Grinding a machine taper with an angle grinder
The challenge is not so much getting it on, but getting it back off. A piece of wood, hit on the back with a hammer (see at left) will do the job.
I had some 16 mm polished steel rod left over from when I made the four-jaw chuck, so I figured I might as well use that to mount the drill chuck. But the problem is that this shaft isn't thick enough for the wide end of the taper.
Before welding, I put a wooden block on the shaft to protect it from splatter, but the splatter ended up going much further than the block, so I should have added more blocks.
The block ended up really stuck, so I had to split it off with a chisel.
The end of the shaft also had a flat spot on it. Because it's barely long enough, I had to use the flat spot as well, so I filled that also.
Looking at the sparks in the photo, they don't have the look of sparks from carbon steel. I suspect now the shaft is stainless steel. No wonder it welded so poorly.
Now I just need to grind that into a precision machine taper. I only put down a bead on each end of the taper. Matching the whole cone to the drill chuck's taper would be too difficult with an angle grinder!
I made a block to fit on the end of the shaft and mounted that in my four jaw lathe chuck. I also mounted a wooden bearing to hold the end near where I will work on it.
Once I could slide the chuck on most of the way, I checked for rocking. I checked with a dial indicator, moving it side-to-side to see where it stopped showing displacement as I rocked the chuck. That indicated which point on the taper it's rocking back and forth about, which tells me where the taper is getting hung up.
I also experimented with slipping a strip of paper in to see where it became compressed.
My goal was to have the taper only make contact near both ends. Getting the whole cone right would be too hard. But it appeared to always make contact at some point away from the deep end of the taper.
Eventually I examined the drill chuck's taper more closely. Only the first three quarters of it were actually tapered, and then it went to cylindrical! The inside-most part of the taper that I was trying to make contact with wasn't actually there! No wonder I kept screwing it up!
Despite my difficulty, and the horrible welding job, I think this technique has potential Given any similar problem, I would certainly try it again!
This technique could also be the start of a drill press build, just for the heck of it. But I'm not convinced building a drill press makes any sense, unless you already have a suitable motor and are looking for a challenge.
Trying it out, it took hardly any time at all to make one, and if I made many, I'm sure I'd get faster and very consistent. No wonder my dad never bothered even looking into buying pre-made knobs. He also had no interest in copy lathes, saying they were too slow.
But this chuck is not ideal for the job. The chuck tightens itself as torque is applied, so it tends to over-tighten on the wood. Also, this chuck is relatively long. A cheaper keyed chuck would be shorter so it would apply less leverage on the taper. Finally, a machine taper is not ideal for side loads, and it can come loose on its own. It came loose on me once.
Table saw fence micro adjusting