Tension Connection for Concrete Filled Steel Pipe Pile

[ramu2krism]

I am reviewing a tension connection detail for 12"OD steel pipe pile filled with concrete.It is common practice welding rebar to the pipe pile for tension this type of connection, Customer wants to avoid welding rebar to the pipe or any type of connection detail which involves welding. They provided a detail with 4-#8 bars embedded in to the pipe pile filled with concrete upto 3 foot with a hook on the other side of the rebar, which would eventually run in to the pile cap.

Since the load transfer occurs completely through bond, they used Eq: I 9-9 from AISC 14th Ed Section I 6.3c to check the tensile strength of the connection, from the load transfer throgh the bond between concrete and inner surface of the steel pipe pile.

- Does the equation mentioned above from AISC is applicable for the condition.
- Has any one ever used or come across tension connection detail for concrete filled steel pipe pile which does not involve welding.

Your response would be truely appreciated.

Sincerely,
krism

 

[emmgjld]

The connection you describe, hooked rebar embedded into the concrete filled pile and then appropriate rebar details in the cap or into the grade beam is what I have seen for the last 38 years+. Welding is uncommon & usually used for oddball connections or to deal with severe mis-alignments.

 

[MacGruber22]

What the?? Since when is it acceptable to rely on transfer of shear solely through bond forces between concrete and structural steel?

It is common practice welding rebar to the pipe pile for tension this type of connection,

Of course, because you can't rely on shear-friction with no mechanical transfer.

"It is imperative Cunth doesn't get his hands on those codes."

 

[Ron]

I'm in an area where uplift and overturning are common considerations in deep foundations. As a result, for concrete filled steel pipe, welding is common between sections.

 

[hokie66]

I think there are two issues here. When driving piles, you have to weld if you are splicing. This welding needs to also provide the required tensile capacity.

In anchorage at the top to the pile cap, I agree with emmgjld that transfer by shear can do the job.

MacG, It seems to me that this is very similar to shear-friction, which relies on bars crossing a plane to provide a clamping force. The steel pipe provides a circumferential clamping force, in my opinion much more reliably so than bars crossing a plane.

 

[MacGruber22]

Good point, hokie. Though, I am not aware of any US design code that accounts for it.

Found a great wiki on CFT load transfer experiments, though.

http://www.northeastern.edu/compositesystems/wiki/Table_of_Experimental_Studies_on_CFT_Bond_Tests

"It is imperative Cunth doesn't get his hands on those codes."

 

[hokie66]

I think most of that research has been done on concrete filled steel columns in compression, to try to define the benefit the concrete core gains from confinement by the tube. A lot of major buildings have been built that way in different parts of the world, but I don't think there is yet actual codification of the methods used, but maybe I just missed it.

 

[emmgjld]

As I said, 38+ years, common to have #5 to #8 bars inserted into the pipe piling 6' to 8' for large steel framed buildings and sign/light posts. I have never heard of a failure of these tension connections. Sorry we are so primitive here in Colorado.

 

[KootK]

Well, it's common knowledge that Coloradans are just cave folk in Goretex.

While those AISC provisions (composite columns) don't distinguish between force transfer in tension and compression, I wonder if they might have been intended for just compression. Like MG22, the thought of such a connection in tension gives me the shakes.

Unless non-shrink concrete is used, I would expect the concrete to pull away from the steel a tiny bit which is not confidence inspiring. In a compression application, I would hope that concrete dilation would counteract that. In a tension application, you obviously wouldn't have that going for you.

If these things are getting used extensively, there must have been some testing done at some point.



I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[hokie66]

Methinks you think too much. Bond between steel and concrete has been used, well, forever.

 

[KootK]

Methinks you think too much.

Surely, this comes as no surprise.

Bond between steel and concrete has been used, well, forever.

Humor me if you will. Name three applications of steel/concrete bond where neither of the following is true:

1) Mechanical "bond" is provided by studs, rebar, channel stubs, knurls etc or;

2) The situation is such that concrete shrinkage would not pull the concrete away from the steel.

I can only think of one... sort of. You can have a concrete encased steel beam without mechanical shear connection. Shrinkage would likely cause the concrete to pull away from some surfaces but not others. Of course, any sane designer would throw some stirrups around the whole thing.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[hokie66]

You named one. The other common one (remember I said forever) is plain bars used as reinforcement. Not sure what the current codes say, but the classic ACI318-63 allowed bond to be one half of that allowed for deformed bars, not to exceed 160 psi. We also commonly use concrete to concrete bond, with different factors than for concrete to steel.

 

[MacGruber22]

They provided a detail with 4-#8 bars embedded in to the pipe pile filled with concrete upto 3 foot with a hook on the other side of the rebar

The glaring issue to me is "up to 3-ft". That is in the vicinity of development length for #8 bars. If this pipe has 150-kips or so of direct tension, I wouldn't inherently trust traditional development length in this situation without code provisions, strong empirical data, or very conservative rationale.

emmmgjld - no one was making an ad hominem attack on you. We don't grow without sharing our ideas as well as critiques. And, besides, 6 or 8-ft (verses 3-ft) of concrete/rebar development into the column could be a huge difference in order to transfer bond to the tube and visa versa . Your initial response to the OP seemed to imply that the 3-ft development was no concern.

ramu - I looked at that Eq. I6-6 (not sure where eq. I9-9 is). You don't get much capacity out of the bond mechanism, certainly way short of the tension capacity of developed 4#8 and probably the 12" pipe. What is the direct tension demand?




"It is imperative Cunth doesn't get his hands on those codes."

 

[KootK]

What the? For rebar, shrinkage is such that the concrete doesn't pull away from the bars, plain or deformed. The bars go along for the ride. And there's no way I'm awarding the concrete-concrete bond as being a valid example. I cry foul due to insufficient similitude.

Here's some kick-ass Canadian research demonstrating that there clearly is some reliable capacity to be had for exactly this setup: Link.

And it seems that I was on to something wrt the dilation business.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[KootK]

@MG22: According to the AISC provisions, you actually have to get the connection done over a distance of 2D. Crazy.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[MacGruber22]

Ah..yes, KootK. I see that in section I4. That is some poor effective transfer length.

2 times the diameter of the pipe doesn't mesh well with developing #8 bars. It seems that if the OP is going to accept this concept, they would want to switch to a smaller bar diameter (and add more bars) that can more easily develop over 2*D. I would still be concerned with the loading type. If it is cyclical, I still wouldn't feel great about it.

"It is imperative Cunth doesn't get his hands on those codes."

 

[MacGruber22]

But, regardless...I still don't see the bond strength equation in chapter I providing the capacity unless the tension demand is not anywhere near the area of steel provided by 4#8.

"It is imperative Cunth doesn't get his hands on those codes."

 

[hokie66]

I'll accept the result of that Australian (not Canadian) research. Monash University is in Melbourne. I can't argue the AISC provision, as I don't have access to it. But generally, I didn't think AISC was involved with piling or concrete.

 

[KootK]

I'll accept the result of that Australian (not Canadian) research.

$%#@!!@#@$!! I change my mind then. Connection no good.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[MacGruber22]

Looks like AISC 360-10 moved the 360-2005 commentary on filled pipe/tube bond stress to the code.

I can't argue the AISC provision, as I don't have access to it.

AISC 360 specifications are free.

https://www.aisc.org/content.aspx?id=2884

"It is imperative Cunth doesn't get his hands on those codes."