Post Installed Anchors in Green Concrete

[KootK]

I could probably find this information in a Hilti manual or by calling a rep but I'm going to start here. I guess I just have a fundamental mistrust of sales literature / people.

Let's say that I've designed a post installed anchor and selected it from a manufacturer's catalogue based on a 30 MPa concrete strength. If the anchors get installed into 20 MPa concrete that hasn't yet reached it's 28 day strength, what does that mean for the anchor capacity in the completed structure? Am I limited to capacities based on 20 MPa? Or will the capacity of the anchors eventually reach that consistent with the 30 MPa assumption? This assumes that the concrete eventually gets there of course.

I'd like comments pertaining to both expansion anchors and drill and epoxy anchors. My intuition is that a drill & epoxy anchor will eventually make it's way to the 30 MPa capacity while an expansion anchor may not.

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

 

[ztengguy]

Do you need the load capacity at 30? If you are ok at 20, then I would say 30 is better.

How does it work for 20, if calcs need 30?

to me its no different than a cracking calc. At 20 mpa it might crack, where at 30 it would be fine.

If you need 30, wait for 30.

 

[precast78]

I think for expansion anchor, it will reach the full capacity at 30 mpa. Why do you think it would not?

 

[Once20036]

If you install a mechanical anchor in green concrete, I expect it would develop the full strength once the concrete cures, provided no loads are applied during this curing.

I would expect an epoxy anchor to fall short of the expected strength. I've heard that ACI 318-11 required concrete to have cured a minimum of 21 days prior to installing an epoxy anchor. As I understand it, it's not related to the strength of the concrete, but rather the higher levels of moisture and how this affects the bond strength.

Whether you believe what they have to say or not, I`m curious what Hilti would say. I would expect to hear, "We haven't tested that" but you never know. You can also ask to speak with one of their engineers rather than the sales staff that typically answer the phone.

 

[KootK]

I do need the 30 MPa capacity.

My thinking with the expansion anchor is that as the expansion occurs, the surrounding concrete might crack locally and otherwise give way a bit to the expansion process. If more of that giving way would occur at 20 MPa than it would at 30 MPa, I would think that it would a) affect the final strength and b) not be recovered as the concrete matures.

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

 

[KootK]

That is interesting about the drill & epoxy anchors. So much for my intuition.

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

 

[Once20036]

To expand on my previous post, ACI 318-11 RD2.2 says, in part,
"The design performance of adhesive anchors cannot be ensured by establishing a minimum concrete compressive strength and the time of installation in early age concrete. Therefore, a minimum concrete age of 21 days at the time of adhesive anchor installation was adopted."

I understand your logic re: local cracking, but I`m not sure that I share your concerns.
If you're installing the anchor in properly cured, old concrete, I can imagine small/micro cracks in the concrete.
If you're installing the anchor in brand new concrete, I can imagine these cracks being slightly larger - but enough to impact the capacity?

In the 30MPa concrete, what's the failure mechanism of the expansion anchor?

 

[KootK]

I bet that 21 days is often not even happening. Good to know about that.

The application is direct tension with lots of edge distance. I assume that means some manner of concrete breakout failure.

As a mental experiment, consider casting an already expanded expansion anchor into wet concrete. When the concrete dries around the anchor, will the anchor possess it's rated capacity? Certainly, the contact stress between the anchor and the surrounding concrete would be less. I feel that the 20 MPa case must lie someplace along the non-linear continuum between fresh concrete and 30 MPa.

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

 

[CELinOttawa]

I understand this is a rather big no-no, as creep and micro fractures both combine to completely undermine strength of mechanical anchors which were installed in early age PCC.

I'll be very keen to learn differently if anyone has sources!

 

[KootK]

@ CEL: in my little scenario, do you accept the 20 MPa capacity as a minimum? Or do you contend that it might be even worse than that?

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

 

[precast78]

In my opinion the strength with the green concrete should be around (SQRT(20)/SQRT(30))*design capacity at 30 mpa. It will reach full capacity once cured even more.

