Steel Beam Supporting CMU Wall

[IngeniumJD]

I have a client that is desiring to support a 9' tall CMU wall with a steel beam below. This CMU wall is an exterior wall that is located over garage space below. We are in both high wind and seismic zone, so we have other load considerations on the beam other than gravity. This wall will be approximately 32' long, but the span of the supporting beam can be broken up into smaller spans with several columns. I am concerned with supporting an entire masonry wall on a steel beam. Deflection (resulting in cracks) and providing connection of cmu reinforcing to beam below are major concerns. Has anyone done anything like this with long-term success? What should my deflection limits be? How do I connect these two systems together? I'm thinking that a masonry beam (or precast) would be a beter solution, but my client is putting pressure on me to use steel beam.

 

[JAE]

Per ACI 530, the deflection limits are L/600 or 0.3 inches for all loads (dead and live) that will be applied to the beam after the CMU is in place....and many engineers include the dead weight of the CMU in this as well - including myself.

Use control joints in the CMU at spacings about 24 ft to 30 ft. on center.

As far as connecting the steel beam to the CMU, what we've done is use 5/8" dia. deformed bar anchors welded to the top flange and extended up into grouted cells 32", lapping with vertical reinforcing and matching their spacing.

 

[archeng59]

I have supported CMU on steel beams before with success. The only difference to what JAE said: I used #5 bars spaced at 24" on center.

 

[UcfSE]

Note that the ACI 530 (1.10) points out that the L/600 or 0.3 inches limit is for unreinforced masonry or empirically designed masonry. Most engineers keep the L/600 for reinforced masonry and drop the 0.3" requirement. Vertical control joints may also help to offset the extra deflection when you drop the 0.3-inch limit. This is all based on engineering judgment so not everyone will agree or do the same thing necessarily.

 

[SlideRuleEra]

I would suggest having a deflection limit that is less than the code requirement of L/600 (say L/1000). Here are my reasons:

1. The L/600 seems to have been established for a "typical", simple-span lintel where the supported masonry is in compression as deflection increases.

2. A continuous beam with multiple supports will have several reversals of the deflection curve. This will cause some the masonry to be in compression, other masonry sections in tension.

3. The geometry of a continuous beam with multiple supports makes it much easier to minimize defection without going to a "ridiculously large" steel member.

http://www.SlideRuleEra.net

 

[csd72]

Sliderules L/1000 suggestion is a good point.

Also what stops the whole thing toppling over in the wind. i.e. lateral restraint of the beam. If you have doors under then this will need to support the top half of the doors.

To connect the ends you could replace a few blocks with concrete and put embedded plates in it (only if it is not exposed).

Regards

csd

 

[ash060]

In my office we came across a similar condition. The existing building (not designed originally in our office) was in a hurricane with a steel beam supporting a cantilevered masonry wall parapet. The wall reinforcing was directly welded and the wall just tipped right over. The welds ruptured along the whole length of the wall. I would be carefully about directly welding the rebar to the flange. Make sure the the welding is done correctly. I think PCI says something about the bar having to be pre-heated.

 

[archeng59]

I have used lateral bracing when needed to resist out of plane wind loads. On a couple of projects, the need for the bracing initiated architectural design changes of the opening sizes so the bracing wasn't needed. In both cases, the CMU was designed to span horizontally for wind load and the steel beam provided gravity load support.

 

[hokie66]

IngeniumJD

Your original post asked for advice on only the gravity load condition, and stated that the other conditions were being addressed.

I think the most straightforward solution would be to design the 9' high masonry wall with adequate reinforcement and grout fill to span the opening itself, as the wall will be much stiffer than a steel beam designed for strength.

Build the wall on a steel beam to carry the weight during construction, or alternatively build it on temporary staging, but in either case, the wall will in the end carry itself.

 

[haynewp]

Consider a control joint at midspan and at each end of the opening where the curvatures are high. Also see:

thread507-176484

If L/600 and 0.3" are applied to all self+dead+live loads then should certainly be conservative. The steel beam will act somewhat compositely with the cmu it is supporting if DBA's or threaded rods are extended up into the cells.

 

[civilperson]

Design the 9' height of masonry units as a 9 foot deep beam with a continous bond beam in the bottom and in the top course. Use vertical reinforcement at 16" spacing with horizontal ladders at 16" vertical spacing. The steel beam beneath will merely be stay-in-place construction support until the masory is completed. Calculated deflection will be small when considering a 108" depth.