We are not concerned with one point and instead we want to get an induced force to one of our parts in the model.Infact we want to draw a load-disp. figure for a part. To get an induced load on part, I tried to sum NFORCES for all nodes of the part but there is a problem with time increments for points and due to differences summation does not work. Even I changed the output to be in exact times( ex. 0.005 s) but again summation of NFORCES is not applicable for a part.
Do you know how to solve this? Any ideas? thoughts?
Regards
I think it won't give result to all part. you have to pick the nodes of your area of interest in the modeling. then plot the graph.
I don't know how you define the global displacement of a deformable part - do you mean the displacement of the center of mass? (But even then, you would have to think about changes to the CoM due to rigid body movement vs. deformation).
To get at global force curves, one way is to use the energy. If your part is deformed by a prescribed displacement, you can plot the variable ALLWK and then use the equation dW/ds=F to get at the momentary force. This is usually much faster than summing over lots of nodes.
Dear laxmikant,
I tried it and faced some problems which I mentioned in my ques. details.
Thanks,
Dear Martin Baeker,
I think your answer was really interesting but I have to evaluate the results and see whether they make sense.however the concept is really interesting. Thanks.
About your answer's first part, I should add that, my part is deformable but as the loading is not severe the deformations inside the part is negligible so the displacement for every point is satisfactory and representative of my part, but of course this displacement and induced load should be in specific direction( Golabal direction).
I will get back to you soon having evaluated the results,
Thank you,
Regards
You can also simulate the clamping by introducing two external nodes and two Spider elements which connect the external nodes to two Groups of nodes belonging to the part (located in the region where the part is clamped). Then apply a force/displacement to one of these external nodes and a constraint to the other one. After the simulation plot the force displacement curve of the movable external nodes. It is a general approach that works fine.
Dear Martin Baeker,
I tried to use your idea and I have to inform you that in some special circumstances it works well however in some other it does not. As far as I realized, for a part with constant disp. for all nodes( like sliding) it can be used as the ds amount is easily obtainable but in other type of displacements where different lateral movements are present for a part in each increment of time, we would face some difficulties in determining ds value.( This was what somehow what you had in mind at first)
Now I am trying to get the average of accelerations for all nodes to get the force. what is your idea about?
Trying to get the average of accelerations for all nodes to get the force.
any ideas about this solution? any thoughts?
- Badges
- Science topic