Improved lathe toolpost grinder/drill

Stuff you made and how to make cool stuff

Improved lathe toolpost grinder/drill

Postby Doug Coulter » Thu Nov 13, 2014 2:43 pm

We're getting to a place where pretty serious accuracy is required. Imagine having to drill .009" holes for example, to make a fusor lens. And oh, they have to be a tight fit on something .009" in diameter or nothing works in the real world - it flies apart under normal electrostatic forces.

My old homebrew drill collet/shaft/bearings simply weren't up to this task. I made it at a time I didn't know what was going to be important, and it has served me well doing all manner of drilling, grinding glass and ceramics and so on - it still works, but I needed something a lot more perfect.

So I went with the high grade bearings as reccomended by Jerry, and a tool steel centerless ground (annealed at the ends) for the shaft, and made a new one that is a whole lot better. I had to make some slight design changes due to the high-class bearings inner race not extending past the outer, but no biggie. I bought a foot long piece, annealed at both ends so I could make 2 if required. I found out just how hard that stuff is. After 5 min in my metal cutting bandsaw, I'd managed to cut the sharpie mark in half, and not even scratch this rod. So I hand-annealed it in the area of the desired cut with a torch, then the saw would cut it fine. Wow...(we're talking top of line bi metal blade and so on here - I've never seen it fail to at least make a scratch on even alumina or quartz, much less metal).

So, first things first. I tapped the collet end about 1" long with a 1/4-20 die. I had issues, as even annealed this took serious torque, and the shaft part I held in the lathe chuck was the hard part of the rod - so it was hard to clamp well, nothing would "bite". But I managed. Moly-D is your friend (too). As I considered this one my "practice run" I then drilled a 1/8" hole about an inch deep too - with a brand new cobalt steel drill - its very first hole. Came out perfect, which itself was a little bit amazing.

I then turned and filed a taper on the very end - there has to be a way to smush down the collet we're making and the way I do that in my design is to make a tapered end try to be forced into a hole smaller than the biggest part of the taper but larger than the smallest part. Seems to work...Once I had my hole drilled and the taper cut. I could then proceed to cut the slit, for which I used a diamond saw blade with about 15 mils kerf, rather than the bandsaw I used last time...it was kind of fun bootstrapping, using the old toolpost tool to help make the new one. I used that same wheel to cut a groove to hold a C clip for longitudinal location. Turns out that due to these bearings having a more or less flat top, I also had to use my punch set to make a tiny 12 mil thick washer, 1/4" ID, 3/8" OD, to space the C clip so it wouldn't rub.
For the other end, I turned down the very end of the drive pulley to get the same effect.

Then I made the all important nut. I used 1/2" OD brass for that, since it won't seize on steel and despite the potential for off-center weight, seemed like a smarter move than using aluminum. I cut it a little over .7" long and faced off the slightly unflat ends on the lathe. I then drilled through with a #23 drill (after the usual carbide centering spot drill). That hits the taper on the end of the main shaft near the very end (it will wear in some over time, I'm sure). Then, all but about .15" were drilled with a #7 drill so I could tap it 1/4"20 to match the rest. Due to a little extra file work, I don't run out of threads before I can smush the taper in - I reduced the main shaft diameter a little further back than required so that would work without issue.

Then I mounted her up. OMG, I have 2 mils peak to peak runout. That won't do. A half an hour making light taps with a tiny hammer...now the error is down to .0003", or 3 tenths as machinists say. Now we're talking, and it's even repeatable with different test shafts chucked into the thing. I could go for perfection, but since all this bendy stuff has a tendency to relax over time (and possibly, heat), I'll leave it as is for now, but keep an eye on it.

I was then going to go for the real gold - one with a 1mm hole, for those teeny drils that have a 1mm shank, but cut .0087 holes (for my 9 mil WC grid rods). But I tried my plexiglass adapter - essentially just a 1" or so piece of plexiglass rod turned to fit a 1/8" hole, then center dilled .039, then slit about half an inch worth - it's an adapter. Guess what - TIR is the same, so I'm done! Yay!

I'm driving this all with an old servo motor whose gearhead broke awhile back, and a couple pulleys, one large on the motor shaft (no gears anymore) and one small on the drill shaft. I can get perhaps 30k rpm on the drill now...but I made the DC supply for the drive motor variable a long time ago, so I can go down to a few hundred RPM too.


So, here's some pix at various stages of this. I didn't document the "simple" stuff any machinist (at all) should already know.
Jig.JPG
Here is is being indicated to I can make it truly straight.
This is with the tiny drill/adapter in it. My first coarser tests were with a simple broken carbide 1/8" shank bit.
Detail.JPG
The gory details. Man, even the dust is big at this scale!


And for completeness, the old shaft it replaced, which is stil alright for many uses, so I'm not tossing it out. There are some flaws. The nut got ground off a little by something and it's off balance. I cut multiple C clip slots as sometimes the 1" sq block this is mounted in gets in the way of the work - not often needed, but nice when you do. As a result, the last big error is that the thing is way too long, and anything out at the end past the drive pulley adds vibration. So, more flexible (literally too - it's 12l14, not harder steel) but not as accurate or reliable for grid making.
OldShaft.JPG
The old shaft. I'll keep it in the lathe tools box for backup.


Now I can explore some different materials and sizes for grid making. We have some .018" kovar (about 40,000 pieces) that might work - who knows? I also found some .009" tungsten carbide rods I want to try instead of the now-standard .020" tungsten welding rods. The new ones should have less electron emission - WC does at low energies with electrons, anyway, and be relatively stiffer perhaps - they're pretty flexible held in the hand, but they return to straight very nicely and will roll down a sheet of glass held just off horizontal, no problems - these things are straight. I'm hoping that by drilling a .0003" too-small hole I can get the "crush fit" required to hold a grid together with these. I know that I'm going to have to make a jig to cut them off t a precise length with the diamond wheel without just shattering them. Heck, it's hard to even clamp something like that straight and tight! That's next I suppose.
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.
User avatar
Doug Coulter
 
Posts: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA

Re: Improved lathe toolpost grinder/drill

Postby Doug Coulter » Fri Nov 14, 2014 3:28 pm

Well. my cute little plexiglass adapter didn't make the grade (it broke), but the idea did. I made another, this time of A2 annealed steel rod (the only other stock I had in the right OD already). Works great. This one, I don't think will break. It was a son of a gun to make, and even with multiple pecks and pulling the drill to air-blast and re-lube, broke one of my two .039" drills (of course, just as the second hole broke through into the first one I drilled to make one long hole), and I had to drive the bit out with a little ground-down .040 tungsten TIG rod. And then polish the innards. Well, what's done is done, so I guess I don't need a spare 1mm drill anymore anyway. I can always use a perfectly straight super hard shank to indicate from if nothing else, but my busted tool bit drawer is getting to be a little too full. Breaks of the game, I suppose. ;)

Edit: While I was at it, I milled a little off the pulley between the setscews, which were steel, and 90 degrees apart, improving the dynamic balance a good bit. As the pulley was Al, they were making that side a little heavier. Can't hurt. Noticeably smoother now.
Posting as just me, not as the forum owner. Everything I say is "in my opinion" and YMMV -- which should go for everyone without saying.
User avatar
Doug Coulter
 
Posts: 3515
Joined: Wed Jul 14, 2010 7:05 pm
Location: Floyd county, VA, USA


Return to Machining and Fabrication

Who is online

Users browsing this forum: No registered users and 4 guests

cron