Urgent - Concrete Beam to Column Connection 1

[STEVE97]

I am designing a 9 story concrete structure that employs a two way slab system supported by 16" sq. columns. At the second floor level a column is being supported by a reinforced concrete transfer beam that is 16"X42". I designed the beam as simply supported and the columns supporting the beam are non-slender and I have the required capacity per ACI318-89 (Code referenced in the current building code) The column that supports the end of the transfer beam terminates at the base of the beam. I am racking my brain to figure out the best way to connect the transfer beam to the column to have it remain a simply suppoted connection and not have eccentricity on the column. Could any of you give me some advice?

 

[msquared48]

Is this a design-build situation with part of the structure already in place?

Why cannot there be continuity of the beam with the column to help control the eccentricity?

Something is missing here. Transfer beams are not uncommon and really not a stresser. Please fill us in further.

Mike McCann
MMC Engineering

 

[STEVE97]

The foundation is in place and they are forming the columns now. Basically my transfer beam sits over top of the column were I have the beam reinforcing bars running over the top of the column and being hooked at the end. The column reinforcing bars run through the transfer beam to develop the bar. My question is if they pour the column and beam integrally will this produce a eccentricity on the column? Should there be a joint between the bottom of the beam and the top of the column to produce a simply supported condition? The column was designed as non slender and there is no requirement for min. eccentricity as per ACI318-89 unless the column is slender. The number work for the axial load but we want minimal eccentricity on the column.

 

[jike]

Do a quick frame analysis and see how much moment is actually transfered to the column. Make sure the reinforcing can transfer the moment.

 

[STEVE97]

jike,

That would assume that the beam to column has a fixed support. The beam was designed as simply supported so the moment at the columns is zero unless the connection that is detailed has rigidity. From what I understand from the code is that the equations checking the axial capacity of a non slender column take into accound a small eccentricity. Is this sufficient?

 

[msquared48]

As I read the situation, even with the simply supported condition, assuming that there is another beam, probably not a transfer beam, framing into the other side of the column, you will get some eccentricity due to the difference in the loads from the two beams.

Had you not accounted for this load difference? Is the transfer beam a last-minute change from the Architect? If possible, I would try to compensate for the ECC with extra top steel in the beam over the column at this point, or increase the column steel if still possible. Assuming a pinned base for the columns, the doweling should still be OK. If not, there is a problem.

Mike McCann
MMC Engineering

 

[STEVE97]

There is no differential loading as the transfer beam is on one side of the column only and bears fully across the top of the column.

 

[STEVE97]

I just want to make sure that the detail that I provide for this connection between the transfer beam and the column will no produce a rigid connection and induce a moment into the column.

 

[msquared48]

"The column that supports the end of the transfer beam terminates at the base of the beam".

"There is no differential loading as the transfer beam is on one side of the column only and bears fully across the top of the column."

Now I am confused here. Sorry for that. I thought that the transfer beam was supported by a column at each end, which is normal.

Is there a way to post a sketch here as what the situation is? At least it would help me.

Mike McCann
MMC Engineering

 

[msquared48]

If no moment is desired, then just provide a minimum amount of vertical reinforcing at the center of the column rather than at the exterior which would develop more leverage. More like a true pinned connection. You will still have to use nominal bars at the corners of the column though per code, so the column will still see some moment, but not a great amount.

Mike McCann
MMC Engineering

 

[ishvaaag]

A more conventional approach would be to include a true hinge atop, ensuring the reaction point remains where wanted. However you may use the X reinforcement of a central section, is quite studied in some books yet less elegant than a true hinge. There are many things to consider to provide details like this, being a transfer bean I am assuming under no circumnstance the beam will get up from the named support, if not or laterally restraints should be introduced by diverse detailed devices.

 

[hokie66]

Pinned connections in concrete structures are very uncommon because of the nature of the material. I would not attempt to do this. You should design the column for whatever moment the transfer beam brings. If the column has to be more heavily reinforced, so be it.

 

[JAE]

STEVE97, I don't think you can achieve a column condition here where there is no moment induced into the column from the transfer beam. In concrete construction a "pinned" connection isn't really feasible without all sorts of gymnastics.

That said, if you modeled the beam and columns together, and correctly accounted for relative stiffness, the amount of moment sent into the columns might be minimial as the beam would be stiffer, not deflect all that much and thus not rotate at its ends.

If you try to create some sort of gap, or hinged joint to mimic a pinned end, you might induce MORE moment into the columns if the rotation creates any kind of eccentricity on the load applied at the top of the column.

For a major transfer girder/column arrangement, with no other redundancy in the system, I'd be pretty careful about designing this without following the full load path through.

 

[BAretired]

A hinged detail could be used, but why? Surely the transfer beam is much stiffer than the column, so why not let the beam transfer a nominal moment to the column? The condition of no moment transfer between beam and column is not particularly realistic. Why are you racking your brain over such an issue?



BA

 

[Ron]

JAE...ya beat me to the punch....exactly

 

[STEVE97]

I appreciate all the input and I am going to check using relative stiffness.

The project is moving so fast and they are building the structure now and will be pouring concrete in the next few days so I just want to make sure that I have all my bases covered thats all. I guess I just went into freak mode and didnt really just sit back and make sense of it all.

Thanks again for all your help!!!!

 

[jike]

JAE:

Thanks for explaining what I wanted to say, but didn't. I was trying to point STEVE97 in that direction, but I did not explain myself!