Large Concrete Anchorage Forces 4

[GalileoG]

I have a beam that frames into a concrete wall, parallel to the length of the wall. This beam will need to transfer about 1200kN (270 kips) axial tension force to the concrete wall. I initially envisioned an embedded plate with weldable rebar (2 vertical rows, 14 bars per row) that are 1200mm long (4') welded to the embedded plate to develop the full strength of the bar. Even with that, I am not getting the concrete break-out capacity that I require. Now I am thinking of welding plate washers to the end of the bars (this should give me the capacity that I require, but I am in the process of confirming it). I am using Appendix D in the code for anchorage design. Do embedded bars with welded plate washers seem like an odd solution to you? Any suggestions or alternatives? Your advice is appreciated.

 

[JAE]

I guess once I use welded rebar to a plate - and lap spliced to rebar in the wall - I'm not using Appendix D at all.

 

[JAE]

One view of this:

 

[JStephen]

If I remember correctly- if you embed bars (or anchor bolts) into the concrete- neglect tension in the concrete behind bars/bolts- and check shear on the surface around the bolts, you pretty much get the concrete breakout strength design from Appendix D.

Regardless, for the original post, are you looking at breakout strength or pullout strength? Pullout will be increased with the welded washers, breakout would be more dependent on the depth and distribution of the anchors. If you satisfy pullout, but not breakout- add reinforcement through the breakout cone, then design for shear beyond that area. If you're adding welded washer plates, you can design them as bolts, and don't need deformed bar.

 

[ztengguy]

If you develop a bar into concrete, why is there a requirement to mess with Ap. D?

 

[JAE]

That's my point too, ztengguy.

 

[msquared48]

Appendix D only applies to younger concrete that is not fully developed yet.

I have used the welded plate option many times, even with the full development length achieved. It just feels better to me, is easy to do, and cheap extra capacity in anchorage design.

Mike McCann
MMC Engineering

 

[JAE]

Appendix D only applies to younger concrete that is not fully developed yet.

Mike, can you clarify where you got this idea? I thought App D was for anchors in all concrete. Not just younger concrete.

 

[msquared48]

Gotcha...

Mike McCann
MMC Engineering

 

[JAE]

OK I'm gotten.

 

[OHIOMatt]

My preferred method is to use Lenton Weldable Couplers at the plate. They come grooved to allow for a high quality weld. The bars are then threaded into the coupler.

https://www.erico.com/products/LentonWeld.asp

While we (or I) am on the subject, you may want to consider Lenton Terminator rebar end anchors. They may work well for your application.

http://www.erico.us/products/terminator.asp

 

[JStephen]

Why mess with App. D: If the rebar going into the slab goes more or less all the way through the slab, then checking for concrete breakout is just a punching-shear check at that location. But, if the rebar only goes partway into the slab, the punching-shear check assumes shear through the entire slab, whereas you could actually have the concrete break-out failure as in App. D instead of a shear failure through the whole slab.

 

[TDIengineer]

At Galileo-
Yes, don't mess with Appendix D with this high of forces. Fully develop the rebar by welding to the plate. Then lap to the rebar for the full length of wall. In high seismic zone when detailing for overstrength, the chance of getting a "headed anchor bolt" to work for anything but the smallest loads is zilch to none.

 

[KootK]

I often (almost always) see embed plates like this detailed with weldable rebar extending from the embedded plate into the concrete approximately the anchorage length of the specified rebar. Usually it's insufficient and indicative of confusion regarding what anchorage length actually is.

Anchorage length is simply the embedment length required to prevent rebar from pulling out of concrete via bond stress failure. Providing anchorage length rebar embedment doesn't mean that you can actually develop the tensile strength of the embedded rebar. To develop the tensile strength of the rebar, you also need to do one of two things:

1) Deal with concrete breakout using some form of diagonal tension check (Appendix D, simplified shear check, etc.)
2) Pass the tensile force to rebar elsewhere in the concrete (laps, strut & tie, etc.)

I apologize to anyone who may find this condescending. I've encountered enough confusion about this that I really do feel that it warrants detailed explanation.

As for the case at hand, I recommend using a simplified strut and tie model to pass the tension force from the concentrated rebar coming off of the embedded plate to the - usually not so concentrated - horizontal rebar in the wall. This takes the form of a suite of compression struts originating at the end of the embedded rebar and fanning back to an equivalent area of horizontal wall steel. All the while, you need to be cognisant of the need to develop the ties and keep your strut angles reasonable. You'll also need a vertical tie in your wall adjacent to the embed plate. Usually, nominal detailing rebar does the trick for that.

When I do this, I usually do a very abbreviated analysis. I don't check node stresses or any of the fancy suff. I just set out the rebar such that I feel confident that I've effectively lapped my embedded rebar with an equivalent area of wall rebar and generally paid homage to STM concepts. Usually, the end result ends up being exactly what TDIengineer suggested: the embedded bars are lapped to matching horizontals and those are carried deeeeep into the wall.

With regard to anchorage plates and lenton couplers, all those do is shorten the tie development length required at the end of the embedded rebar. I usually omit them as they cause some congestion and don't decrease the overall embedment length all that much. For an extra fun version of this problem, tie an axially loaded beam into a wall up near the top. I find this requires even longer embedded rebar lengths and a serious vertical tie at the end of those bars, usually involving U-bars lapped to the vertical ties.

The greatest trick that bond stress ever pulled was convincing the world it didn't exist.