At the office the other day we got into an interesting discussion after a contractor told us he does not remove shims [used for leveling] underneath column base plates.

That situation sort of concerns me because even if you did pack grout underneath.......in my mind that is still partial bearing underneath (i.e. just the bottom of the base plate to the shims). I've had a few people express the opinion to me that it's not an issue as long as non shrink grout fills the voids. They say that before anything bad would happen the shims would yield and everything would reach a happy medium (i.e. it would all share the load). Indeed, I found this on AISC's web site:

Quote:

Grouting Base Plates

Question:

"We were told by a steel erector that they typically do not remove shims from column base plates after grouting. They also informed us that they do not back off leveling nuts below the base plates during the grouting process. Are these practices acceptable?"

Question sent to AISC's Steel Solutions Center

Answer:

"Yes, shims are left in place underneath the base plates. From a construction standpoint, those shims or leveling nuts hold the load while the grout cures. Their presence after the grout is structural and does not reduce strength. And from an economic standpoint, removal would needlessly increase cost.

Axial compressive forces from the column will be almost evenly distributed as bearing forces on the shims and non-shrink grout. Even if the shims were to take the majority of the load, the assembly will deform in a self-limiting manner through localized yielding or crushing of concrete as the force-distribution model assumed in sizing the base plate is attained."

Bill Liddy
American Institute of Steel Construction

Posted on July 1, 2004

And I don't buy that for one second......the way I see it either the concrete or the shims are going to win that fight as to what supports the bottom of base plate. Either the shims (that are [in this case] relatively small compared to the overall plate area) will yield to the point where the grout takes over or nothing happens and the shims carry all the load on a few points (which can't be good for the bearing of the plate [as far as concrete stresses go]; or stressed in the plate, etc.). I don't believe we will all hit a happy equilibrium and it will all be shared equally. They can't be possibly be compressed at the same rate.

Then again: the paragraph above assumes a infinitely rigid plate on the bottom. You would think that the plate could deform and distribute the load like that......but the plate is a couple inches think.......so I don't know. Thoughts?
 

If you removed the shims, what would take the column load?  Just follow the AISC advice.  Think of the principles of ultimate strength design...it doesn't matter whether the shims or the grout take the load first, as in the fully loaded condition, both will participate.  The structure will decide on its own how to take the load...we can't dictate to it.

If the shims yield, they still support the force that caused them to yield!

This has been the practice, successfully used for a very long time. You will need a good reason or a money saving process if you want to change it.

Michael.
Timing has a lot to do with the outcome of a rain dance.

Quote:

If the shims yield, they still support the force that caused them to yield!

What if they don't yield [nor does the plate deform] and nothing above them is damaged but the concrete the shims bear on is cracked/damaged?

BTW, thanks for the replies so far.

What would be your failure mechanism?
1. Shims take all of the load and the column rotates over the shims - The rotation will eventually be resisted by the grout.
2. Shims take concentrated point load into the foundation - Foundation is properly reinforced and will yield unitl the load redistributes through the grout.
3. Column base plate yields due to shim concentrated load - yielding redistributes load to grout.
 

This is one of the reasons I like to use leveling plates and leveling nuts.
As discussed on this forum ad nauseum, the leveling plate (1/4" usually) can be installed on leveling nuts and grouted with virtually no load on the plate prior to column installation.
The only problem with either method as I see it, is if the leveling nut can force the anchor road to take a compressive load that is usually not designed for in terms of concrete anchorage.  

That's interesting Toad, that's why I wouldn't use leveling nuts for a grouted base. Switchyards and steel transmission poles are generally not grouted so we used leveling nuts and designed the bolt for the load.

Michael.
Timing has a lot to do with the outcome of a rain dance.

Pad-
To be clear, I DO design for the compressive load.
My point was this is not normally a design consideration.
The only other way I know of is to use jacking bolts and grout holes in the base plate.

IMO, the compressive loads on leveling nut are not a concern.
I try to use jamb nuts for leveling instead of heavy hex to mitigate the problem.
 

I think leveling nuts are exactly the same as shims if placed under grouted column base plates.  If leveling nuts are used with a leveling plate, then the base plate has to be square with the column, which is not likely, so some yielding is required with that solution as well.

What's a "jamb nut"?

Its a nut with about half the thickness of a heavy hex nut

...usually used as a lock nut....simply tightened down onto the main nut or "jambed" against the main nut.

