Is it possible to obtain the statically indeterminate parts and the statically determined parts of the prestress by integrating with the Sofistik Decreator? The “normal” result is the combination of both parts.
I know that it is possible with SIR, but I wonder if its possible with the Decreator too
Hi Tobias,
no, this is not possibly directly. But is should be possible when you use load cases 15xxx and 16xxx from CSM, attached you find a little example.
But be aware: there are some limits when working with Design-elements, see also attached pdf-file.
Best regards,
Jost
csm31_design.dat (42.8 KB)
csm31_design.gra (1.7 KB)
Bridge_and_Design-Elements_english.pdf (84.4 KB)
Many thanks for your response. I wasn’t very specific last time. I need the statically indeterminate parts and the statically determined parts of the prestress for quad-elements. After that I want to integrate with the decreator. Is this possible? I don’t manage to implement this in csm for Quad-Elements.
Best regards,
Tobias
Hi Tobias,
the statically determinate part of prestressed quad elements can be determined for simple systems using a trick: by calculating the pre-stress on the statically determinate system. I have done this in the new appendix between lines 626 and 657. There are no quad elements in this system. If there were any, the two components would still be present separately in the two load cases 25011 and 26011.
But beware: Because a tendon in a quad element simultaneously loads adjacent quad elements, the statically determinate component is not necessarily the prestressing force multiplied by the lever arm to the center of gravity.
This example only works fine in version 2024!
Best regards,
Jost
Thanks a lot. It works fine.
But I have one last question to this part of yours:
“But beware: Because a tendon in a quad element simultaneously loads adjacent quad elements, the statically determinate component is not necessarily the prestressing force multiplied by the lever arm to the center of gravity.”
It’s true. I have a slightly higher statically determinant component than expected, although the bridge have an oblique angle, so it’s quite difficult to verify this by hand. Is it a problem to use the sofistik output or do you simply want to warn me to not be concerned if its not the exact same value and I can use the result for further calculations?
Hi Tobias,
I am trying to point out that the proposed workflow cannot provide such a clear division into statically determinate and indeterminate parts as you would get in a system with bars only. However, this is in the nature of things: a quad element that is part of an overall system can no longer be calculated as “statically determinate” as the complexity of the overall geometry increases.
Best regards,
Jost
You can get statically determinate part of prestress by running a loadcase without any supports but with weak dynamic stiffness:
STEP 1 dt 999
Why dt 999? It leads to slightly higher results than dt 1, which is Jost’s recomendation. I can’t understand the difference.
Why you use dt 1 instead of dt 999 and whats the difference?
Hi Tobias,
STEP N… DT… is normally used for dynamic-time-step-calculations. In our case we can use it to calculate the statically determined prestress because: when we deactivate all supports we get this statically determined prestress. But in a static (not dynamic) calculation the system must be supported, otherwise you get an error message. And therefore we use a dynamic calcuation with STEP N… DT…
But the hint from mico is sensible and correct: with STEP… the prestress is actiavted suddenly and all masses are moved in the system for only 1 second which is not enough time. With STEP 1 DT 999 the masses have 999 seconds to find there final position in the prestressed system. In this long period the system has enough time to swing from the sudden activated prestress and gives you at the end correct values.
Make a comparism with DT 1, 5, 10,… There will be a value where the changings are only very small.
Best regards,
Jost
1 Like