I just learned working with RFEM 6. I'm using a pre-stress load case and i keep getting errors that nodes and nodes in meshes are instable.
When i seperately calculate the dead load(without initial state from the prestress) there are no stability issues. But when I add the initial state i get the error : The stiffness matrix is singular | The structure is unstable | FE mesh node No. 86, in direction Z, increment 1.
I played around with the speed of converge which this solves the issue of the error but then i get displacements of e+10.
It looks like the edge members of your structure are not connected to the membrane surface. As a result, the membrane has a free surface edge and cannot maintain the prestress along that boundary.
Free edges in a membrane model typically cause the surface to contract or “neck” inward, because the prestress has no structural element to anchor to:
I added cables to to the outer edges of the membranes. Now i still errors, only it is not in a mesh node but in a node which attached cables to a beam. Canopy Bus station.rf6 (6,8 MB)
How can i check if everything is properly connected?
Although you have defined edge cables, they are currently connected to truss members. In this configuration, truss members cannot transfer the loads further as intended, which may lead to unexpected structural behavior.
A good approach is the following:
Start by modeling the connected members as beam members.
Then, introduce member hinges where needed to represent the desired releases.
This allows you to control the load transfer more precisely and avoids unintended hinge chains.
You can find more details in the RFEM 6 Online Manual:
Free edges should not occur in this type of membrane model, otherwise the prestress cannot be transferred correctly.
