I've seen that process, and similar, described on the HAMB. About the best you can hope for is to make it oval in a way that the kingpin won't rattle around. It's not a "fix". I have to believe there are good axles close by you for little money.
It takes a great deal of skill, experience and understanding to "shrink" the boss as you describe. There is a much better chance of stretching the boss instead, making it useless, and/or changing the temper of the metal. As said, axles are cheap, and most are good.
Well I got the spindles back from the machine shop, and they charged me $60 which I thought was fair considering he used a honing machine so they came out about as good as they get. I still have the hourglass shaped axle bosses to deal with. I read about heating the axle boss, then inserting an old but good pin in the hole then peening the upper and lower portions of the axle boss to effectively shrink it. Anybody done this?
I agree with AX . . . It sounds very dangerous. You stand a good change of embrittling the metal on a very critical part. As Ross said, just replace the axle.
Heat isn't necessary. You would have to get it really bright-red hot to change the metallurgy. The axle I posted was done cold. If it were done hot enough for the metal to work easily, the shape would be changed dramatically with one good shot, and not in a controllable direction.
I found a very nice axle at a swap meet. Nice tight holes (!)
Are these axles symmetrical, allowing installation either way, or is there a king pin inclination? Hard to tell by looking at it on the garage floor...
To check using the broomstick method I guess I could try to get it into a vise until the mounting pads are level. I was putting the pin in there and eye-balling it against a speed square but it looks very close to 0 degrees that way. I guess an inclinometer would be helpful. Is 2 degrees enough in a stock spring set up?
I replaced the pins and bushings in the springs and frame so now the springs don't move sideways anymore. Funny how this king pin job snowballed on me. Shoulda known...
Anyway, how much inclination is built in to the axle? One needs to know how to recognize that to determine which way of the beam faces forward? Or is there another way to tell?