PROGRESSIVE FLOOR COLLAPSE FEA SIMULATIONS




I decided to take a break from the core FEA and fiddle around with a 60 foot truss, dropping it to see the effects. These are simple models without the bridging truss. The truss is resting on the wall brackets without bolts, gusset plate, etc. The first two videos are fully elastic.



YouTube VIDEO 1



YouTube VIDEO 2


The next two videos show the collapse with plastic or yielding included. Note, there were some issues with the meshing, that is why you'll see some of the connections "stretching". Again these are preliminary simulations. The next ones should include bridging trusses.



YouTube VIDEO 3



YouTube VIDEO 4

from this post



I suppose concrete fracture is out of the question... (only talking about one slab impacting another, probably need a lot of elements)
For now, it is out of the question. Although we should never have a direct slab to slab impact. The steel from the truss will interfere.

I wonder if it is possible to run this same simulation again but amend the initiation by unclipping one side at the top rather than both. It would certainly resolve some issues.

It is something I would like to explore as well, but as will be shown, I need the simulation to drop a few more floors before any real meaningful conclusions can be drawn.


With that said, the next simulation shows what would happen when one floor drops onto the next in a purely elastic form.



YouTube VIDEO 1

YouTube VIDEO 2

I took data from the top of the upper floor.



Displacement:

Large Image

Velocity

Large Image

Acceleration of one point


Large Image



The next simulation incorporates plastic deformation or the yield in the steel.


YouTube VIDEO 3



YouTube VIDEO 4

Again, I took data from the top of the upper floor.


Displacement:

Large Image


Velocity


Large Image


Acceleration of one point


Large Image


The round area is where the model exited with an error (element distortion). It was longer than my first run with a much coarser mesh. This model took nearly 10 hours just to get to this point. I am going to need to look into using beam elements to speed up the process.


Large Image


Also note, I rested the truss on the ledge. If I were to bolt them in place and add the upper welded gusset, the times will only increase.


Large Image

from this post