Transfer column on a thickned slab 2

[slickdeals]

Folks,
One of my colleagues is designing a building where a column is offset by 24" from the column below (architects !!!!). This is an edge condition. They don't want a transfer beam and will only allow a thickened slab beam (20" thick).
Column below is a rectangular column (24x48) and column above is circular 18" diameter.

I was suggesting the use of an embedded steel beam to transfer the shear at the interface. The axial load from the column above is of the magnitude of 700-750K (ultimate).

Has anyone had a similar situation to deal with and if so, what details did you use?

 

[slickdeals]

@BA,
You the the picture correct. There is a 60" wide by 20" deep slab beam. The rest of the slab is a 10" post-tensioned two-way slab system

 

[slickdeals]

*you have the picture

 

[hokie66]

It looks like to me the band beam would just work, with a lot of shear stirrups and 6000 psi concrete. But I would still like to deepen the section under the column, say the first 60" from the outside face, to 24". Not too much to ask.

 

[hokie66]

Actually, that answer is only for the column load. We really need the total band beam reaction to know if it works or not in beam shear.

 

[BAretired]

This design is a little outside of my comfort zone. Considering one way shear of the 60" x 20" beam, the round column is almost totally within the distance 'd' from the edge of the rectangular column, so most of its reaction does not have to be included in beam shear by code.

For punching shear, part of the critical section overlaps the rectangular column, lessening the probability of punching shear failure but definitely muddying the waters of understanding.

Perhaps the strut and tie model is the best way to deal with the problem.

I think I would insist on some changes in geometry as suggested in earlier posts. The consequences of being wrong are simply too great.

BA

 

[rapt]

I do not think you could justify using all of the 60" wide band beam to transfer an 18" diameter column over this length (3" gap) to a 48" wide column below. I would find it hard to justify much more than the width of the column above, possibly 24" maximum.

I would tell the architect to use an extended circular column about, or a blade column, to get a reasonable overlap and take out the thrusts in the 2 floors as I mentioned above is more depth is not available.

 

[hokie66]

I disagree with you, rapt, as to the effective width of the band beam. The reaction is 48" wide, so I wouldn't have a problem with distibuting this over a width of 60" in beam shear. 60" bands are commonly supported on smaller columns than that. Punching shear has to be checked for both the circular column above and the rectangular edge column below, and those calculations are complicated by the proximity of the load to the reaction.

I agree that it is a strut and tie problem, but believe it can be resolved in one level if the architect comes to the party with a bit more depth in the band.

 

[apsix]

Looking at BA's sketch it would appear that the only way failure could occur is by crushing of the strut between upper column base and lower column top, or bending in the beam due to the offset. The strut is inclined therefore there must be horizontal forces to resist.

I can't see that shear over the whole beam width is relevant. I also don't think that it's a classic punching shear problem as about half the upper column lies within the critical shear perimeter.

 

[hokie66]

Strut failure will not be an issue, as the strut is confined within the 60" wide band. Anchorage of the bottom tie bars is the most important consideration.

 

[rapt]

hokie66,

That is your prerogative. I would agree for a normal slab/band without the transfer.

However, with the face of the column above only 3" from the face of the column below, the stresses do not have sufficent time/distance to distribute sideways to make the full width effective. The only way they will distribute significantly sideways is if there is a shear failure at the column which defeats the purcpose!

 

[BAretired]

A steel shearhead such as shown in the attached sketch might be the way to go.

BA

 

[Prestressed Guy]

You can also use shear friction (ACI318 section 11.6) to greatly increase the shear capacity. With shear friction you can increase shear capacity to 0.2f`c*Ac or 800Ac. That is 6 times more allowable shear capacity Vc. Shear capacity is so low because of the unpredictable nature of shearing cracks and the catastrophic consequences of a brittle failure. Shear friction provides ductility to any potential shear failure so you can go deeper into the capacity. The precast industry uses shear friction all the time and has done extensive testing in both lab and real world conditions.

The Avf bars need to be fully developed at where they cross the assumed crack line but since you are using the top and bottom mat of reinforcement that is already there this is not usually a problem.

 

[ron9876]

Shear friction is for direct shear not diagonal tension. In my opinion doesn't apply here.

 

[csd72]

All this talk of shear ties and shear heads.

20" thick slab with only 5" between column faces. This will Not fail in shear!

Strut and tie is the correct way to analyse this and also leads to the simplest solution.

Now you can get onto the hardest bit of figuring out how you develop those column bars.

 

[hokie66]

csd72,

If this wouldn't fail in shear, how do you think it would fail? It might not be typical beam shear or typical punching shear, but shear is the problem. Strut and tie is just a design model for dealing with shear by truss analogy.

 

[slickdeals]

I agree that a compression strut will form between the two columns and all the shear will flow through that strut. But what reinforcing needs to be provided across that strut. The code asks for minimum steel crossing the plane of the strut.

 

[Tomfh]

Personally I feel the strut is working too hard, i.e. the concrete is approaching ultimate shear in this zone.

 

[BAretired]

The green shaded areas in the attached sketch represent struts. The red lines represent ties.

Anchorage of the upper tie at points 'a' and 'b' are essential.

BA

 

[rapt]

BeRetired,

You did not nominate the width of the struts. Under the 18" column above where the compression strut starts, it will be 18", and this controls the strut design.

The compression over this width and the splitting forces/stresses in this strut will govern the whole design, not the compression over a 60" width which is the width of the band beam.

 

[BAretired]

rapt,

I am not entirely sure of your meaning, so why don't you tell us how the strut and tie model should be drawn? It is not entirely clear to me and perhaps we can all benefit from your insight.

BA