Flat Slab Load Dispersion Behaviour 1

[struggle67]

Hi All Good Day
If I visualize flat slab as two way slab with beams in two direction - column strips as band beams and middle strips as two way slab. First I consider, whatever load applied into the middle panel (highlighted in green) will be carry to the column strip (beams) by middle strips half load in each direction. Then column strip will carry the load to the columns. That is why I usually used 70/30 % moment distribution between column strip and middle strip for both mid span and at support. I feel right and easier to visualize the load dispersion for me. Am I right?

Or should I just stick to distribution based on stiffness approach as recommended by most codes. At mid span the moment distribution between col/middle strips can be almost 50/50?

 

[WARose]

Or should I just stick to distribution based on stiffness approach as recommended by most codes.

ACI gives you the option of using the "Direct design method" which is prescriptive. You have to meet certain criteria that they spell out for you (in terms of spans, # of bays, etc.). The alternative is something like a FEA.

Me personally, I've always found it faster to use a FEA approach.

 

[rapt]

struggle67

The column/middle strip logic is as per most design codes. But the distributions you suggest do not match FEM results.

The 75/25 assumption everywhere would assume that the "beams" (column strips) are rigid supports for the middle strip slabs, and they are not. It works that way for PT load balancing but not for distribution of load effects throughout the slab.

That is why design codes suggest 75/25 at supports or 60/40 (ACI and AS) or 50/50 (BS) for maximum positive moment.

 

[struggle67]

RAPT, WARose Thanks

In Eurocode2, it is allowed to go up to 70/30 distribution at mid span. That is why I was wondering.

How should I approach/visualize the distribution of load if there is a heavy concentrated point load at the middle of the panel(in middle strip)? Do I still need to carry full load in both direction?

Thanks

 

[rapt]

Some codes allow figures like that at ultimate strength,but they require redistribution to achieve the distribution.

That redistribution is not available at service, so you then have to design service and ultimate for different distributions.

And when you do crack control checks, you will have much more trouble satisfying crack control on sections moment has been redistributed from.

My preference has always been to reinforce the slab the way it wants to act and in my opinion that is the elastic response, so I try to use the elastic distribution for both service and ultimate.

 

[struggle67]

Hi Rapt,

Thanks I see now. 70/30 at mid span requires redistribution.

BTW any advice on heavy point load at mid span middle strip? At mid-span middle strip take 100% moment in both direction? How about distribution at support and column strip?

 

[rapt]

No,in the 2 way strips logic the middle strip takes 50% of the point load in each direction as a 2way edge supported slab where the edge supports are the column strips. This is "deposited" on the column strips and the column strips then carry into the supports.

This is the same logic as for the UDL for PT tendon distribution. The middle strips are 2way slabs supported by the column strips which then transfer the loads to supports.

But if you look at it on finite element software you will find that more of the negative moment would tend toward the column strip as it does for UDL loading. The effect of the concentrated point load tends disperse out wars as you move away from the point of application. I would use the same distribution factors as for UDL loading if the point load is in the middle strip. Maybe make the +ve moment distribution would be more like 50/50 than 60/40.

If the point load was in a column strip, I would put the full effect of that point load in that column strip only, with none in themiddle strip.