tensile bond strength for bonded structural topping that includes isolated repair areas

[ajk1]

What tensile bond strength should be specified for bonded structural topping where the topping contains rebar and is to act compositely with the base slab? The topping includes but is not limited to the top steel repair areas in the base slab which is a two-way flat slab.
CSA A23.1 specifies 0.9 MPa "minimum" for bonded toppings (see attached) but is vague about what this really applies to, leaving it up to the owner (i.e the structural engineer) to decide if more than the minimum should be specified.
0.9 MPa sounds like what might perhaps be appropriate where the topping is not structural, but seems low these days when 2.0 MPa can be achieved with a shotblast and good bonding agent. If the topping is bonded at 0.9 MPa, I wonder if that is good enough for the punching shear area around the supporting columns of a two-way flat slab. So my question is:


What tensile bond strength should be specified for bonded structural topping where the topping contains rebar and is to act compositely with the base slab?

 

[KootK]

1) I dunno. What follows is just musings.

2) Secondly, without having run any numbers myself, I suspect this is largely a non-issue. You're counting on the the bonding between layers to transfer horizontal shear. And the area over which you'll be doing this is rather large.

3) It's tempting to just reinstate the concrete modulus of rupture strength so that you can say that you've stitched the concrete back together as if it were monolithically cast. I imagine that this would result in a relatively high required pull of bond stress though.

4) I've proposed an evaluation method below. It's based on two fundamental principles:

a) There should be enough horizontal shear transfer capacity available to develop your tension reinforcing over normally assumed development lengths.

b) The pull-off tension stress can be used in place of transverse rebar to achieve something like a shear friction load path.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[ajk1]

to Kootk - an interesting idea. I will give it some thought and see how it works out. Thank you

 

[ajk1]

To Kootk:

2 way waffle slab with ribs at 24" centres, each rib in column strip has 1#7 bent (truss) and 1 straight bottom. Adding the bent bars from the span each side of the column gives 2#7 top bars. In addition there are 5#7 straight top bars within the column strip, over the support. This gives 2#7 + 5 x2/12 = 2.83 #7 bars in 24" width.

As = 2.83*0.60 = 1.70 square inches of top steel per 24" width.

If we take fy= 400 MPa = 58 ksi, then Ld = 575 mm ± = 22.6 inches ± , for #7 bar (interpolating from tabe in CAC Hnadook for metric bars -- I do not have the latest edition at home so may not be exactly right).

µ for concrete placed against deliberately roughened concrete = 0.60 (from CSA A23.3, for shear friction provisions).

σ_pulloff = As fy / (s Ld µ) = 1.70 x 58000 / (24 x 22.6 x 0.60) = 302 psi = 2.08 MPa.

CSA A23.1 minimum required bond strength for topping is only 0.9 MPa.

It seems to indicate that the CSA minimum is insufficient. Any comment?

(Note also that this is a 2-way system supported on concrete columns, so that may complicate things even more, but let's just deal with a one-way system to start).

 

[KootK]

It seems to indicate that the CSA minimum is insufficient. Any comment?

1) What you did is exactly what I had in mind. It's interesting to see the result and commendable that you've chased it down.

2) I'm sorry that the result has worked out unfavorably. I wasn't expecting that.

3) There are some other mechanisms that could be engaged. One is the fact that each "patch" being dislodged could grab on to its side neighbors for more averaging. Another is that, where you have more than Ld available on each side of the peak moment, you've got a patch length greater than Ld to work with. Unfortunately, both approaches result in one needing to look at all of the development length conditions on a case by case basis which seems a bit impractical.

4) 2.0 MPa is sounding better all the time.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[ajk1]

To Kootk;

Regarding your point #3, perhaps instead of the development length, I used the length from face of column to the bend down point of the bent (truss) bar. For 24 foot span, that would be in the order of 0.15 x 24 x 12 = 43"±. That would reduce the σ_pulloff to 1.09 MPa. This is still 20%± greater than 0.9 MPa minimum given in CSA A23.3 but getting close. However, it assumes a uniform distribution of shear stress over the 43" length, whereas I would expect the shear stress to be much larger at the face of support.

Does BARetired have any comment?

Is it worth my writing to CSA about this?

 

[KootK]

I would support the approach that you've suggested ajk1. I wouldn't sweat the uniformity of shear stress bit. Modern codes abandoned the strict application of the bond stress concept a while back so I think that we can too.

I was thinking about this over the weekend and two things struck me:

1) How does this work without requiring a check for monolithic concrete? I've seen some pretty densely reinforced slab top mats.

2) It's always bothered me that one way shear checks exist in two way slab design as it never governs and the lateral distribution of one way shear isn't uniform at all (yes, there would be some redistribution).

I wonder if the one way shear check on two way slabs isn't really intended to check horizontal shear like we're discussing here. And that would imply that you could spread your horizontal shear over the entire design panel.

Another avenue might be to check with a bonding agent supplier and see if you can figure out how much shear stress capacity normally accompanies a particular pull-off stress. The shear friction concept may not be the best way to go about establishing horizontal shear capacity.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[ajk1]

You have given me lots to think about, Kootk. I will give it more thought. One thing does not seem logical is to adopt the CSA 0.9 MPa minimum bond strength requirement for toppings, as being all that is required irrespective of the load on the slab. Yet I suspect that is what some engineers or engineering technicians may be doing.

