Removing existing columns 4

[hoshang]

Hi

Please find the attached file.
The object is to remove the two columns to have a larger open space. The thicker walls are stone masonry built about thirty years ago; the thinner walls (extension) are hollow concrete blocks built about ten years ago. I have the thought of introducing permanent W12 beam supported by two W8 columns spaced 9.8m clear distance located in the the thicker and thinner wall interaction (before cutting off the columns). The columns are to be supported on single footings. I have a limited depth for the beam of 30cm due to the clear height required for the ceiling. My concerns are
1- The lateral- torsional buckling of the beam
2- Constructability due to the enclosed area. I suppose the beam to be cut into two pieces and spliced.
3- Immediate deflection of the slab
4- Should I use nonlinear analysis?

 

[hoshang]

a W12x170 is 14" deep, as is the European HE 320M, which would be the rough equivalent.

Thanks. maybe HEM320 the best choice.

If you're going to have something fabricated, it may become advantageous to add some type of shear connectors, if the roof deck can be used as a composite top flange

Please read the original post carefully. The slab is already existing. What do you suggest as shear connectors for this case? Provide a sketch, please.

Otherwise, providing lateral bracing at a few location should become easier.

How providing lateral bracing? In early posts I suggested using Cap Channel on the W12 beam. What do you suggest for lateral bracing? Provide a sketch, please.

https://www.screencast.com/t/FimMajImEN

What about using 3 pieces of the beam (2.5m on each side and 4m on the center) and splicing them (instead of using one heavy and long piece) as a construction method due to the enclosed (limit) space.

 

[ukbridge]

You don't need lateral bracing if the beam isn't susceptible to lateral torsional buckling. LTB is only a concern where the top (compression) flange is susceptible to lateral displacement and rotation, the mechanism which causes a beam to fail due to LTB. I think so long as the slab fits snugly against the slab soffit, and the fact that the beam is roughly equally loaded both sides its acceptable not to consider lateral-torsional buckling even if theres no composite action between the beam and slab.

I would absolutely not use a channel, the last thing you want to do is introduce any more potential lateral stability here.

Ultimately its a bit of a judgement call, but unfortunately makes a massive difference in the capacity of your beam. I think @HotRod12 makes a good point, if theres any doubt in your mind use an angle or something to fix the beam to the slab or at the least spend a bit of time reading about it if nothing else for a bit of robustness. If you can get a beam that works for Lcr = 10 within the construction depth, then happy days just use that and move on, its a single beam at the end of the day and its peace of mind for you.

Regarding splicing its basically a calculation isn't it to see whether or not its viable. It comes down to what you can actually fit into the building, if 2 pieces can fit, and you can get away with a single splice at midspan then great.

gl

 

[BridgeSmith]

"The slab is already existing."

I assume that means it's concrete? Making it composite may not be an option worth considering then, as it would likely require chipping out a significant amount of concrete while leaving the reinforcement and then grouting the "pockets" around the shear connectors. Currently, the maximum spacing for groups of shear connectors in the AASHTO bridge design spec is 2 ft. Based on new research, there is a proposed change to increase it to 4 ft, but you would have to comply with your governing design code in that regard, anyway.

"How providing lateral bracing?"

I don't know what the configuration of the roof system is, so I wouldn't know what to propose. What I can tell you is that it doesn't take anything particularly strong to prevent LTB, so long as it doesn't allow any significant lateral movement before it engages. In other words, as long as the beam remains straight, there is very little lateral force exerted. Generally, even small angles placed tight against the edges of the flange, and attached solidly to the roof diaphragm, should provide adequate lateral support. Your governing code should be consulted as to the capacity requirements for the beam to be considered braced. I would not recommend attaching the beam directly to the roof diaphragm, as the resulting composite action could induce large shear forces in the connection, unless provision is made to allow the beam and roof to move independently in the longitudinal direction while preventing transverse movement (i.e. slotted connections).

 

[JLNJ]

During the houseparty that floor will be jumpin'.

 

[BridgeSmith]

"I think so long as the slab fits snugly against the slab soffit, and the fact that the beam is roughly equally loaded both sides its acceptable not to consider lateral-torsional buckling..."

You can't count on friction to prevent LTB, nor does the loading being equal help. If the beam is susceptible to LTB, then it must be rigidly braced or the design capacity will be reduced. However, given the large flange sizes that are required here, and the short span, the capacity reduction due to LTB may be small or nonexistent. at most, one braced point at midspan would likely eliminate any LTB capacity reductions.

