Modify Development Length for Column Dowels in a Footing

[smith2727]

Let's say column was designed with (x8) #6 bars (A_s = 226 in^2). If half the verticals were switched to #8 bar for constructability (A_s = 314 in^2), we can reduce the specified development length of all the dowels by (226/314 = 0.72)? Is this correct? Or does this reduction only apply to the #8 bars.

 

[hokie66]

So now you have 4#6 and 4#8? I would say stick with the development length of the specified #6 bars. Why quibble?

 

[smith2727]

Thanks Hokie! Yes I have 4#6 and 4#8. Didn't know if there would be any reason to need a larger development length for the #8 (even though they are in excess of the original design).

 

[TLHS]

You're probably fine. Two potential problems that jump to mind with upsizing bars.

1. You could technically run afoul of the code provisions on shear friction, that tell you that you need to fully develop the bars. If you're using shear friction at that joint, you may not technically be in compliance.

2. Seismic applications where you may be required to do capacity design.

 

[hokie66]

More quibbling, TLHS, I think.

 

[Ingenuity]

#6 bars (A_s = 226 in^2). If half the verticals were switched to #8 bar for constructability (A_s = 314 in^2)

Curious, how did you calculate your A_s?

8#6 = 3.52 in²!!!

 

[smith2727]

Ingenuity. Good Catch.

A_s (designed): 3.52 in2
A_s (actual): 4.91 in2

 

[TLHS]

It's quibbling, but worth having in the back of one's head when making this kind of decision, especially when we don't really know any context.

Basically, it's good to know the potential issues and then decide that they aren't a problem in a given case. The alternative is to not recognize the potential issues and miss them on the one percent of cases where they could matter significantly.

 

[hokie66]

As to the shear friction issue, the whole concept of shear friction is to develop a clamping force. With larger bars of the same embedment, surely the clamping force is increased. The dowel component is also increased. How could there be issue?

As to the seismic capacity design, I don't know enough to criticise your statement, but can't imagine how increasing the size of footing dowels for constructability would be detrimental.

 

[TLHS]

I'm not getting into the shear friction argument. There's certainly arguments both ways, but the code provisions clearly say that you need to develop the bar fully if you're going to use shear friction. I recognize that people dispute that, and they might be right to, but it is a code compliance issue that people should be aware of when making decisions.

The seismic issue is related to ductility. It has to do with dowels and with general strength of the column. Reinforcing details are a big deal for concrete structures detailed to seismic systems with higher R values. Increasing reinforcing can shift the assumed locations of plastic hinges or screw up capacity design related requirements for seismic design.

Additionally, I don't know the ACI requirements for seismics that well, but CSA A23.3 has a requirement that you need to fully develop the reinforcing of columns and walls that are part of the seismic force resisting system into the footing or other foundation type to ensure that your failure mechanism is reasonably ductile.

 

[KootK]

My thoughts:

1) One drawback of partial development is that your anchorage failure mode becomes non-ductile. As such, if you guess the demand for development wrong on any particular bar, you may loose the capacity of that bar entirely as part of the redistribution process.

2) Is it just the starter dowel sizes that have been altered or is it the column verticals as well? Because of issue #1, I believe that it's important to design each bar for the anchorage demand associated with the force expected in that bar from the column design process. For pure axial load and vert sizes matching dowel sizes, I agree with OP's proposed method. For load cases involving moment, it may not be a suitable approach.

3) It's well known that creep results in column rebar absorbing more compression that we assume in our typical column design calculation algorithms which do not account for time dependent effects. For this reason, it's tempting to recommend designing the dowels for the yield strength of the associated verts regardless of the calculated axial force in the bars. Again, if you go with partial development and guess the demand wrong, you risk losing the contribution of the associated bar altogether.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[dik]

The cost savings is pennies... and the results are less than clear. Might be best to not consider the application.

Dik

 

[Nor Cal SE]

I agree with those who say it's possible to overthink this, especially if the column is providing only vertical load resistance. In this case, I think all the basic recommendations already provided are fine. If this column is resisting lateral load, such as a column in a moment frame, then I would just double-check the ACI code provisions associated with the particular lateral load system at hand.

I would have a difficult time buying that if the column was adequate with 8-#6 dowels, suddenly it would be inadequate to have 4-#6 dowels and 4-#8 dowels. However, I don't design many concrete moment frames.

 

[Shu Jiang PEng SE]

The development length of the rebar can not change if it is calculated per #6 bar, the development length of the rebar can modified by As(#6)/As(8) if it is calculated per #8 bar. refers section 12.2 and 12.3 of ACI-318.