On Monday 18 April 2005 02:35 pm, Alan King wrote:
> LOL, I think it's more that 95% of people don't have a good clue what
> they're doing. It takes extremely accurate axial alignment of the drill to
> the motion to use small bits. Most people just don't even see how poorly
> aligned their systems are. If the angle alignment isn't exactly parallel
> with the motion you're going to snap bits very often.
>
> #80 bit in the drill. Hold a 1/64ths ruler behind the tip and move the
> bit up and down making sure the shaft of the drill moves exactly in the
> same line. Not just to the 64th but over the same part of the line of the
> ruler. Then do the same 90 degrees around. That will get it close. Then
> drill a hole, and see which way the bit tends to bow. Adjust until you've
> gotten the bow to the minimum you can and you're done. Then it's aligned
> accurately, and most bits will last. I get plenty of holes per bit even
> with very tiny bits, and that's with a cheap dremel knock off or flex tool
> handle. To tell you the truth I didn't even realize how crap my alignment
> was myself with the .039 etc bits until I started trying the #80s.
This is interesting stuff, and I even have bits as small as #80 here to try
that with...
> If the angle isn't exact, and the drill makes a hole but by the time the
> drill is all of the way in, if it's even 1/10th of the bit diameter over
> you're putting a heck of a strain on the bit for every revolution and they
> will not last long at all.
Hm.
> Another thing is most people don't realize how little cutting a small
> diameter bit does at say 10,000 RPM, and they feed too fast. Bite off too
> much and it's going to snap in an instant. 20 or 30 K RPM isn't even too
> fast for a small bit with many materials, and you still can feed too fast.
> And your hand generally isn't accurate enough for feed on very tiny bits,
> small jerks in feeding on a drill press means you'll break many more small
> bits by hand than on an automated system.
So, would feeding a z-axis with a stepper be a real problem, then? If I'm
using 1/4-20 then one whole rotation is 1/20th of an inch, and one step is
1/200th of that, so maybe not. :-)
> Trivial to make enough automation to do say an IC tube with a line of
> holes too. Go to Lowes and get 2 of the KV drawer slides. Mount a
> platform between two and put the rails on a base. That's your X to move
> the tubes. Then another board, and put on two door hinges. Two 4' pieces
> of wood on the door hinges, coming out to a vertex. Basically a frame to
> get to 4' over from the hinges, far enough away and circular motion is
> basically linear. Shift the alignment of the hinges until there is very
> little play by having them not exactly in line, almost binding even. Then
> mount your drill at the point 4' away from the hinges. That far away the
> motion is close enough to linear, at least for a thin drilled item like an
> IC tube.. Oh yeah they are three rail slides, take off the outer part and
> only use the inner and middle rails.
>
> One stepper motor and 1/4"-20 threaded rod to move each, use 1" long
> connecting nuts for the motion they have less play.
I'm planning on building my own small mill, and haven't gotten very far with
it, all sorts of other things keep coming along to distract me in the
process. I have thought that 1/4-20 rod was a good way to go with it, but
that's about as far as I've gotten. No idea what I'm going to use for
couplers, or whether I'm going to need a bunch of bearings to keep axial
load off the steppers, or how I'm going to rig things for those nuts. Where
did you find those connecting nuts you nention, for exampe? How are they
mounted to the moving bits?