Deformed Bar Anchors (DBA) used in seismic and cracked concrete

[Prestressed Guy]

I have an EOR questioning the use of Deformed Bar Anchors (DBA) for uses in seismic connection and cracked concrete. This is based on the Nelson Stud Welding Inc. ESR-2907.
Have any of you run into this issue?
Under 3.1 General: it states "...The use of these anchors is limited to installation in untracked normal weight concrete."
In Conditions of use: "5.10 use of anchors to resist seismic loads is beyond the scope of this report"

Quite frankly, I have never before thought to look for an EAR for DBA's and have used them for years for connections directly out of the PCI Design Handbook for seismic connections.

Anyone ever run into this issue before?

 

[Agent666]

Generally standards allow it, for example ACI318, but with some reductions in capacity compared with uncracked concrete.

Sounds like they are saying its outside the scope of their report, meaning it wasn't something they actually looked at specifically. It does not say you cannot do it by comparison and there is a lot of literature by the likes of PCI and ACI which offer guidance on both cracked and seismic design of headed studs.

 

[Agent666]

Actually strike out what I said, you're not talking about headed studs!

 

[Prestressed Guy]

This question is dealing primarily with DBA's made from Deformed wire A1064. The EOR had originally called for 3/4"ø D45 wire. He then found that wire sizes over D31 are prohibited by 318-14 20.2.1.7.2 so he looked up the ESR-2907 which limits them to non-fire rated, non-seismic, uncracked concrete applications and came to the conclusion that DBA's are bad and should not be used.
None of these limitations appear in 318-14 for D4 through D31 so...

 

[Agent666]

Reading that clause (20.2.1.7.2) it seems like it's more referring to being applicable to use of the referred welded wire reinforcement. The preceding clause notes for the D45 mesh that it has issues with achieving tension development, but in the context of the clause as a mesh as I read it which is understandable given the bar size.

Seems to me short cast in welded bars of the same material is a whole other field. If the material is a limitation you can always weld normal bars instead as it would be unlikely you're governed in your connection by the steel failure strength. Typically the brittle concrete failure mechanisms would govern.

Part of the reason cold worked wire is not good in seismic scenarios is that it generally lacks the ductility required, and has limited elongation performance prior to fracture.

 

[Brad805]

It is interesting how it was interpreted that "outside the scope of this report," turned into "do not use". I have not run into this exact problem. I find the design of DBA's frustrating, but our clients would put in 2 - 3 times the area of DBA's vs welding a bar. I would be interested to hear Koot's thoughts as I know he works for some precasters.

 

[TLHS]

What's the reasoning behind the use of the deformed bar anchor? Is it the higher strength?

You can use them in cracked concrete, but you still either need to demonstrate the load path via lapping principles, strut and tie or by getting it into the mass of concrete using the Appendix D provisions.

 

[r13]

I think you shall simply follow ACI or PCI provisions for embedment design. Nelson is very reputable but quite conservative, back to the days (before ACI app. D) that required a safety factor in the range of 2-4.

 

[KootK]

I would be interested to hear Koot's thoughts as I know he works for some precasters.

1) My precasters heavily favor the DBA as well. Obviously, that's about the automated welding and easier QC.

2) I also find DBA design frustrating. As far as i can tell, there are two paths:

a) It develops like rebar so use it like rebar with RC concrete design principles, lapping it to other reinforcing etc, and not relying on concrete in tension other than concrete splitting resistance. I feel this is legitimate excepting potential ductility issues with cold worked "wire" in applications where that's important. Obviously, in this category, concrete is allowed to crack because there would be no point to the reinforcement were the concrete to not crack.

b) Use it truly as a deformed bar ANCHOR in the same sense that headed studs and wedge bolts are anchors. Just a device for mobilizing concrete tensile resistance locally. In most of the D2L types of applications, I feel that we're in this category. Also, in this category, I feel that limitations on concrete cracking and connection ductility would apply until the requisite testing has been undertaken to establish performance capabilities in the more challenging environments.

I regularly hear people say that they've gone to DBA to solve an anchorage issue that didn't work out using other technologies. "I just developed it instead". This concerns me because there seems to be the impression that developing a DBA, without lapping it to other reinforcement, somehow gets you out of having to worry about the concrete breakout issues that Agent666 mentioned. It'll certainly help but it's by no means a done deal.

When I'm dead and gone, I want my legacy on planet earth to be DEVELOPMENT <> ANCHORAGE. And maybe the kids... Yeah, them too.

Circling back to OP's original question, I do not believe that a DBA can currently be used as an anchor in cracked concrete owing to the absence of appropriate testing.

 

[Prestressed Guy]

Thanks Koot. Your above stated views reflect mine exactly. In this case the project is an Intermediate Precast shear wall structure and the EOR (who has several details with 3/4" DBA's in his details) has suddenly "inferred" that DBA's cannot be used for any seismic load based on table 20.2.2.4a. In this table, it states that A1064 deformed wire is "not permitted" for "Special seismic system".

If this project was a Special reinforced concrete shear wall with R = 5 that I would agree that A1064 is not permitted. In this case, R = 4 so that loads are increased by 20% and the connections are not so "Special" The connections need to be designed per 18.5 to have a steel or rebar yielding element and all other elements of the connection have a required strength of 1.5Fy the yielding element.
The connection design in question is for horizontal foundation to wall panel and
vertical panel to panel connections The DBA's are used to anchor a steel embed plate into the concrete which then has a A36 loose plate welded to the embed in the adjoining member. The DBA's are at a length greater than l_d and considered the "yielding" element and all welds and structural steel elements are designed to 1.5F_y the DBA.