Problems with J or L bolts?

[mecheng19777]

Hello, I came across this thread when researching L bolts to secure a 2x4 wall to a concrete block foundation for a single car garage project I am working on.

I don`t understand the issue with L bolts, I thought they would be superior to a regular bolt with washer. CSA shows you can use them.
Thank you

https://www.eng-tips.com/viewthread.cfm?qid=416314

 

[jayrod12]

The two largest issues are the bearing stresses of the concrete at the inside of the bend and potential for straightening of the hook/leg.

 

[KootK]

...I thought they would be superior to a regular bolt with washer.

You thought wrong. As we all did twenty years ago.

CSA shows you can use them.

You can use them. There are capacity equations specifically for them.

CSA doesn't know anything about anchorage that it didn't pilfer from ACI, so don't sweat that.

 

[MechEng1977]

Thanks guys, so the American spec banned them or is there a reduction factor if I use them?

I am assuming a bolt with a washer welded or
Two nuts sandwiching a washer is preferable?

What about just threaded rod alone? I don’t see this as a recommended solution

 

[phamENG]

They're still allowed, but they have a lower capacity. I don't think it's a reduction factor, but a different equation as KootK mentioned.

Welding to an anchor bolt is usually a terrible idea, though some have special compositional properties that make it feasible (dual certification as A36, or the Supplement 2 for F1554 Gr. 55). Many of them undergo various forms of heat treatment, and welding can undo that treatment and weaken the bolt.

A washer isn't really needed. A standard washer is so thin that it's very flexible compared with the nut below it and concrete around it. By the time the assembly has moved enough to mobilize the resistance of the thin washer, your anchor is already on it's way out of the block of concrete.

Threaded rod alone wouldn't have any resistance. The whole idea is based on the top of the nut (or out-turned leg of the J or L bolt) bearing on the concrete above when the assembly above is in tension. The only way it works without anything to bear on the concrete above is if you use an epoxy anchor adhesive.

 

[strucbells]

You can make the washer a 1/4" plate washer sandwiched between 2 nuts to deal with the flimsiness issue of standard washers, my old company used to spec that all the time.

Now I usually just specify headed anchor rods.

 

[Tomfh]

Plain threaded rod is fine *provided* it’s embedded deep enough. You need extra length to match a headed bolt because the failure cone won’t generally reach the tip of a thread rod, as it does with headed bolt.

Regular headed bolts do very well. You get even bearing around the bolt, as opposed to L bolt which bears eccentrically.

 

[phamENG]

Tomfh - can you explain how plain threaded rod is fine? What is engaging the concrete and forming the cone? The threads? Or are you referring to bending it into an L?

strucbells- I still don't like the plate between nuts detail. See the sketch below. The top nut is going to be a lot stiffer than even a 1/4" plate washer. Take a 3/4" anchor rod - the nut is going to be about 1/2" thick, and the shortest distance from the edge of the rod to the outer face of the nut is only going to be about 1/4" for heavy hex. Compare that the flexural stiffness of a 1/4" plate with a diameter that puts the outer face of nut to outer edge of washer at an inch or more. So while the washer will resist some load, the vast majority of the resistance will occur at the nut. I think the nut failure cone is the failure that would happen, though perhaps at a slightly higher load than would otherwise be predicted. The big issue comes when somebody uses the washer diameter as the bearing area, since I think that overestimates the capacity of the connection (both by overestimating the effective area and underestimating the embedment depth, albeit by only about 1/2").

 

[r13]

Use threaded plate washer.

 

[dik]

Use them all the time... but spec F1554 S1 for weldability.

From Portland Bolt:
"When an F1554 Grade 55 anchor bolt has been specified to meet the supplemental requirement “S1”, the fastener must then be designated by a white paint mark on the side of the bar near the end to be encased in concrete. This is to designate that the anchor bolt can be welded. Since F1554 Grade 55 anchor bolts need to be specified as being weldable, it is important to confirm that they do in fact meet the “S1” supplemental requirement prior to any type of welding. Because of this, Portland bolt has all of our raw material used to manufacture F1554 Grade 55 anchor bolts melted and rolled to meet this requirement whether the “S1” supplemental requirement is needed or not. As a result, you will never need to question the safety of performing a weld on any of our F1554 Grade 55 anchor bolts."

