Concrete Compression Members and AS3600

[OzEng80]

Hi

I would really appreciate some assistance to confirm some fundamentals and sort out some nagging uncertainties that I have with respect to concrete compression members.

Firstly, I tend to define a column as a compression member that utilizes compression reinforcement and a wall as a compression member that doesn’t.

Columns
It is my understanding that a column subject to a load takes the force in the concrete and steel with stresses in each proportionate to the modular ratios. Creep and shrinkage further reduce the stress in the concrete and increase it in the steel.
Q – Given that all column calculations include the vertical reinforcing how are situations/zones where the reinforcing is not developed handled? I am referring to column connection situations where either:

- A compression splice is not provided with the starters (vertical column bars not developed in compression until about 20db).

- Less starters bars (diameter, number etc.) than the vertical reinforcing are provided or the starters are not developed in compression themselves (reduced fy) or are in a different position (which they have to be) than the design cross section.

- Negligible connections (shear dowels, precast connections) etc are provided.

Given the requirement for developed (in compression) reinforcing it seems that the critical cross section (and therefore the design cross section) should be taken at the top or the base of the column (<20db) for all columns that don’t have fully developed starter bars providing compression splices with all of the vertical reinforcing. The design cross section for the most extreme situation (shear dowels) is effectively an ‘unreinforced’ cross section (a wall?).

It seems possible that higher bearing stresses could be permitted at these local end zones, but I cannot find any justification or design methodology for addressing the stated situations. Help?

Walls
Despite walls tending to have a much lower reinforcing ratios than columns they still carry load in the exact same fashion as columns (ie stresses proportionate to modular ratios etc..) . I am therefore confused by the arbitrary restraint requirements for the vertical reinforcing. AS3600, CL 11.6.4 requires restraint for walls ‘designed as columns’ when N*<0.5phiNu. I design my walls, and most of the time my wall end zones as walls, not as columns (what a sentence!). I still end up with situations (tension loads, wind loads, lifting in precast) where walls end up with a very high percentage of steel – not for compression.
I assume that chapter 11 has been written on the basis that a given cross section cannot be loaded such that the stress in the steel exceeds the buckling stress (anyone know what that is?) and restraint for the vertical reinforcing is therefore not required.

Q- How can the above assumption be justified without a specified maximum reinforcing ratio? Isn’t it possible for heavily reinforced (not for compression) walls have the bars buckle and spall? Or is the fact that the bars are not heavily loaded (below the buckling stress) eliminate this problem? (Note that the draft code waives restraint requirements for walls ‘designed as columns’ for either - N*<0.5phiNu, concrete strength is <50MPa and vertical reo is not used in compression, % reo is not greater than 0.02 with min horiz of 0.0025).
I have had (precast) walls that have been designed close to their limit as walls that have ended up with close 0.01 of steel in them due to lifting requirements trimmers etc.. I ended up specifying ties (not quite complying with Cl 10.7.3.1 ‘to every alternate bar’) just because it ‘didn’t look right’…

Q- Are reinforced shear walls subject to the slenderness beam limits (Cl 8.9) as they are ‘cantilevering deep beams’?

Q- Is a reduced design cross section required for precast walls with chamfers (say 20x20) at the ends>

Thanks in advance for your time! It is much appreciated.

 

[hokie66]

A shot at your first question. The applicable section for bearing surfaces is 12.3.

Are wall vertical bars required to be restrained just because the amount is larger than nominal for other than compressive reasons? I don't think so if the concrete is designed to take the load.

I don't believe 8.9 is intended to apply to shear walls, but then the Code doesn't do a good job of describing what is required for shear walls.

I have always ignored the chamfers, and I think that is standard practice.

 

[OzEng80]

Thanks hokie.

‘A shot at your first question. The applicable section for bearing surfaces is 12.3.’