 

[KootK]

Link
Link

Here's some more to think about. The Lehigh testing seems to suggest that you should be able to get shear values closer to 4 MPa. The Loov analysis seems to be pretty consistent with what we've been discussing.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[KootK]

It's also interesting to note that Loov advocates a VQ/I approach similar to your proposal above. I'm not sure what to make of that near columns in two way slabs where shear stresses are very high and very localized.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[SkiisAndBikes]

ajk1,

From my days as a junior engineer, I completed a fair number of bond tests using a LokTest machine. I believe our specifications called for a minimum 1.0 MPa bond strength, but admittedly this was for concrete repair patches and bonded toppings that weren't necessarily required to be structurally bonded. In addition to the failure load, the location of the failure was recorded (repair material, bond line or parent concrete). Often the failures were in the parent concrete which we considered a good indication of the repair work completed. While it was not difficult to achieve greater than 1.0 MPa, I cannot remember too many results that exceeded 2.0 MPa. If you were to conduct a sufficient number of bond tests on the existing structure, prior to work commencing, would your specified bond strength need to exceed the values you obtained for the parent concrete? Could a testing program provide you with additional guidance to your calculations?

 

[Ingenuity]

For FRP strengthening to concrete elements we commonly undertake bond tests of the laminate to the substrate, similar to what SkiisandBikes describes.

We regularly obtain pull-off results of more than 400 psi (2.8 MPa). We have had values as low as 100 psi for a historic concrete structure built in the late 30's.

The following three samples had a range of 424 to 557 psi, with an average of 484 psi.

Similarly, a recent bridge overlay with a substrate that was 30+ years old obtained values of 300+ psi.

But, the rate of loading applied by the technician (usually a hand-crank system) comes into play with the results, such that one manufacturer has discontinued their old hand crank systems for an automatic loading...and a cost of US$5k +. Gulp!

 

[ajk1]

to Skis and Bikes - good points. But at least one of the 2 low strength tests (out of 5 tests) was at the bond line. But your suggestion of checking some before the new topping is placed, is a good one. Thanks.

to Ingenuity - interesting. We have done lots of bond tests on urethane waterproofing traffic topping membranes and with some manufacturers' membranes we fairly consistently get above 1.7 MPa, and with one manufacturer;s membrane we fairly consistently got above 1.9 MPa, at least on on 2,000,000 square foot garage where there was a lot of these tests done on the membrane.

 

[ajk1]

There must be a umber of experienced and thoughtful engineers on here who have designed structural repairs to parking structure floors and the like, where the concrete is chipped out to about 25 mm below the top rebar and the rebar is cleaned and new concrete placed (including supplementary rebar where necessary). I am interested in hearing from you as to what minimal bond strength you specify for such repairs. Do you specify the 0.9 MPa bond strength given in CSA A23.1 (and I assume in ACI corresponding standards) or do you specify a higher bond strength? My initial opinion is that for a structural repair, as opposed to a leveling course type of topping, a higher minimum bond strength should be specified.

 

[Ingenuity]

200 psi min., typically, I have seen bridge specs use 300 psi min.

 

[MacGruber22]

ajk1,

For me, 300 min psi for my direct tension test for non-structural overlayment and thin patch structural repairs. That said, a number of manufacturers don't publish the direct tension bond values, rather bond is shown for ASTM C882 Slant Shear. Slant shear bond values are much higher of a value because of the nature of the test. For thin structural repairs (up to 1/2*db, I want the slant shear to be greater than 2500 psi if there is no direct tension bond test.

am interested in hearing from you as to what minimal bond strength you specify for such repairs

Forget direct tension bond for those deeper repairs. Slant shear is more representative of deeper repairs. You won't find many manufacturer's who test direct tension bond on those products. I want at least 1000 psi slant shear for those deeper repairs.

"It is imperative Cunth doesn't get his hands on those codes."

 

[ajk1]

Thanks you Ingenuity and MacGruber 22. Those are the direct tension strength ranges that I have been thinking of as being appropriate.

I am curious MacGruber as to:
a) why you say to forget the direct tensile strength tests since they are so widely carry out on site to measure the bond strength of concrete repair areas and the test method is given in CSA S231.2, and
b) is it practical and economical to carry our multiple slant shear tests on site?

 

[MacGruber22]

a) I am really speaking for selecting bag mix repair material for deep repairs, because that was your description. If you are performing a field test to verify, you can still use the direct tension pull-off test, and compare to the 200-300 psi criteria, but you will have to create a more shallow repair so that you can perform the test.
b) I never saw anyone perform the slant test in the field as the cylinders are placed in the same machine to test cylinder compression strength.



"It is imperative Cunth doesn't get his hands on those codes."

 

[Ingenuity]

I have only used slant shear testing on epoxy adhesives, and the test is a pain to conduct, even in a lab setting.

I did find this reference from ACI in my files: Evaluation of Procedures for In Situ Tensile Bond Testing of Concrete Repairs by J. E. McDonald and A. M. Vaysburd, with the following recommendation pertaining to tensile pull-off testing:

 

[MacGruber22]

I am interested in hearing from you as to what minimal bond strength you specify for such repairs

I have only used slant shear testing on epoxy adhesives, and the test is a pain to conduct, even in a lab setting.

I was trying to gear my comments around ajk1's question of specifying bond for deep repairs. For instance, here are a few bag mix concrete repair materials for deep repairs:

_________________ASTM C882 MOD______ASTM C1583
USCP HP___________2600 psi___________Not Published
Sikacrete 211________1500 psi__________Not Published
RapidSet DOT________7000 psi__________Not Published

I guess my question is why the pull-off test isn't published for these types of products.

"It is imperative Cunth doesn't get his hands on those codes."