Contact Mechanics - Friction Test

Numerical methods and mathematical models of Elmer
kevinarden
Posts: 2237
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: Contact Mechanics - Friction Test

Post by kevinarden »

Your are restraining the cube in the Z direction, so the restraints are taking the gravity load and not the contact surface. In order for the contact surface to feel the weight of the cube (the normal force) then you would have to let the cube displace in the Z direction and rely on the contact surface to hold it up.
maike151
Posts: 41
Joined: 10 May 2019, 10:00
Antispam: Yes
Location: Cologne, Germany

Re: Contact Mechanics - Friction Test

Post by maike151 »

Thank you so much for your effort!

I changed everything according to your explanations, but now somehow no more contact is recognized and the contact surface does not hold the cube up, the cube penetrates the floor and just falls through it. I attached two pictures.
Attachments
Friction_test_step25.png
(230.33 KiB) Not downloaded yet
Friction_test_step1.png
(231.71 KiB) Not downloaded yet
kevinarden
Posts: 2237
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: Contact Mechanics - Friction Test

Post by kevinarden »

repost SIF and mesh, I will see if I get the same result
maike151
Posts: 41
Joined: 10 May 2019, 10:00
Antispam: Yes
Location: Cologne, Germany

Re: Contact Mechanics - Friction Test

Post by maike151 »

Thanks!
Attachments
case.sif
(4.93 KiB) Downloaded 219 times
mesh.zip
(91.65 KiB) Downloaded 210 times
kevinarden
Posts: 2237
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: Contact Mechanics - Friction Test

Post by kevinarden »

I had the same result until I deleted this line, then contact started working

Projector Max Distance = Real 1.1e-6
contact.png
(96.97 KiB) Not downloaded yet
kevinarden
Posts: 2237
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: Contact Mechanics - Friction Test

Post by kevinarden »

Further investigation it still isn't contacting, although the output from ElmerSolver started reporting contact pairs found for the projector, the cube still fell through the base.
kevinarden
Posts: 2237
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: Contact Mechanics - Friction Test

Post by kevinarden »

it appears to be contacting but the cube is still falling through the base
output 1 with the Projector Max Distance = Real 1.1e-6
output1.txt
(1.76 KiB) Downloaded 210 times
output1.png
(55.58 KiB) Not downloaded yet
with it deleted
output2.txt
(1.79 KiB) Downloaded 200 times
output2.txt
(1.79 KiB) Downloaded 200 times
maike151
Posts: 41
Joined: 10 May 2019, 10:00
Antispam: Yes
Location: Cologne, Germany

Re: Contact Mechanics - Friction Test

Post by maike151 »

Thanks again for your effort and sorry for the late answer, I had to call in sick.

So I tried the same thing and got the same output. I am really confused by this, because once contact is recognized, usually the non-penetration-condition of contact (contact gap >= 0) should hold, shouldn't it? However, until now I cannot find it anywhere in the Source code. Before I started this test, I once tried to displace a cube in z-direction ONTO the floor instead of along the floor and it worked well, it did not penetrate the floor but was pushed into the floor and deformed it (see picture in attachment). So maybe it is a problem of the tangential contact?

I also tried to give the Solver a little bit more time to recognize the contact by only prohibiting the displacement of the cube in z-direction for the first 0.05 seconds by using the "Displacement 3 condition" as variable of time:

Displacement 3 = Real 0.0
Displacement 3 Condition = Variable time
Real
0.0 1.0
0.05 1.0
0.06 0.0
3.0 0.0
End

So my intention was that the cube cannot fall in the first 0.05 seconds and is "released" after that, but somehow this doesn't even work.

Another idea would be that contact only works well if displacement is the "cause" for contact and it does not work with forces being the cause of the movement of the objects, because in the Source Code it says
! The variable to be constrained by the contact algorithm
! Here it is assumed to be some "displacement" i.e. a vector quantity

and also all the output variables are called "displacement contact load", "displacement contact active", and so on...

Sincerely
Maike
Attachments
Normalcontact.png
(265.68 KiB) Not downloaded yet
kevinarden
Posts: 2237
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: Contact Mechanics - Friction Test

Post by kevinarden »

I took out the lateral motion of the upper cube and just let it drop due to gravity and had the same result. It is confusing. I have gone through all of the test problems and did not find anything helpful. There is no documentation of the options to reference.
maike151
Posts: 41
Joined: 10 May 2019, 10:00
Antispam: Yes
Location: Cologne, Germany

Re: Contact Mechanics - Friction Test

Post by maike151 »

I changed "Normal-Tangential Velocity = Logical True" to "Normal-Tangential Displacement = Logical True" in both the Master and the Slave BC and deleted "Normal-Tangential Velocity Rotate = Logical False" and now the cube does not fall through the floor anymore and contact is recognized! But unfortunately, the output data is still the same for friction coefficient 0.2 and 0.01.

In the source code there are the lines

[...]
Fstatic = Fstatic + mustatic * ABS( NodeLoad )
Fdynamic = Fdynamic + mudynamic * ABS( NodeLoad )
Ftangent = Ftangent + ABS( TangentLoad )
[...]

(NodeLoad being the normal load)

and I would like to output exactly Fstatic and Fdynamic. How could I get this?
Post Reply