 

[Once20036]

Regarding failure modes - there could be a couple for an anchor in direct tension with lots of edge distance:
1) Steel failure
2) Pullout
3) Concrete failure.

I have a hard time believing that steel failure (of course) or pullout would be affected by the strength of the concrete at the time the anchor is set.
If one of these controls your design - maybe you have nothing to worry about.

If your failure mode is the concrete cone/breakout - we're back to square one with the above discussion.

 

[CELinOttawa]

My gut says that once you install into 20MPa concrete, that's what you'll be able to rely on from then on. Creep characteristics in particular mean that you can generally only count on the 28 day strength, no matter the age of the structure. General rule, but only to be varied from with due caution.

 

[Once20036]

I`m curious if when the load is applied makes a difference in the conversation.
I`m picturing this as being an occasional tension load, created by wind or seismic such that there wouldn't be long term tensile loading subject to creep. If this is the case, I`m still leaning towards being able to use the 30 MPa.
If the tensile load is constant from the date of installation, maybe the creep plays a more significant role.

Has there been a call to the manufacturer?

 

[KootK]

Thanks for all the advice everyone. I submitted this question to Hilti via email. My hope is that they'll respond via email and I'll be able post that response here.

@Once: I expect that when the load is applied would make a difference. I'm hanging some equipment with these anchors. The equipment won't show up until well after the 28 day strength is reached.

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

 

[CELinOttawa]

I am willing to bet this is going to vary greatly on the type of anchor... Some anchors put their stress on the concrete immediately (undercut) and others the stresses increase with load application (wedge and drop-in for example).

If the maximum stress is immediate, you're going to have more trouble as the majority of concerns should revolve stressing prior to design strength. Where the higher end and peak stresses occur after load is imparted onto the concrete fastener, you're far more likely to be achieving full strengths despite the early install.

I would love to have a simple chart of anchor type and stress effects... Maybe I've just thought of a little side project to create a FAQ.

 

[KootK]

A phone call to Hilti ended up being the way to go. There is relevant information in section 2.1.2 of their Anchor Fastening Technical Guide. Here's what I learned. I'm going to reference 20/30 MPa values here for clarity as that's been the context of the discussion so far.

For mechanical expansion anchors, installation should not take place until the concrete has aged at least seven days. If you let the concrete mature to its design strength before loading it, you get the full capacity (30 MPa). If you load the anchor before the concrete reaches design strength, you're stuck with whatever capacity would be associated with the strength of the concrete when the anchors were installed (20 MPa). Hilti feels that this last statement is probably conservative which is fine by me.

Apparently the rules for adhesive anchors are identical to those for mechanical anchors. The Hilti representative that I spoke to wasn't familiar with the 21 day requirement mentioned above (ACI 318-11 RD2.2). Hilti feels that adhesive anchors can be installed at seven days without reduction. That being said, they have wet install factors that do need to be accounted for.

Power-actuated fasteners, which we have not yet considered here, are different. You can install them after seven days but you are permanently stuck with the capacity associated with the concrete strength achieved at the time of install (20 MPa). It seems that power-actuated fasteners achieve a sort of fusion weld to the concrete just like they do with steel. The quality of that fusion weld is a function of the concrete strength at the time of install of course.

I invited my new Hilti friend to join our discussion here, at Eng-Tips. If he turns up, we can pepper him with more arcane questions then. It sounds as though Hilti has tested most of this stuff already which is reassuring.




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

 

[CELinOttawa]

So for clarity:

Hilti feels that adhesive anchors can be installed at seven days without reduction. That being said, they have wet install factors that do need to be accounted for.

Is to say that you get whatever strength the concrete has achieved without a further reduction, correct? So in this case (for our example values above) we would have strength based on 20MPa concrete?

 

[KootK]

No, you'd get the full strength of the anchor CEL (30 MPa). The reduction factor that Hilti specifies is for adhesive anchors installed in a wet hole. And that's wet as in it's raining or somebody's spilled their big gulp. Hilti is unconcerned with any free water coming from partially aged concrete which seems to be the gist of ACI 318-11 RD2.2.

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