Hello All,
I am modeling a dome roof for an aboveground storage tank. The model consists (very generally) of arching beams pinned at both ends with a plate connected to the top flange (both bracing and as a liner for a concrete roof).
I feel that I am having some issues with the compatibility between my beam elements and plate elements. When I output the results, I see that there are discontinuities at each of the nodes along a beam. At any given "j" node of a beam, the value for shear force, in particular, does not match with the corresponding "i" node of the subsequent beam. This is essentially giving me a "saw tooth" effect, where there are two different values of shear at any given location on the beam. I have noticed that if I reduce the stiffness of the plates to 0% Young's Modulus that this effect goes away. I also speculate that the arching geometry of the beams plays a role. The magnitude of the loading may or may not magnify the discontinuities. Has anyone see this type of output for a beam-plate connection?
I believe this issue arises due to the fact that the beams are modeled in STAAD as quadratic/cubic elements, whereas plates are linear elements. This is just my thought, and may not be the only reason or the reason at all. My question(s) are therefore whether or not the plates can be modeled as quadratic/cubic elements or if the beams can be modeled as linear elements? If I am able to match the element types and see the saw tooth effect disappear, then I know the solution to my issue. Is my thinking correct that there are compatibility issues between the beam and plate elements?
Please feel free to respond with your thoughts. I welcome and appreciate all feedback.
Thank you,
John