Precast Culvert Shear Connector Design - CSA S6-14

Mmmm, glad we don't have that harsh of a requirement from AASHTO. I usually rely on the shiplap shear key between sections with some threaded rods to hold the joint from spreading. I'd start by looking to see if a grouted shear key can work as I don't believe you're going to get 60 kN/m out of a normal shiplap joint.

I've never tried actually detailing a shear connection. Maybe weld tabs or something like that could be used effectively.

Professional Engineer (ME, NH, MA) Structural Engineer (IL)
American Concrete Industries

It sounds like you just want to make sure the male and female joint is strong enough to resist 60,000N/m. I think that is doable. If you have that value, you do not need shear connectors. That is how I interpret it anyway.

DenverStruct said:

It sounds like you just want to make sure the male and female joint is strong enough to resist 60,000N/m. I think that is doable. If you have that value, you do not need shear connectors. That is how I interpret it anyway.

Click to expand...

I interpret it similarly.

Professional Engineer (ME, NH, MA) Structural Engineer (IL)
American Concrete Industries

Thank you for the responses. I do agree that providing an adequate shiplap shear key should theoretically preclude the use of steel shear connectors, but providing a shear key is not an option in this particular scenario. This is because the goal is to "extend" an old box culvert by adding a new precast piece, so the simplest solution to satisfy the joint shear transfer clause would be to provide steel shear connector plates connected to the old culvert by means of post-installed anchors (a hypothetical sketch of the intended arrangement is attached).

My question remains, assuming you have a 1-meter long joint with the minimum of two shear connectors provided (as shown in the attached sketch), how would you go about analyzing the steel plates and the forces in the anchor-bolts? How much tension do you think each of the 4 bolts on each plate in the attached sketch would experience due to the code's 60,000 N/m per joint?

It seems like the 60,000 N/m is applied on one side of the joint. The intent is to simulate the differential settlement caused by a wheel on one side of the joint and no wheel on the other (as per the associated commentary clause, attached). I ask because I was not able to obtain any references/examples to validate the approach I am taking to designing this element.

Attachments:

As an alternative, what about doweling rebar into the end of the existing culvert and using that rebar with a closure pour to connect to projecting rebar from the new culvert. Similar to below:



Professional Engineer (ME, NH, MA) Structural Engineer (IL)
American Concrete Industries

Hmm, that is pretty hard to know what you are supposed to do. But before you go through all of this and use your engineering judgment, have you considered:
1. Putting 150mm slab right at this transition only?
2. Is this extension at the end of the run? Will traffic go over it? Most of the time the end of a box culvert will only have soil or sidewalk and you are 2 joints away from traffic.

agree with TEH

typical dowel connection for box culvert

Thank you all for the responses, my comments to your responses are as follows:

TehMightyEngineer said:

As an alternative, what about doweling rebar into the end of the existing culvert and using that rebar with a closure pour to connect to projecting rebar from the new culvert.

Click to expand...

DenverStruct said:

Hmm, that is pretty hard to know what you are supposed to do. But before you go through all of this and use your engineering judgment, have you considered:
1. Putting 150mm slab right at this transition only?
2. Is this extension at the end of the run? Will traffic go over it? Most of the time the end of a box culvert will only have soil or sidewalk and you are 2 joints away from traffic.

Click to expand...

cvg said:

agree with TEH

typical dowel connection for box culvert

Click to expand...