Coping beams so they bolt up in the air, not on the phone

The beam comes off the saw at 26 ft 0 in, lands on two sawhorses, and the cope at the top flange is still a soapstone line with one punch mark in it. The drawing shows a W16x26 roof beam into a W21x44 girder on a school gym job. The cope detail reads 4 in long by 1 1/2 in deep, 1/2 in radius at the inside corner, 1/2 in clearance to the girder flange. Every one of those numbers gets checked here, on the horses, before the beam ships. The same check in the air costs a man basket, a grinder, and the first hour of an erection crew's morning.

What gets measured at the bench, and with what

The tools are a 25 ft tape, a 12 in combination square, a radius gauge set, and the erection drawing for this end at 3/4 in to the foot. Everything is taken off the cut end, so that face gets squared first. The cope length is taped from the end of the beam back to the heel of the cope, and the cope depth down from the top of the top flange.

The tape read 26 ft 0 1/16 in overall, against 26 ft 0 in on the cutting ticket. Cope length read 4 in at the north end and 3 15/16 in at the south end. Cope depth read 1 1/2 in at both ends, taken from the top of the top flange.

Does the cope clear the member it lands on?

The cope length has to clear whatever the beam lands against, which is the check that decides whether a beam goes up or comes back. Here it lands against the girder's top flange, 6 1/2 in wide, putting 3 1/4 in of flange on each side of the girder's web centerline. The drawing notes 1/2 in of clearance, so the smallest cope that works is 3 3/4 in. The detail calls 4 in. The tape found 4 in and 3 15/16 in, both over the minimum and inside the ±1/8 in this shop works to when drawings are silent.

The cope depth answers a second question. The top flange of a W16x26 is 0.345 in thick, and the cope cuts 1 1/2 in down, so the flange is gone and about 1.155 in of web goes with it.

The corner that starts the crack

Protocol here is the radius gauge, run into the inside corner of the cope at both ends. The detail called 1/2 in. The gauge read 1/2 in at all four corners, and a 3/8 in gauge would not seat in any of them.

The radius is not decoration. A square inside corner at a cope is a stress riser, and the AISC Specification's provisions for copes and cut edges exist for that reason; the detailing columns in Modern Steel Construction have made the same point for years. What the gauge does not check is the cut face. A flame cut edge with a nick behaves like a notch, so the same hand runs a gloved finger along the cut for a gouge that catches a fingernail. Two copes on the previous run were ground out for that.

Where does the kerf go, and who eats it?

Protocol: the cut was made with a hand plasma torch on 3/8 in material, soapstone line left visible until the torch passes it. Kerf on this setup ran 3/32 in. A track torch on the same thickness runs wider, closer to 3/16 in, and that number belongs in the note because it changes the offset.

The line is the scrap side. Run the torch down the line and the cut face lands half a kerf outside it, so the piece comes out small. On this beam the torch ran just outside the line at the north end and the cope taped 4 1/16 in instead of 4 in, inside the ±1/8 in, so it shipped.

How do you know the gage line on the angle is right?

The cope is half the joint. The other half is the angle that ties the beam's web to the girder, punched at the bench, so its gage gets taped too. Protocol: tape from the back of the angle to the center punch of the hole on the leg facing the beam's web, one reading per hole, on every assembly in the run.

These are L4x4x5/16 angles with 3/4 in bolts. The AISC Steel Construction Manual carries the standard gages for angle legs, and for this leg the gage is 2 1/4 in off the back. Three assemblies taped 2 1/4 in each. A fourth taped 2 3/16 in, inside the ±1/8 in, and it went through. Hole size was 13/16 in for a 3/4 in bolt. A gage off by 1/8 in at the bench is a bolt that will not start in the air.

What ships with the beam

Before the beam goes on the truck: cope length and depth taped at both ends, radius gauged at every inside corner, cut face checked by hand, overall length against the cutting ticket, piece mark painted on the web, cope confirmed on the correct end. A coped beam flipped end for end is a beam that rides back to the shop. The traveler carries the readings, the date, and the initials of whoever held the tape.

What this note does not cover

This note covers two copes on one beam, read with a tape, a square, and a radius gauge, against one drawing, on one day. It does not cover the connection design: whether those angles and bolts carry the load is the engineer of record's work, and the RCSC Specification governs the joint itself, hole size, bolt tension, and the faying surfaces. It does not cover the weld that attaches a shear tab to the girder's web, which belongs to the WPS, the procedure qualification, and the welder qualification under AWS D1.1. It does not cover the fit-up in the field, where a beam that taped clean here can still want a drift pin. It certifies no frame. The AISC Code of Standard Practice sorts out which party answers for what when a piece does not fit, and it reads better before the disagreement than after.

What to do with this on your own job

If you run a shop: put the tape on the first beam of a run while it is still on the horses, and write the cope dimensions on the traveler in the same hand as the reading. If you are the erector: take a tape to the first two beams off the truck and check the cope length before the crane lifts, because finding it on the ground costs a conversation and finding it in the air costs a basket. If you are the GC or the owner's representative: find the cope detail on the erection drawings before the first beam is cut. If that detail is missing, or the top of steel is flush and no cope is shown, that is an RFI to the engineer of record, not a field fix.