 

[hoshang]

In other words, as long as the beam remains straight, there is very little lateral force exerted

Thanks
What's the purpose of the new beam? It takes loads from the existing slab (by tributary area or whatever method). This load induces moment in the new beam (compression stress on top flange and negative stress on bottom flange). The compression stress in the top flange will tend to buckle the top flange while the tension stress in the bottom flange tends to straighten the bottom flange. This will induce LTB in the new beam. Every beam strength formula is based on the restrained length along the beam (whether it's at third point, quarter point, center point,etc.). So, LTB is the main concern for the new beam. This is my understanding of LTB.

 

[Sawbux]

You could put beam above the roof, then put rods down through roof to pick up slab. Done that a couple of times. It will remove head-knocker at ceiling. Just need to make weather tight and brace for LT buckling.

 

[hoshang]

What about using 3 pieces of the beam (2.5m on each side and 4m on the center) and splicing them (instead of using one heavy and long piece) as a construction method due to the enclosed (limit) space.

What about this splice arrangement?

https://www.screencast.com/t/gegFBiOCp

 

[dik]

Best to talk to a fabricator to see what problems there are.

Dik

 

[BridgeSmith]

hoshang, the only thing missing from your description is the mechanism by which LTB occurs, that being the eccentricity. The lateral moment in the compression flange which causes the buckling is a P-delta effect - axial load multiplied by the eccentricity. No eccentricity = no moment = no buckling. It's similar to the ponding effect, where the farther the flange displaces laterally, the more moment there is, and the more moment there is the farther it displaces. Limit the unbraced length, and you limit the lateral displacement.

"So, LTB is the main concern for the new beam."

Depends on the lateral moment of inertia of the compression flange. The wider the flange and the shallower the beam (the displacement is restrained by the bottom flange which wants to stay straight), the less susceptible it will be to LTB. Your beam may be shallow and wide enough that the capacity reduction may be very small or zero, even if unbraced. You'll have to do the calculations for the beam configuration you choose, see what the reduction in capacity is, and decide if it's worth the effort to brace it.

 

[hoshang]

Best to talk to a fabricator to see what problems there are.

Thanks. What problems do you expect with this detail? Mine I would suppose leveling two heavy pieces of the beam.

 

[hoshang]

Hi

I have other concerns. Now, it's about interfering of the new foundation for the 2 steel columns supporting the new W12x170 beam with the existing 2 single column foundation. These 2 existing foundations can't be removed before the construction of the new steel frame is completed. So my thought is that we don't remove the existing footings and drill holes and insert reinforcing bars into the existing footings so that it spliced with the reinforcing bars of the new footings.

 

[BridgeSmith]

Drilling into existing concrete to install dowels is fairly common. A couple of words of caution based on our experience, though:

1) Proper installation of the adhesive anchorages for these doesn't happen without strict oversight. If there isn't someone watching the installation carefully, the holes don't get cleaned thoroughly, so the epoxy doesn't adhere to concrete. Also, the two-part epoxy often doesn't get mixed properly and doesn't cure.

2) Because of the questionable installation practices, we don't count on splices with dowels to transfer moment. If we need the connection to transfer moment, we require chipping away the concrete to expose enough of the existing reinforcement to install mechanical rebar splices.

 

[hoshang]

based on our experience

Thanks
based on your experience, Placing the new steel column foundation on top of the existing column foundation (using adhesive anchors for connection between the two foundation) is better? Or placing the new steel column foundation at the same level as the existing column foundation (using adhesive anchors for connection between the two foundation) is better?

 

[BridgeSmith]

Not sure what you're asking, but essentially my caution is to not rely on tension capacity from a bar set with an adhesive anchor. They should be ok for shear.

 

[hoshang]

my caution is to not rely on tension capacity from a bar set with an adhesive anchor. They should be ok for shear.

Thanks.
If the new and existing footings are at the same level, adhesive anchors (on sides) can be beneficial for resisting differential settlement between the new and existing footings.
I'm sorry, I think I should switch to Foundation Engineering thread if you (Structural engineering) agree, since the discussion is now about foundation.
The discussion about the steel frame was very helpful.
The discussion ended with using W12x170 for the beam (as dik suggested).
Thank you all, especially dik, HotRod10, etc.