Steel Beam Sitting on an Existing 6" CMU Wall 3

[dursundrsn]

Hi Folks,

I do not want to generate a high compressive stress on an (e)6" cmu wall for that reason instead of having my HSS 10x4x1/2 beam sit with an end bearing plate on the (e) 6" cmu wall I am planning on placing another stiff HSS 10x5x1/4 beam fully on top the (e) 6" cmu wall, and connect HSS beams together. This is the first time I am dealing with this type of a condition, and I would like to get your thoughts on the behavior of the beam and cmu wall for the described condition.
Here is the question. How do you guys think that the beam will behave on the (e) 6" cmu wall ? Do you guys distribute the reaction from the ends of the beam with a 45 degree along the height of the wall even though the beam will be distributing the vertical load uniformly along the length of it ?
Also, how do you guys approach to design of the anchors between the beam and (e) cmu wall ?
Attachment is indicating what I am planning on doing ..
The reaction is 15kips.

*(e)= existing

 

[DaveAtkins]

I am OK with spreading the load along the entire length of the tube, as long as you design the tube for the bending induced (just like you would a bearing plate). I am not sure it will really behave that way, but if the CMU at the location of the beam reaction started to fail, the load would travel along the tube to other parts of the CMU wall.

DaveAtkins

 

[OldBldgGuy]

Maybe a dumb question, but why on earth would you use an HSS 10X4X1/2 for a beam concealed in a ceiling? The cost and efficiency compared to a W section makes me tremble.

 

[dursundrsn]

@OldBldgGuy Most of the time weights of W sections equivalent in stiffness to HSSs are almost identical plus you do not need to deal with bracing your flange with HSS members.

 

[kipfoot]

Most of the time weights of W sections equivalent in stiffness to HSSs are almost identical

I'd guess that this is never true.

HSS 10x4x1/2 W= 42 lb/ft. . Ix= 129 in^4
W10X26 . . . . W= 26 lb/ft. . Ix = 144

Roughly equivalent stiffness in 3/5 the weight. Are there any examples where it's close?

 

[BAretired]

I suppose it would be close if the web thickness of the W section is twice its flange thickness but that would be rare indeed.

BA

 

[struct_eeyore]

you say that your beam is on a bearing plate, but your sketch shows attaching with an end plate with embedded anchors - so are you attaching into a tie-beam or bond beam? I would assume the load is redistributed to 3-4 cells, place rebar in them, epoxy to tie-beam and foundation, (grout solid of course). then check that particular section of wall for axial load and moment from the eccentric connection (again, I'm assuming youre coming into the face of a tie beam, and not bearing directly on top of the wall). Assuming your end plate is of sufficient thickness, I would take the anchors to be in pure shear - usually use 3/4 powers wedge bolts or something similar.

 

[BAretired]

How do you guys think that the beam will behave on the (e) 6" cmu wall ? Do you guys distribute the reaction from the ends of the beam with a 45 degree along the height of the wall even though the beam will be distributing the vertical load uniformly along the length of it ?
Also, how do you guys approach to design of the anchors between the beam and (e) cmu wall ?
Attachment is indicating what I am planning on doing ..
The reaction is 15kips.

A beam placed over the wall to distribute load cannot deflect because of the wall below, so it will spread the load over a very short length of bearing. The wood joists are existing, so the 1.5" dry pack grout and the HSS over the wall will require temporarily supporting the joists while the work proceeds.

It might be better to confine the disturbed area as much as possible. You could use a long rectangular base plate to spread the load over, say 24" of wall, i.e. 10" each side of the HSS. This would result in a bearing stress of a little more than 100 psi which should be okay. If the existing joists are bearing on a bond beam, the plate can be anchored into the bond beam each side of the HSS. If there is no bond beam, cells of the CMU wall under the plate should be filled down to foundation level.


BA

 

[Ron247]

I think I am understanding this thread correctly about the beam on the wall. I kinda agree with BAretired, if the beam cannot deflect because of the wall below, you get less distribution. In that case, distribution is more related to how tall the beam is. I would consider grouting under the beam for some distance (example 6") and then leaving a gap (example 24") and the doing another 6" grout. That gives me 2 locations for my load to start down while keeping their overlap from occurring so quick. But I am not sure from your sketch what I am seeing. I cannot make out all the handwriting. Were you a doctor in a previous life?

 

[BAretired]

Doctor in previous life? Good line, Ron247.

BA

 

[dursundrsn]

@kipfoot I wish that you analyzed a simple beam before making a bold statement like " never true " --
Alright, have a simply supported beam and apply 100 plf DL, 120 plf LL uniform load on it. You will find out that the smallest wide flange that you can use is W 5x16, and the lightest HSS member is HSS 4-1/2" x4-1/2" x 5/16" (16.96 lb/ft), and believe it or not actually HSS 5x3x5/16 (14.83 lb/ft) is the lightest but it is barely making it in deflection.