# Modeling of Friction – Coefficient of Friction

**URL:** <https://community.dlubal.com/t/modellierung-von-reibung-reibbeiwert/3326>\
**Category:** Product Support\
**Tags:** modeling, friction, laminated-timber\
**Created:** [April 15, 2026, 12:18pm UTC](https://community.dlubal.com/t/modellierung-von-reibung-reibbeiwert/3326 "2026-04-15T12:18:26Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![AlexITI](https://community.dlubal.com/user_avatar/community.dlubal.com/alexiti/32/2125_2.png) [@AlexITI](https://community.dlubal.com/u/AlexITI)\
**Post date:** [April 15, 2026, 12:18pm UTC](https://community.dlubal.com/t/modellierung-von-reibung-reibbeiwert/3326/1 "2026-04-15T12:18:26Z")

</div>

Topic Friction in RFEM6,

We are currently conducting orientation tests to determine the coefficient of friction of CLT elements.  
I am modeling the physical tests in RFEM6 and trying to obtain the measured values correspondingly through the numerical approach.  
I encountered the following problem:  
V1) As a multilayer surface model, friction representation works quite well, but I do not get any information about the shear stresses in the core layer. These are of particular importance due to roll shear failure of individual CLT lamellas.

V2) As a volume model, I do get the shear stresses, but I cannot represent friction through the type "surface contacts" because both friction surfaces lie in one plane, causing the corresponding error message to appear.

Can someone help me with this problem?

 ![Screenshot 2026-04-15 at 14.03.48](https://community.dlubal.com/uploads/default/original/2X/6/62dc08abd5d3e7c109cc422eaac29918555d4a07.jpeg)

---

<div class="post-metadata">

**Author:** ![eike.hartmann](https://community.dlubal.com/user_avatar/community.dlubal.com/eike.hartmann/32/94_2.png) [@eike.hartmann](https://community.dlubal.com/u/eike.hartmann)\
**Post date:** [April 16, 2026, 3:03pm UTC](https://community.dlubal.com/t/modellierung-von-reibung-reibbeiwert/3326/3 "2026-04-16T15:03:13Z")

</div>

Thank you very much for your detailed description.

The behavior in the volume model is related to the fact that surface contacts in RFEM cannot be defined if both contact surfaces lie exactly in the same plane.

To circumvent this, there are two possible approaches:

- Introducing a minimal geometric distance between the contact surfaces
- Alternatively, using [surface releases](https://www.dlubal.com/de/downloads-und-infos/dokumente/online-handbuecher/rfem-6/003587)

From my point of view, the approach with a very small distance is preferable. This allows targeted control of the decoupling. When using surface releases, the verification is somewhat more complex, especially checking whether nodes and lines have been correctly released.

I hope this helps you further. If you have any questions, please feel free to contact me.

---

<div class="post-metadata">

**Author:** ![AlexITI](https://community.dlubal.com/user_avatar/community.dlubal.com/alexiti/32/2125_2.png) [@AlexITI](https://community.dlubal.com/u/AlexITI)\
**Post date:** [July 8, 2026, 4:32pm UTC](https://community.dlubal.com/t/modellierung-von-reibung-reibbeiwert/3326/5 "2026-07-08T16:32:07Z")

</div>

Thank you very much for your reply - I have a follow-up question regarding friction in RFEM 6 using contact volumes

We have conducted a small test series with CLT cubes (10x10x10cm) that we currently want to recalculate in RFEM.

The FE model (Figure 2) also produces affine distributions, but with too low a load level (Figure 1).

Higher friction coefficients do not solve the problem (and otherwise do not represent the physical test).

The contact volumes consist of S355 steel and were currently chosen with a thickness of 5mm because relative displacements of 1.2mm are imposed, and the contact volumes would otherwise experience too large distortions. Excessive distortions lead to the error message "Jacobi matrix is not positive definite". The FE mesh was chosen with 1cm and a mesh refinement to 5mm was defined in the area of the contact volumes.

With this configuration, the model computes, but with too low system stiffness.

Could it be due to the thickness of the contact volumes (5mm)?

Reducing it to 2mm and adjusting the mesh refinement to 2mm leads to an increase in computation time by a factor of 30 – which I want to avoid…

Are there any ideas regarding this?

 ![Figure-1](https://community.dlubal.com/uploads/default/original/2X/b/b8442e4d5eca4bb21a00c15e4f087e6924220ebc.png)  
 ![Figure-2](https://community.dlubal.com/uploads/default/original/2X/f/fe4d19e516979c1229e18ad4307fffb51ed209a9.png)