From: Doug Jackson <actionj@io.com>
Subject: Chinese 9" x 19" lathe (LONG & detailed)
Date: 24 Aug 1999
Newsgroups: rec.crafts.metalworking

    There have many threads on this NG about the various Chinese lathes,
asking about quality & things to check for if you've bought one.  I
thought I'd share my experiences with the Grizzly 9" x 19" that I bought
about a year and a half ago.
    I believe this is the same lathe as the Jet and Lincoln (as seen in
the Rex Supply catalog) and others, except that mine has a 39mm x 4mm
spindle thread while some others apparently have the 1-1/2" x 8 TPI
thread (which is much to be preferred, as this machine doesn't come with
gears to cut 4mm threads.  It does come with 8 TPI gearing, so with the
others you can thread new backplates & other spindle tools).

    When I first received it there were many problems: I almost sent it
back, but I'd already hacked apart the packing crate (I didn't have room
for it, since my landlord wouldn't appreciate it sitting out back) so I
decided to make the best of it.  I guess I got what I paid for, but with
a lot of work I've made it into a pretty good machine.

        Here's the *short* list of problems I had with this machine:

    The cross-slide leadscrew came seriously bent (the cross-slide feed
handle had been touching the side of the shipping crate).  I fixed this
with an arbor press.  The carriage leadscrew was also slightly bent.  I
haven't been able to fix this yet.  The bent leadscrew causes noise
inside the carriage gearbox and causes a very slight ripple pattern on
finish cuts because it pushes hard against the carriage on every
revolution.  I'm open to any suggestions on how to straighten this
without running the risk of breaking it; it seems to be quite hard &
therefore possibly quite brittle.

    Many of the nuts & bolts were of the "offset allen-head,
self-stripping" variety that seems to be so popular with the Chinese.
The bearings seemed to have been lubricated with dirty grease scraped
off the factory floor. Also lots of sharp burs perfect for slicing
fingers open.

     When I got the lathe I cleaned it up and spent a few hours checking
it out.  When everything seemed to turn fairly smoothly by hand I turned
it on.  After about 2 minutes of run-time it suddenly started making a
hideous loud noise in the gearbox.  I investigated and found that the
bearing on the idler pulley had seized (this is the "low speed" pulley
driven by a toothed belt from the motor pulley).  Fortunately there is a
spring-loaded ball-bearing clutch that attaches the tooth-belt pulley to
the bearing, and this prevented damage to the toothed belt or the motor
(the hideous noise was actually the sound of this clutch ratcheting).
When I took apart the bearing I found that it was steel-on-steel journal
bearing, with rough-turned surfaces and insufficient clearances.  I
fixed that, then took the rest of the lathe completely apart to look for
other such problems.  In the process I smoothed out all of the other
small steel-on-steel journal bearings in the headstock and carriage
gearboxes by chucking the journal shafts in a drill press and smoothing
them with a file and emery cloth.  I used a dremel tool with a small
polishing wheel for the bearing sleeves.  So far I have not had any more
trouble with these bearings.

    After that I ran the lathe for a few more minutes, then one of the
ball bearings that supports the quick-change gears seized (This is a
6202 ball bearing just below where the leadscrew comes out of the
headstock).  It had green paint in it.  It also seems to have been
hammered in place, as evidenced by punch marks on the races.   I
replaced it along with its opposite number on the same shaft.  I had to
make a special jig to press the bearings in properly.

        When I finally got the lathe in a condition where it would run
without eating itself, I found that it left a "ripple" pattern on any
pieces I turned (deeper than what the bent leadscrew still causes
anyway).  I checked the preload on the spindle bearings and that was
OK.    I made a jig to press the spindle bearings in & out and replaced
them with Timken roller bearings.  This solved the problem but cost
about $50 for the new bearings.

