Rock achor to concrete 1

[joseantonio1965]

Hi everyone,
There is this rock anchor design (vertical) for which the contractor wishes to determine the length of the actual anchor sticking out of the rock and embeded in to the concrete foundation to be able to whithstand the uplift load (225kN). The foundation will be poured on top of the rock and the anchor will be embeded. It is a passive anchor (no prestressing), and the foundation is footing with pier on top all reinforced.
So the lengh is function of the concrete-anchor interaction, right? how can I find that value? Can the failure cone be neglected since there are rebars paralel to the anchors and ties for those rebars?

Would be the embedment length 0.45*k*Fy*Diam/sqrt(Fc) from the code be enough, considering there is transverse reinforcing across the potential failure cone (in there is one)?

Any help greatly appreciated.

Regards

 

[Lion06]

Is the rock anchor a bar with threads that conform to rebar spec?

 

[joseantonio1965]

Yes it is, it is a continuous threaded rod that you can install a nut to.
Thanks

 

[Lion06]

If the deformations meet the requirements for rebar then you should only have to embed it into the footing by a length equal to the bar's development length.

 

[joseantonio1965]

Thaks for your respose...
This anchor bar's Fy is 1035 MPa, as opposed to the 400 for regular rebar. If I calculate with that, the length is way too high, the thing is that the design load for this anchor is 225 kN which divided by 819mm2 of net area, I get a Fy of 276 MPa, no even close to where it starts to elongate, but if I use 276 as Fy to enter the equation to calculate ld then I get a pretty decent ld, on the other hand, I'm not sure if the ratios strain-stress would be the same...

 

[Lion06]

I don't know what code you're using, but ACI has a provision where you can reduce the development length based on a proportion of As,req'd/As,provided. I don't know that I'd be comfortable reducing the development length based on that, however, given that it's a critical connection. If you're using ACI, I would look at the actual development length equation taking advantage of everything you can (most notably the (cb+ktr)/db factor, which can reduce your development length significantly). If you have a single bar in a sea of concrete you shouldn't have a problem getting the cb factor to 2.5. This will effectively reduce your development length by 60%.

 

[joseantonio1965]

Another approach: would it be too crazy to consider the shear resistance of the concrete, since the threads exceed by far the spec for regular rebar, times the contact area of the anchor bar? the threads pitch are 8 per in which abides for any A325 nut to be snugged in.

I appreciate your input

btw I'm using Canadian A23.3

 

[hokie66]

Why not put a nut on the rod near the top of the footing? If you want, use 2 nuts with a plate between.

 

[joseantonio1965]

Yup hokie, that could make the trick, but the thing is how do I calculate the resistance? Actually there are two anchors per footing, so I could have a plate bridging from anchor to anchor, but I have to have a justification to that, I mean numbers, i.e. how do I calculate the resistance without using the cone principle and taking into account all the rebar is imposing on the footing and pier?

StricturalEIT:even if I use the cb factor of 2.5, the Fy value of 1035Mpa sends the ld value to the roof.

 

[hokie66]

The US guys seem to use just a nut or headed bolt for uplift, based on that Appendix 6 thing they are always arguing about. I like a bit more anchorage myself, thus the suggestion of the plate.

 

[Lion06]

ACI has App. D which deals with anchoring to concrete with nuts and headed studs.

The nut or plate would have to be designed for bearing against the against the concrete (using a max bearing stress of 8f'c based on the high local confinement).

Do the deformations on the anchors you're using conform to the spec for rebar deformations? We recently used very high strength bars (over 150 ksi) on a job in a similar application. These bars were threaded rods, not rebar, but the threads conform to ASTM that specs rebar deformations. Based on that, we treated them as rebar for development length. They were almost 3" in diameter and very high strength. Using 8ksi concrete, the development length wasn't completely unreasonable (somewhere around 7'-8', I believe). I agree that is alot if you're going into a footing. What kind of development length are you getting for your bar?

If the threads conform to the rebar deformation spec then the nut (and plate) is (are) really just belt and suspenders. If the threads don't and the nut is necessary I don't think you're going to get away from the thick footing. When the nut/plate load the concrete locally, that will kick you into App. D. Then you'll need rebar (fully developed) on both sides of the failure plane in the footing. This will cause the footing to be very thick.

 

[a2mfk]

This is like an upside-down uplift connection into a footing the way I am imagining it?

Along the lines of what Hokie suggested, I like to use an embed plate at the end of 4 rods for columns with high uplift. Then you design for the tension in the rods and punching shear of the plate. And local stresses also I suppose, but don't skimp on the plate and that should not be a problem. I usually throw a couple of rebar through the punching shear zone and beyond for good measure and cheap insurance.

Maybe that or something similar will work in your situation...?

 

[delagina]

SEIT, pardon my ignorance. i maybe asking a dumb question. hehe

but your last sentence

"When the nut/plate load the concrete locally, that will kick you into App. D. Then you'll need rebar (fully developed) on both sides of the failure plane in the footing. This will cause the footing to be very thick."

why rebar needs to be fully developed. you can also reduce it based on Asreq/Asprov for tension right? Also you may add more rebars there to reduce the length right?

 

[joseantonio1965]

a2mfk: it's not like that, what we are tring to find (and actualy I did already with some help, lol) here is the embedment length into concrete of a rock anchor. The contractor and I are assuming it's for uplift i.e. tension (could there be the need for rock anchors to work on compresion in the case the rock load bearing capacity is not enough?).
StructuralEIT and hokie66, Thanks for your imput guys, it was valuable and helped me clearly understand how to approach the solution.
Cheers

 

[hokie66]

jose,
a2mfk described what I was suggesting. So if his comment does not apply, neither does mine.

 

[joseantonio1965]

hokie66, I got confused when he mentioned "you design the rods for tension" because we were pased that stage already, but I guess you are right, the principle is the same, thanks for pointiing that out, my apologies.
I ended up calculating the development length of the anchor taking into account the reinforcement around and position of the anchor (dcs+Ktr) and it turned out good, so no need of any other type of afixing.
Thanks a lot guys