Concrete Beam Pocket - High Loads

[bhiggins]

Howdy Ya'll,

I've got a high-end residential project. The perimeter is constructed of 12" 4000 psi concrete walls with footing which support a concrete over metal deck with steel K-joist and beam framing. I'm designing 6.5" deep x 7.5" wide beam pocket subject to a relatively high point load (34k LRFD). My colleagues are saying this is a purely compression connection, and if beam bearing is OK the connection is good. I'm worried about a shear failure occurring at a 45 degree angle from the interior face of the beam pocket (see attached).

A rough calculation: phi*2*sqrt(f'c)bd = .75*2*sqrt(4000)*6.5*7.5 = 4.6k NG, not even close.

I know horizontal ties are probably the best choice at the connection, or even hooking the vertical bars inwards, but I feel like the contractor will give me flak for that sort of detail or just not put them in all together.

I'm just wondering if my assumption of the shear failure plane is valid, and what is the best option to make the connection work in the most construction friendly manner.

Thanks!

 

[Shotzie]

http://www.eng-tips.com/viewthread.cfm?qid=299199

This thread has some good discussion on a similar problem if you haven't seen it already.

 

[KootK]

This never seems to get checked, or cause problems, but I agree that it's a real failure mode. I've seen it evaluated exactly once and it was similar to the PCA method mentioned in the other thread but without the assumption of horizontal rebar or a T&S load demand. Went like this:

1) Chose a potential failure plane.
2) Calculate the load induced shear on the plane.
3) Calculate the load induced normal force on the plane.
4) Treat the normal force as faux shear friction reinforcing.
5) Use #4 to estimate shear capacity parallel to the plane.
6) Try a few different planes to ensure that you're covered.

It is debatable whether or not one should go with a design estimating no temperature induced forces. Kinda depends on the situation.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[bowlingdanish]

If you can't trust a contractor to put in a U bar that's a bit of a worry. Although I don't particularly share you concerns, I don't think that's a valid failure plane. The load will form a compression strut to where it's supported. If the wall is designed for the eccentricity of the load and there is a load path to get the bending into the wall (U bar!) then I don't see a problem. But yes, check for bearing.

What Shotzie posted is talking about bursting forces from a high point load placed near the edge of a wall (i.e. close to the edge out of plane from yours). That's more of a possoins ratio type situation I think.

Edit: Damnit KootK disagrees. Back in my box. I might buy it for an unreinforced wall with a sliding bearing but otherwise I'm still convinced it's a non issue.

 

[bhiggins]

This is just what I was looking for, thanks! There's a pilaster design example in that thread from PCA. They used 20 degrees as the failure angle and designed using shear friction. Is there any significance to 20 degrees (other than just "feeling" reasonable).

I'm just going to specify closed stirrups and hope they install them correctly. You never know if they will follow the details with residential projects though.

Another avenue I was exploring is to use an angle as the beam seat and bolt the horizontal and vertical legs to the concrete for more even load distribution (see attached). My gut says not to rely on this mechanism though. I'm probably going to do both the angle and stirrups. I'm also going with long slotted horizontal bolt holes to neglect temperature forces. Thoughts?

 

[bhiggins]

Attachment broke. See below:

 

[calvinandhobbes10]

Deeper anchors = another element crossing your assumed failure plane.

I'm used to residential basement walls being masonry not concrete. If your failure plane assumption is valid, yikes.

 

[KootK]

Damnit KootK disagrees. Back in my box.

Ah..I laughed out loud at that. Yes, back in the box, with breathing holes.

I'm really not so sure about it myself. Particularly in regard to the difference between a corbel and pocket. With a corbel, there is a diagonal strut tending to push the corner off and thus a clear demand. With a vertical compression path from load to footing, there's kinda not.

For me, it comes down to the validity of these postulates which I hold to be true:

1) Any chunk of concrete needs to be in equilibrium when examined with an FBD. The afformentioned Bhiggins Wedge is no exception.

2) The Bhiggins Wedge cannot be in equilibrium without some shear forces working on the proposed crack surface.

3) 1 + 2 = there's something to check for the Bhiggins Wedge.



I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[TehMightyEngineer]

Keep in mind your shear plane you drew will "scoop" out a part of the wall and this will cause the horizontal bar in your shaded area to be crossed twice by the failure plane. This will undoubtedly add a lot of capacity to your connection.

I'm going to agree with both KootK and your colleagues at the same time.

I agree with KootK and yourself that this seems like a potential shear failure plane and I've always wondered why we never see it in many design examples and standards. Any dapped connections or other bearing details always seem to show the crack either vertical or back into the support or dapped member; it rarely seems to be shown with a crack sloped forward toward the free edge.

That said, the PCI manual and a number of other references I have don't seem to regard that this needs to be checked. For example, on a bearing surface the PCI handbook recommends that the bearing strength be checked by the usual formula. Then, if this strength is exceeded then they provide details for reinforcement of the bearing concrete.

That pilaster design is interesting. I need to read up on that more.

P.S. Your second PDF comes up blank for me.

Professional Engineer (ME, NH, MA) Structural Engineer (IL)
American Concrete Industries

https://www.facebook.com/AmericanConcrete/

 

[KootK]

Whenever I look at something involving monolithic pour shear friction, the answer always seems to be that it works out. So much so, that I've come to believe that the shear friction provisions may in fact be calibrated to make that the case. This example may well be another of those cases.

It's easy to mistakenly come at structural engineering assuming that what we do is to check everything that could possibly go wrong. We don't. We check everything that has actually gone wrong in the past. It's very much a reactive technology that way. Failures occur and we pivot to prevent them in the future. Terrifying but true. This issue strikes me as one that's kind of on the bubble. It probably doesn't need to be checked. But then it's just there, staring us in the face begging to be addressed.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[bowlingdanish]

1) Any chunk of concrete needs to be in equilibrium when examined with an FBD. The afformentioned Bhiggins Wedge is no exception.

I agree. But there is just so many things working for bhiggins here.

Take the most extreme example imaginable - the failure plane is already cut, two steel plates are attached to each side of the plane, and the steel plates are greased. Even for that to collapse, the thrust from the connection would have to be enough to deflect the wall laterally by over 6 inches assuming it didn't snap off. In reality there is the bolt passing through the failure plane, the beam acting as a tie, the strength of the unreinforced concrete, the reinforcement in the wall acting as a tie, etc. I think you could analyse it the same as a corbel, but the strut angle is that from the centre of load to the centre of wall base ('e'). The resulting strut angle would be a couple of degrees and requires tie force tiny.

 

[KootK]

I agree with all that. It's difficult to envision anything going awry without an axial tension in the beam. And I've no doubt that's why PCA covers that case.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.