Using rebar for anchor bolts

[smvk3]

I want to use rebar for anchor bolts, specifically, ASTM A615 Gr. 75. My reactions involve rather high moments and tension loads and I am therefore embedding the rebar into a drilled pier (and developing the bar). What failure modes do I need to consider? For tension, do I just need to make sure that I develop the bar and have enough steel area? For shear, can I just use the ACI shear-friction method to transfer the shear load to the drilled pier?

 

[KootK]

Be careful with this. Engineers often use rebar in place of anchor bolts because they think that it's a workaround for Appendix D. For the most part, it's not. The only major failure mode that rebar addresses is side face blow out. Other than that, it's just an anchor bolt. In fact, it's an anchor bolt that actually develops more slowly than a common rod with nuts or an anchor plate.

If you can manage to get your pier rebar to lap directly with your rebar anchor bolts, then you can skip the anchorage voodoo. That's pretty rare though.

The greatest trick that bond stress ever pulled was convincing the world it didn't exist.

 

[KootK]

I got so focussed on the warning that I forgot to actually try to help. If you go with rebar anchor bolts. I'd recommend looking at it as a strut and tie joint. The end result will be long-ish rebar-bolts and some concentrated ties at the top of the drilled piers. Essentially a well thought out non-contact lap splice.

The greatest trick that bond stress ever pulled was convincing the world it didn't exist.

 

[smvk3]

I don't see how you would breakout the concrete in tension; the rebar is developed fully into the drilled pier and is pretty close to the drilled pier rebar cage. As long as cover and development length requirements are met, you won't breakout the concrete in tension.

 

[KootK]

That's the classic misunderstanding smvk3: assuming that embedding the rebar Ld means that you've achieved anchorage to the concrete. You've only got anchorage to the pier in this scenario if:

1) You effectively lap the bolts with the pier cage as I've described above or;
2) You make the Appendix D style anchorage checks work (they won't).

The greatest trick that bond stress ever pulled was convincing the world it didn't exist.

 

[hokie66]

I have often used reinforcing steel for anchor bolts, but never had to deal with "Appendix D". When using threaded deformed bars, be careful of the force used, as the outer skin which is removed in threading is the strongest part of the bar.

 

[KootK]

@Hokie:

1) You didn't use app D but you did take measures to effectively lap the bolts and the pier verts, right? And that required more than Ld? Before strut and tie, we would speak of a 35 to 45 degree failure surface with bars lapped on both sides of it. Sometimes, we still do.

2) I wasn't aware of the higher strength material lost at the outer surface of the rebar. How do we account for that?

The greatest trick that bond stress ever pulled was convincing the world it didn't exist.

 

[hokie66]

1) Yes, appropriate measures are taken in the detailing, depending on the loads involved.

2) See the last page of the attached.

http://www.reinforcing.com.au/~/med...einforcing_Bars_bending_rebending_welding.pdf

 

[smvk3]

I don't see how you need to lap the rebar with the drilled pier rebar cage; if you develop the rebar into the drilled pier, the load will be transferred into the drilled pier through the friction between the rebar and the concrete. What you are talking about is the method ACI appendix D uses (which they call "anchor reinforcement") where you have to lap an anchor bolt according to a a 35 degree tensile failure cone.

For deformed rebar, you don't need to do this as the force is transferred through friction between the rebar and concrete. The design tensile strength of the rebar (LRFD)would be 0.75*(Ane)(Fy) where Ane is the net area of the rebar (which takes into account the area removed from the threads).

 

[smvk3]

For shear transfer, I use the ACI shear friction method with a coefficient of friction of 0.7, and a rebar area of Ane, where Ane is the net area of the rebar (which takes into account the area removed from the threads).

 

[KootK]

@OP: I strongly disagree with your 23:31 post. Consider this extreme situation: Link. If simple anchor bolt/rebar development length got the job done, it would be fine. Clearly, however, it's not fine. The remedy is to run the vertical cage up so that the bars can be effectively lapped, as mentioned previously.

Your question about the use of shear friction here is an interesting one. I'm curious to see what others have to say about that. I don't know the answer.

What you are talking about is the method ACI appendix D uses (which they call "anchor reinforcement") where you have to lap an anchor bolt according to a a 35 degree tensile failure cone.

Not so. This is a long standing rational method that predates appendix D. See AISC design guide number one.

The greatest trick that bond stress ever pulled was convincing the world it didn't exist.

 

[KootK]

@Hokie: thanks for that. It's interesting that ductility is compromisde when, often, that's exactly what designers are attempting to achieve with turned rebar anchor bolts. In a previous thread, didn't you mention that all AU rebar is weldable grade?

