Column min. reo when Asc*fsy>0.15*N* (AS3600)

[rscassar]

Does anybody have any issues with the clause in the code regarding column design which says:

The cross-sectional area of the longitudinal reinforcement in a column shall-

a) be not less than 0.01*Ag except that, in a column that has a larger area than that required for strength, a reduced value of Asc may be used if Asc*fsy>0.15*N*.

Just say I am designing a column (N40) with 1% reinforcement (fsy=500MPa). The design squash load (which doesn't get designed for because of minimum eccentricity requirements) is calculated as follows:

0.6*(0.99*Ag*(0.85*40MPa)+0.01*Ag*500MPa) = 23.2*Ag [MPa]

If we solve what component of Asc*fsy this is we get over 20%.

0.01*Ag*500/(23.2*Ag)=0.2155

So even when columns are designed to work at squash load (which we cannot because of minimum eccentricity requirements) for 40MPa concrete Asc*fsy will always be greater than 0.15*N*.

Am I interpreting this provision of the standard correctly?

All opinions are welcome and encouraged.

 

[rowingengineer]

Kipfilp,
The comparison is to N* not ultimate capacity. But you have lost me a little.


An expert is a man who has made all the mistakes which can be made in a very narrow field

 

[rapt]

kikflip,

Asc * fsy has to be equivalent to at least 15% of your applied axial force in this case. It is not related to capacity.

 

[rscassar]

Take a 400 square column with 40MPa concrete and 1% reinforcement. The design squash load (phi*Nuo) is approximatley 3700kN. Say my N* is 3700kN,

Asc*fsy=1600*500=800kN.

15% of 3700kN is 570kN<800kN so I am allowed to design for less than 1% reimforcememt as long as my N* is less than phi*Nuo.

 

[rapt]

kikflip,

Yes, if your applied load was 3700KN, then this is what the code says you can do. It will actually be lower as you have to allow for minimum moment as well, you can never have Nuo as your applied load, so .15 of N* will actually be even smaller.

 

[rscassar]

I guess what my argument is is that the it can always be proved that Asc*fsy>0.15*N*. I'll explain through the use of an interaction diagram later tonight.

 

[rapt]

Kikflip,

But it gives you a minimum area of reinforcement required. You can only reduce Asc to the point where Asc*fsy >= .15 * N*. In your example the minimum would be 570/800 = .71%.

 

[IDS]

I think Kikflip has a valid point. Since the 0.15Asc*fsy requirement will always give less than 1% steel for 50 MPa concrete or less, and since columns should always be at least a little bit bigger than required for the design load, it seems that the 1% requirement is redundant except for high strength concrete.

I don't know if this was the intention of the code writers, but if it was it could be worded more clearly.

Doug Jenkins
Interactive Design Services

http://newtonexcelbach.wordpress.com/

 

[rscassar]

Thanks Doug, that's the point that I was getting at. I have produced an interaction diagram for a 500x1000 column (65MPa) concrete and reinforced with 16-N20's (1% reinforcement). Calculations show the maximum N* for the section considering minimum moment is 14,773kN, but Asc*fsy/0.15 is 16,755kN, so the 1% minimum reinforcement may be waived.

I don't design columns with less than 1% reinforcement, my rules are:

- Minimum 1% reinforcement
- Vertical bars to have a minimum clear spacing of 100mm
- Use 250MPa ties where possible unless higher grade is required for shear or core confinement

The reason I post the question is that I have in the past needed to defend myself for having the minimum 1% reinforcement in columns despite the fact that the code says that a reduction in steel can be used.

 

[hokie66]

I'm relying on my memory here, so somebody pull me up if I'm wrong. Before AS3600 (AS1480), the minimum requirement was 1.0%Ag, but if the section was larger than necessary for architectural reasons or whatever, you could reduce that percentage, with 0.5% of the larger area as the absolute minimum. For some reason, that simple provision was changed, perhaps not for the better. I don't believe that ACI318 has this provision, but rather sticks with 1.0%.

Enough with my memory, will have to look it up when I get a chance.

 

[IDS]

I found a detailed report on the minimum reinforcement question at:

http://www.michigan.gov/documents/MDOT_RC-1473_161865_7.pdf

Skipping to the conclusions the answer seems to be "it depends".

Maybe we should keep the 1% minimum, with a reduction allowable if a rigorous analysis including creep and shrinkage is carried out.

Doug Jenkins
Interactive Design Services

http://newtonexcelbach.wordpress.com/

 

[hokie66]

You must have pushed the wrong button, Doug. That report is about salt.

 

[IDS]

Sorry, try this one:

http://www.utexas.edu/research/ctr/pdf_reports/1473_S.pdf

Doug Jenkins
Interactive Design Services

http://newtonexcelbach.wordpress.com/

 

[rapt]

KikFlip,

I have asked the committee for clarification.

The old absolute minimum of .5% has disappeared too!

 

[rscassar]

thanks everyone for their time and contribution to this thread.

 

[rapt]

Kikflip,

Still waiting for answers on this.

One answer so far is that the reduction is only allowed if the reinforcement is not needed at all, ie the column as unreinfroced can carry the load by itself. But this is not how the code rule actually reads. Unfoirtunately, Standards are rearranging their website and minutes of meetings at which this was decided are not currently available so it is proving hard to egt answers.

Interestingly, British code BS8110 allows a minimum of .4% reinforcement.

 

[IDS]

ie the column as unreinfroced can carry the load by itself. But this is not how the code rule actually reads

It doesn't seem to make a lot of sense to me either. This would mean that a column loaded to 11% of its capacity under axial load, but requiring reinforcement for the bending moment, would need at least 1% reinforcement, whereas a column loaded to say 90% of its axial capacity but with a small bending moment could have less than 1%.

It seems to me that either the conditions when the 1% rule applies need to be better specified (as they are for confinement reinforcement), or it should be stated that 1% is a deemed to comply provision that may be relaxed if a more detailed analysis is carried out (if that is the intention).

Doug Jenkins
Interactive Design Services

http://newtonexcelbach.wordpress.com/

 

[degenn]

The reason for the 1% minimum comes from the ACI318 code (10.9.1). If you have 1% steel then the steel can resist shrinkage/creep stresses caused by the concrete. With less than 1% the steel will yield, which may in turn effect straightness and buckling.

The allowance to go below 1% is if the column is large for non-structural reasons, where presumably this is not a problem.

This is, I think, often not appreciated by engineers who design blade columns as walls with less than 1% reinforcement. ACI318 is more clear than AS3600 in defining a column.