Non-Contact Lap Splice in ACIP Pile - Bar Spacing Exceeded 1

[humanengr]

I have a case where tension in the ACIP pile must be transferred from the rebar cage to a center bar which
continues to the bottom of the pile. The rebar cage terminates approximately half way down the length of the pile.

The spacing between the cage rebar and the center bar exceeds 6in. (exceeds maximum for non-contact lap splice).

Since there is a considerable over-lap between the cage and the center bar (approx. 46 ft.), my thought is a combination
of shear stress and diagonal shear or diagonal tension will occur between the outer bars and the center bar. The interface
for these stresses will be approximately at the surface of concrete bonded to the bars.

Due to the large over-lap length, the tension load transfer (and therefore, the shear stress) will be spread out
over a long enough length so as not to exceed permissible stresses.

I would like to avoid extending the center bar into the pile cap - otherwise, the center bar tension would be applied directly.

There is also a theory "out there" that concrete between the bars will be in compression rather than tension, in which
case the distance between non-contact lap splices could exceed 6 in. (2nd sketch).

Just wondering if anyone has a critique of this approach or any comments.

thanks.

 

[SlideRuleEra]

I won't comment on the code aspect of your question, but agree that the rebar cage of the 24" diameter pile shown will effectively transfer tension to the central rebar.

http://www.SlideRuleEra.net

http://www.VacuumTubeEra.net

 

[BridgeSmith]

I'm having trouble locating it at the moment, but I read a research report a while ago that indicated that the lap length of a non-contact lap should be increased by the distance between the bars. It seems their research and testing showed that the 45 degree shear failure plane of concrete was applicable to to the situation of a non-contact lap splice. It would appear you have plenty of overlap.

 

[Trenno]

This has got KootK's name written all over this.

 

[dik]

HR: I would think that no increase was necessary... the hoops on the outer bars would ensure the concrete was confined and a regular lap length would suffice... alternatively, you can look for an embedment in plain concrete... but, the confined model would be more correct.

Dik

 

[BridgeSmith]

Dik, it seems what they saw was a combination of a pullout failure of the end of the embedded length of the center bar (or inner concentric cage) and a cone failure to the face of the concrete between the center bar and the outer cage.

 

[Agent666]

As a reference in support of hodrod10s approach, NZS3101 requires a standard development length plus an extra 1.5 times the spacing between the bars for an offset lap. This assumes a 33 degree compression field between the bars. This increase is only required when the spacing between the bars is greater than 3 bar diameters. There is no 6" max spacing limit like noted in ACI318.

Our concrete code was historically based on ACI318, so perhaps something similar is present in ACI.

 

[JLNJ]

In my mind, the symmetry of the assembly helps with the transfer mechanism. Some of those offset bar requirements are probably to mitigate issues with bars in a single plane. You have a nice circle.

 

[KootK]

This has got KootK's name written all over this.

Truly. I have so much to say on this that I've let the perfect be the enemy of the good and delayed my response far too long. I'd originally wanted to post a couple of articles, a bunch more sketches, and some counterpoints to the comments already posted. I'm going to streamline things a bit though and, instead, let the two sketches below speak for themselves. I think that they communicate 80% of what I'd wanted to say anyhow. Effectively: I think that tension transfer should be considered to happen locally and the reinforcement disposition should be made to reflect that assumption.

 

[Agent666]

That solution makes perfect sense to me. Keep the long bar and put the additional confinement (if its required over and above the minimum provided) for a REAL GOOD = FEEL GOOD^TM solution.

That's the one potential downside of S&T, it can accurately address any loadpath that you choose. But if you choose a model that is at odds with the way that the forces naturally want to go in the first place and reinforce for this, the forces will mobilise your chosen loadpath eventually, but there will be some unintended cracking/distress felt where the forces would rather naturally go. Theres a bit of an art to recognising where forces would rather go in some arrangements.