Horizontally Spanning CMU and vert control joints 1

[chrislsnider]

I searched the forums and nothing came up, so I apologize if this has been asked.

I have a tall CMU wall that spans horizontally 12' to pilasters. Wall lengths are in the neighborhood of 116' each. The wall works spanning horizontally and the pilasters work vertically. My question is about the vertical control joints. Since my wall relies on the horizontal continuity of the bond beams, can I use greased bars or something similar to transfer the shear across the joint? Is this a standard practice? Do these dowels need to fully lap with the horizontal bars?

Thanks for your time and input.

 

[Michel60]

Good morning Chris,
That is an interesting problem...I would think a shear key or unbonded dowel could be used to create the joint separation. Maybe also consider having some sort of "double pilaster" with a control joint in between?

regards,
Michel

 

[hawkaz]

If you put a control joint at midspan, could your wall span 6' horizontally as a cantilever?

 

[chrislsnider]

I'm pretty sure the wall could cantilever 6' fairly easily. If the architect will buy that joint layout, this would make the problem almost go away. Do you think I would still want the unbonded dowels to ensure there is no differential out-of-plane movement of the walls? My gut is telling me yes...

 

[hokie66]

I would reinforce the wall continuously through all control joints. With the wall laid in running bond between control joints, shrinkage cracking will occur at the joints as desired.

 

[eng003]

Chris - can you further describe how you plan to reinforce the wall?

 

[chrislsnider]

eng003 - not exactly sure what you are looking for, but I've got 12" CMU w/ (2) #5 horiz @ 48" o.c. and (2) #5 vert at __ o.c. (Sorry, I'm at home right now and I don't remember for sure.)

 

[eng003]

chris-so horizontally you have fully grouted bond beams @ 48" o.c. that span between pilasters? your talking about control joints not expansion right? can you show a sketch? i would think there is no reason you cannot maintain flexural and shear capacity through a control joint provided you have adequate roughness or a key way... I am interested in your question too...

 

[chrislsnider]

I guess our detail looks more like what the MDG shows as expansion joints. Either way, it is a weakened plane in the wall. The attached pic from the MDG is close enough to what we are doing.

 

[hokie66]

That detail works fine for a control joint in a vertically spanning wall, but not for one that spans horizontally. Where the joint filler is shown, that head joint should be fully mortared, and the bars should go through.

 

[chrislsnider]

So would there then be essentially no vertical control joints in my wall? I liked the earlier suggestion of putting the joints midway between my pilasters and having the wall cantilever to each side. Numbers-wise that works, but is it not something that would be standard?

 

[hokie66]

As I described it, they are control joints, but not "expansion" joints. The purpose of control joints is to make shrinkage cracking occur in a straight line. If you build the wall in normal running bond between the vertical control joints, the cracking will occur at the desired locations, most likely. This is a case where, with conflicting requirements of strength and serviceability, strength takes preference but serviceability is respected.

 

[concretemasonry]

chris -

The approach you have is very different from used on many thousands of buildings around the country. Most are industrial/commercial office warehouse buildings using 12" CMUs and are 16 to 24' high with pilasters at about 20-24' O.C. and control joints at one side of each pilaster. The pilasters are 16-24" deep and are turned to project inward or outward, depending on the functional interior use of the buildings and architectural appearance. More information on your design challenge is necessary, since you may be using a concept based on a different geometric requirement.

Details are a different story and construction of a control joint that can effectively transfer the loads well enough to be considered as a moment connection is very, very difficult and costly. At the face of the wall, it is much more reliable and more pleasing architecturally, especially if the CMUs must have an architectural appearance. Turning the pilaster so it projects into the interior is very common to build a wall with just joints at the pilaster and no projections outward

How tall are the walls?

Is there any vertical loads on the top of the walls.

Is there any lateral support at the top for the roof system/deck?

Are the 12' pilaster spacings determined by architectural or the goal to force the wall to span horizontally between the pilasters?

Dick



Engineer and international traveler interested in construction techniques, problems and proper design.

 

[hokie66]

I disagree with Dick that providing control joints in horizontally spanning walls is difficult and costly. Run the bars through the joint, fill the collar joint with mortar as the wall is built, rake the joint to half the depth of the joint width, provide a bond breaking tape rather than a backer rod, sealant as required.

This may not be as common as spanning the wall vertically, but I have done it many times. That type joint also works in fully grouted walls.

 

[BAretired]

The walls are 116' long. In extreme climates, control joints may not be enough to avoid unsightly cracking. Some provision for expansion and contraction may be advised. Smooth greased steel dowels at the joints would seem to be prudent.

BA

 

[concretemasonry]

I did not find that the wall in question had a brick veneer, so there is no collar joint joint to fill as shown in the MGD detail. That is nice detail IF you have the luxury of a veneer that can move independently of the 12" CMU, which can be accomplished without a filled collar joint the makes the wall a composite wall.

Much depends on the part of the world you are (wind, seismic) and the traditional methods that contractors can do quickly and easily to a high quality if this is a typical box like a gym,retain store or office warehouse. If it is "one-off" type structure like a school gym, auditorium or similar.

Engineer and international traveler interested in construction techniques, problems and proper design.

 

[chrislsnider]

It seems like I have started a good debate. This is an highschool auditorium in Missouri. The exterior of the building will be the CMU - some smooth, some split-face. I have no need to transfer the moments through the joints. Numbers-wise, the wall can span 'cantilevered' off of the pilasters assuming the joints are mid-span between them, i.e. 6' cantilevers. If I put the unbonded dowels in, that would simply ensure any deformation compatibility, right?

 

[abusementpark]

We usually instruct them to run the bond beams continuous thru the control joints. No problems yet.

 

[hokie66]

I misspoke about the collar joint. It is just a continuous head joint, and that is what is shown in the MGD detail.

I wouldn't like to have 6' sections of wall cantilevering each side of a pilaster. By all means, put the joints central between the pilasters, but run the horizontal bars through.

Various opinions, now it is your call...

 

[chrislsnider]

I think the general consensus, and best sounding practice, is to run the bars through, hokie66. Do you have a detail that you wouldn't mind sharing of how to prepare the joints better than what I've got? What I've got is basically the same as I had posted earlier...