Re: Z-axis CNC conversion on a mill-drill: a challenge (long)

Hello,

Jeff's design is excellent and his plans are worth the cost. BUT...

I wouldn't choose either of these over the method Dan invented and
which I have also used a few times. Here's a big part of the reason:

Have another look at that side view picture you just posted. Now
imagine that the drill chuck shown in the pic is no longer there cuz
it gives a false sense of things. Next, imagine a SHORT endmill
holder (and perhaps the endmill too is short). We know that anything
we do to reduce the overhang/length of a cutter past the lower
spindle bearing is a VERY good thing. Basic Good machining practice
demands this attention to reducing leverages wherever possible.

Okay, with that in mind, and looking at the side view, do you see at
least a couple problems?

One: a short cutter--means stiff setup--will be inhibited in cutting
multi depths because of that HUGE "link block" from the screw to the
quill. You may have even more problems when you factor in the
already limited 5" travel typical of most Mill/drill quills.

And that link block itself has to be so huge because of
the "overhang"/cantilever of the drive point to cutter point.
Another "lever" that is better left off, IMO.

Two: the quill itself now has its effective diameter/size much
increased at the point closest to the work. Most tool designers will
attempt to MINIMISE the machine's potential interference with the
cutter; this design does the opposite.

Three: (not based on the pic) There have been literally hundreds of
machines converted using the worm and quill drive. Some with the
preload up springs you mention. Some with another means to control
backlash. Some that just "deal with" the .020 or so. Point is, these
systems using the "normal quill path" DO work.

Adding Dans mod only makes that situation even better. The downward
force is ON the quill, so no "excess" leverage. No additional nose
increase in size. Great speed and force available. SIMPLE mod and
motor mounting.

Now let's address your concern that Dan's mod is "hard". The
only "no return" portion is the reduction of the quill pinion shaft
to accomodate the added eccentrics. SOooo... You have this done at a
jobshop (probably for less than the cost of plans for another
method) if you know how to talk to the guys doing the work and
perhaps you buy them some afternoon libation... (I've been truly
amazed over the years at the value of this after-hours offering
of "goodwill" and its effect on quoted price...)

Barring that; you can do the "turning" of this shaft WITH your mill
drill. Set the shaft up vee-blocks, and mill, then Rotate and re-
clamp. Repeat. Tedious, but simple and effective. Hint: Convert the
XY first and both this and the next step's made easier...

But... The eccentrics can be made manually using a boring head in
the mill too. You don't have to have CNC to do XY "arc" work!
Coordinates is coordinates, and SIN/COS work whether there's a
computer doing the handwheel turning or the wetware of your brain<G>

BTW, the cost of that shaft you're modifying is pretty inexpensive
as a replacement part from Grizzly. So buy one, and you will STILL
have the ability to get back to your old setup if it doesn't work
out... (This extra shaft is again probably less than the cost of a
competing set of plans.)

One last point: If you do decide to use a link-bar setup--and you
don't use Jeff's excellent design... I'll suggest that you mount the
motor at the FRONT of the machine power head, instead of to the
rear. Why? shorter link length, easier to be aware of and clear
something in the front where you can see it than in the back where
you can't. Look again at the side view and imagine all the kinds of
work you may be doing on that machine. If these aren't all basically
flat things, IMO you're looking at adding some serious limitations
to an already limited tool...

Hope this helps. YMMV,

Ballendo

P.S. I was once able to reduce the Z backlash by simply rotating the
QUILL's relationship to its pinion by using a modification to the
dog point setscrew that fits into the "don't rotate" Quill keyway.
NO, it's NOT the best way--no way near as good as Dan's method. But
it is/was simple cheap and fast as a way to reduce z axis backlash
on a mill/drill. For awhile, until something better is done.

>In CCED, "arquibaldo2001" <nmachado@...> wrote:
>Thank you for your contributions, R Wink & Dan Mauch. By the
>way, I am aware of Dan Mauch's brilliant eccentric-bushing-
>modification, but I think only an experienced machinist should try
>to implement it.
>Let me explain a little: The reason I decided against working on
>the rack/pinion backlash itself is twofold:
>
>First, it involves changes deep inside the machine, and if I
>screw-up bad enough (I am inexperienced) I end up with nothing,
>while WT's modification is external; if things go wrong, I still
>have my old machine and can try again...
>
>Second, even if I managed to reduce the rack/pinion backlash, one
>problem still remains: where to power the Z axis: if directly on
>the pinion, some other reduction would have to be implemented since
>the pinion movement is too coarse (80mm per turn instead of 2.5mm
>per turn for the X & Y); if on the worm shaft (part 347 in Grizzly's
>model G1006/G1007 parts list) we have another link to worry about
>(worm shaft-to-pinion) and this one involves friction and the risk
>of slippage (although I don't really understand how this link
>works...)
>
>Please refer to the two photos in message #90489 to see WT's design:
>http://i14.tinypic.com/344unnm.jpg
>http://i14.tinypic.com/2m5ge90.jpg
>
>With respect to Jeff Davis's plans, I followed the link but couldn't
>find any details about his Z axis design. If anyone knows the design
>and can confirm that it is similar to WT's one, I will gladly part
>with the $40 to find out how to implemet a rotating ballnut.
>Nelson