Shear Friction Through Grout Under Baseplate

[r-struct]

For a steel column baseplate connected to the foundation by anchor bolts with, say, 1" to 2" of grout under it, acted on by both tension (or compression) and shear at the same time, can we use the following shear friction expression as the shear capacity of the base connection?

V,capacity = phi*mu*(Ab*Fy - T)

where

phi = friction resistance factor
mu = coefficient of friction
Ab = total cross sectional area of anchor bolts
Fy = yield strength of anchor bolts
T = tension force acting on the anchor bolts (positive if tension, negative if compression)

This is assuming that the anchor bolts could reach their yield strengths prior to other failure modes occuring (concrete breakout, anchor pullout, or sideface blowout). Do you think, assuming only the shear friction mechanism works, this allows us to not needing to check the bending of bolts due to shear, or having to have some other form of shear transfer mechanism such as shear lugs or separate anchors specifically for shear?

Any thoughts from the more experienced engineers will be appreciated.

 

[hokie66]

My thought...I would never use the concept of shear friction in this application. Why? Engineering judgment.

 

[r-struct]

Hi hookie66. Would you care to share the reasoning behind it?

 

[P205]

First off, this seems to be a heavily debated issue, so expect varying opinions.

Secondly, do you have access to this book:
Foundation and Anchor Design Guide for Metal Building Systems - Alexander Newman

Newman discusses this topic in good detail. Although he advises against this method of shear resistance, he admits that shear friction needs to be overcome in order for the anchor bolts to act in bending.

Thirdly, most of the buildings I design, I'm able to bury the column baseplates below the surface of the slab and can take out the shear by bearing against the concrete.

Lastly, I've personally have had situations (not often/not many) where I had no real other choice.

 

[WARose]

The only time I've ever transferred shear through the base plate grout (i.e. by friction), it was because there was plenty of (compressive) axial load to make it happen. Remember: grout is unreinforced, thin, and typically cracks.

I'm not sure where this formula comes from but if you involve the anchor bolts, the transfer with them will be by bending of the bolts alone (with the bolts as a cantilever). No friction will help anything. It's sort of all or nothing deal.

 

[hokie66]

r-struct,
I thought I made it clear that it was my judgment that didn't allow me to use friction, shear friction or otherwise, in resisting substantial forces. Shear friction is a concept developed relatively recently, and has been adopted by ACI, but not by many other code authorities. It is still controversial, and not well enough supported by theory to suit me. It is supported by testing for fully developed reinforcement, but not for anchor bolts. In concrete, the equation which has been incorporated in ACI depends partly on friction, partly on dowel action. I prefer just to use the dowel action.

 

[r-struct]

@P205 - Thanks for the reference, I will try to check out that book.

I have thought about that case too, where the baseplate is buried under the slab: my concern was that on how to keep the slab in contact with the steel column if the concrete around cracks or shrinks away. I am guessing additional details around the column will need to be added to preclude that from happening.

@WARose - Thanks for the feedback. That formula follows the interface shear friction formula from the concrete code with cohesion=0 and the term inside the parentheses is like the "net normal force" that holds the base plate, grout, and top of foundation together.

 

[WARose]

[blue](r-struct)[/blue]

That formula follows the interface shear friction formula from the concrete code with cohesion=0 and the term inside the parentheses is like the "net normal force" that holds the base plate, grout, and top of foundation together.

Ok....but I don't see how that is your "net normal force"......unless your bolts have pre-load in them.

And to build on what hokie was talking about.....another problem with shear friction is getting a number (i.e. the coefficient of friction) you trust for that. Most codes don't address it.

 

[r-struct]

@hookie66 - Thanks for expounding on that. I was just curious to see what other engineers in this forum think about it and also the reasons why.

 

[BridgeSmith]

Usually, shear friction in this type of application is ignored for several reasons:

1) Typically, the shear capacity of the anchor bolts is adequate to carry the shear.

2) Grout shrinks; even supposedly "non-shrink" grout shrinks a little. With a smooth base plate, no pressure = no friction. Unless you can be assured that the nuts on the anchor bolts are going to be retightened periodically, you cannot be confident there will be friction.

3) In the applications I've seen, a grout pad is employed where leveling nuts are used on the underside of the base plate. This again is a system where there is no pressure between base plate and the grout pad, hence no friction.

Generally, if more shear capacity is needed, the solution is to increase the size or number of anchor bolts.

Also, in the AASHTO Bridge Design spec. the bending of the anchor bolts due to shear is not considered where the unsupported length of the anchor bolt is less than the diameter of the bolt. We just always make sure the distance from the leveling nuts to the concrete is less than the bolt diameter so we can design for the straight shear capacity of the bolts.

 

[cooperDBM]

The shear friction method in ACI, and many other codes, assumes concrete with aggregate. The aggregate interlock causes a slight opening of the crack as it shifts which mobilizes the clamping force(Ab Fy) in the steel across the crack. Grout without aggregate won't develop a clamping force. You'll have the column force but no contribution from the bolts for friction. Not sure if the coefficients in the code would be valid for grout.

 

[dik]

HotRod10: Many grouts slightly expand...

Dik

 

[BridgeSmith]

Dik: Initially they do expand during hydration, but many shrink slightly over the long term as they dry. If the bond strength is sufficient, the shrinkage cracks will develop internally, but the bond with a steel plate is generally not sufficient for that to occur.