Embed Plate with Welded Rebar or DBA's 1

[slickdeals]

My understanding is that the use of a welded rebar or a Nelson DBA in an embed plate gets us out of the ACI 318 Appendix D mess. This was based on the failure modes of anchors not applicable to deformed bars. A reinforcement bar fully developed in concrete can carry it's tension load capacity as long as adequate cover is provided. Assume that cover is say 3", the code will let me develop full tension capacity of the reinforcement.

I was reviewing the ICC-ESR for Nelson DBA's and they seem to have much tighter restrictions for edge distance and spacing.

http://www.icc-es.org/Reports/pdf_files/ESR-2907.pdf

Has anyone used the values in these tables or assumed that the full tension capacity of an embedded deformed bar can be achieved?

 

[Lion06]

I often use embed plates with welded rebar when I can't get headed studs to work with App. D.

The DBA have different requirements per 12.6. The development is a function of the deformations AND the mechanical anchorage, so manufacturer testing is required.

 

[slickdeals]

Thanks. So the deformations of the Nelson DBA are different from a standard deformed reinforcement bar and the Nelson values are derived from testing? Seems like Nelson's stuff isn't too useful when you are trying to get out of the Appendix D pickle and might be worth just going with ASTM A706 welded to embed plates.

 

[KootK]

I believe that the use of rebar or DBA's gets you out of one, maybe two, of the Appendix D failure modes. I've assumed that using rebar eliminates side face blowout concerns associated with stud heads and nuts. On occasion, I've also assumed that shear friction at the weld plate / concrete interface could be used to transmit shear directly to the concrete. All of the other failure modes still apply in my opinion. In particular, unless you've effectively lapped your DBA's to other tension reinforcement, there's still a tensile concrete breakout mode to contend with.

Obviously, test result trump everything else. I'd be very surprised if test results for welded rebar didn't closely match those for Nelson's DBA's.

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

 

[dcarr82775]

I asked a similar question a few weeks back. In my way of thinking, you must check breakout of the concrete, simply looking at development length may not be enough. I decided that it is similar to an epoxy bar per App D. If the concrete can breakout for an epoxy bar, it seems unwise (IMO) to think a similar breakout cannot occur simple becausy it was cast in instead of glued. ACI318 simply doesn't address the situation.

 

[Lion06]

It's not that the deformations are different, it's that the length of bar you have is often less than a full development length. That means you're counting on the deformations AND the head for anchorage.

kootk - My comments are specifically related to rebar and not the DBA, but I believe using welded rebar on the embed plate completely eliminates App. D checks.

 

[KootK]

D2L's don't have heads do they?

I've got a challenge for you Lion, if you're up to it. Draw a free body diagram of one of these embed plates under a tensile load. Include the weld plate, the DBA's, and the concrete immediately surrounding the DBA's. Do not, however, include anything akin to concrete breakout or supplemental lapped rebar.

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

 

[DCBII]

I don't see how DBA's are all that different from rebar. Appendix D was intended for anchors that bear at the end (like headed studs, bolts, or j-bolts) and epoxy anchors. DBA's and rebar both transmit stresses gradually through deformations along their length instead. I think the development length concept is more applicable than Appendix D. Results of manufacturer testing should obviously be taken into consideration as well. I could see an argument for Appendix D if DBA's were transmitting shear, but they're made for tension loads.

 

[Lion06]

kootk - As stated in my post, my comment was specific to welded rebar (A706), not DBA. In my mind this is no different than any other rebar. The discussion was also related to getting the load from the plate into the parent concrete. Obviously, the parent concrete element must be properly reinforced to take the loads from the embed plate, but that wasn't the topic of the discussion.

I don't see the difference between a rebar welded to an embed plate and any other rebar. In a beam for example, the bottom rebar at midspan is no different than an embed plate with welded rebar under a tensile load. All the rebar sees is tension. Provide proper development length and cover to develop the force into the parent concrete and properly reinforce the parent concrete for the applied loads.