The question was more specifically: what is the design cross section for a column without fully developed starters; and secondly; how do you design it?
The bearing anology would be applicable to the column/base interface but certainly cannot be utilized above that.

’Are wall vertical bars required to be restrained just because the amount is larger than nominal for other than compressive reasons? I don't think so if the concrete is designed to take the load.’

Agreed but if there is no defined upper limit on reinf. % how much concrete is there?

’I don't believe 8.9 is intended to apply to shear walls, but then the Code doesn't do a good job of describing what is required for shear walls.’

I think if the end zones are within the axial capacity as either a wall (or if reinforced accordinly) a column there won’t be local buckling.

’I have always ignored the chamfers, and I think that is standard practice.’

I remember seeing some reference to ‘acceptable recesses’ somewhere. It gave maximum lengths and depths for vertical and horizontal ‘grooves’ without having to reduce the design capacity. Seems likely that along the base would warrant a waiver.

 

[hokie66]

1. I think if the column bars are not developed adequately at some point which is not braced, you must assume a hinge or a cross-section with the reduced reinforcing applying.

2. I would use the gross area of the wall.

3. I think we agree.

4. You asked about corner chamfers, which I ignore. As to horizontal grooves, I decrease the assumed thickness if the grooves are anywhere other than at the floors. At the floors, I use bearing stresses, so they don't control.

 

[OzEng80]

Hokie/others; I think a hypothetical would be useful:

Say a 300x1000, N40 concrete column with 10N20 verticals. Say for whatever reason that 4N32 starters are placed centrally at the base (instead of 10N20); the vertical column bars continue and cog into a 250 thick slab at the top.

What is the design cross section, assuming pinned connections for the design?

At the base, the 10N20 are not splicing adequately with the N32 starters and are therefore not developed until 32db - 640mm up the column. The base design section is therefore either the 4N32 centrally located or some ratio of partially developed N20’s and reduced fy N32 (somewhere between 0-640mm).

At the top, as previously established 640mm is required to develop a bar in compression. The bars terminating in the slab are effectively only (250-50cover)/ 640 = 31% developed (cogs don’t count in compression). Therefore the design cross section at the top of the column should be using a reduced yield stress of 500x.31 = 155MPa.

In the past I have tended to treat pinned column connections as pins, with little consideration to the actually ‘development of the column reinforcing’. Given that I have never heard of any design procedure (in any text books, uni or in practice) for partially developed reinforcing in columns, I have assumed my methodology has been correct. Isn’t the buckling load based on a cross section mid-height of the column? Perhaps this is a reason?

So……What is standard practice? What justification can be used not to use the reduced cross sections listed above? Are other standards more specific with ‘design cross section’ notation?

And, what about the other scenarios; precast column with dowels or just a mechanical connection; starters offset to miss verticals (not joggled)?

As this seems to be a very basic concept that I have gone a bit off the rails with, some ‘straightening out’ and examples of other engineer’s design practices would be much appreciated.

Thanks

 

[asixth]

OzEng80,

What can I say, that is quite a question and shows that you are thinking hard about your designs, which is good.

I recently designed some columns of very similar dimension and reinforceing to what you described. I am a relatively inexperiences engineer however I will list my approach to the design.

1. I would always match starter bars with the column reinforcing, it just seems illogical to me to provide anything less for the starter bars.

2. I also assummed the columns were pinned, therefore the maximum moment was at the column-beam junction for both gravity and lateral loads, therefore the critical design section is indeed at the top of the column. No, I didn't consider the developed stress at this region, I just assummed that the bars where fully developed and I can say every engineer in my office would do the same because no-one picked me up on it. Our standard has been to provide a consistant horizontal and vertical lap length, I don't think there is any rationale behind these values, I hate to say it but that's the way our splices have always been and they seem to be consistant with industry standard (not necessary the Australia Standard AS3600).

I cannot comment on the concrete walls because I haven't designed any loadbearing walls. All walls in the building I have been designing are just masonry partitions.