Loading according to the expected deformation

Replies to This Discussion

Permalink Reply by Vadim Shapiro on October 15, 2010 at 11:10am

My guess is that you are not getting any replies, because it is not clear what the question is. It sounds like you are trying to solve a difficult "inverse problem" by guessing which boundary conditions to apply so that you can simulate a particular deformation -- is this right? Generally speaking, this is a tall order. But if the problem is something else, maybe you want to provide more detailed information. What do you know about the restraints, the loads, and the solution from ANSYS? ... and what is it that you need to figure out?

• ▶ Reply

Permalink Reply by Lubos Polak on October 18, 2010 at 2:53am

Thanks for notice, i'll try to explain it better (but sorry for my English, i don't use it often).
I'm trying to solve a problem with tearing of the material, see on this picture:

Generally: we are a producer of some sports equipment (this is only our hobby) and sometimes we have a defect on it. This defect is in the same place, so we have to solve it. The method of "trial and error" is not good for us, so we found and contacted expert in mechanic. With this expert i agreed procedure, which can show us, where can be our problem and we agreed, that i try to do a simplified model and he makes the calculation - the result of this calculation is picture with output from ANSYS. But, because this procedure (let to calculate and finding an final solution) is too expensive for us (as i wrote, it is only our hobby), i let to cout only a few modifications and because Your SnS promises a simple procedure to get a result, i'm trying to repeat the same calculation with the same result, asi i can see in ANSYS.
The calculation in ANSYS is based on the measured strain on a physical model. The estimate deformations are on this picture:

Now i'm trying to load my model in Rhino to match the measured strain. Going back to the output picture from ANSYS, in my model should be the restraint in Z-axis deformation or loads and restraints should be such as to get the expected deformation.
The results from ANSYS i can use to comparsion and controll of output from SnS. When i'll be able to simulate this problem in SnS right (with the expected result), i plan to do further calculations with the modified model. After that we will try it in practice on the physical model.

So, my question is, how to set up a restraints and loads to get expected deformation - the biggest problem i see in restraint of deformations.

I hope, that my text is understanable.

• ▶ Reply

Permalink Reply by Vadim Shapiro on October 18, 2010 at 1:34pm

Lubos, thank you for explaining the problem. If I understand it correctly, you want to apply observed deformations (displacements) to the model and solve for stresses -- to see where the material may break.

Unfortunately, the current version of Scan&Solve supports only "fixed" restraints, which is to say that the displacements are zero at all indicated locations. It seems that you are trying to simulate the same conditions by guessing which force loads produced the observed displacements -- but this is not likely to work -- or at least, I am not aware of any method to do this systematically.

Future versions of Scan&Solve will allow more general types of restraints and non-zero displacements, but we have to figure out how to add this functionality without loosing simplicity and intuitiveness of the user interface. Thank you for your patience.

• ▶ Reply

Permalink Reply by Lubos Polak on October 19, 2010 at 2:18am

Yes, i want to apply observed deformations to the model, this is, what i need to figure out.

It is unfortunate that the current version of SnS isn't able to do it, but i know, that development needs a time, we experience the same thing in our work - we are trying to make the best product, we can do, we deal with this for many years and we still see, that there is many questions to be answered. So i keep my fingers crossed for further development, you are doing a great job!

If i can afford a note:
in the software used in construction (for example SCIA) we can define restranit with non-zero displacement or bending moment in a certain axis - if i understand correctly, this is the way, you want to go.

Maybe there can be other ways.
Maybe you can use the power of modeling in Rhino to define some "support unit", modeled in Rino, which would act as a strong barrier with with zero surface friction.
Or you could try to make another method of loading - i saw it in ANSYS. In ANSYS can be established a virtual point out of the model and the given surface is loaded with focus in this point.
But, as i wrote, i'm not an expert in this, maybe this my note is irrelevant ...

thanks

• ▶ Reply