Interior CMU Partition Wall 1

[PittEng88]

Hi All,

I am currently designing CMU partition walls (non-load bearing) that are 16' high, for an existing building in Puerto Rico. The architect has proposed a steel support frame for behind the walls to provide bracing at mid-height and the top of the walls, see attached image. The length of the long wall is 30' and the short wall length is 6'. The only loads that I anticipate for these walls are their own selfweight, a live load of 5 psf and their seismic loads (SDC = D). Due to the walls being installed in an existing building, I would like to reduce the amount of loads induced on the existing structure as much as I can. Therefore, I would like to avoid attaching the walls to the roof and to consider the connection at the bottom of the wall as pinned. I am planning on using either 8" or 10" block.

My question is, would I be able to eliminate the steel frame and be able to use just bond beams at the top of the wall and mid-height to provide enough stability for the wall? I.E. for in-plane loads, design the wall as a two story shear wall with the bond beams acting as the chords to distribute the shear force into the wall, and for out-of-plane loads, consider the bond beams as simple supports for the walls.

I hope I explained myself clear enough. However, if you have any questions or need more clarification feel free to ask. Any and all suggestions will be welcomed. I am a little green when it comes to designing w/ masonry.

Thank you for your time,
Mike

 

[hokie66]

What would support the bond beams if you eliminate the steel frame?

 

[Brad805]

Seems you have three options:
1. cantilever design (pilasters possibly) if foundation feasible
2. bond beams supported at ends via a connection to the existing. 30' seems problematic for a normal bond beam.
3. steel frame to support simple masonry design

Each has pros and cons that I would judge based on trade availability.

 

[PittEng88]

I kind of figured 30' would be a little problematic for a bond beam. The more I think about it, the more the steel frame seems like the better option.

Brad805, when you say a connection at the end of the bond beams, what exactly did you have in mind?

 

[jike]

Do the walls actually need to be 16 feet high? It not, I would suggest that you only build them as high as you actually need to be above the drop ceiling. This will reduce your dead load. Then brace the top of the wall with bond beam at top spanning between diagonally brace angle frames up to the structure. Make sure you detail it where the structure can move vertically without imposing load on the wall and thus the floor below.

If you need to close off the space between rooms for sound or air flow reasons then provide drywall between the top of the masonry wall and the structure above.

 

[Brad805]

It really depends what you have to connect to. This could be embedding dowels into an existing concrete wall or a steel bracket that bolts to the existing wall. Obviously, that assumes you have something reasonable to attach too. This will impart a load on the existing structure, so you will have to use your engineering judgement to determine if you want to pursue that.

The steel frame is an easy solution, but it is not without its own little problems. That top beam will need to be connected as well unless you have a moment base connection. I suspect the all masonry solution will be an easier option to detail and avoid steel shops, but either options work.

 

[msquared48]

Seems to me you should be able to eliminate steel verticals with the 10" CMU wall, just relying on the reinforcing steel to span vertically to a bond beam and the slab below. I would still include the horizontal stuts back to the columns to support the bond beam. If the bond beam still proved to be insufficient, I would add a steel beam at the top of the wall between the horizontal struts.

Mike McCann, PE, SE (WA)

 

[PittEng88]

Brad805, thanks for the input. The existing structure, where the bond beam would be attaching, is concrete. However, I am not really sure how thick of a wall it is but I can ask. Good point about the steel shops! I am trying to avoid any moment connections as much as possible with this wall.

Jike, that's a great point you bring up about the "required" height of the wall. If I am able to reduce down that height, it will definitely make things easier on me. The roof is made up of prestressed double tee beams, so I want to try and avoid inducing any unnecessary forces on them.

Msquared, I like your idea with the 10" wall and steel struts to brace the beams. If I do find that the bond beam is insufficient and needs additional capacity, would you suggest that I use a C-Channel turned on its side to "straddle" the wall?

 

[KootK]

I recommend running your masonry wall to the roof and installing a connection that allows vertical movement of the double tees but restrains the top of the wall laterally. Here's why:

1) Double tees are heavy and will generate large seismic diaphragm loads on their own. The extra load that your wall will add won't affect the system much.

2) If your support frames will be attached near the tops of your columns, as shown in your sketch, the lateral loads on the wall are going to wind up in your roof diaphragm no matter what. And adding transverse seismic loads to columns is generally undesirable.

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.

 

[KootK]

Forgot to add: if you attach to the double tees, do it in a way that won't mess with the prestressing strands.

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.

 

[msquared48]

PittEng88:

Yes. A channel could work well, depending on the connection used for any horizontal struts.

Mike McCann, PE, SE (WA)

 

[PittEng88]

I was planning on just using steel angles for the struts.

 

[msquared48]

That works too... no worries.

Mike McCann, PE, SE (WA)

 

[Brad805]

I like Mike's channel idea. That will be easy and the steel should fall into the miscellaneous category.

 

[msquared48]

Wasn't my idea. Was the O P's.

Mike McCann, PE, SE (WA)

 

[hokie66]

A channel on top of the wall is a common solution, but I wouldn't turn it toes down. It probably won't fit that way, and is best bolted to the wall with toes up.

 

[Brad805]

Mike, I see that in the post a few up now, my error. This internet thing is serious business. haha.

 

[PittEng88]

Well my name is Mike too, so no worries haha

 

[jike]

Rolled channel will not work without cutting / chipping block. It is common to use a bent plate channel to fit the block dimensions.

 

[hokie66]

If you place the channel toes up and bolt it to the wall, why won't it work?