Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.
Friction in Frequancy Domain Studies
Posted 7 mars 2016, 23:39 UTC−5 3 Replies
Please login with a confirmed email address before reporting spam
Hi Everybody,
Has anybody applied friction force in a frequancy domain study before? I am trying to apply the friction using the friction subnode under the contact node. However, I am not sure if I do it right.
I use the penalty method for both friction and contact pressure analysis. However, the below error message always appear when trying to run the study.
"The dependent variables comp1.solid.cmX_p1_old, comp1.solid.cmX_p2_old, comp1.solid.cmX_p3_old, comp1.solid.cmX_p4_old, comp1.solid.cmX_p5_old, comp1.solid.cmY_p1_old, comp1.solid.cmY_p2_old, comp1.solid.cmY_p3_old, comp1.solid.cmY_p4_old, comp1.solid.cmY_p5_old, comp1.solid.cmZ_p1_old, comp1.solid.cmZ_p2_old, comp1.solid.cmZ_p3_old, comp1.solid.cmZ_p4_old, comp1.solid.cmZ_p5_old, comp1.solid.contact_p1_old, comp1.solid.contact_p2_old, comp1.solid.contact_p3_old, comp1.solid.contact_p4_old, comp1.solid.contact_p5_old are missing in the segregated solver steps. Add them to existing or new steps.
- Feature: Stationary Solver 1 (sol1/s1)"
Please help,
Has anybody applied friction force in a frequancy domain study before? I am trying to apply the friction using the friction subnode under the contact node. However, I am not sure if I do it right.
I use the penalty method for both friction and contact pressure analysis. However, the below error message always appear when trying to run the study.
"The dependent variables comp1.solid.cmX_p1_old, comp1.solid.cmX_p2_old, comp1.solid.cmX_p3_old, comp1.solid.cmX_p4_old, comp1.solid.cmX_p5_old, comp1.solid.cmY_p1_old, comp1.solid.cmY_p2_old, comp1.solid.cmY_p3_old, comp1.solid.cmY_p4_old, comp1.solid.cmY_p5_old, comp1.solid.cmZ_p1_old, comp1.solid.cmZ_p2_old, comp1.solid.cmZ_p3_old, comp1.solid.cmZ_p4_old, comp1.solid.cmZ_p5_old, comp1.solid.contact_p1_old, comp1.solid.contact_p2_old, comp1.solid.contact_p3_old, comp1.solid.contact_p4_old, comp1.solid.contact_p5_old are missing in the segregated solver steps. Add them to existing or new steps.
- Feature: Stationary Solver 1 (sol1/s1)"
Please help,
3 Replies Last Post 9 mars 2016, 02:06 UTC−5
Please login with a confirmed email address before reporting spam
Posted:
9 years ago
Hi,
Since a frequency domain study is a linear study type, it is not possible to use friction (or any other inherently nonlinear feature) there.
Regards,
Henrik
Since a frequency domain study is a linear study type, it is not possible to use friction (or any other inherently nonlinear feature) there.
Regards,
Henrik
Please login with a confirmed email address before reporting spam
Posted:
9 years ago
Thank you Hernik for your fast response,
Do I have then to use the time dependent study to simulate my HarmonicLoad problem?
I am only interested in the displacement amplitudes over a wide range of frequancies so the frequancy domain study seemed very suitable. Is there any backway to simulate the effect of friction in this type of study?
Another question please, I want to simulate a concrete cylendrical object inside a domain of an elastic material. However, this concrete object is surrounded by a very thin film of a weak material. The question is how to simulate this thin material film? If I used a very thin domain around the pile the problem may not converge due to the very small mesh size in this location relative to the rest of the model.
Do I have then to use the time dependent study to simulate my HarmonicLoad problem?
I am only interested in the displacement amplitudes over a wide range of frequancies so the frequancy domain study seemed very suitable. Is there any backway to simulate the effect of friction in this type of study?
Another question please, I want to simulate a concrete cylendrical object inside a domain of an elastic material. However, this concrete object is surrounded by a very thin film of a weak material. The question is how to simulate this thin material film? If I used a very thin domain around the pile the problem may not converge due to the very small mesh size in this location relative to the rest of the model.
Please login with a confirmed email address before reporting spam
Posted:
9 years ago
Hi,
Since you cannot get a harmonic motion if friction is present, the only solution is a time dependent analysis. Unless of coarse the friction force is sticking, in which case the surfaces can be considered as 'bonded'.
The the thin layer, check out the Thin Elastic Layer feature, see for example
www.comsol.com/blogs/structural-analysis-with-thin-elastic-layers/
Regards,
Henrik
Since you cannot get a harmonic motion if friction is present, the only solution is a time dependent analysis. Unless of coarse the friction force is sticking, in which case the surfaces can be considered as 'bonded'.
The the thin layer, check out the Thin Elastic Layer feature, see for example
www.comsol.com/blogs/structural-analysis-with-thin-elastic-layers/
Regards,
Henrik