Pre-stressing Strand as Mild Reinforcement 1

[Lion06]

I have a strong recollection that there are limitations on using pre-stressing strands as mild reinforcement (not prestressed), but I can't find it anywhere in ACI 318. Am I way off base, or just looking in the wrong location?

 

[slickdeals]

Some discussion in this thread
thread507-173910

 

[Lion06]

Thanks, slick.

It looks as though the limiting factor for this is the ability to develop the breaking strength before the concrete crushes.

I can see how that's a problem in typical members, but if you have a very deep member, you can easily develop the breaking strain in the strand before the concrete reaches a strain of 0.003.

rapt, are you around? Does that sound about right to you or am I missing something?

 

[rapt]

Lion06,

I do not see how the depth of the member affects this. It is related more to the amount of tension force being developed comapred to the depth.

But this all comes out in the calculations if you do not make the assumption of yield, instead do the calculations by strain compatability and you will know what stress is developed in the strand. To ensure a balanced design you will have to limit the neutral axis depth much more severley than the default limits in codes.

The biggest problem still is how much stress can be developed based on bond and developemnt lengths. In the tests I mentioned in the previous discussion on this, they show, as expected, that it is not possible to develop full yield stress if the strand is not prestressed.
In areas where stress/strain changes gradually along the strand (positive moment areas with UDL loading), you might be able to develop up to 1200-1400MPa.
In areas where the stress/strain rate of change is more severe (negative moment areas, changes in section, point loads, etc) only 800-900MPa might be able to be developed.

So, no you cannot simply use strand without stressing it and assume full yield strength.

 

[Lion06]

The neutral axis depth is always the same for a given tensile force. If you have a very large d (say 40'), then the strain in the steel at crushing of the conrete will be high enough to fail the cable, no?

 

[rapt]

Yes, but that happens with low ductility steel or FRP also if you do the calculations properly. Even Class N steel in Australia and europe (Normal Ductility 5% peak strain) theoretically has this problem for lightly reinforced sections. For normal reinforcing steel calculations most designers simply ignore this check and do not cfalculate steel strain, as the codes do not specifically limit reinforcing tensile strain, though logically they should. In normal situations it does not have a significant effect on ultimate member capacity and is not easy to calculate without computers so codes have ignored it.

The correct solution is to limit the concrete strain to less than .003/.0035 to reduce the steel strain to less than the peak strain. This will result in a deeper neutral axis depth which is the only thing codes really limit, for ductility (this does not mean codes are correct, just lazy).

RAPT gives the designer the option to do this in a design if desired.

 

[Lion06]

ACI assumes concrete crushes at 0.003 - well, that's the usable strain that we're allowed to assume.

Just humor me here for a minute. Looking at this from only a strain and subsequent force standpoint for unstressed prestressing strand - say you have a 20' deep member (let's say it's 100' long so we're not concerned about bond length) with two 1/2" diameter, 270ksi strands. The member is 10" wide with 5ksi concrete. The neutral axis depth is assumed at ((2*0.153*270)/(0.85*5*10))/0.8 = 2.43". With the depth of the strands at say 19' = 228", the strain in the strands when the concrete reaches 0.003 is 0.278. This is much higher than the strain needed to reach 270ksi, which would be 270/28500 = 0.0095.

Is that all there is to it, other than the bond length, of course?

Do you happen to have any literature on the subject? My searches have turned up little.

 

[RWW0002]

What about serviceability issues? Regardless of bond, if grade 270 steel is sized to have the same moment capacity of a section with mild steel, the curvature at service-load moment would be significantly greater than that for mild steel (due to less steel in the cross section). Attached is a sketch of some moment-curvature diagragms of the same section with grade 60 and 270 steel. If high strength steel is used, deflection would govern designs and effectively eliminate the advantages of having high-strength steel.

 

[Dinosaur]

I have not designed a concrete member in quite some time but I seemt to recall there was a maximum YP allowed for mild steel, which was something around 75ksi. Is this still in the code or has it really been that long?

 

[yakpol]

Lion06,

In your example, the strand will yield at strain 0.9(270)/28500 = 0.0085 and will rupture at strain = 0.05 well before concrete compressive concrete strain reaches 0.003. As rapt said, prestressing strands perform poor if not prestressed.

 

[Lion06]

yakpol-

How do you figure that the concrete is reaching a strain of 0.003 before the cable yields or ruptures? As I noted above, the strain in the strand when the concrete reaches a strain of 0.003 is 0.278. This is well above the two strains you just listed, which means that teh steel is straining before teh concrete crushes.


RW002- Point well taken.

 

[yakpol]

Lion06,
The strand will rupture first. According to your calcs the strand tensile strain 0.278 at the time concrete compressive strain is 0.003.
The ultimate strain is near 0.05 for prestressing steel and 0.12 for A706 mild steel (both less than 0.278). The 20-foot deep section is grossly underreinforced, so steel yields and ruptures before strain in concrete reaches 0.003.

 

[Lion06]

yakpol-

That's exactly what you want to have happen to develop the strength of the cable. If the concrete crushed before the breaking strain was reached, THEN you wouldn't get the full strength of the cable. I'm not following the point you're making.

 

[Lion06]

yakpol-

For a typical RC member, you simply check the strain of the steel when the concrete is at 0.003 and if the steel strain is above yield, then you use the yield strength of the bars in the calcs. Why would (from an purely analytical standpoint) this condition be the exact opposite?

 

[rapt]

Lion06,

But what if the strain is above fracture?

 

[Lion06]

Why is that not what you want? You don't need the concrete to reach 0.003, that's just the max that the code allows you to use. If the steel reaches yield BEFORE the concrete reaches 0.003, then all is good, right?

 

[yakpol]

FRACTURE - not yield.

 

[Lion06]

Who cares? The required moment is reached before then. The required moment capacity is reached b
BEFORE anything fractures. The only thing the strain diagram is telling you is that the concrete doesn't crush first. I've never seen any requirement to check steel strain against fracture.

The point I'm trying to make is that the steel will not see the strain associated with fracture, because the moment capacity (moment associated with steel strain reaching fpu) is achieved long before that.

Your point is well taken, but I'm really only concerned with understanding if it's possible to develop the full breaking strength of a non-prestressed prestressing cable. The use is chord reinforcement in a diaphragm. It's common in precast construction to use very little steel (far below code minimums for flexural members) for chord reinforcement. I've always seen mild steel, but I have a guy who wants to use non-prestressed cable. I just want to make sure he's using the right cstrength for his calculation - i.e. 100% of fpu, not like 25% of fpu.

 

[RFreund]

Lion's point IMO is well made if the steel yields before the concrete crushes then the design is ductile. You then limit the capacity of the member to the yield strength of the steel (and verify that the steel yields first).
I don't understand the counter-argument. Is it that the steel is not ductile and will fracture instead of yielding therefore making this a 'brittle' failure state similar to the crushing of the concrete?

EIT

 

[Lion06]

Another way of verbalizing my point is that at some point before the concrete crushes (just at some point, I don't really care when) the steel will reach its full fpu. Then in the nominal moment capacity I can use the full breaking strength.