Anchoring to Concrete - Large Shear Force

[LSPSCAT]

I am working on designing several large brackets to anchor to an existing concrete structure. The brackets are anchored to a wall near the top edge with design loads ranging from 50 kip, 100 kip, and 200 kip. Load is applied at approximate 12" eccentricity to the face of wall. Anchors must resist tension and shear. I am leaning towards design with a shear key to transfer the shear load; as I cannot get reasonble anchor bolt groupings and associated bracket size using bolts to transfer shear. Some contention at that juncture because we do have evidence of older designs which simply used anchor bolts to transfer shear force.

Design specification is to use post installed adhesive anchors. Cannot thru-bolt the anchorage.

The design work is preliminary at this point and we do not have existing drawings to indicate existing wall reinforcing.

Current design references are AISC Design Guide 1, Design of Monopole Baseplates, and Appendix D.

Does anyone have any other strategies to tackle this situation? Using Appendix D really seems to limit the shear breakout condition. Looking back at older PCI design guides it appears they could readily resist 100 kip shear loads with embedded nelson studs. I know this topic has been discussed before just seeing if anyone has any updates or thoughts. Also just two points of clarification below. (Thinking out loud)


Point of clarification:

1.) It will always be conservative to simply use the bolt group moment of inertia in determing the bolt forces rather than performing an elastic analysis of the anchor bolts using a reinforced concrete beam analogy. Just looking for a confirmation.

2.) I have also seen calculations performed where engineers assume the baseplate pivots about an edge with a linear strain distribution through the anchors; this would produce unconservative results.

 

[JAE]

Those are very big shears to take at 12" eccentricity at the top of a wall.

I'd also be concerned with the top of the wall failing in flexure as the vertical wall reinforcement may not be developed past the point of cracking.

Supposedly, older bolt methods are not as accurate as ACI 318 App. D. However, App. D. is quite vilified by many engineers - some I know choose to ignore it because they say it is way too conservative.

For the size of loads you have, I'd wonder if a hanger plate assembly, draped over the top of the wall might take the shear. You'd still have the moment/tension forces to take in the bolts.

 

[slickdeals]

What is the thickness of the wall?
What will the bracket support? Steel beams?

Can you use a saddle type connection with stiffeners? This would mean that your shear is in bearing on top of the wall and the stiffeners will take out the moments.

 

[msquared48]

These loads are screaming for new columns. No way to do that?

Mike McCann
MMC Engineering
Motto: KISS
Motivation: Don't ask

 

[LSPSCAT]

Wall and structure can resist the loads. (Wall Thickness ~ 48") Basically a very large concrete vessel, the load is applied vertically in the (+) direction, picture a jacking type arrangement.

I agree about Appendix D regarding large anchor forces, if you call Williams or DSI they can satisfy these requirements; it just conflicts with Appendix D.

 

[Lion06]

I was at a local university recently where a
Grad student performed multiple tests of multiple anchor configurations and compared the results to App. D calcs. I was surprised to see that App. D was a little UNconservative for anchor groups with large numbers of anchors with overlapping failure cones.

It was a little conservative with fewer anchors and anchors without overlapping failure cones, but I wouldn't be so quick to dismiss it as overly conservative.

 

[LSPSCAT]

Point well taken Lion6 and thats why am on the hunt for a solution! Like I said, in the past and even now I sometimes just default back to ACI 349 which was the old code for anchorages in concrete for nuclear work. (1980's) Too this day Williams still swears by it and they can be viewed as authorative for large anchorages. Just wanted to run this up the flag pole and see what others are thinking. I do not have any real senior "mentor" type of engineers to work with so I like to hear peoples thoughts.

 

[JAE]

Lion06 - I'd be interested to know whether there was any reinforcement in the concrete at all. I have always suspected that Appendix D was based on an ideal piece of concrete, never mind the bars.

I know there are factors for reinforcement but it seems like they have a way too small effect on the numbers compared with what I would think they would provide in terms of helping the capacity.

I understand that fairly shallow anchors wouldn't be helped by a bar some distance away based on the failure cone, etc. but when you deepend your anchors per App D such that the cone would be crossed by, say 8 #5 bars, I just don't believe the failure load that is predicted.

 

[LSPSCAT]

Based on what this is for I am assuming setting the first anchor row 8"-10" away from the edge I will be inside some large horizontal bars - #8 - # 11.

Based on assumptions I can almost talk myself out of a detailed look at shear breakout at the top of the wall.

I am just getting the general question "..well if they did it this way in the past..". Need to jusify an economical and sound solution as we have a multitude of these to install.

I am probably going to require that a shear key is cut and grouted in for the largest anchorages and resist the moment with tension in the anchors.

Less than 80 kip I can manage with a reasonable number of anchors and bracket size.

 

[racookpe1978]

Rather than a shear "key" (implies a square or rectangular bar or plate or insert into the face of the concrete ( = difficult to cut the notch and get a clean joint to mates fimrly with the key), why not drill through the wall (or into the wall 20+ inches) with a round rod 1-1/2 to 2 inches in diameter?

Use the rod as your shear key, and if deep enough - or through the whole wall though 48 inches is a loooooooooooong way - then use the rod to resist moment as well as shear.

 

[Lion06]

JAE-
The tests were done with completely unreinforced, uncracked concrete.

 

[bones206]

Using adhesive anchors for loads that large would make me nervous.

You would also require continuous special inspection of the anchor installation and that might make the design less economic.

I like slickdeals' idea.