry. It was 5.245mm.

Message 24 of 37
zhuangs
in reply to: pugAA

It seems the last qustion, but indeed influnece the resuts, whether the upper roll roates and the two lower rolls move at the same time.  That is to say, when the upper roll roates, the two lower rolls move upward, based on the two load curves (t=0;Multiplier=0)-->(t=1s;Multiplier=10) and (t=0;Multiplier=0)-->(t=1s;Multiplier=5.245).

I just want to make sure in case that you want the two lower rolls move upperward 5.245mm first and keep still, and then the upper rolls rotats.  If like this, the two load curves should be changed.

Message 25 of 37
pugAA
in reply to: pugAA

In the reality the two movements happen simultaneosly. While the top is rotating, the other two move upward till 5,245mm and then only the Y-movement finish. The top roll rotates until the whole sheet is deformed.

Message 26 of 37
zhuangs
in reply to: pugAA

Some results are based on your original load curves using a modified model.  Just to clarify whether you want the two lower rolls to rotate as shown in the figure.  Then I will use the updated load curves to run the modified model.

Message 27 of 37
pugAA
in reply to: zhuangs

From looking at the image, I may see that the two bottom rolls can freely rotate and that's ok, while the top one is the drive roll.

A small recap:

     - The top cylinder has the spin move (0.632RPM).

     - A metal sheet is supported by two other cylinder that are free to rotate on their axis.

     - The top cylinder should drag the metal sheet by friction contact.

 

Thank for your help

Message 28 of 37
pugAA
in reply to: pugAA

Hi,

have you got any good news for my model?

 

Regards

Angelo

Message 29 of 37
zhuangs
in reply to: pugAA

I did get some better results. I copied the modified model the FTP site (whic I sent to you via email) without results considering the file size, and you can run it again.  I gave some significant modifications on the model and some contact settings.

The most modifications on the model are:

  1. Part 5:  The center of beams is changed (and should be changed) to be in the same line with the axis of the upper roll.  Otherwise, the rotation distorts the upper roll.
  2. Part 5:  Two beams are added along Z-direction to execute the moment.
  3. Prat 6 and Part 7:  These two parts are disabled.  Since the two lower rools are not static stable arlong Z-axis.  Part 6 and Part 7 makes the two lower rolls hard to rotate.  I don't know why beam part 6 and beam part 7 are using.  If the two beam parts just work for controllong rotation, you can set the two low rolling as 3-D kinematic element, and contraint the center nodes only except Rz (this is the best way based on my understanding).  If you have to use the two beam parts, I would recommend that you can try to reduce the section area or material strength, which can help reduce the possible numerical instability from beam parts on this model.
  4. Contact setting: see the model for details.
Message 30 of 37
zhuangs
in reply to: zhuangs

I tested with the two beam parts 6 and 7 with smaller radius 1mm, the model works.  I found that the origianal is 1000mm, which is too huge and might influence on the stability of total stiffness matrix.  Please also provide a reasonalble radius for beam part 5.

 

To get better frictional results, you can change the "Tangential stiffness ratio" from 0.01 to 1 for each contact pair.  While this might make convergence slow.

 

-Shoubing

Message 31 of 37
pugAA
in reply to: pugAA

Hi Shoubing,

I used beam elements in part 6 and 7 because there's also the Y-movement (load curve 2) and I don't know if it's possible with kinematic elements.

I reduced the diameter of the beam from 1000mm to 1mm and I've got the rotation but it's uncontrolled: the metal sheet advances more than the value set for the top roll. How can I work on it?

Message 32 of 37
zhuangs
in reply to: pugAA

I did find the problem.  As I mentioned previously, the beam part 5 should also have a smaller radius.  If I just changed its radius from 1000 to 100 (100 might be still large if compared with the mesh size of part 5: ~22mm), the result is much better.  A better and more reasonable radius might still need to be figured out.  I uploaded the model with results via FTP.

 

-Shoubing

Message 33 of 37
zhuangs
in reply to: pugAA

I gave the model more modification and the results should be much more reasonalbe (the model with results is available on FTP).  Two beam parts are added, based on the beams I previously added to beam part 5 and beam parts 6 and 7.

 

Beam radius of part 5: 10 mm

Beam radius of part 6: 1 mm

Beam radius of part 7: 1 mm

Beam radius of part 8: 100 mm

Beam radius of part 9: 10 mm

 

Hope this updated model solves your problem.

 

-Shoubing

Message 34 of 37
pugAA
in reply to: pugAA

Hello,

I downloaded all file from FTP. The latest, Forza_bruta3 is improved a lot. In particular the contact between the metal sheet and the upper roll is quite perfect. The contact between the metal sheet and the lower rolls works fine but still is quite far from reality. In fact the lower rolls rotate more than the upper rolls (about a double angular rotation).

 

If I want to realize a simulation of 2 minutes I'll got huge calculations time. Do you know how can I speed the calculations?

 

Thank you so much Shoubing for your help.

 

Angelo

Message 35 of 37
zhuangs
in reply to: pugAA

First of all, I use "Distance and Length" function to get

   radius of upper roll: 750 mm

   radius of lower roll: 645 mm.

Thus the ratio rotation angles of lower roll to upper roll should equal to 750/645=1.16.

 

Second, since the two lower rolls move upperward using the same load curve with the rotation of upper roll, the contact pressure keep changing in the process, which might result in imperfect contact state, especially for MES contact because of penalty method.

 

 Third, to spped the simulation, I would recommend the following 2 methods:

   (1) Coarse mesh on the two lower rolls.

         For example, two layer of mesh along Z-direction will be fine.

   (2) Displacement tolerance can be changed from "Automatic" (0.005) to "0.01".

         From the log file, we can see that most iterative rasidule is less than "0.01".  "0.01" should be small enough.

         But your original setting "0.1" is too large, that is why I changed it to "Automatic" (0.005).

 

-Shoubing

Message 36 of 37
pugAA
in reply to: pugAA

Hi Shoubing,

the Z-rotational displacement result for the upper roll is 3.79° and -8.43° for the lower rolls. The ratio is bigger than 1.16.

 

Message 37 of 37
pugAA
in reply to: pugAA

Hello,

I've obtained very nice results today.

Thank you Shoubing. You did a really good job.

 

Can't find what you're looking for? Ask the community or share your knowledge.

Post to forums  

Autodesk Design & Make Report