Dik

 

[Tomfh]

Tomfh - can you explain how plain threaded rod is fine? What is engaging the concrete and forming the cone? The threads? Or are you referring to bending it into an L?

Straight bar. The threads engage directly with the concrete. I've seen it a few times with proprietary lighting cages, with ~800mm threaded rods lapping with the pier cage.

 

[phamENG]

dik- thanks for the correction.

Tomfh- interesting. Might be an Australian thing? I don't think ACI lets us do that in the states.

 

[mecheng19777]

Ok so I gave the L bolt some more thought, is the issue the following: when it is loaded in tension, it will naturally want to torque and in doing so it could break the concrete.

So what you guys are recommending is a simple anchor bolt, and if I can get a 1/4 inch plate sandwhiched between two nuts even better,
correct?
I just thought the head of the bolt area is very small to provide adequate pull out resistance.

Regarding the threaded rod, I always thought the threads provide pull through resistance when the concrete fills them.

 

[KootK]

Tomfh- interesting. Might be an Australian thing? I don't think ACI lets us do that in the states.

My read on Tomfh's method as it's been described is that, in our North American parlance, it would not actually be considered "anchorage". At least not in the Appendix D sense. Rather, it's essentially using reinforced concrete principles to effectively lap the anchor rods to the pier verticals. And I believe that we're also allowed to do this in North America so long as the eye's are dotted and the tee's crossed. Similar methodology exists in AISC's DG01.

 

[phamENG]

mecheng19777 - That's part of it. It has more to do with the stiffness problem I've been harping on, though in a different way. When loaded in tension, the resistance along the length of the leg of the L bolt is not uniform. It's going to vary along the length proportional to the stiffness of the anchor at that point. So the very tip will be doing essentially nothing, while the portion of the bolt adjacent to the bend will be doing the heaviest lifting. As a result, you get a localized crushing failure in the concrete. When that happens, the out-turned leg is now in bending to resist the tension in the vertical leg. It wasn't designed for bending, and will likely give up. As it starts to deform and straighten, it'll progressively pull out of the hole and come out as a more or less straight piece of steel. This can happen with limited observable damage at the surface of the concrete.

As long as you treat the plate as a belt and suspenders and only design for the bearing area of the head, you should be fine. The plate washer really isn't needed, though.

KootK - could be. Can't say I'm inclined to trust it, but that may just be lack of experience with that situation.

 

[XR250]

I don’t know how they build things in your neck of the woods, but around here, the top course would be pulled off way before the “j” straightened out.

 

[phamENG]

XR - thanks for bringing us back to the original point. I got lost in theory there. In an unreinforced block wall, you're absolutely right. If I need more than the tension capacity perpendicular to the bed joint, I reinforce the foundation wall with a hook at the top of the bar (which most residential contractors tend to ignore since it's a bit of a PITA).

 

[KootK]

KootK - could be. Can't say I'm inclined to trust it, but that may just be lack of experience with that situation.

How could you not? It's effectively RC concrete design. One would expect that to be vastly more reliable than any of the APP D stuff. Right back to not using concrete in tension other than splitting stresses etc.

 

[mecheng19777]

I just look at a regular hex bolt and the head has so little surface area, but I guess you are saying that is all that is needed and the head of the bolt is stiff and therefore won`t bend?

 

[Tomfh]

Kootk, Pham,

Yes using a headless bolt and lapping with the reinforcement in the the footing is the same concept as lapping reinforcement.

By all means use headed anchors if you prefer. We do on our standard anchor details.

I just look at a regular hex bolt and the head has so little surface area,

It doesn’t pass normal bearing checks, but apparantly it’s enough to yield the bolt. Confined concrete is pretty strong.