PSI Convergence for Stiff soils

I am encountering the "System Error " error while I am running inplace analysis using PSI. In the lst file I see a number of warnings saying "PILE STIFFNESS CONNECTING JOINT 102P IS NEGATIVE OR ZERO IN THE DY-RZ DOF FORCE COMPARISONS" and error saying "ERROR-STIFFNESS MATRIX IS NON-POSITIVE DEFINITE FOR LOAD CASE 2501 "

The fact is that soil at the top layers is actually very stiff. I made a trial by removing the P-Y lines for the first 2 meters and the analysis gave no error. But that is not the realistic case for  me. I need to take the effect of the top layers.

Using the TABR lines didnt help.

Appreciate your advice to overcome this error.

Thanks in advance


  • Typically this type of error occurs when there is a soil failure and the soil is simply not stiff enough to prevent the pile from moving. This is usually fixed by extending the pile or increasing the diameter of the pile, but since removing soil layers improves convergence it seems that there is something else going on. It kind of sounds like there may be an issue with the transition from the stiff layer to the softer layer. What happens if you extend the softer layer to the mudline instead of the stiff layer? Does it still converge?