Instability type 3 and large deflection for hollow slabs with linear releases

[Jummybear]

Hi,

In my simple model I get instability of type 3. No loads applied yet, only DL. Tryed just for test with some horisontal forces, displacements seems to be ok. Can you confirm that I can ignore this jumping from one point to another instability of type 3? Model attached.

Thx.

@Jummybear Artur Kosakowski edited the topic for better findability

[Rafacascudo]

you can do it yourself. Create a load case with a unity load apllied on the reported instability node.
If it is a translation xyz , apply a concentrated load on the correspondent direction. If it is a rotation , apply a moment.
Run it. If the displacements are huge , then you have a problem that needs correction

[Jummybear]

Something wrong with my middle hollow core slabs on 2/3/4 floor, but I cant figure out what...

Displacements are not huge but quite big, much more than expected (~150mm with less than 6m span). I think the reason is hollow core slab but I have no ideas at the moment.. I will be very aprreciated if you have a look or someone from autodesk.

PS If I cut internal walls then displacements fly to 2500mm 🙂 Looks like middle panels dont work appropriately. Question is why? Panel orthotropy direction is Y which should be right one.

[Artur.Kosakowski]

Try to switch off disregarding stiffness of transversal bending for ortho panels with defined linear releases.

If you find your post answered press the Accept as Solution button please. This will help other users to find solutions much faster. Thank you.

[Jummybear]

I cant. We were discuss it in neighbor topic. The loads in this way will transfer not like it should be in reality. I add one pic, simple model, 2 plates, one with option (dstb) on and another off. Results in bending moment of beams is about 2,5X different.

[Artur.Kosakowski]

I see. In such case you may consider deleting the linear releases (as you already reduced the corresponding stiffness in the definition of the panel thickness). There will be no instability and the bending moments at the previously releases edges will be small.

If you find your post answered press the Accept as Solution button please. This will help other users to find solutions much faster. Thank you.

[Jummybear · Accepted solution]

I have already tried that. That doesnt help with "problem" places. It reduce deflection, but only a little bit. 122mm is quite big number, but why this place is such dramatic I still cant realise. Similar places on opposite side as on the left and right side with beams are ok, deflection in 10mm, not more.

I will continue my invistigation, but I have to move forward with my project. I found only one solution, or better say another solution for modeling. Its a modifed panel calculation mode: NO FE, Stiffening diaphragm with partial stiffening in XY, transfer of loads in one direction. Then results looks like closest one to those which i expect.

But case with FEM hollow core slabs is unsolved for me...

[Artur.Kosakowski · Accepted solution]

I'm not sure if I understand your concept of the core slab strong and  no stiffness directions.

If you find your post answered press the Accept as Solution button please. This will help other users to find solutions much faster. Thank you.

[Jummybear]

Feel ashamed... Thx Artur.

I thought that ortho dorection is according to local axis. Thats why I was turned my middle panels to the walls... As usual, try to find mistake in simple things, I couldnt figure it out. Thx again!