Concrete adhesion to steel? 1

[kidCivil]

Hello everyone,

I'm working on a design and need some pointers on either some literature or people's experiences with concrete adhesion to steel (not ribbed reinforcement).

The design calls for steel screw piles (steel circular hollow section) to be installed under a conc floor. The piles need to resist uplift forces on the floor. Once installed, the circular section is filled with concrete and one or two engagement bars are cast into it. These engagement bars are ties to the slab reinf.

The screw pile resists uplift in it's own right but I have to prove the load transfer capacity from the reinf engagement bars to the pile THROUGH the concrete. If the reinf bars are cast into tie concrete, what is to say the uplift force doesent pull a plug of concrete out of the steel hollow pile?

What I am looking for is literature that proves the chemical bond between concrete and steel. I have been looking on the net WRT Aust steel manufacturers but cant find anything useful.

Anyone?

thanks in advance

-John

 

[kenvlach]

Concrete adhesion to steel depends upon mechanical as well as chemical bonding, which is why rebar is hot rolled with deformed pattern. So, only the overall adhesion is (normally determined and) reported.

The chart "Concrete Reinforcing Steel Bond" at

http://www.galvanizeit.org/showContent,278,322.cfm#strength

shows bond strengths of 350-550 psi for both black steel and galvanized steel deformed rebar.

Bond strength testing can be conducted per ASTM A944-05 Standard Test Method for Comparing Bond Strength of Steel Reinforcing Bars to Concrete Using Beam-End Specimens

http://www.astm.org/cgi-bin/SoftCart.exe/DATABASE.CART/REDLINE_PAGES/A944.htm?L+mystore+ttgt4192

The test results for a given size rebar are reported as loaded end elongation per applied load. Some test data for #6 rebar (3/4") are shown at

http://www.cortecmci.com/Mailing/AET 05-01412 b.pdf

The bond strength was sufficiently high that failure at 23,600 lbs. load was via concrete cracking rather than debonding from the steel. I've never tried converting this type of data before, but as it appears to me from Fig. 2 of ASTM A944 that the test bar bonded length is 12", the total #6 rebar bonded surface area (nominal) is 28.27 sq. in, and hence the bond strength > 23,600 lbs/28.27 sq. in. = 835 psi. Probably though, the concrete-to-steel loading is higher near the applied load end of the test bar, hence the bonding may be also.

It's important to note that the bond strength [μ] = k[√]f[sub]c[/sub], so you can increase the bond strength by x1.414 by doubling the (compressive) strength of the concrete.

Hope that helps.

 

[kidCivil]

What if the mechanical properties were ignored? That is, chemical bonding to the smooth walls of the section?

 

[SlideRuleEra]

Ok... Here is what should be a workable answer - obtained, and documented, in a round-about way:

1. The buoyancy of a cofferdam can be restrained by a concrete seal.

2. This force to resist uplift can be developed from two sources, the weight of the concrete plus the friction of the concrete against the flat sides of any steel H-pile encased in the seal.

3. The Indiana DOT Design Manual addresses both issues in Chapter 66 "Foundations". Consider only the friction component.

4. Paragraph 66-2.11 "Concrete Foundation (Tremie) Seal" states that "the allowable service load bond stress between the steel pile and the seal concrete should be taken as 250 kPa" (250kPa = 36 PSI)
Here is the link

http://www.in.gov/dot/div/contracts/standards/dm/Part 6/Ch 66/ch66.htm

5. The analogy between the above situation and your application sounds close enough for this approach (resulting in a value of approximately 36 PSI) to be reasonable.

As an aside, I know from personal experience that this friction force can be very large - without going into details, I was present on a jobsite in 1981 when a 40' long, 2' diameter rock-socketed caisson was accidentally pulled out of the ground (totally intact) because of this (smooth)steel/concrete friction.

Best Wishes

http://www.SlideRuleEra.net

 

[MLMoring]

I think the essence of the question revolves around the adhesion ("bonding") of the concrete to the hollow steel pile. I assume the reinforcing has been adequately sized and it's development length verified. I also assume that the embedment length in the pile is sufficient as to preclude a cone type pullout (i.e., the failure cone intercepts the sides of the pile and not the upper surface of the concrete).

Pullout of the plug will not occur if the adhesion factor times the inner circumference of the pile times the length of embedment is less than the applied uplift. (I know this is not precisely accurate because of factots such as tensile strength of concrete, the length of embedment is less due to the cone effect, etc.)

The actual issue can be complicated by many conditions. Some include: The condition of the surface of the steel pile (rough, smooth, coated with something which would affect adhesion); The amount of shrinkage the concrete may have undergone during curing. Some of the best and earliest studies of this were done by the late Professor George Winter.

Therefore, look up resources on "bonding" of concrete to steel and try to find conditions described to approximate your situation. Check with the ACI.

 

[DaveVikingPE]

I am never comfortable with relying soley on the bond between concrete and structural steel - unless there's shear connectors/studs involved. For the circular screw piles, I would check the advantage adding some shear studs around the outer perimeter would give me.

http://www.nelsonstudwelding.com/

has some great info and you can check it out in the AISC manuals, too. What say you, SlideRuleEra?

 

[kidCivil]

thanks guys

much appreciated

-John

 

[SlideRuleEra]

DaveVikingPE is right, anything to provide a mechanical anchorage (compared to relying only on friction) is an improvement.

However a pinned, external sleeve (in addition to the concrete fill) may be easier. Adapting the concept shown in this photo to your application should be straightforward

http://www.screwpile.com.au/a_tens.asp

Also be sure to check the downloadable drawings at that link (under the "Design" tab) for other ideas.

http://www.SlideRuleEra.net

 

[MLMoring]

I agree that the studs are positive anchorage, but I'm under the impression that the concrete and reinforcing bars are inside the hollow pile. This makes it impractical to install the studs.

It may be feasable to weld the reinforcing to the inside wall of the pile at the top. The length of weld required will be significantly less than the development length and if the pile isn't too small in diameter, could be done easily. If so, be sure to check the compatability of the welded metals as well as the stresses in the pile wall.

The detail indicated on the "screwpile" website is certainly workable, but it sure looks expensive.

 

[apsix]

Couldn't you look at using a length of rebar threaded horizontally thru 2 holes near the top of the pile tube, and then bent up into the floor slab. Corrosion protection could be provided by a small pile cap.