Discussion Forum

Hi

Sorry but I do not fully understand your model, could you pls precise a little...

if you have a rectangle in 2D axi, I assume parallel to the Z (vertical) axis, so you are either simulating a full cylinder (left vertical edge is on the axis r=0), or a hollow tube = left vertical edge is to the right of the axis hence r>0, and the right vertical edge is somewhat with r>>0.

You fix the right (vertical ?) edge, OK, you apply a -Z (?) force on a point on top of the left (vertical?) edge (not the best practice, it forces a "point" singularity, but depending on what you want it might well work OK, so why not). The refernce to the canteliver I do not catch, but OK.

To be sure your 2D axis is as you expect, have you tried to do a "revolve 360 " and generate a 3D geom of it ?

Now you want a plot of force applied versus displacement of a point (?), or an average displacement of the edge (?) (and not the by-default prepared one: displacement versus parameter (and then edit the plot legends to transform param to force)

Is it so ?

For the last point there are a couple of threads already on the forum discussing this.
Clear, to use Matlab is one way to play as you want with the data.
Another is to make a default plot "displacement versus Parameter", save it as ascii data and use EXCEL or matlab again, on the data.
The last one is to make your displacement a global variable, i.e. an "integration coupling variable" such as W=w for a point or you integrate the length on an edge as Length=1 and then the average "Z" displacement as W=w/Length over the edge. You use then a Postprocessing domain plot parameter and specify Param as variable and W as "X-axis Expression. Note: you do not get a "line" on your plot, these are only small dots as the lines are perpendicular to the paper, so you should turn on the line markers in the plot preperties , and increase the line width. You can also use (better) a Global variable plot, set Param as the expression and W as the X axis expression, this time you get a "line" and you have the force versus displcement, you could add the tick marks and edit the legends in the plot mptoperties palette.

do not forget that in 2D axi the horizontal "x" direction is renamed "r" with displacement "uor" and the vertical "y" direction is renamed "z" with displacements "w", this is sometimes forgotten

Hope this helps
Have fun Comsoling
Ivar

Hello,

I have attached my model files to explain myself better.

I am using Hyperelastic material models.
As you said, this would be a hollow cylinder problem. A membrane to be precise, with very low thickness.

Now, referring to the model (model_load.mph), say I solve it and I want to plot force vs displacement at point (say) (r,z)=(7e-3,0). How can I do that?

Also, can I do a similar plot for the model (model_disp.mph)?

Thanks
Aseem

Hi

A couple of comments on your model_disp:

1) understand you enforce a boundary condition bloxing X and increasing Y wih a parameter sweep

2) in the "Solve - Solver settings" or "Physics - Properties - Analysis type" your model is still "static", I would have used parametric since you select the parametric solver and want the "parameter" to be used for the solutions.
3) you have the reaction forces you can measure on the fixed part, (integrate over the edge, just check that the units are correct an that you use, if appropriate the 2*pi*r multiplier (Compute surface integral), does not apply to the RF but to the others
4) as you have an emforced displacement you can also calculate the reaction forces over the "inner vertical edge", just hit the "plot" to see the force versus diplacement (adapt your Para or change the axis legend (plot palette). If you want to invert the graph you need to calculate the surface Traction force as a gobal variable (integration coupling variable i.e Taz = Taz_smaxi*2*pi*r and Solve - Update Model) "Postprocessing - global Variable plot" Expression= -0.015*para, then ">", in x-axis data select Expression and typr Taz finish with an "apply"
Note: do not use reaction forces (reacf() or the RF...) in the integration coupling variables
5) you have really small steps there, I'm not sure you need all of them, try starting with just a few, its much quicker to sole, then you validate at the end for detailed results
6) turn on the "postprocessing - General tab - Deformed mesh" to "see" the deformations

Hope this helps
Have fun Comsoling
Ivar