annealing question

Just put it in the kitchen oven at about 500 degrees with a good steak for starters. I'll bet that bolt is something like 52100 or 4140. Hardness may not be the problem, though. It may be just that there is a real sharp edge right there that concentrates the stress or vibration at that one spot.
Pete Stanaitis -------------------
Grant Erw> I recently made a guillotine style (hinged) fuller which used a socket

Heat up to cherry red and bury it in a pile of powdered lime. Should take a few hours to cool down.
Powdered lime is available from garden stores.
I used to anneal knife blades that way before final heat treating.

Aw crap I didn't mean to ask how to anneal it I mean how to TEMPER it ..
Ernie Leimkuhler wrote:

Heat to cherry red. Quench in mineral oil. Polish up real clean. Bake in oven or wash gently with small propane torch until light blue color is reached. That is soft spring hard, and should be as hard as that bolt should be.

Sounds like a shear load failure. Go to a softer bolt, like a grade 5. Socket heads are usually grade 8 and are designed for high tensile loads. Go to a longer bolt, one with an unthreaded area for the bearing surface. Deburr the holes in the parts so there are no sharp stress concentrators. Use washers to spread the tensile loads to a larger area, grease the joint to cut the drag. Don't over-torque the bolt, then upset the thread to keep it from shaking loose. If this doesn't fix the problem, then redrill the holes for a larger diameter bolt and try that.
Charly

These all make sense. I did temper the replacement socket head shoulder bolt to a blue color. If this one fails then I will try your methods. The holes are already deburred and there aren't hardly any tensile forces. I can't upset the threads in this design because it is designed to be movable to any of several holes depending on what thickness I want to fuller to. But I can use a longer bolt and use 2 lock nuts to hold it in place. I don't know why I got the idea that socket head bolts were tough AND hard. They now seem just hard to me.
Grant Erwin

If the design is (I'm guessing here, based on the descripton and the failure) with the shoulder of the bolt in the hinge shear-plane (how to say without pictures - screwed through the moving part, screwed flush to the face of the stationary threaded part), that is a stress concentration that would tend to snap it. Better to bore through and use a pin of one diameter when it breaks again, or at least counterbore the stationary part so that the shoulder is not right in the shear plane. If quick-change is desirable, you can use a pin retained with cotter or lynch pins, or grooves and circlips, rather than a bolt, for faster changes.