Cantilevered concrete slab benefits 2

[milkshakelake]

In a concrete flat slab building, is it good to place columns a few feet/meters away from a slab edge (if the architectural layout allows it)? I believe that I read somewhere that it'll reduce the unbalanced moment. I also personally think it helps with punching shear. But I just can't find the textbook or code reference where it said that.

It's also something I don't see any other company in my area doing, like WSP/DeSimone/others, so I might be making a fundamental mistake in this assumption.

 

[KootK]

1) Yes to a load balancing benefit but I've found that pretty inconsequential in most cases.

2) Yes to a punching shear improvement if the offset is enough that we consider the usual three sided punching perimeter to have been converted to a four sided perimeter. Usually this is around [4 x d]. ACI says something about this but I'm fuzzy on the exact multiple of slab depth.

3) While the offset has benefits, it's not as though it's a "must". My experience of work in NY is the same as yours regarding column placement and I think that may be a function of the incredible value of square footage there. That creates a big incentive to bury as much of the columns as possible in the exterior walls.

 

[milkshakelake]

@KootK Thank you! This is very helpful. I agree and I think that square footage trumps all other concerns.

 

[rapt]

Cantilevers of about .3 - .35 * internal span will give the most economic structure. And solve many punching shear problems with increased punching perimeter and less moment transfer.

It is a matter of convincing the architect. And that gets down to overall building layout. If the cantilever was 3m long, the column is normally far enough from the edge to not worry the architect as much.

 

[Trenno]

I've seen and applied this strategy, especially if you're locating blade columns within partition walls perpendicular to the slab edges.

In my experience, all your really aiming for is to hit the resi default 200mm / 8in PT flat slab and arrange the columns to suit.

 

[KootK]

Another thing to consider is that, while a healthy setback behooves slab strength and punching shear, it may have a detrimental impact on slab edge deflection. Few things are worse than having to fend off a cantilever popping up in the air when you're trying to keep cladding plane deflections < 3/8". In this respect, I prefer true edge columns as your cladding deflection becomes reduced to just that which occurs along your slab edge strips.

 

[slickdeals]

@Kootk - Not to forget drainage. There was an article on this subject a while back.

https://www.structuremag.org/wp-content/uploads/C-PractSolution-Rogge-June121.pdf

 

[rapt]

If you really want an edge column because of facade connections, using an end span length about .8 * internal span length will give similar efficiency in strength and deflection, as well as reduced load and moment on the edge column to reduce punching shear problems. My preference has always been for an edge beam also to stiffen the edge and help punching shear but many builders do not like this detail these days.

 

[milkshakelake]

@rapt That's a great rule of thumb. In case of a 3m cantilever though, I'd be worried about deflections. I try not to go more than like 1.7m for typical slab depths of 200mm.

@Trenno Glad to hear that someone else has applied this strategy. I'm in the same condition of partitions perpendicular to the slab edge. I work in an area where PT is expensive so nobody does it, but I think the same idea applies.

@KootK Agreed, too much cantilever will pop up the facades too much. I can probably design something between the pop up cantilever and sagging. FEM can make good work of it.

@slickdeals That's pretty concerning. I haven't considered dead load only in long term deflection. Now I'm wondering about the projects where I didn't do that. Thanks for bringing this up.

 

[HTURKAK]

Dear milkshakelake (Structural),

I did not look the previous posts and i hope my post will not be duplicate of one of them..
IMO, the best practice for flat slab systems is, the use of dropped beams at perimeter rather than placing the columns a few feet away from a slab edge to get some cantilever action ..

The advantages of perimeter beams are ;
- Punching risk at perimeter columns will be eliminated,
- The deflection will be limited so the facade wall will not suffer..
- Better , enhanced frame action..
- Better space use..

ETC..

 

[rapt]

milkshakelake

You would not have a 3m cantilever with a 200 slab based on .3 * internal span if you used sensible L/D ratios. That would be a 10m internal span. With a flat plate slab, for PT the slab would be about 250-275mm. For RC 350+mm for 10m internal.

 

[dik]

I had one project about 30 years back in Vancouver where the architect did not want to see the columns exposed on the exterior... 15' cantilevers for a parkade in a seismic zone... one of the two most difficult projects I did... had to use the perimeter upstands to stiffen the plate for seismic response...

Rather than think climate change and the corona virus as science, think of it as the wrath of God. Feel any better?

-Dik

 

[milkshakelake]

@HTURKAK Nobody in my area (NYC) uses perimeter beams, it's all flat slabs. It might be a different market from yours.

@rapt Not using PT though. It's too expensive here.

@dik I didn't really understand. Can you explain it again? I know you're a superhero in these forums, but you need to dumb it down for people like me.

 

[MSUK90]

had to use the perimeter upstands to stiffen the plate for seismic response...

I always wonder how load transfer happens if we give up-stands(which are generally without support). Can you shed some light on it please?
Also, how did it help in seismic response?
Would also like to look at floor plan and sections of the design you did of that particular structure..

 

[dik]

It was a 6 storey flat slab parkade; I don't recall what the interior was now, but there were no exterior columns... the slab was stiffened with the spandrel barrier and cantilevered 15'... had a heck of a time analysing it for seismic. It helped with the seismic response in forcing a mode of vibration that I could analyse... so far back; I don't recall the particulars... but, it was difficult and prior to desktop computers... I did it before I was married, so over 40 years ago... I think electronic calculators were in their infancy. I have no recollection about what the core was, other than it was central and significant.

Rather than think climate change and the corona virus as science, think of it as the wrath of God. Feel any better?

-Dik

 

[HTURKAK]

Yes.. I am at a different region with different traditions.. But still perimeter beams could be an option.
If the façade system / walls are sensitive to deflection , you should think about it.

Providing a cantilever portion at all perimeter will increase the behavior and punching capacity. However, the cantilever width should not be more than 5-6 ft. or 1/4 of interior span.

 

[milkshakelake]

@dik That is absolutely terrifying. I'd be anxious to design a 15' cantilever.

@HTURKAK Spandrel beams is not an option because nobody else does it. I agree that cantilever shouldn't be more than 5' or 6', anything more makes me physically uncomfortable.

 

[MSUK90]

It was a 6 storey flat slab parkade; I don't recall what the interior was now, but there were no exterior columns... the slab was stiffened with the spandrel barrier and cantilevered 15'... had a heck of a time analysing it for seismic. It helped with the seismic response in forcing a mode of vibration that I could analyse... so far back; I don't recall the particulars... but, it was difficult and prior to desktop computers... I did it before I was married, so over 40 years ago... I think electronic calculators were in their infancy. I have no recollection about what the core was, other than it was central and significant.

Thanks for your reply dik.
Would have been lovely if I could have hands on that design.. But, ahhh, luck...