Dear Dlubal community,
I have a question regarding the modeling of friction in volumetric models.
In my opinion, friction works well in surface elements in RFEM 6. As I understand it, exceeding the static friction can be well represented by means of virtual nodes and visible node displacements in the friction surface.
I think friction works differently in volumetric elements. Since friction between two parallel surfaces lying in one plane cannot be generated via surface contact, I have resorted to contact volumes. In contact volumes, the connection of the nodes in the shear surface is established via the contact volume. When the static friction is exceeded, there is a displacement of the nodes, which is represented by a distortion of the contact volume.
Have I basically understood the integration of friction in RFEM 6 correctly?
In this context, I have a question regarding the correct input of the values for the contact volume. For example, I can define a static friction coefficient, but I must also assign the material a stiffness using the modulus of elasticity. On one hand, how does the modulus of elasticity affect the transmission of local stress peaks between volumetric elements (is a lower modulus of elasticity better so that local peaks are not absorbed), and on the other hand, how does it affect the transferable shear forces in the shear joint (is a higher modulus of elasticity better so that distortions are influenced only by the friction coefficient and not by the modulus of elasticity)?
Can someone help me regarding this?
Best regards
Alexander Gerger
Hello Alexander,
First, I would like to address the different types of contact modeling for the connection of surfaces: Here, besides the rigid connection, surface releases, surface contacts, and contact volumes are available. These differ in their approach and thus also in their effects on the structural system.
- Surface releases release the degrees of freedom under certain conditions where these share a location (overlap). We are currently working on the implementation of nonlinearities such as the friction contact described here.
- Surface contacts establish a connection of the degrees of freedom between surfaces with a gap.
- Contact volumes function geometrically like surface contacts. However, the stiffness of the volume from geometry and material is incorporated here.
Since friction between volumetric bodies should be assumed to occur at the contact surfaces, the influence of the contact should be minimized. This can be achieved by a very small gap and, in the case of a contact volume, with high material stiffness.
The typical use of contact volumes is, especially historically, the connection of two surfaces modeled at their centroids. The recommendation here would be to choose the material with the lower stiffness of the connected elements. Another use case is the connection using an adhesive/polymer, for example for volumetric bodies, where the material properties of the bonding material should be applied.
To test the influence of the stiffness of the contact volume on the friction connection, I have prepared and attached a small model. The same condition is modeled in two ways here: on the left by means of a contact volume and on the right by means of a surface contact. Additionally, the material of the contact volume was exchanged via the design situation. The deformation in the horizontal direction is shown in the following image. From left to right, the results with a low and a high modulus of elasticity of the contact volume, as well as their difference determined by result combination, are shown. As this shows, the influence is not negligible. In this example, the pure friction displacement is 1.85 mm, whereas the deformation due to the stiffness of the contact volume is 7.65 mm.
Test_KV-nr.rf6 (1.3 MB)
Best regards
Marc
Hello Marc,
Thank you very much for your response! This clarifies the influence of the stiffness of the contact volume. I have already tested the different options for modeling contact & friction in RFEM 6, but have not yet found a completely satisfactory solution.
-
Surface releases (do not yet have friction)
-
Surface contacts have a problem with the contact condition in my specific case, as the solids begin to slide into each other even at low load steps (0.14-0.16). Regardless of the number of load steps (tested up to 500). I have attached the model.
- Contact volumes allow the correct representation of contact through "prestressing" (no sliding of the solids into each other).
However, when transferring friction I encounter a problem. When I apply shear forces as a "displacement-controlled test" through imposed nodal displacements, I get results, but the static friction range (before sliding) cannot be represented correctly because I am forcing a displacement. I have attached the model.
In the case of a "force-controlled test," I get no results because my system becomes unstable. I am still troubleshooting this.
What would be your recommendation for this specific case? I've also attached a photo of the test setup again.
Best regards
Alexander
Reibversuche-Dlubal-Flächenkontakt.rf6bak.zip (1.4 MB)
Reibversuche-Dlubal-Kontaktvolumen-verschiebungsgesteuert.rf6.zip (2.5 MB)
