Discussion Forum

Hello Ramgopal Balijepalli

Your Discussion has gone 30 days without a reply. If you still need help with COMSOL and have an on-subscription license, please visit our Support Center for help.

If you do not hold an on-subscription license, you may find an answer in another Discussion or in the Knowledge Base.

Hi,

Have you find answer for your question? I also have a same problem and I would be happy to know how to consider adhesion in contact element of surface.

Yours,
Marjan

Hello,

I haven't received any response till now. I even contacted the COMSOL support it was also not that helpful. It seems there are not many possibilities to deal with Contact modelling or Adhesion by using Comsol.

In my opinion there are a few people who are using Comsol to deal with Solid mechanics part and when we talk about Contact modeling, Comsol offers very less to deal the contact problems. This is what I understood as an intermediate user of Comsol and this is the reason for posting my query here and waiting for the response of professional users of COMSOL..

Note: There is a new feature called Thin Elastic Layer introduced in Comsol 4.2 version and it might solve the problem.

Best Regards,
Ramgopal

Hi,

I was wondering if the thin elastic boundary condition helped you in the problem.
I am doing similar research on adhesion of surfaces.
Do you have other approach?
Thanks in advance.

Best regards,
Andres

Hi,

Using a Thin Elastic Layer is the easiest method to simulate this. You can then enter the spring constant or the spring force as a function of the relative displacements between the surfaces. A Thin Elastic Layer can be used either on an internal boundary, or as a pair condition in an assembly.

The variable for the relative deformation between the two surfaces is named "<interface>.uspring<N>_<tag>", (e.g. "solid.uspring2_tel1" ). N is the displacement direction (1, 2, 3) and <tag> is the tag of this Thin Elastic Layer.

If it is a glue layer, then the stiffness of the (undamaged) layer would be given by the physical properties and thickness of the glue.

If it is more like "contact with adhesion", then the stiffness should be high enough to give negligible penetration between the surfaces at the maximum expected compressive load.

Note that the spring_force vs. relative_displacement function should be continuous in order to ensure convergence. Problems like this can still be unstable depending on the boundary conditions, since locally (at the crack) there is a "negative stiffness" as the force decreases with displacement. A displacement controlled model has better chances of success than a load controlled problem has.

This technique will usually be appropriate when the adhesive is torn apart. Since it is an elastic model, it is will follow the same path back in compression. This may or may not be physically acceptable.

Regards,
Henrik

hello henrik
i am trying to simulate the peeling of an ahdesive foil from a surface. colleagues of mine are running such simulations with ANSYS and ABAQUS, using the mentioned "cohesive zone model".

so it was very intersing to read your reponse in this discussion thread, it looks like there might be a way to do this also with comsol.

could you provide an example model that shows how to use the thin elastic layer approach for a glue/adhesive layer with known material properties?

- stefan

hello henrik
i am trying to simulate the peeling of an ahdesive foil from a surface. colleagues of mine are running such simulations with ANSYS and ABAQUS, using the mentioned "cohesive zone model".

so it was very intersing to read your reponse in this discussion thread, it looks like there might be a way to do this also with comsol.

could you provide an example model that shows how to use the thin elastic layer approach for a glue/adhesive layer with known material properties?

- stefan

Hi Stefan,

I have uploaded a simple 2D model (version 4.3b) illustrating the principles of how a cohesive zone can be modeled in COMSOL Multiphysics. The model is not solved, but it only takes a minute or so to do it (and the animation during solving is nice).

Important features are:

- The expression in the thin elastic layer and the constitutive function it references.
- The extra weak expressions added for keeping track of the maximum opening experienced in a certain point. This is not necessary in a case where there is no unloading but then the constitutive relation would have to be written in a different (simpler) way too.
- The setup of the solver steps.

The uploaded plot shows the stress distribution (red) and a damage parameter (green).

The model is run under prescribed displacement. The reaction force plot shows why this is necessary. The released energy during cracking/delamination is so large that it would be unstable under a prescribed load.

