Question for Tank Designers - ACI 350-06 - Equation 10-4

[CivilPipe]

The result of this equation is the maximum allowable reinforcing steel stress in a one-way member to control cracking. Center to center spacing of rebar has a significant effect on the result. The closer the spacing, the higher the allowable stress and the less steel required. The code and the commentary for this section does not appear to address alternating bar sizes.

Does the code make any differentiation between these two options?

Option 1 - No. 9s at 12 inches on center alternating with No. 4s at 12 inches on center resulting in 6 inch spacing and a provided As of 1.20in^2/ft.

Option 2 - No. 7s at 6 inches on center resulting in a provided as of 1.20in^2/ft.

It seems that Option No. 2 would be preferable from a stress distribution standpoint but I am interested to know if the code addresses this anywhere.

Thank you!

 

[bones206]

I do not believe that alternating bar sizes are explicitly addressed for determination of fs,max. If I was doing a design with alternating bars, I would conservatively determine fs,max for the #9's @ 12" and separately for the #4's at 12". I can't see how the alternating bar approach would have any advantage over uniform moderate spacing with moderate bar size of equivalent area per foot.

 

[TehMightyEngineer]

Agreed with bones, I would halve the loads and determine fs,max for the two bar sizes separately. I also agree that I see no real advantage over uniform spacing of a uniform bar size.

Ian Riley, PE, SE
Professional Engineer (ME, NH, MA) Structural Engineer (IL)
American Concrete Industries

https://www.facebook.com/AmericanConcrete/

 

[CivilPipe]

Thanks for the response bones206.

"I would conservatively determine fs,max for the #9's @ 12" and separately for the #4's at 12". This would definitely be conservative because it would completely ignore any increase fs,max. The assumption would be 12" spacing instead of 6".

"I can't see how the alternating bar approach would have any advantage over uniform moderate spacing with moderate bar size of equivalent area per foot." It is an economics issue. In tanks, moment can vary significantly at different locations within tank walls. One approach for large tanks is to calculate the minimum steel (regular steel) required for the selected wall thickness. Extra steel is then placed in the corners and other locations where required. The most economical selection for the extra steel is usually a different bar size than the regular steel.

 

[CivilPipe]

Thanks for the suggestion Mighty.

"I would halve the loads and determine fs,max for the two bar sizes separately." The bar size has very little effect on fs,max. For the purpose of this discussion, fs,max is approximately the same for both a No. 9 and a No. 4. The stress in the bars is also approximately the same for the same bending moment or tension meaning that the load (moment or tension) carried by the No. 9s is significantly higher than the load carried by the No. 4s not 1/2.

 

[TehMightyEngineer]

load (moment or tension) carried by the No. 9s is significantly higher than the load carried by the No. 4s not 1/2.

You're correct, I was too hasty in my reply.

Regardless, what I've typically done in the past is use the same size reinforcement and just halve the spacing at high stress locations.

So, you might have #7 @ 6" o.c. at the corners and then at some point it switches to #7 @ 12" o.c.

If I need more refinement then I usually use WWR or will just use a staggered lap splices between differing bar sizes/spacings. Mixing bar sizes in tanks or walls introduces a higher likelihood of error during construction. I generally only mix bar sizes in beams and other small elements where errors are easily identified and corrected without much difficulty. You don't want to receive the email where they tell you they accidentally put in only #4 bars at 6" o.c. before the inspector noted the error and they need advice on how to proceed as they're already behind schedule.

Ian Riley, PE, SE
Professional Engineer (ME, NH, MA) Structural Engineer (IL)
American Concrete Industries

https://www.facebook.com/AmericanConcrete/

 

[CivilPipe]

Never had any problems alternating bar sizes as long as there are two sizes difference in the bars and it's clear on the plans. I appreciate the opinions on other issues; however, I would still like thoughts on the original question concerning the code.

Thanks!

 

[JoshPlumSE]

There used to be code commentary on the reason behind this provision, I believe. Or, maybe I'm thinking of the PCA notes. Regardless, the concept is that you're better off for crack control using smaller bars at tighter spacing. Very important for ACI 350 applications, which is why ACI 318 and 350 now have different procedures for crack control.

You could create a transformed section including both bars and use that to calculate stress in each bar. Honestly, I don't think the stress will be all that different. If they're at approximately the same depth, then you'd get a similar strain in the bar.... hence similar stress. The #9s would have a much higher force, but about the same stress.

 

[CivilPipe]

JoshPlum,

Everything you have said is correct.

My question goes back to equation 10-4. Is it appropriate to use s=6 to calculate fs,max for Option No. 1 in the original post?

 

[bones206]

This would definitely be conservative because it would completely ignore any increase fs,max. The assumption would be 12" spacing instead of 6".

Agreed. I tend to be a lot more conservative than usual when it comes to concrete tanks. I also try to make the design simple for construction and minimize variation. If someone was pushing me to use the alternating bar size scheme based on economics, I would probably be more inclined to push back than to try to finagle the code to suit the more economical design scheme. I guess it would depend on who was pushing and how hard.

On the other hand, Section 10.6.4.4 allows us to use a less precise approximation for the strain amplification factor. That implies that equation 10-4 itself does not absolutely require total precision in determining fs,max. If we all agree that's the case, then one could argue that using an averaged bar diameter of 0.8" at 6" spacing is an acceptable approximation in equation 10-4. It would be hard to definitively justify but I can't reject that logic outright either.

If you have experience successfully using the staggered bar sizes at corners without any adverse effects, then there is probably nothing inherently wrong with doing that. I just don't think it's explicitly addressed in the code. If it is a common enough practice, maybe that's something that could be brought up in the ACI 350 committee for consideration in future code editions.

 

[JoshPlumSE]

Absolutely, I think it is okay to use s=6" to calculate fs,max. It is the distance from the current bar to the next one that will help with crack control.

It would be more complex if you had 8" to the bar on the right and 4" to the bar on the left. But, even then I'd interpret it more like a trib width of concrete for this one bar for what it has to resist for cracking.

 

[CivilPipe]

Thanks everyone for their input.