Ties in compression members

[shaneelliss]

My question is specifically concerning ACI 318 section 7.10.5.3 which says "Ties shall be arranged such that every corner and alternate longitudinal bar shall have lateral support provided by the corner of a tie...and no bar shall be further than 6" clear on each side along the tie from such a laterally supported bar."

Anyone have a good idea why this is required? In the commentary of this section it notes that tests have been performed comparing fully tied columns and columns with no ties at all and there were no appreciable differences in column strength. If this is the case, why have the requirement for all compression members if it is not needed for shear and it does not help in compression?

 

[frv]

I may be pulling this out of where the sun don't shine, but I seem to remember my concrete professor mentioning it having to do with actual construction. In other words, if the bars weren't tied, there was a good possibility that they would not be located properly within the column by the time it was finished.

 

[JLNJ]

I remember seeing some of pictures from the Kobe earthquake in Japan of single stanchion elevated freeway supports. If I remember correctly the veritcal bars in the piers were not well-tied and there were a lot of buckled bars (and stanchions lieing on the ground).

That's what I picture in my mind when I think about column tie bars.

 

[shaneelliss]

Thanks for the responses frv and JLJN.

I live in earthquake area (Utah) and have a decent understanding of why lateral support of the longitudinal bars is a good idea for earthquake design (especially in plastic hinge zones), but I didn't think chapter 7 of ACI 318 dealt with earthquake design, which is what confuses me. Chapter 21 has other requirements for ties which deal with earthquakes.

I am not sure I completely buy the idea of the ties being needed for construction, though I suppose it would help somewhat. But if you have a tie that loops around the four sides and connects the four corners, then I would think you would have enough lateral support on the middle bars to keep them from bulging much between the main ties at maximum spacing. But maybe not. That is something to consider anyway.

 

[DST148]

As per the ACI commentary the tests were conducted on full size bars - bars without splices. The commentary only mentions about the ultimate strength comparison of tied column and column with no ties. No mention is made about the concrete strain capacities for the two cases.
Seismic and construction issues apart, the lateral ties do come in handy for calculating the splice lengths. (in our office we usually ignore that provision)

 

[DST148]

My previous post, last line: what I meant was ties come in handy for reducing the splice lengths.

 

[BAretired]

Tied columns tested to failure show that vertical reinforcement tends to buckle outward. The purpose of the ties, apart from holding the bars in position, is to provide a reasonable unbraced length so that the vertical bars will not buckle prematurely.

BA

 

[DST148]

Vertical spacing of column ties, thread 167-155640 - This thread discusses in detail the abstract of the reference quoted in the ACI commentary. The referenced paper was published in 1964. I have ACI 318-05 and the commentary still uses that as a reference.
Shaneelliss - You did mention that provisions of Chapter 21 do not apply but still....
Reference - Instruction material complementing FEMA 451: Confining reinforcing can improve behavior in two ways. First it can enhance strength by restraining lateral strains. Second it can increase the usable concrete compressive strain well beyond the typical value of 0.003. Also published literature on full-scale FRP confined reinforced concrete column point to the same conclusion.

 

[shaneelliss]

I get that heavily loaded compression members, especially those subject to earthquake forces can benefit from having the longitudinal bars supported by ties. I understand that strength and strains in confined concrete can be much higher than unconfined concrete. I guess what I don't understand is why the ties are required for all compression members period. Often I see concrete pedestals for steel columns that have relatively low compressive forces for their size and that are not expected to be ductile in an earthquake event and that do not have these ties as the code requires. As a plan reviewer, I often point out this code reference and tell people it is required to have all those ties and the other engineers always just comply without comment, but I feel a little guilty about adding the hassle and expense to the engineer and contractor to add those in when I don't really know if it is doing any good. So I was wondering if in these cases there was justification for me to ignore that code provision and not require the change.

 

[ToadJones]

In pedestals the ties also help to confine the anchorage, no?
I dont really deal with concrete columns, but I have always used the tie provisions for piers and pedestals for column footings.
If nothing else, as others have mentioned, the help in construction.
They also act as stirrups in piers with moment, no?>

 

[DST148]

Just because a member is vertical and carrying some axial load does not mean it has to be designed using the provisions of a comression member.
This is what I have done in the past for lightly loaded concrete columns, piers, and pedestals which are not in high seismic zone, and not part of a lateral load resisting system. I use the criterion given in ACI 318 -02, section 21.3.1.1, to distinguish between a flexural member and a compression member: The factored axial compressive force on the member shall not exceed Ag. f'c / 10. (I use this definition given in Chapter 21 for classification purposes only.) If the vertical member qualifies as a flexural member using this definition, then I just follow the provisions for a flexural member. Let us say the vertical member is subjected to shear. If Vu < 0.5( Phi . Vc) then no shear reinforcing is required per section 11.5.5.1. Ties may be provided from construction point of view only.