Column base plate with large moment load 4

[calemadsen]

I'm affixing steel columns to an 8inch concrete wall to support a roof structure. The columns will see a rather large moment compared to the compression load because of the span that is involved--about 350 kip inches in moment compared to 13.5 kips axial (compressive) load. I've selected W6*25 columns to connect to the base plate (which is welded at the top to a W12*19 beam), and am having trouble deciding the plate thickness. I used the AISC procedure (

http://vietcons.com/uploaded/download/AISC-T-DG01-Column Base Plate.pdf)

for column base plates with high eccentricity, but the problem is that because the base plate is restricted in size in one dimension (the 8" width of the wall) the 'critical section' about which the bending is to occur is outside of (or closer to the face of the wall than) where the anchor bolts should be to maintain the embedment I'm hoping for, i.e. 1.5 inches. This seems like a problem to me--is it?

The numbers otherwise work out and seem intuitively reasonable. However, I don't feel confident in the base plate thickness that is given because it seems that the moment for an inch wide strip is likely incorrect.

 

[JAE]

I can't imagine that ACI 318 Appendix D would give you any anchor bolt configuration that would work with an 8" wall taking moment about an axis parallel to the wall

How are you fastening the plate to the wall?

With four bolts spaced at 4" o.c. to get about 2" clear - you get about 40 kips of uplift on a single bolt after including the minimal 13.5 kips of compression.

 

[JedClampett]

Listen to JAE.
You'll need to build a pilaster about 16 inches square into the wall to get enough edge distance/spacing for those embedded anchors. I'm not too crazy about the 30 ft.-kips (you don't say if it's factored or not, but either way) in an 8 inch wall either, but in a 16 inch column, no problem.

 

[calemadsen]

Thanks for the replies.
JAE: I'm planning on casting the anchor bolts into the top of the wall before the concrete sets using a pre-made template--and then torquing down a nut onto the bolt. How did you find the 40 kips uplift? I got something closer to 25 kips per bolt after extending the width of the base plate (along the length of the wall) to 16 inches (also that is with a slightly longer lever arm since I assumed the 1.5" clear).

If the equations given in the design guide are reasonable to use here (you're right of course, there doesn't seem to be any example/procedure that deals specifically with such a narrow landing area, but the procedure given for finding the tension that will be seen in the anchor bolts don't seem to be compromised by the geometry--it's only with the base plate thickness that there seems to potentially be a discrepancy) then even the range for uplift you gave would be achievable if I for example added another bolt to the tension side, which I would have room for with a wider base plate. Do you think that would be a viable solution?

If I were going to try to check the stress seen in the base plate, about which point would you recommend taking as a fulcrum for the uplift load the bolts will see? I guess I don't really know what the critical section physically implies here. Or more specifically, why the point around which the plate acts as a cantilever is based on geometry only (the formula given for that was the [(width of the plate perp. to axis of bending) - ((.95)*depth of beam)]/2

Jed: The moment that I used is factored. Also, this may be a stupid question; if the compression side of the base plate doesn't exceed the capacity of the concrete and the tension side doesn't exceed the capacity of the bolt/pullout strength, why would that moment cause a problem in the 8 inch wall?

Using a 16" pilaster would certainly make me rest easier. I guess I'm just trying to avoid the extra concrete/labor/forms involved.

 

[calemadsen]

Also, Jed: That WAS a stupid question. Should have thought about what you meant a little further. I think the wall should be capable of carrying that moment though, as I'll be adding two lengths of vertical rebar in addition to the 12 O.C. that is already there--so the area under each 16" plate (if that is the way I end up doing this) will have about 1.5^2 inches of reinforcement.

 

[JAE]

The only way I could see this work is to embed your base plate into the concrete and use welded-on rebar or long deformed bar anchors to transfer the uplift on the "bolts" into vertical rebar in the wall.

Using simple 8" deep anchor bolts isn't going to work in my first-blush general view.

The column would then have to be directly welded to the base plate and you'd then have to design through the load path (column to base plate to embed plate to embedded rebar to wall reinforcement).

 

[JAE]

I just ran a quick HILTI App. D check on an 8" wall with 8 - 3/4" dia. anchor bolts 14" into the top of the wall. Spaced at 4" apart to give 2" edge distance.

These were overstressed to over 500% of capacity (using 36 ksi steel). Using 105 ksi steel didn't help. - the concrete fails miserably.

 

[structSU10]

I recommend you try an anchor bolt program, like Hilti Profis, to check you baseplate. I put the loads you specified on there, and it is greatly controlled by concrete breakout. If you design the reinforcing in the wall for the tension, then it may be possible from an anchorage standpoint.

 

[calemadsen]

I played around with Hilti for awhile and finally got one configuration to work, but the bolts ended up having to be so close to the midline that I'm a bit leery of the possibility of getting enough tension capacity with reinforcement. I'll try to come up with some numbers for this though, will update on where that goes.

I'm going to start looking into the embedded plate route--thanks JAE. Luckily the steel manufacturer is happy to weld the base plates to the columns.

Also, would saddle (U?) mounts be an option, where a wide plate of steel would compress into the face of the wall as moment is applied?

 

[paddingtongreen]

I don't understand why you are working so hard to take an unnecessary risk!

I always looked to totally develop the top concrete with the bolt above and the rebar below. I don't think pullout is your main worry here, the whole top of the wall may pop off if it is not fully secured by the rebar. I say this because I have seen it when short bolts are used in a narrow base or wall.

I consider that I want the equivalent of a lap joint between the bolt and a developed rebar of equal or greater strength.

Michael.
"Science adjusts its views based on what's observed. Faith is the denial of observation so that belief can be preserved." ~ Tim Minchin

 

[structSU10]

Along the lines of what paddington is saying, developing the anchors with the rebar is important. attached is a paper that goes through a design procedure for the reinforcing. You may need a fairly long anchor.

 

[theonlynamenottaken]

Any possibility of having the flange of the W-column lay flat against the vertical face of your wall? If you have a a few feet of wall to work with you can spread that moment connection out, minimize fastener tension, use through-bolts to attach the flange to the wall, and avoid any welds on the base of your W-columns (unless you need flange stiffeners...). That would just leave you with the flexure in the wall itself to tackle.

 

[TXStructural]

Paddington has it right, Appendix D only addresses the issue of the anchors themselves, and doesn't necessarily get the forces into the reinforcement. Since these will be cast-in-place, do not be afraid to do a very long embed. The steel cost is insignificant, and the risk is too high to use shallowly embedded anchors.

You could easily saddle the wall and through bolt, but make sure the wall reinforcement is developed where the bolts anchor the base of the column.

Also, I may have missed it, but have you determined that the wall is capable of taking the aggregate of the forces imposed on it? (Thinking about the base of the wall and all of the roof columns, as well as local at each column base.)