problem with multiple displacement load step

[jgallegosanchez]

I'm performing a two load step analysis. In the first step I'm pre-loading the structure at two key points (points 1 and 3) by applying a displacement condition. In the second step, once the structure is preloaded, I'm actuating the structure from a third key point (point 4), again by a displacement condition (no force is applied). The problem is the following, during the first load step, point 4 suffers some displacement from its initial position, now, during the second load step I would expect the point 4 would displace from its actual position (solution given by first load step) to the given displacement condition, but instead, ANSYS snaps the structure to a configuration where point 4 is again in the initial position and then applies the given displacement, while the pre-stress is still in consideration. This ANSYS files attached as well as a figure, where the displacement of point 4 is plotted against the pseudo-time. the upper figure shows the actual result, the lower figure shows, what I would be expecting and I'm not getting.

 

[Drej]

What you describe is exactly how it should occur. Remember that enforced displacement (using the "d" command) is always relative from a node/s initial (undeformed) position.


------------
See faq569-1083 for details on how to make best use of Eng-Tips.com

 

[jgallegosanchez]

Thanks Drej for your answer.

Do you have any idea, how can I make then, that the second load step applies the displacement condition, starting from the deformed position after the first load step?

Or this is just not possible?

 

[Drej]

In your first step, you apply UX displacements to KP 1 and 3, and since KP 4 is unconstrained in the UX it is allowed to move by -2.19 units. Then, you attempt to apply a displacement of -1.5 units to KP 4. Since you did not constrain KP 4 in the first load step, ANSYS will immediately bring the KP back to its original position and THEN apply the UX of -1.5 units. I don't know the details of your model, but if you want to bring KP from -2.19 to -1.5 without the snap back to zero, you will need to apply a displacement of -2.19 at KP 4 in the first load step and then in the second load step apply UX at KP 4 of -1.5.

As I say, I'm not sure of your model details. Maybe you need to consider applying a UX constraint of zero at KP 4 in the first load step, and then applying your actuation at KP 4 in the second?


------------
See faq569-1083 for details on how to make best use of Eng-Tips.com

 

[jgallegosanchez]

Now, I've run the analysis including a displacement condition of -2.19 for kp4 at the first load step, keeping conditions at kp1 and kp3, and kp4 is behaving as I want, but now the problem is that I'm getting a force reaction different to zero for kp4 at the beginning of the second load step.

Should be zero, don't you think? (because displacement of -2.19 is the displacement for equilibrium at kp4 just with pre-loading at kp 1 and 3)

Thanks again for your help!

 

[Drej]

Not really. This is a non-linear model and is therefore path dependent, so it is perfectly reasonable that the reaction is non-zero. How much is non-zero?

The important thing is that your boundary conditions represent those in your physical system.


------------
See faq569-1083 for details on how to make best use of Eng-Tips.com