For first my model: axial column
I faced kind of problen. Watched the movie NL Buckling and tried to re-create the setup on different model.
So the case is that for the column I applied Axial Load = 20kN.
Linear Buckling sequence SOL105 (without lateral perturbation loads) identified the buckling factor of about: 0.613 (so critical load for the column = 12.26 kN )
Then I created two non-linear (SOL 106) solutions (with perturbation loads) and LGDISP activated.
First solution had applied Axial Load of 20 kN (as per linear solution). As per results it looks that 'something' starts to happen with load increment factor about = 0.47 (so at load about 9.4 kN)
Second Non-linear solution applied redued axial load = 20 kN * 0.47 , (LGDISP is ON) and analysis type in load-step is switched to Buckling with default Lanczos options.
And now, what is strange: all the results in second SOL106 solution had suffix 'sec' and no buckling modes were calculated. What is wrong with it ? What I missed ?