Horizontal joint reinforcing between wall and column

[skeletron]

I have a grouted masonry wall going up against a new concrete column. This is a repair, not a new build, so the two dissimilar materials are required. Column will be poured first, then the wall will be built up.

My question is:
Do you tie the masonry wall to the concrete column through horizontal post-installed bars? And if so, how do you determine how much steel needs to be there for restraint between the two dissimilar materials? Or, do you purposely let the joint float?

My concern is that:
A) They can't cast in the horizontal bars because of formwork and sequencing requirements
B) I'm looking at the proposed detail and it looks like they are tying in every 16" which makes me worried about chewing up the column and/or difficulty actually getting the bars to not hit the ties.

These seem more like prescriptive requirements revolving around typical .0015 or .002 ratios.

 

[hokie66]

My approach would be to leave a joint which can be sealed with backer rods and sealant. No ties to the column unless the column provides support to the wall, or the wall and column work together in resisting lateral forces.

 

[r13]

I would provide masonry bond beams to tie the column and wall together. Cast the column with dowels for splices.

 

[STrctPono]

I guess the answer would depend on how the system is designed to behave. If the wall is needed to brace or restrain the column then some ties would be required. From the way you describe it though, they sound like independent systems and should have a joint between them.

 

[Lomarandil]

Not at my desk, but MSJC (Appendix B, maybe) has some prescriptive requirements for doweling infill walls to confining elements as needed to transfer out-of-plane loads.

How the dowels and joint are detailed differs between infills designed as participating or non-participating for in-plane loads, as the others have alluded.

The post-installed dowels are ordinary practice around here. If you hit a stirrup, patch it and try again a few inches away.

----
just call me Lo.

 

[skeletron]

Great. Thanks for the responses!

 

[KootK]

I guess the answer would depend on how the system is designed to behave.

That one. Given that you practice in a (very) high seismic environment, we really need to know both what your wall is doing for you and what the column is doing for you (other than the obvious). Here in aseismic middle earth, common solutions will be:

1) No connection.

2) Dovetail anchors installed in the column.

3) Dowels epoxied into the column.

4) Dowels installed into form saver couplers installed in the column.

5) Vertically oriented clip angles bolted to the column at intervals up the height of the wall, wrapped in paper, and buried in the cells at the ends of the wall.

But yeah, any of these solutions can wind up wildly inappropriate depending on what you're doing with the wall and column.

 

[rowingengineer]

I think hokie idea is best unless we know more, which is what KK is alluded too. tieing things together makes them stronger most of the time unless you create captive elements.

 

[dik]

can you use dovetail masonry anchors?... cast a strip into the face of the column?

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

-Dik

 

[dik]

hi lo...

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

-Dik

 

[skeletron]

The wall isn't an infill wall. It appears to be part of the original wall/pilaster masonry exterior wall system. So, it's definitely taking the lateral load. Except that the structure on the opposite side of the joint is storefront, with some meagre little steel framing.

I think you need to tie into the column. Still questioning whether I want to do it consistently all the way up (2 storey) or just at the bond beam (where I know the wall is stiff). No openings to deal with, so that's good.

Thanks for the captive column effect note.

 

[KootK]

Even with the wall and column ostensibly disconnected, it will likely be difficult to prevent your column from acting like the boundary chord for the shear wall and, thus, increasing over strength capacities etc where those capacities would be used for the design of other elements. If this is your thinking then, yes, I would shear connect the wall to the column for in order to develop exactly that boundary chord axial force via shear friction etc. As an incidental bonus, I suppose that this would restrain differential, out of plane deflection between the wall and column as well although I'd not think that terribly important in such an application.

I'm imagining a multi-story structure with concrete slabs that would tend to transfer axial load from the wall ends to the nearby columns. If it's something else, my recommendation may well be bunk.

 

[skeletron]

@KootK: Your first paragraph parallels some of my thoughts. The structure is wood-frame inside, so the second floor is almost like a mezzanine style floor, ledger'ed into the exterior wall. I see the load path to as directly coincident with where the ledger ties into bond beam and the bond beam then ties into the pilaster new concrete column. I want to provide confinement there, and maybe not as critical above and below that joint.

 

[KootK]

I see the load path to as directly coincident with where the ledger ties into bond beam and the bond beam then ties into the pilaster new concrete column. I want to provide confinement there, and maybe not as critical above and below that joint.

I'm not seeing it. In my mind, you want to keep your seismic shear in the wall and not the column. If you're your going to engage the column at all, I'd think that a vertical shear connection best accomplished along the entire length of the joint. Seismically, dumping lateral load at a concentrated location at the top of the column is going to create that same captive column effect where you're pounding the column in shear.

 

[skeletron]

So, transfer the shear across the joint as opposed to at a singular point?

My real frustration is understanding how the original masonry design wall/pilaster was determined to originally behave.

 

[rowingengineer]

Skeletron,
Often it is because they aren't using the same design effects due to age of the code at time of design. I agree tie all the way along the joint keeps the shear demand in check.

 

[r13]

Do not confused with the different modes of failures.