 

[DCBII]

One other thing. Has anyone else noticed how much we've used the word obviously in this conversation?

 

[dcarr82775]

DCBIII

ACI318-12 Appendix D covers epoxy anchors, so it is no longer 'limited' to bearing type anchors. How is a DBA on an embedded plate that different from an epoxy dowel under the same conditions? The bond stress of the epoxy to the steel and concrete is rarely the controlling condition.

 

[slickdeals]

Guys, then why does Nelson's ESR have such stringent requirements that we wouldn't use with rebar. I haven't thought twice about using #9 bars at 4" o.c., but look at Nelson's ESR.

 

[slickdeals]

Looks like we use DBA's and then lap rebar to it. So why use DBA's, why don't we use double studs and lap rebar to it?

 

[DCBII]

slickdeals,

Appendix D is for specialty inserts. Tests have shown that epoxy anchors actually behave quite a lot like a headed anchor. The tension breakout calculation is almost identical to that for a headed anchor. Stress trajectories in epoxy anchors concentrate near the top of the concrete, rather than dissipating down the length of the anchor. When an epoxy anchor fails, it often brings a failure cone with it at the top of the anchor while the bottom comes out clean.

When rebar has (brittle) failure it typically splits the concrete perpendicular to the bar (hence the Ktr calculation). I can't imagine a DBA behaving any differently, the stress transfer mechanism is nearly identical.

 

[DCBII]

Here's some good pics showing what I was describing.

 

[DCBII]

Let's try that again... Here's some pics

 

[KootK]

@ Lion06:

I only brought up the "heads" because you did in your initial post. I guess I don't understand the first paragraph of your initial post. Neither rebar nor the Nelson DBA's appear to have heads. Could you elaborate?

Regarding your comments:

...I believe using welded rebar on the embed plate completely eliminates App. D checks

AND

Obviously, the parent concrete element must be properly reinforced to take the loads from the embed plate...

The first statement from your first post is quite incomplete without the second statement from your second post. Unless I'm badly mistaken, the crux of this thread is whether or not one can count on As x Fy in the absence of additonal, lapped rebar. This may be obvious to you, and is certainly consistent with my views, but it is a point of frequent contention on this forum.

Several of the people participating in this thread feel that it is enough to develop the DBA's / rebar into the parent concrete, regardless of how that parent concrete is reinforced. If the rebar/DBA's welded to the embed are directly lapped to other, appropriately proportioned rebar in the parent material, I don't think that anyone disputes the validity of the load path.

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

 

[KootK]

@ DCBII: For me, the scariest thing about those photos is that the failure cones didn't extend to the ends of the dowels. I live in fear of exactly that outcome.

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

 

[Lion06]

kootk- That's my fault for not looking at the picture more closely. I looked up Nelson DBA and saw a small plate on the end and assumed it was a head embedded into the concrete.

In either case, I think the statement "...I believe using welded rebar on the embed plate completely eliminates App. D checks" stands on its own. App. D is about the anchorage to concrete (i.e. getting the loads into the parent concrete) and has nothing to do with the parent concrete resisting the applied loads (that is handled with other chapters in ACI 318).

 

[KootK]

Yeah, I thought the exact same thing when I opened up the link. I've always thought the advantage of DBA's was that they can be shop welded into place by some efficient process akin to how regular nelson studs are installed. I never expected any kind of performance benefit over weldable rebar.

I guess that the issue that I have with your statement, Lion06, is that many who read it will take it to mean that Appendix D checks can be considered to be satisfied or unnecessary if welded rebar is used. That is only true for the one, trivial case where the welded rebar is directly lapped with properly detailed and proportioned rebar in the parent material. For all other cases, the welded rebar is simply another local anchorage device, just like headed studs or anchor bolts. Unfortunately, Appendix D doesn't have much to say about rebar/DBAs used as anchorage despite their popularity. There's the bit about post installed rebar which seems to have mostly just muddied the water even further.

I would encourage anyone interested in this thread to check out these ones as well:
Link
Link



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