Strange Slab Deflection

[toha_mh]

Hi,

Please see the attached model. Here the deflection of the lower story slabs are 18-22 mm, which I guess is all right. But as I go higher the slab deflection increses with a maximum deflection of 49 mm on the 19th floor. Isn't it strange? Why is this happening? I tried to delete all the loads and applying new loads of DL=3KN/m2 and LL=4KN?m2 for all the slabs. Still the same problem. Why the deflection is increasing upwards? Did I do any mistake?

[Artur.Kosakowski · Accepted solution]

This  is caused by shortening of columns and walls that support slabs. Checking the displacement of nodes of the bottom slab you can see the effect of shortening of one 'level' of supports (columns and walls defined below it) whereas for the 19th story this effect increases as there is 18 levels of walls and columns that get shorter.

For this reason the RC Slab module checking the deflection of slab includes this effect and calculates the deflection with the respect of the relative reference level being the smallest vertical displacement of its supports.

I think that for calculation of deflections of upper slabs of multi story building you should consider use of the sub-structure that includes the selection of the slab you want to design plus columns and walls defined directly below and above it. Applying boundary conditions to the 'outer' ends of these walls and columns you will be able to calculate results for vertical loads without the effect of supports' 'shortening'.

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.

[peter-rc]

Hi Artur,

I also have same problem.

my project consists of 10 floors, but i seperated 2 floors and created a new small model. I checked the deflection from small model, and they were ok. then I added one more floor to my small model but deflections were become 2-3 times bigger then old model. I did not assign any new load. Do you also think, reason is "shortening" or what it can be? if yes, could you please explain the solution more details(screenshot etc.)

thanks

[Artur.Kosakowski]

I'd expect such effect for cracked deflection when you scale the results obtained for linear static analysis but as far as I can see you compare the results of elastic displacements. If so at first glance the difference in values seems a bit too large unless you assigned elastic soil under the 'bottom' panels. In such case you may have much larger vertical displacements in areas directly under the walls that support floors comparing to the other locations. If that's the case try to define rigid vertical supports and compare results again.

If that is not the case I will need to look at both models.

[peter-rc]

thanks for your quick responds. yes, I compared elastic displacements

I've already defined rigit vertical supports. (but  I also get deflection in column zones, it is interesting   )

here my models..

Trinity_R1_12 with 2 floors

http://www.2shared.com/file/2-nybRKV/R1_12.html

Trinity_R1_13 with 3 floors

http://www.2shared.com/file/FqXc2x_8/R1_13.html

[Artur.Kosakowski]

Duplicated self weight load records in the model 13.

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.

[peter-rc]

Artur,

thanks it works, now deflections are ok...

also do you have any idea about deflection in column zones, with rigit supports it musn't be...

[Artur.Kosakowski]

Which node of the column to check the displacements for? Which load case?

[peter-rc]

Hi Artur,

it has been long time, I wish you didn't forget my problem...

here some screen viwes to explain the situation.

to remind the situation; I had assing fixed supports to columns and walls. but I received displacement in Z(global) direction. Do you have any idea about it?

thanks in advance

[Artur.Kosakowski]

To make sure I understand the pictures correctly. The vertical displacements at the bottom of a chain of columns is 0 and then its gets larger towards the top of the chain. If so that seems to be the effect of shortening of the columns under the applied load.

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.

[peter-rc]

hi,
also this is the reason of deflection in column zones in slabs?
is there any way to ignore "shortening", or not to take in account?

thanks...

[Artur.Kosakowski]

For the whole model there is not or at least I can't think of any easy solution. You may consider definition of elastic support with very small KZ value and define imposed upward displacements that correspond to values of 'shortening' under self weight for each of them or you can save the story as a substructure (separate model) and use such model for calculations of deflection of slabs for vertical loads.

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.