RE; ProfileSPDLinDirectSolver::solve() - aii < minDiagTol (i, aii): (73, -5.54971e-13)

[ryancwl95]

Hi admin,

While running a consolidation problem of a soil column using Linear elastic material and SSPbrickUP Element, I encountered this warning/problem.

Screenshot 2023-10-30 124850.png (8 KiB) Viewed 5207 times

Could you explain in detail what the warning means? and perhaps some troubleshooting measures. Thank you:)

[STKO Team]

it's saying that your model is singular.
double check your boundary conditions

[ryancwl95]

Hi,

Thank you for your reply.

When generating sp fixities in tcl file, and the input script describes (for the same node) 2 fixity inputs, which takes priority, or do they merge?

For eg,
fix 211 0 0 0 1
fix 211 1 1 0 0

When opensees reads the tcl file(example), does the fixity for node 211 follows the first input of "0 0 0 1" or the second input "1 1 0 0", or does merge to be 1 1 0 1.

[kesavapraba]

Please be aware that if you attempt to apply a single-point (SP) constraint (fixity) to a degree of freedom (DOF) of a node where an SP constraint is already specified, OpenSees will generate an error

Domain::addSP_Constraint - cannot add as node already constrained in that dof by existing SP_Constraint

In your situation, SP constraints should be combined since you try to apply the constraints at the DOFs that have "0". So, No error would appear and they should be combined together.

[ryancwl95]

Also, just an additional question,

Is there a function to reset the displacement before executing a new analysis? or is it automatic, i.e displacement of the 2nd analysis is independent of the 1st analysis.

[kesavapraba]

I am sorry that your question is not clear to me. To what state do you want to reset the displacement of the system, Initial? Anyway, you can use reset command to bring your system to its original state (although there are some limitations, please check the link below).

https://openseespydoc.readthedocs.io/en ... reset.html

Moving on from first stage to 2nd stage of the analysis (as you pointed out), the displacement and loads are carried forward and loads are maintained constant for the 2nd analysis if you enable loadConst with pseudoTime=0.