CMU bearing/shear wall on beams 3

[Engin1]

Have anyone done cmu block bearing/shear wall on steel/concrete beams? We have a client wants to build a 4-story hotel with basement parking and he wants to use block and plank. This is my first job with CMU. I have a concern how to connect the cmu to steel beam. I have tried the analysis using software, the software treats the cmu wall as a beam and has big difference in deflection as beam. Most of the loads go to the end of the cmu wall. Any ideas how to do the analysis? Does this structure make sense? I prefer to use steel columns, beams and wood on top, but my boss insists using block and plank. Please help.

 

[jayrod12]

Run the beams by hand. In reality, the wall will likely act as a big beam and the steel beam supporting it will see negligible load, but if you run the steel beams by hand assuming the wall doesn't act like a beam then you're double covered.

Discussions have been had previously on modelling this, if I remember correctly you'd essentially have to run two separate models. One with the true stiffness of the walls accounted for in order to accurately depict the lateral resistance, a second with the stiffness of the walls cranked way down so they apply the load to the beam.

The connection from the beam to the wall is easy, just weld rebar dowels to the steel beam to start the wall reinforcing.

 

[JLNJ]

A CMU wall of any decent height will be much stiffer than the beam you set it on. You will probably left to use your judgment as to how and how much load is carried by the beam.

Sometimes I will size these types of beams for a tight deflection criteria based on the loading of a shift's worth, or maybe one story, of wet block. I rarely use the traditional triangular distribution because I'm not convinced that this represents how things actually work.

One very important factor is the location of joints and openings in the CMU wall. A wall with a door opening in it might transmit all of the load from above directly into the beam, maybe at mid-span, and you will want that beam to be sufficiently stiff. While the arching/beam action of a CMU wall is a real thing, you will need to intentionally detail the wall to ensure it functions as you designed it to work.

There's a balance to achieve between a beam stout enough to avoid wall cracking, yet not so heavy and deep that the architects cant make it work with their floor plan and ceilings.

 

[Engin1]

Thanks! @jayrod12 @JLNJ
Good point. I will design the beams considering beam takes the uniform load from wall as well as point load (arching action) from the wall.

 

[KootK]

I've done similar things myself but I find this system difficult to love because the state of stress in the parts is so hard to assess. A whack of your wall vertical bars are going to start their service lives in a yielded condition as they attempt to prevent the top of the steel beam pulling away from the bottom of the wall. This exacerbates the compression at the ends of your wall in a prying sort of way. Conceivably, you could even wind up with sections of your beam experiencing bottom flange compression. And, of course, getting your bar couplers welded to the tops of the beam such that they're aligned with the block cores is no mean coordination feat.

Is it safe to assume that these block walls are also shear walls?

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.

 

[ChipB]

First, Engin1, write down all of your thoughts based on your experience and guidance you have received from here or elsewhere, and take them to your boss on how to approach it going through your thoughts. Ultimately, he's the one that has to sign it, and if he thinks you have been chasing a rabbit down a hole, he will likely not be too happy.

Typically, I don't consider arching action of the CMU wall if the span is over 8'(the usual height of the first bond beam).

I have never considered a composite action taking place between the CMU wall, and the steel beam. I put shear lugs on the beam to transfer the shear from the wall, to the beam. If you want to check how much gravity loading would be carried by the wall, and how much would be carried by the beam, look up "Flitch Plate Beams". It would be the same design concept using different modulii of elasticity and different moments of inertia.

I always make the first coursing of CMU a bond beam, if it's at a floor or not.

Hopefully, your design isn't in a high seismic area, as your steel beam is now your collector element, and will need to be designed using the overstrength factors...

Also, your beam is a collector element, no matter what. You'll have to design the beam for combined axial forces and bending.

Just my 2 cents.

 

[dhengr]

Engin1:
In situations like your’s, I’ve often started the wall, atop the beam, with a couple coursed of bond beam blks., so I could get some longitudinal rebar in the bottom of the wall. This was to prevent some vert. cracking in the wall due to the stiffness difference of the wall and the beam below. To make the wall act like a deep beam by itself. As mentioned by JLNJ you do want these beams to be pretty stiff. To address a couple of KootK’s concerns, you can prop or shore the beam while you lay up the first height of wall, and let the wall cure a bit before removing the props. While you can’t ignore the location of the vert. rebars coming up off the beam, you just bust (or cut) a hole in the bot. of the bond beam blks. at the rebar, and then the mason has a couple courses to adjust its location slightly to match the blk. cores above. As for the masonry arching action, the biggest issue is that you have some solid wall at each end of the arched area, so that you can distribute the trust from the arching action.

 

[KootK]

Is it safe to assume that these block walls are also shear walls?

I see now that the title of the thread makes it clear that these are shear walls. Question withdrawn.

As for the masonry arching action, the biggest issue is that you have some solid wall at each end of the arched area, so that you can distribute the trust from the arching action.

Unless I'm misreading the building typology, I think that it's a pretty safe bet that there is no buttressing thing to push against. You've either got to:

1) tie the wall internally in such a way that that node width jives with bearing requirement or;

2) Provide some kind of "stopper" element attached to the beam so that the beam itself can be the tie for the arching action.

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.

 

[masonrygeek]

Engin1, why are you resistant to just designing this as a loadbearing masonry building? Block and plank structural systems work well, especially in a 4 story structure. Eliminate the steel beams and/or columns and rely on the masonry to do all the work. You don't mention about the size of the floor plate or if there are any long span elements as that will dictate if masonry can be used. Often, engineers aren't taught structural masonry in school, so they are reluctant to design with it, but it can be an economical choice and you don't have to jump through hoops to make a steel and masonry connections work. Just a thought... and there are resources out there to help you. Where are you and the project located?

 

[Engin1]

Thanks guys!

I think the cmu walls need to be shear wall since no other structures to take the lateral load.
masonry.geek , Block and plank structural system works well. But for this project, it needs parking space under the building, the block cannot go down to the parking space based on arch design. I have to have steel structures to take all the loads. Each bay has about 27ft.

I think the basement will be steel structure and block and plank on the steel structure(4 story+parking basement). The block wall will be bearing wall as well as shear wall. Please correct me if I am wrong.

I also have a concern that all the lateral loads go the steel structure, however, I cannot use bracing in the parking space. I have to use moment connection, not sure if the moment connection will be enough for this case.

Welcome to join this discussion!

 

[jayrod12]

You may want to design your main floor as a large transfer diaphragm to drag your lateral loads out to the basement walls. I assume those will be poured concrete?

 

[KootK]

I also have a concern that all the lateral loads go the steel structure, however, I cannot use bracing in the parking space. I have to use moment connection, not sure if the moment connection will be enough for this case.

I could see that being an issue. Soft story, heavy framing, P-delta... If it works, it works though. No shafts that could be block / CIP?

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.

 

[Engin1]

jayrod12 : Actually the floor will be 8" precast hollow plank bearing on cmu walls. It's 240 ft long building, I don't think all the lateral load just go to the end of the building to the basement wall. If it only goes to the basement wall, I expect there will be huge tensile load at the end of the wall.

KootK : Yes, it seems an issue, I am not sure how to deal with it now. No shafts will be block/CIP. I hope the openings through the plank are small enough for plank itself to handle.

 

[jayrod12]

a 240ft long multi-storey building, with no block/CIP stairwell or elevator core accessing the parking? that seems somewhat unbelievable. I would be pushing for that, or bracing. I think moment frames would be enormous in order to be stiff enough.