Shear Wall Hardware Manufacturer Claims ACI 318 Anchorage Requirements Don't Apply 15

[D.E.N.]

TL;DR:
A shear wall hardware product manufacturer is claiming a steel threaded rod with epoxy anchored to concrete does not need to comply with ACI 318 anchorage provisions. I disagree. If ACI equations are used to check the anchor’s capacity, the product would have a reduced capacity and wouldn’t work as they have designed.

Full explanation:
I am the EOR on a large 5-story light-framed wood apartment building. I designed and detailed all of the shear walls using “typical” hardware, such as coil straps floor-to-floor and holdowns with threaded rods in epoxy at the base. Recently, I received a hardware substitution request from the GC, proposing a different product. Initially, I assumed this was a request to substitute hardware directly from a different manufacturer. However, the submittal proposes a completely different shear wall system.

The proposed system is a proprietary continuous hold-down assembly that connects to the building at the roof level only and anchors to the concrete at the bottom. Its tension capacity is based on testing and comes with a “code approval” report from a certified testing agency. This report, a Technical Evaluation Report (TER), is similar to an Evaluation Service Report (ESR) from other agencies. Due to the proprietary nature of the assembly, no published calculations or test result data are available; only a single allowable tension capacity is provided. The TER specifies installation conditions that must be followed to achieve the published tension capacity.

I have several concerns about this proposed system's ability to resist the required shear wall overturning forces and the proposed load path. Although I have been working with the manufacturer to address these concerns, one aspect of the design remains problematic. Both the manufacturer and the testing agency assert that the system's anchorage into the concrete does not need to be checked using ACI 318 anchorage provisions. The proposed shear wall system consists of a steel cable assembly with steel threaded rods at each end, with the bottom rod embedded in an epoxy-filled hole in the concrete. In my opinion, this anchorage design is not proprietary since it involves a steel threaded rod with epoxy in concrete, which is clearly defined in ACI 318. I would understand the argument if we were talking about a Simpson “LSTHD” Strap-Tie Holdown or MiTek “LSTAD” Foundation Strap, which is a truly proprietary anchorage design since it is just bent steel embedded in the concrete. But this is a steel threaded rod embedded in epoxy that just happens to be attached to the end of a proprietary cable assembly.

The manufacturer has a proprietary epoxy with an ESR report, stating to follow ACI 318-14 Chapter 17. However, the testing agency and manufacturer claim that the ESR for the epoxy is not valid when used with the cable hold-down system. They argue that the ESR only applies when the epoxy anchors steel threaded rods to concrete. It just so happens that the bottom of the cable hold-down system has a steel threaded rod attached to it…

Problem:
Using ACI 318 equations to check the anchorage capacity, I find that the assembly's allowable tension capacity is about 60% less than the manufacturer’s published values, primarily due to concrete breakout. This calculation includes product-specific variables and factors from ACI 355.4 tests, accounting for any proprietary epoxy behavior, which are published in the ESR.

I believe the proposed system’s anchorage must be checked using ACI 318 provisions, as the design closely matches the scenarios covered in the standard. The difference between the published assembly capacity and calculated anchorage capacity is a significant concern that I want to resolve before allowing the product to be used in my building. The testing agency and the manufacturer are telling me not to worry about it since the submittal and the TER are both sealed by PEs in the project state. They claim that since it is a delegated design, I don’t need to worry about how it interacts with my building. I would like to reject the entire system and stick with the original design, but the contractor really wants to use the cable system.

Question:
Have you ever dealt with something like this before? I don’t see how I can allow this anchorage “design” to bypass the ACI 318 requirements just because they are calling a standard anchorage method “proprietary.”

 

[XR250]

I have never heard of the certification agency they used. Dr.J?
I wonder how hard it is to become certified?

 

[D.E.N.]

XR250 - This is from the cover page on their TER. I am not familiar with the process though.

 

[WinelandV]

A quick google search is showing ASTM E3121 as possible testing standard (Standard Test Methods for Field Testing of Anchors in Concrete or Masonry).

One area of concern, and a possible out, is that I'm unaware of a test set-up for the sustained tension. Pretty sure no one is going to be willing to keep a sustained load on there for 1 or 2 years as part of a test.

I don't like that the manufacturer is so quick to hand wave away the sustained tension on the epoxy. Pretty sure none of the big epoxy manufacturers (hilti, simpson, etc) has any data on what effect, if any, the confinement from the pre-compressed studs might be on the tensile stresses in the epoxy.

RE: Cable - I've spent time designing temporary guy wires, and every time someone suggests wire ropes/cables, I have nothing but questions: connection hardware, proper D/d ratios, correct rope/wire callout, required pre-tensioning load, pre-tensioning procedure, removal of construction stretch, and on and on and on....

Edit: The epoxy they are calling out specifically says it has to be designed per ACI 318-19. Hang your hat on that and reject it.

 

[bones206]

From the cover sheet:

Sounds like a group that expects to be challenged and had a lawyer craft this verbiage to shield them.

 

[bones206]

Also, installers of adhesive anchors under sustained tension must be certified to do so per ACI requirements. Part of the response to the Boston tunnel tragedy, I believe.

It would be good to check with the GC to see if they have any workers who have completed this training and certification process.

The long term creep of the adhesive would exacerbate the loss of tension in the cable as well. Personally I would go for some kind of a cast-in anchor for this system.

 

[D.E.N.]

WinelandV - Thank you for the link to the field testing procedure. I am not sure how to address the sustained tension capacity. ACI 318-14 Section 17.2.5 Commentary says in part "R17.2.5 For adhesive anchors subjected to sustained tension loading, an additional calculation for the sustained portion of the factored load for a reduced bond resistance is required to account for possible bond strength reductions under sustained load." I can't find a "reduced bond resistance" value given in the epoxy ESR. But even if it was in the ESR, they are claiming that the ESR doesn't even apply (and by extension ACI 318) to this application since it is "not a steel threaded rod" and is instead "a proprietary connector".


bones206 - That's kind of what I was thinking as well. That statement from the cover makes me uneasy for some reason.

