Attaching Interior Shear Wall Bottom Plate to Post-Tensioned Slab on Grade 2

[TopKnot]

I have some high load interior shear walls I need to attach to a proposed PT slab on grade foundation. (550 PLF) Finishing concrete around placed anchor bolts on the interior of the slab will be a nightmare. I do not want anything nailed or shot into PT deeper than 1/2 in, which is basically nothing.

Can I use construction adhesive? I have looked and looked for data or testing on this scenario, and I can't find it. Does anyone have any references to testing data on this? ICC-ES hopefully? Thanks in advance.

_________________________
TKE

 

[hokie66]

No chance you can depend on glue for that purpose. Why are you against post-installed (drilled in) anchors? You just need to locate and miss the strands, and that should be simple enough.

 

[TopKnot]

This is a long, skinny, four story wood structure with a lot of interior shear walls every 35'. Slab strands at 3-4' on center each direction, drilled anchors would be a nightmare. Most other similar plans I've seen use friction and PAF that penetrate concrete by 1.5". NO WAY I can specify that, without xraying the slab everywhere, and those are not very strong.

Why do you say "no chance"? We rely on epoxies for long term concrete bonds. We rely on glue in glue-laminated beams for long term bonds. Have you seen a code limitation for construction adhesive? Is there data that shows long-term delamination or relaxation? There is another argument for the glue increasing the friction coefficient.
-
I see 28-day shear strength of Liquid Nails multi-purpose wood adhesive listed at over 450 psi. Even trowel-applied polyurethane wood flooring adhesive lists shear strength at 150 psi. There is practical strength there. Anyone that has removed a glued bottom plate knows the wood shears, not the glue or the bond. Is there a code limitation that prevents using that strength? I'm not trying to be combative. Someone must have previously studied this one way or another. I'm looking for references. Thanks for responding.

_________________________
TKE

 

[hokie66]

I suppose I should have said "No chance I would depend on glue...", and now that I know it is a 4 storey structure, I am even more sure, notwithstanding what any code says. You may get different answers from others.

 

[TopKnot]

Has anyone seen glued wood-to-concrete connections fail long-term? If so, do you have thoughts on why?

More stories means a higher the sustained dead load for friction resistance. The math says that 60% of the sustained dead load resists the base shear in friction, if you believe 0.6 friction coefficient. Would you also neglect friction resistance and size drilled anchor bolts alone for the base shear?

_________________________
TKE

 

[msquared48]

If you have problems with drilling in the sill plate anchors that are relatively flexible in location, you must be mortified at the thought of drilling in the holddowns with their inflexible locations. Can't glue those sucker.

These details should have been specified in the plans prior to construction, not after the slab is in place.

Mike McCann, PE, SE (WA)

 

[dhengr]

TopKnot:
The way I see it.... The codes don’t say you can’t use adhesives for an application like this, but they don’t explicitly say you can either. It’s for you to prove the sufficiency of what you want to do, and then acceptance is up to the local AHJ, who may not know adhesive from shinola. And, he has no incentive to stick his neck out, even though never held responsible as you could be. There are plenty of adhesives which seem to perform pretty darn well, but their makers and suppliers won’t guarantee them for 50 years, and we don’t know what various environmental situations and various material interactions might do to the bond joint over that time. So, the code writing people say it is an unproven system, and may not be able to be relied upon, long term. It should be an ideal system for distributing that shear load in a uniform, low unit stress fashion; but less certainty in tension visa-a-vi a concrete slab. That’s an ideal stress distrib. for that application. It shouldn’t creep because the loading is only occasional, but it might age poorly and embrittle.

GlueLams were developed at a time when code scrutiny was much less stringent. They had the backing of major players in the wood industry. I’ve been designing with them since the mid 60's and we had plenty of design info. at that time, and very few problems with them. They are manufactured in a controlled environment, by a trained crew of workers, doing the same thing over and over, with the same materials. They certainly went through plenty of early tribulations in the way of glue formulations, etc., and consistency of product, susceptibility to moisture, etc. Anchor bolts (A.Bs.) are pretty well tested and proven, but they are somewhat susceptible to creep, very much dependant on the quality and care taken during installation, etc. We have certainly had some real disasters with their misuse and poor installation or selection of the proper epoxy. Again, they have had strong developmental backing from some major players. You’ll have to get some adhesive makers interested in backing this use of their product. So there is someone who will supporting long term testing and research to prove the system. MikeMc is right, these walls, their A.Bs. vs. the PT’ing. should have been coordinated on the plans and specs. Do you have enough slab depth for A.B. embedment?

 

[TopKnot]

It's under design. Finishing a slab around 700 placed anchor bolts in the interior is tough. Also tough to get them in the right place on 20,000 SF pours. X-raying a 44,000 SF slab for 700 drilled anchors is a pain. And the likelihood that an anchor is hit is too high for me to consider. I'd like to use some strength from adhesive and friction to reduce the bolt placement. I'm not advocating using adhesive alone.

Does anyone use sustained dead load to account for friction resistance under the bottom plate? I have enough load to take the shear in friction alone.

Maybe the best thing is to lose the PT and drill anchors after. That increase cost by $1.20-$1.40/SF. What have you seen done elsewhere?

