Resultant Anchor Shear Question on Ledger 2

[JD P.E.]

I am designing a platform that is supported on two sides of a concrete pier via a cantilevered ledger. Regarding the top member axial shear, would it be a fair assumption that all three anchors share it equally similarly to a regular bolted connection design?

(Also, ignore some of the other reaction directions it was just a quick sketch I made to initially understand the problem, I know they are wrong.)

 

[bugbus]

Realistically, probably yes. But the edgemost anchor (depending how close it is to the edge, and therefore, how ductile its failure mode is) might not have the ability to deform sufficiently to allow the load to spread evenly across the three anchors as assumed. There are various requirements, depending on what code you're working to, that can force you to assume that the edgemost anchor takes the full load and needs to be designed for that.

 

[JD P.E.]

Realistically, probably yes. But the edgemost anchor (depending how close it is to the edge, and therefore, how ductile its failure mode is) might not have the ability to deform sufficiently to allow the load to spread evenly across the three anchors as assumed. There are various requirements, depending on what code you're working to, that can force you to assume that the edgemost anchor takes the full load and needs to be designed for that.

Thanks. I am going to assume full load on the edgemost anchor. Would you happen to know general code references for forcing this assumption? I am working with AISC and ACI 318.

 

[bugbus]

It's mentioned somewhere in ACI 318 Clause 17.5.2, possibly also in Appendix D, but I almost never use ACI 318 so others can hopefully chime in.

Another reference is SA TS 101:2015 Clause 4.2.2.2 in Australia

 

[icebloom]

SA TS 1-1:2015 in Australia has been replaced by the AS5216 standard now. As far as I know all codes around the world (ACI, Eurocode, Australian standards) use the concrete capacity design method (CCD) for calculations involving concrete anchorage, and as codified the CCD method makes you assume that for edge breakout your shear force acting towards an edge is resisted by the anchors closest to that edge. Endplates or baseplates for cast in or post fixed anchors typically have larger clearance holes than bolt holes for steel to steel connections, which means there is the chance that the anchors closest to the edge of concrete will take all of the load if the other anchors in the group are not in contact with the fixing (both due to the larger holes, as well as potential for installation related issues such as a slight angle in the anchors, etc). Concrete cracking is the failure mode the CCD method designs for, and as such if your most critical anchors take the full shear and your concrete cracks before the load can be redistributed to the other anchors, your connection will fail.

Can't remember what the ACI says about it but EN 1992-4 (and if you want to go more in depth you can look at fib bulletin 58 guide to concrete anchorage) allow you to share the load amongst all anchors provided that there is a method to fill the annular gap between an anchor and its fixture. For typical shear for an endplate or baseplate towards edge of concrete this method is the shear taken by back anchors method, where all anchors are assumed to be effective and your breakout prism begins at the topmost anchors in the connection rather than the closest to the edge. For post fixed connections, HILTI have a method in their software called either filling the holes or SOFA method which for certain anchor configurations (limited by the guidance given in the fib-58 guide) allow shear forces to be resisted by all anchors, not just those closest to the edge. In practice, HILTI produce washers for anchors up to an M24 which are specifically designed to allow the injection of epoxy into the gap between an anchor and a fixture after the washer is installed, which would ensure load sharing between all anchors (google HILTI filling washers).

 

[Eng16080]

I am going to assume full load on the edgemost anchor. Would you happen to know general code references for forcing this assumption? I am working with AISC and ACI 318.

As mentioned by bugbus, I would look at ACI 318, Section 17.5.2. Specifically, the commentary Section R17.5.2.1. The commentary is not technically the code, so I'm not sure you would be forced to follow it, but it seems like good practice. That section gives 3 potential cases to check, one of which is assuming all the load is on the edgemost anchor (Case 3). Personally, I would probably design it per this assumption (most conservative) and see if it works. If not, then see if you can justify a less conservative case. If you can adjust the edge distance and spacing between bolts, I think you should be able to justify following Case 1 or Case 2 which should give higher capacity.

One minor clarification per bugbus's post, the new versions of ACI 318 contains the anchorage requirements in Chapter 17. This used to be in Appendix D. Currently there is not an Appendix D.

 

[TRAK.Structural]

Depending on the type/size of anchors you are using, you may be able to use (2) rows of anchors. The C12 is deep enough, but with concrete limit states its hard to tell by just looking which configuration will produce a higher capacity.

 

[JD P.E.]

Depending on the type/size of anchors you are using, you may be able to use (2) rows of anchors. The C12 is deep enough, but with concrete limit states its hard to tell by just looking which configuration will produce a higher capacity.

Thank you for your feedback. I went the route of 2 rows of anchor bolts (4 total) because there was room and "why not". I don't love just relying on two anchors in this type of design, but I'm sure others do it. The concrete edge distance is the limiting factor here with the minimum dimension being about 10" from the edge of concrete.

One more follow up on this (see attached dwg), there is a portion of the platform that will have knee braces - to resolve the axial force to the anchors I am treating it like a t section for prying. I feel good about that approach, just wanted other opinions. Seem reasonable?

 

[EngDM]

It's mentioned somewhere in ACI 318 Clause 17.5.2, possibly also in Appendix D, but I almost never use ACI 318 so others can hopefully chime in.

Another reference is SA TS 101:2015 Clause 4.2.2.2 in Australia

I am wondering if anyone is familiar if the CSA A23 have a similar requirements?