Re: [CAD_CAM_EDM_DRO] Sherline CNC retrofit

OK, Gang! Bill says go ahead, so...

Backlash and lead-screws: First, gawk at Photos in Album #17 (first Album on
p.2 of PHOTOS). I grant you these snapshots were NOT made with the intent of
showing details of the lead-screws, per-se, but these two items WERE 100%
home-brew. Now the lathe, as you can see, is controlled by a Tandy 1500HD
(1991!) laptop running MSDOS GWBASIC. NO mouse nor Windows nn! (See further
comments on electronics below!)

As I had built an earlier similar lathe in '84 using a long-discontinued CBM
PET (which also is still going, in use!), I already had a "feel" for BSA's
ACME lead-screws and their Turcite nuts. This is a fiber-glass-impregnated
PTFE plastic, so it has "controlled" friction and is sturdy. However! The
fiberglass in it has a peculiar property! For my use of this machine, most
of the "Z-travel" (that'd be "X-travel" in Cartesian-speak!) is within the
same 4", so that portion of the lead-screw where the Turcite preloaded nut
bears over that part of the screw is highly-polished, and the friction there
is NOTICEABLY less (when the screws' handwheels are turned by hand,
stepper-windings OFF). That is, the glass fibers impregnated in the Turcite
act as "potted abrasive", and have actually POLISHED the screws in that
portion!

These screws are 4 mm/turn ACME, 20 mm dia., stainless steel. Those Turcite
nuts (bright red plastic) are made in TWO halves, each of which have two "90°
wide" fingers which, when properly meshed as they are sequentially screwed
onto the screw-stock, "lock" one-another so that if either turns, the other
must follow; no angular "play" between them. Then, a rather heavy
coil-spring fits between these two halves, pressing them apart, so one half
presses on the "right side of the threads" and the other half presses against
the "left side of the threads". ONE half is also externally threaded so that
it screws into a black-anodized aluminum flange-ring which makes bolting the
whole assembly to the mechanism (in this case, the "saddle" of the lathe).

Now these screws and nuts are BSA products, and are fine stuff, if pricey.
But all considered, having these lead-screw parts made by a good job-shop
charging, say, $75-$150 per hour would cost at least twice as much, assuming
it would even be POSSIBLE to get the same quality/consistency.

The "friction load" of these nicely-hefty nuts on the 20 mm. dia. screw-stock
is much more than one would like to have on any kind of manually-operated
lathe's lead-screws, but the hefty BIPOLAR Size-34 steppers clearly don't
care at all! So, with no "play" (aka "backlash"), and steppers which manage
fine, this is a "slop-free" arrangement.

I built this lathe simultaneousely with a "twin" which got shipped to a
customer in Ohio (he supplied is own laptop) in the latter quarter of '99. I
had not yet heard of Gecko, and though everything is working perfectly on
both machines, I perhaps wish I had, as I would have saved considerable
$heckles using the G210 drives (were they available in '99?)! I used "black
boxes" by the stepper-manufacturer, Superior Electric. These "black boxes"
contain TWO bipolar-chopper drives AND a 120/220-240 VAC-to-DC power-supply
which have opto-isolator inputs, "All Windings OFF" inputs (to remove DC from
the motors for set-up, etc.), and a sturdy "important-looking" box. The only
"electronics" left for me to do was to build a simple "buffer-interface"
between the laptop's "printer port" and the opto-isolator inputs on the
stepper-drive box. I used a 74HCT541 Octal non-inverting buffer, a pair of
74HC4538 monostables to "set the pulse-width", and a ULN2003 Darlington
driver to "light up" said opto-isolators. OH, I had to make a tiny 5VDC
supply for that, as this laptop did not have a USB nor "game-port" to supply
the +5VDC.

The first two bits of the parallel-port are for "Z", the next two for "X" (on
a lathe, "left-right" is the Z axis, and "north/south" is the X axis. There
are more than one "Now, lessee..." stories as to how that came to be, and I
must admit I use standard Cartesian "X" for East-West, and "Y" for
North-South, but DON'T TELL ANYONE!. Then the next two bits are XOR'ed to
operate the electropneumatic (100 psi "shop-air") tail-stock valve, and the
final two bits are XOR'ed to operate the spindle-motor SSR's. Three-phase,
so three SSR's!

These Superior Electric "black boxes" are their series "SS2000---" They cost
me over $700 each. I am sure I could have built power-supplies for $20, and
a pair of G210's would have been less than $300. So, I could have saved over
$350 per lathe, had I only known of Mariss' fine product! But the SS2000
things work fine, so... One thing it took me a day or so to "get over": I
had built several stepper-motor things using old UNIpolar systems with big,
hot "L-R ballast resistors", so I had a subconscious "need" to think the DC
supplies had to be able to supply the "rated current" to the motors. That
is, if each of two windings wanted 4 amps, two motors, the DC-supply would
have to supply 16 amps. But with Bipolar Chopper Drive, if your supplies are
going to be six times the "motor voltage", they need supply only a fraction
of the CURRENT, as no power (that's Watts!) is wasted as HEAT in those
ballast resistors! So, for two Bipolar motors rated at 3.5 amp per winding,
you would need NOT 14 amps, total, but only about 3 amps, total! Really!
Blew me away when Mariss explained this to me in an e-mail about a season
ago!

Jan Rowland


[Non-text portions of this message have been removed]