 

[phamENG]

How has Simpson not sued these people into oblivion yet? Everything but the continuous hold down system appears to be lifted straight from the Simpson Catalog and online CAD library. Some of the product identifiers are even the same!

 

[bones206]

Simpson probably no longer holds patents on those particular products.

 

[Brad805]

The proprietary argument may hold if there was some aspect unique about the question being asked, but here we are talking about a simple concrete embedment problem. Presumably they embedded their anchor into concrete and pulled until it broke. I believe we have all seen that before, no? The overall system may be proprietary, but it seems made up of standard parts we have all seen before.

Has the contractor suggested the potential cost savings of the system for the project? My decision to reject without further review would depend on that.

 

[XR250]

Honestly, the testing agency seems as sketchy as a 1-year online MBA.

 

[bones206]

I am not sure how to address the sustained tension capacity.

17.3.1.2 covers the sustained tension capacity. Essentially a 0.55 reduction factor applied to the bond strength.

 

[D.E.N.]

Brad805 - I agree regarding the standard nature of the actual embedment itself. It is just a steel threaded rod in epoxy. The fact it is attached to a steel cable should have nothing to do with the anchorage design in my opinion.

The contractor said there would be a cost savings of over $75,000. But this doesn't consider all of the extra end chord studs required for the proposed system (re: "upward compression"). We are also talking about a system that doesn't really "work" as currently specified in my opinion, so the cost comparison would need to be adjusted I think. If they had to limit their assembly capacity to be controlled by the ACI concrete breakout capacity they would have to add a lot more cables and close the gap on the cost difference even more.

 

[D.E.N.]

bones206 - Wow, can't believe I missed that. Thank you for the reference.

Something else I am still hung up on too is how much tension the anchorage will actually experience and how the pre-load/post-tensioning amount is determined. The TER says that the tensioning amount is determined by the RDP (registered design professional) which I believe refers to the delegated engineer designing the cable system in this case (which is the manufacturer). The TER says the pre-tensioned level shall not exceed the allowable tensile capacity of the cable assembly. The only info I received from the manufacturer about how much pre-tensioning to expect was a screenshot (shown below). The majority of the cables they have specified for this job have an allowable capacity of about 6.5 kips. The submittal/delegated design has several cables loaded to 100% of this capacity. If the cable needs to be pre-tensioned approximately 140% of the design load (6.5k in this example), the cable would need to be tensioned to 9.1 kips. This number exceeds the allowable tension capacity of the cable assembly and violates the TER, right? So does the anchorage calculation need to use 9.1 kips instead of 6.5 kips (these would need to be converted to LRFD first of course)?

 

[XR250]

Thats's a lot of sustained load thru 5 stories of wood. Can't imagine creep won't play a part.

 

[phamENG]

I imagine they'd save even more if they removed the shear walls entirely. I'd give that about the same amount of consideration.

 

[driftLimiter]

I think you've got a lot of good discussion, and are on the right track rejecting this thing outright.

However I am curious how your original design was meant to address shrinkage? Were you going to use any shrinkage take up devices?
Perhaps add the shrinkage displacement to hold downs at ultimate load and design for it?

Cause in a way I see here a 5 story building that you say you designed typical straps and hold downs for, then I see the contractor proposing a method that accounts for shrinkage.
Seems there is a bit of a hiccup there.

IBC requires that we need to explicitly address shrinkage by code requirements for this 5 story wood shear walls. (e.g. IBC 18 2304.3.3)

 

[dold]

Seems like a shrinkage take-up device wouldn't even work on a system like this. The system relies on continuous pretension, no? So when the wood shrinks it loses all pretension. I guess then the takeup device could take out any slack, but I can't imagine it would be able to maintain the 6.5k over the life of the structure.

Also, if this thing is pre-tensioned to 6.5k, the compression studs would have to accommodate that pre(compression) plus the compression due to overturning. That'd be a ton of studs (which will be butchered by the electricians installing light switches and what have you.

 

[phamENG]

Typical straps and hold downs can accommodate the shrinkage that occurs in a single floor. It's the multi-floor setups that require a more direct approach to compensating for it.

 

[D.E.N.]

driftLimiter - This is a really great point. Our original design does not explicitly address shrinkage considerations. We do have notes on the drawings that say something like "install the top part of the straps first, then nail the bottom once the roof framing is completed". I personally don't think this is sufficient to address the shrinkage concerns though. I have been told that shrinkage is not really a concern for our firm's buildings as long as the straps get nailed after the roof framing is completed. I disagree. I would like to see us transition to using a continuous rod system with takeup devices for our projects moving forward. I've been told "no" by management previously when I have suggested their use before due to cost concerns.

But you are right, the proposed system would likely do a much better job addressing building shrinkage than my original design would. The strap and hold-down system (although probably the most common with the most history of use) is far from a perfect system, but at least I can prove it works (unlike the cable system). If I had to choose between accounting for shrinkage and a proper/verifiable load path, I chose the load path. Now, that being said, I am planning on using a continuous rod system with take-up devices for our next project. Because I am still not happy about the shortcomings of the traditional strap and hold-down system.

 

[D.E.N.]

dold - Yeah, they don't use a take-up device at all with the cable system. They are relying exclusively on "pre-loading" the cable. The way they explained it is that the cable is over-tensioned (~140%) and as the building shrinks the "pre-load" reduces until it has about 5% sustained tension (to address longer-term shrinkage I guess?). Yes, there are way more studs required for this system compared to others.