Unusual embed plate

[dozer]

We're thinking of trying something a little goofy and I was wondering if anybody has an opinion. We're going to install some steel embed plates in a large concrete beam to support a large piece of equipment. There are several plates and we want them to be at the same elevation. The beam deflects about 1" at the center.

Anchor bolt (threaded rods) will be cast into the concrete, then after the beam is set in place, leveling nuts will be threaded on to the rods and plates with holes in the appropriate places will be slid over the rods. The plates will all be leveled.

Okay, so far nothing weird, but here's the rub. Ideally we would like the top of the plates to be smooth, no anchors with nuts sticking through. So here's the idea (ok, ok, it was my idea). The holes will be beveled (think funnel). The annular space around the anchors will be welded. The anchor cut flush with the top of the plate and ground smooth. Other than a heck of a lot more labor than cranking some nuts down to hold the plate, does anyone see a problem.

One that's been suggested to me is that the concrete might act like a giant heat sink and cause some problems with weld quality. I'm not sure about that. It seems to me that the heat transfer would be limited by the anchor which will be holding the plate about 2" off of the concrete.

Any thoughts would be appreciated.

 

[UcfSE]

Can you install another plate on top of the first with bigger holes to cover the nut within the thickness of the 2nd plate as though the nut were countersunk and then weld the two plates together? Extra plate material and new bearing height to coordinate, but it might be cheaper than fancy details.

 

[aggman]

dozer,
While not exactly doing your application I have done plenty of embed plates with anchors as they are very common in the mineral processing industry. What we typically do is drill holes in the embed, and let the anchor stick 1/2 way through the plate. We then run a fillet weld around the anchor rod and plug weld the hole and grind it down smooth. Once painted, it leaves a nice, clean finish. I do caution you to check the weld capacity very close and think about what forces will be applied to the weld due to shear loads. The anchors will act like a sway column fixed at the top, applying a moment to the weld connection on each anchor. Plus the weld quality will likely be less than what you expect so be cautious. In dynamic applications, these types of connections fail in fatigue all the time, but seem to do well in static applications. We typically weld the anchor to the bottom of the plate and then set the plate at the proper elevation, but I think your idea will work.

 

[JAE]

dozer,

How much uplift or other loading is applied to these plates?

One idea would be to cast sleeves into the top of the beam using PVC or even a small pipe. Cast the beam and then have embed plates with all-threads welded to the bottom sides. Insert an adhesive resin into the sleeves and drop each plate onto the top with the all-threads penetrating the adhesive-filled sleeves. The adhesive may squirt out a bit unless you control the quantity placed in the sleeves.

You could use leveling nuts underneath to set the levels - or you could survey - shoot the top of the concrete beam and develop an exact stand-off for each plate to create a level condition - sort of like shooting shim shots on bridge girders to set the underside elevation of a bridge deck.

 

[apsix]

What about casting in threaded inserts and using countersunk bolts to fix the plate?
As for the concrete acting as a heatsink when welding, it wouldn't be worse than when welding to a large steel beam. Steel has about 50 x the thermal conductivity of concrete.

 

[jheidt2543]

Two suggestions:

1. The precast industry uses all types of embedded plates for various kinds of connections and anchorage. Their PCI Design manual has a lot of typical details that you could reference regarding your application.

2. Dayton-Superior is a company that makes embedded inserts and plates for the CIP and precast concrete industry. Try their website for a look at their product line and data. They may have something that could work so you don't have to reinvent it.

Just a thought!

 

[DRC1]

What I have seen on embed plate is shear conectors welded to the plate and the pate cast into the beam. Properly designed, it gives you a very good connection.

 

[Ron]

dozer...what you are proposing will work. We have used similar over the years. The concrete will not be an issue with the welding, though you certainly can affect the concrete with the welding heat...might cause localized spalling. You might want to create a small annular "dip" in the concrete using a styrofoam or rubber washer.

As JAE said, be concerned about the uplift loads. The weld joint here is pre-qualified here, but make sure the welder understands the need for appropriate fusion, penetration, and effective thickness of the weld. I assume you will analyze these as partial penetration welds.

Having said all that, JAE's process is much easier and will nicely.

 

[dozer]

Thanks for all the input everyone. This is still a few months away from reality. If anything curious turns up I'll let you know. In the mean time if anybody else has any ideas, keep 'em coming. There is a fairly significant uplift but rest assured this will be checked carefully.

 

[henerythe8th]

The "localized spalling" that Ron is speaking of can be dangerous in some circumstances. The spalling can occur due to the aggregate in the concrete exploding due to rapid expansion from the weld heat. I saw it happen years ago in high school shop class when the teacher made a rather large weld on a piece of plate that was in direct contact with the slab floor. It actually made an exploding sound that couldn't be explained until the plate was moved and a substantial (about 4 inch diameter X 1 inch deep)hole was found!