Toad,
I was just having a go at your spelling.  The word is "jam".

TJ...some good points mentioned on designing the AB for the compression load.
Many times the compression load will be greater than the tension load.
Any consideration to the stability of the bolt in compression?
I agree that the steel will yield eventually and transfer the compression load to the conc. But what happens in the meantime to the AB in compression before this distribution takes place.
Have designed many AB's and have never thought of considering this which is an other good reason to NEVER skimp on AB or fdns.

@SAIL3, The grout will have cured before the bulk of the load is built onto the column.

Michael.
Timing has a lot to do with the outcome of a rain dance.

This argument bears some similarities to the transfer of stress to the reinforcement when the concrete in a column creeps over time (Reference section ACI 318-10.9 commentary).  In the commentary, ACI seems to think that yielding the rebar over time is a bad thing and should be limited (not sure why).

Perhaps yielding of the shim stack or anchor bolts is a bad thing (even long term creep of the grout).  But, no one has illustrated what the failure mechanism would be should the shim stack, anchor bolt and grout not share the loads properly.

Due to the mostly elastic, redistributive nature of steel and reinforced concrete, I am unable to think of a scenario of failure.  We aren't talking about glass.

Teguci...that's true it's not glass...however:
Headed AB..does not depend on concrete bond....what about the  
           stability of the AB in compression?, assuming an  
           unbraced length from bottom of base pl to the bottom  
           of the head of bolt.
Threaded AB...depends on concrete bond...if development
              length was based on tension loads, this load could
              be exceeded by the compression load.

Admittedly, those scenarios are abit of a reach, but in light of all the nit-picking and extremely low probability of some of the events ever happening, that nowadays are given a long and serious
analysis and attention by the current codes and engineering studies, I think the above possible failure modes should be given prompt attention and may even warrent a new revision of the codes.
Sorry, can't help it!!.
                     

hokie-
I've seen it spelled both ways....apparently by mistake.


dhengr-
Really never have a clue where you are coming from...not that I give a damn.
Your smart a$$ reply was actually somewhat insightful however.
I have hung more "jambs" then I have spec'd out "jam" nuts. (hanging the door jambs actually pays better still).
 

 

SAIL-
Thanks for the reply.
I honestly have never considered buckling of the anchor since it was embedded in grout.  
There has been some discussion on this forum from a someone who mentioned that the utility pole industry uses some criteria that goes something like "as long as the ungrouted bolt between the foundation and the base plate is no more than two diameters then no consideration need be..."

 

Sail3 - thank you.

The buckling of the AB should be considered.  But, the concrete embedment would limit the braced length to between the base plate and the top of foundations (like a piled foundation in soil).  Arguably, the grout will further reduce the braced length to 0.

The loss of concrete bond would be a concern under a load reversal situation.  Failure of the bond would occur under compression and under tension would be a reduced amount of resistance.  This should be addressed by excluding bond resistance for the tension design (use headed or undercut anchors).

Listen up!

The column is set on the shims or the leveling nuts, then next level or two of steel might be erected or not, but either way, when the column is plumb, the grout is placed, then erection of the steel continues, the floors are placed, the walls are built, and the other trades get into the structure. The column does not see the bulk of the load until after the grout is cured and ready to take it's share of the load. The grout continues to cure as the load increases.

It is because of this process that I have no worries about the questions asked here.

 

Michael.
Timing has a lot to do with the outcome of a rain dance.

Pad-
I may be repeating myself needlessly, but I have always had leveling plates set on the nuts and grouted prior to steel erection.  

Sorry Toad, I wasn't aiming at you. I was trying to point out that the shims or the leveling nuts only see a small part of load before the grout takes over.

I don't like leveling plates myself, the miss-match that could be caused by the tolerances on level and the out-of-squareness of the baseplate bothered me. One of those personal idiosyncrasies.

Michael.
Timing has a lot to do with the outcome of a rain dance.

I agree there can be problems there.
When done properly ahead of time, man, it can make steel erection go fast.
Our guys would usually set columns and tighten the nuts with a few soft strikes with a sledge and a wrench. After a level or two of steel was erected and roof trusses/beams/ bracing were in place, they plumb the structure, then tighten the nuts more.
I never noticed any baseplates that were not in good contact.
We did have a foundation contractor that was excellent.
He called me in panic one morning when an anchor bolt group was 1/2" off center on a 480' long industrial building.