The greatest trick that bond stress ever pulled was convincing the world it didn't exist.

 

[smvk3]

Kootk,

In that situation, yes, obviously you would not fully transfer the tensile load into the pier. However, what I am saying is that as long as you develop the rebar into the pier cage (which would have the cage extend up to the top of the pier in your picture), you would fully transfer the tensile load into the pier.

 

[hokie66]

KootK,
The statement about ductility probably being compromised is, I think, a reference to the stress risers created by the threads. I didn't realize that anyone was attempting to achieve ductility by using threaded rebars. That would be misguided. Yes, the Tempcore 500 MPa bars are weldable. However, there is some imported stuff which is not as good, so care is required.

 

[KootK]

I didn't realize that anyone was attempting to achieve ductility by using threaded rebars.

@Hokie: Usually only in the sense that designers are attempting to circumvent brittle concrete breakout failure modes. That's all I was getting at above.

I don't see how you need to lap the rebar with the drilled pier rebar cage; if you develop the rebar into the drilled pier, the load will be transferred into the drilled pier through the friction between the rebar and the concrete.

what I am saying is that as long as you develop the rebar into the pier cage (which would have the cage extend up to the top of the pier in your picture), you would fully transfer the tensile load into the pier.

@smvk3: In my opinion, these two statements contradict one another. If the first statement is true, then it wouldn't matter we stopped the pier cage in my example. If the second statement is true, then it implies some form of load transfer between the anchor rods in tension and the pier vertical reinforcing. And that load transfer would obviously be something akin to a lap.

The only thing that development length accomplishes is to prevent a reinforcing bar from pulling out from the concrete. Development length does not prelude the concrete cone surrounding an embedded reinforcing bar from breaking out from the main body of the concrete element. If you only embed your anchor rods Ld, and those rods are not directly lapped to matching pier verticals, you will have failed to develop the full capacity of your anchor rods. If you don't believe me, check out AISC design guide one. Or Widanto's now famous paper.

The greatest trick that bond stress ever pulled was convincing the world it didn't exist.

 

[smvk3]

Kootk,

Per your original picture, the failure cone would not be at a 35 degree angle (since the rebar does not have a bolt at the end). It would propagate straight towards the edges of the drilled pier. So if the pier rebar above this crack is developed, how does the concrete separate?

 

[KootK]

Per your original picture, the failure cone would not be at a 35 degree angle (since the rebar does not have a bolt at the end). It would propagate straight towards the edges of the drilled pier.

I don't agree with this. In my sketch, the failure plane was horizontal because, given where I stopped the pier cage, the failure mode would be direct tension, not breakout. Breakout style failures happen at an angle because they represent concrete compression struts that are not sufficiently restrained. With headed anchors, which develop almost instantaneously, we assume that the struts travel at 35 degrees which is fairly aggressive. With rebar, the angle must be assumed to be steeper, on the order of 40-55 degrees per ACI strut and tie provisions. This is why I claim that it often makes little sense to use turned rebar in place of anchor bolts. The turned rebar only addresses side face blowout. Other than that, turned rebars are just slow developing anchor rods. Turned rebar generally requires deeper embedment.


The greatest trick that bond stress ever pulled was convincing the world it didn't exist.

 

[smvk3]

I guess we will have to agree to disagree. I am well aware of widanto's method of anchor reinforcment, however, I don't believe you can treat rebar and headed anchor bolts the same way. You would not have such pronounced diagonal concrete struts with the rebar; the tensile force is transferred through the bond stresses between the rebar deformations and concrete throughout the length of the rebar. For a headed anchor bolt, all of this transfer occurs at the top of the anchor bolt nut, which creates the concrete strut shown in the picture above.

 

[KootK]

Good call smvk2. I was getting ready to suggest that we stop at a gentlemanly disagreement as well. For what it's worth, the sketch below shows what I think ought to be done here. It's really just basic reinforced concrete design: passing tension from one bar to another.

The greatest trick that bond stress ever pulled was convincing the world it didn't exist.

 

[dcarr82775]

I think it can work. You need to extend the anchor bars far enough into the pier such that a 35 degree (App D) line off the bottom of the anchor bars intersected the pier verticals at a location to provide at least ld embedment on the pier verticals above that line. Thus you are providing steel to resist the breakout cone per Appendix D. Anything less than that and you are into Appendix D or something akin to it. I have seen a couple posts on this website where people think rebar cast in concrete doesn't have to worry about concrete breakout. I couldn't disagree more on that.