_________________________
TKE

 

[hokie66]

What is so hard about locating tendons? You know where the ends are, and they should be in straight lines. If you can't control that, you have no chance of controlling the quality of field applied glue.

 

[Canuck67]

TopKnot,

I have easily been able to relatively accurately locate post tension cables in up to 8" thick slabs using a covermeter (rebar locator). Over the years I have located and specified work around 1000's of post tension tendons. I never used X-ray for 'large' situations, it was too impractical and not cost effective for widespread location. With tendons located 3' to 4' apart, and given a reasonable tolerance around the locations where tendons were identified, there should be ample room to drill and install anchors. The Contractor needs full time supervision by technical personnel who understand tendon layouts.

Pretty much any product adhered to concrete eventually debonds. Toppings, waterproofing membranes, coatings, etc... I could see using an adhesive in conjunction with post installed anchors as a 'belt and suspenders' type situation. Also, perhaps some limited percentage of adhesive use could be justified to bridge areas where tendon locations prevent the installation of the specified anchors.

 

[TopKnot]

Thanks for the responses. Tendons are not hard to locate. Just tedious with 1800 interior anchor bolts. That is a lot of tendons to miss when drilling past. I may end up coming full circle and shooting down plates with PAF. Those will be less likely to damage tendons. Contractor can choose to locate tendons or rely the concrete guy got the tendon deep enough.

No one commented on my question about using a portion of the available friction from sustained dead loads. 0.6*0.6*DL of sustained dead load seems like a real resistive force that would reduce required mechanical attachment. Has anyone used this? In a four story building with 3/4" gypcrete on all floors and perpendicular joists, that is a lot of available load for friction. If anything, the adhesive will help improve the coefficient of friction.

I see stories about 7,8 & 9-story wood framed structures. I imagine they are not relying on shear anchorage alone to resist base shear. But maybe those all have concrete stair/elevator cores?

_________________________
TKE

 

[msquared48]

OK...

Friction, to me is very unreliable in this case, particularly when the load to generate it can be very problematical and in a high seismic event with a lot of vertical motion, as well as horizontal. During the vertical motion downswing, the gravity forces that would generate the friction would be much less to zero to resist the lateral forces. I say no. If you want to provide the glue as an unquantifiable extra, OK, but do the primary with bolts.

Mike McCann, PE, SE (WA)

 

[TopKnot]

Lateral is not controlled by seismic. Wind only.

_________________________
TKE

 

[TopKnot]

Do you ever consider vertical movement for seismic design? If horizontal accelerations are 0.3g-0.6g, vertical movement is 1/3rd of that. Which is why vertical movement is generally ignored. It is not enough to impact a structure's vertical design. But I am no seismic expert. It's an interesting consideration though.

_________________________
TKE

 

[msquared48]

Different fault types generate different loads. Lateral slip produces lateral forces, and strike/slip faults produce uplift as well as lateral. Just depends where you are and what you have.

Mike McCann, PE, SE (WA)

 

[hokie66]

Back to locating the tendons...the simplest way should be to mark the tendon straight lines on the slab surface. Then just miss the lines when you drill. No need to x-ray or use a covermeter.

 

[Ingenuity]

hokie66, for a suspended slab, marking the forms at the tendon locations before concrete placement works well, but the OP has a PT slab on grade. Form marks nor x-ray will work to pre-locate the tendons in a SOG.

 

[hokie66]

Maybe you can explain that for me. Aren't the tendons straight?

 

[TopKnot]

Some tendons will have minor deflections to work around conflicts, slab depressions, etc. But the slab is huge. There would likely be too much variance in location to risk marking a tendon from the end. They are criss-crossing everywhere. 2'-10" on center each way.

Everyone has different opinions, but in the absence of publish data or unknown code issues, my opinion is as follows:
1. Drilling 1800 anchors 2.75" into a PT slab on grade is way too much risk. I would never require that in good conscience. I'd rather make the concrete guy finish around placed anchor bolts.
2. I prefer the PAF attachment, with tendon location by a covermeter. The fastener will theoretically not hit the tendon even if placed directly above it. 1/4" is too close for comfort, but there's a lot less risk in that than drilling an anchor past a live tendon.
3. The only reason given to ignore friction effects from sustained dead loads is seismic, and this is not a seismic area. Even if it was, the vertical effects of gravity will act to reduce the sustained column dead load by 10-30%. Which is still within the 0.6 of sustained dead load. If the ground dropped away from the plate, you had better keep your column unity check under 0.4, because your static gravity loads just went dynamic. I wouldn't want to rely on friction for more than 50% of the base shear. But using friction effects to reduce attachment makes sense to me. I was hoping someone could point me to a published article addressing this one way or another.
4. Construction adhesive should only be considered as a belt and suspenders approach. If anything, it ensures that you could at least obtain a friction resistance of 0.6*0.6*Sustained DL.

For whatever it is worth, I'm going RC for this one. The cost benefit is not worth PT in this case. But I could see other squishier sites having a big savings with PT. I will check back to this over time to see if any other experienced PT slab on grade guys can comment on the issue. Thanks for the responses.

_________________________
TKE

 

[hokie66]

You are probably choosing the right course to use conventional reinforcement. You would also largely avoid the issue of drilling if you used bonded PT, but I suppose you are in the US.