Regards,
Henrik

Hello Henrik,

I thank you very much for providing the model. Unfortunately I only have Comsol 4.2a and it would be very nice of you if you could provide a model file compatible with 4.2a.

Best regards,
Ramgopal

Hello Henrik,

I am trying to use thin elastic layer in modelling adhesive layer in piezoelectric devices to attach PZT to the aluminum plate. From the discussion and the information available about the thin elastic layer for me it looks like a fantastic approach but there isnt any tutorials and information on its use. i didnt go thru your attached model because i have 4.3 version of comsol. KINDLY SHARE SOME INFORMATION ABOUT THIS IN COMSOL 4.3. We need to understand this tool to use it for our work.

cheers,
Pramod

Hi pramod,

Mine is also a similar case. Please let me know if you find anything.

Many thanks.
mohammed.

I'd like to see this model; however, it fails to open in C4.4. I get the following error: "COMSOL Multiphysics model file is damaged or not valid." I'd appreciate a re-post.
Many thanks,
A

Hi Andrew,

I tried opening it in 4.4 and it works for me. It seems like some problem with your download then. If you cannot solve the problem, try asking support to send it.

Regards,
Henrik

Hi Andrew,

I tried opening it in 4.4 and it works for me. It seems like some problem with your download then. If you cannot solve the problem, try asking support to send it.

Regards,
Henrik

Hi Henrik,
i learn carefully about the model you uploaded. I have a problem as following:
what the "umax" is ? is it a internal variable in comsol?

Hi,

umax is a user defined field defined along the crack in Weak Contribution 1 -> Auxiliary Dependent Variable 1. It represents the largest opening that the crack was ever subjected to as a function of the position along the crack.

You can try adding a Line Graph along line 4 and plot umax. If you plot it for all parameter values you will see it growing. If you do it for parameter values 0.01 and upwards it has a constant distribution along the line, since that is the unloading phase.

Regards,
Henrik

Hi Henrik,

Is there a way to alter the weak contributions/auxiliary dependent variables to work in a time dependent problem?

Obviously there's no "Previous Solution" node that allows you to track the value of umax_old in a time dependent problem.

Thank you,
-Joe

Hi for all,
sorry if my problem is far from Comsol, but the theme of discussion is important for my work.
I'm newcomer in the programation with Ansys, but I 'm working in contact problems, adhesion, so my main work is to deal a contact adhesif between Plan/Sphere to get out Force/displacement, note that both materials are gold.
Could someone can help me to solve numerically with ansys?, because analytically the problem is not easy, so,
any help in this regard would be welcome.
Thanks in advance.

Hello Henrik,
I do not understand how you find the table graph of applied force.
Sincerely
Filippo

Hi,

The table graph of the applied force points to Table 1.

Table 1 is generated by the Line Integration node under Derived Values. That integration sums the reaction forces on the boundary where the displacements are prescribed, which is the applied load.

Regards,
Henrik

Hi Henrik,

Is there any other tutorials or more details about using the Thin Elastic Layer tool in modeling delamination. This will encourage more people interested in this type of problems to turn from Abaqus and Ansys to COMSOL. Currently there is no information.
Many thanks,
Peter

Hi Peter,

There is a model in the Application Libraries, called "Mixed-Mode Debonding of a Laminated Composite". In the Application Gallery it is available as www.comsol.com/model/mixed-mode-debonding-of-a-laminated-composite-19961 .

More important however: In the next release, both Adhesion and Decohesion will be available as built-in functionality.

Regards,
Henrik

Thanks Henrik for your prompt reply,
Yes, I actually found this model. This will be great in the next release.
Thanks a lot,
Peter

Starting at version 5.2a, the Structural Mechanics Module offers pre-implemented capabilities for modeling Adhesion and Decohesion. See www.comsol.com/release/5.2a/structural-mechanics-module .
Best,
Jeff

Hello Henrik.

How do I model decohesion due to moisture (drying effect).

I've worked on my model but the problem comes in defining a global equation to convert the hygroscopic forces into a displacement.

In this case I get stresses on the hyroscopic layer but there is no debonding.

Thank you.