        At some point I got around to checking the parallelism of the
ways.  I had to shim the headstock to make the spinde axis parallel to
the ways, but fortunately the ways themselves were not detectably warped
(I checked this by turning a long 2" piece of stock and measuring it in
the middle and at each end; an "hourglass" shape would have indicated
warped ways.)  However, I found that the tailstock ways were not
parallel with the saddle ways; this became obvious when I adjusted the
side-to-side position of the tailstock to turn a 5" rod straight, then
found that for any other length rod it would cut a taper.  You can check
this quickly with a dial indicator mounted to the cross slide with its
tip held against the flat side of the tailstock.  With the tailstock
butted up against the back of the carriage, move the carriage along the
ways with the tailstock pressed down against the ways and also pressed
up against the carriage.  It shoud indicate no movement of the tailstock
in relation to the carriage.  Doing this test I found that at the height
of the spindle axis the tailstock moved about .003" for every 5 inches
of carriage travel.
        I don't have a surface grinder so I made a jig to attach a
Foredom handpiece to the cross-slide and used this with a 1" grinding
wheel to grind the tailstock ways parallel to the saddle ways.  Of
course this lowered the tailstock a bit.  But this was no problem
because I had already machined a few thousandths off the bottom of the
spindle housing to get the spindle parallel to the ways (I hate to use
shims when I can play with my milling machine instead!)   Now the
tailstock stays aligned with the toolpost within .0002" along the entire
length of travel of the carriage.

        The 3-jaw chuck had serious runout.  I found that the "ledge" on
the chuck's back plate was about .015" undersize in diameter compared to
the the back of the chuck where it mates.  To correct this I removed the
chuck from the back plate and clamped it onto a 1" rod centered in the
4-jaw (I turned a few thousandths off the 1" rod first so it was
concentric with the spindle along its entire length).  I turned the back
of the chuck concentric with the rod and also faced the back of the
chuck.  Then I turned a new ledge on the back plate (mounted normally in
the spindle) to match the new I.D of the back of the chuck.  Now the
chuck is good to about .0005 TIR with 1" stock and around .002 for most
other sizes.

        The MT3 spindle taper was out of parallel with the spindle axis
by about .005" at the spindle nose.  I ground a new taper on it, using
the Foredom handpiece mounted on the toolpost and a taper attachment I
built for the lathe.  The Foredom handpiece doesn't have the
super-high-precision bearings that good toolpost grinders have, but it
was adequate to grind a pretty decent finish on the taper.

        As it comes from the factory, the cover on the headstock gearbox
is latched only by an allen-head screw.   If you close the cover after
changing speeds and don't bother to grab a wrench to tigten the screw,
the cover will swing open by  itself when you turn the lathe on.   Bad
news for careless fingers.  I built a spring-loaded latch for it.

        The tailstock on this lathe is tightened in place along the ways
with a 14mm nut, which is very inconvenient.   You have to turn the
wrench about 3/4 of a turn to free up the tailstock, but you can only
turn the wrench less than 1/2 turn before you have to shift it to turn
it the rest of the way.  You have to do this twice every time you move
the tailstock.  This eats up a lot of  time if you are doing many
operations with the tailstock, so I ended up modifying the tailstock
with a lever/cam arrangement of my own design.

    The speed reducing belt/pulley arrangement uses a *very* thin
v-belt; it is only about 3/16" wide.  I'm sure they did this to minimize
vibrations, but it won't handle much tension => you can't get much
torque out of it.  At low speeds the belt slips very easily (well below
the 3/4 horsepower output of the motor).  This limits the machine to
very shallow cuts in large-diameter pieces, and makes threading very
slow.  I have a replacement belt that Grizzly was kind enough to send me
for free, so I'm going to try to put a heavier tensioner spring on it to
increase the belt tension and see if I can get some decent torque
without breaking the belt.  I've also heard that some people have
replaced the whole motor/belt/pulley arrangement with a variable-speed
treadmill motor.  If anyone knows of a Web site showing this I'd love to
see it.

     A few people in this group seem pretty satisfied with this lathe,
so I guess mine was just a particularly sour lemon.  As it is, I've
spent a LOT of time on this thing, but I had fun doing it and now I have
a halfway decent lathe that fits in my apartment living room (As you
might guess, I am single.  Gonna have to get a house with a separate
shop *before* proposing marriage :^)  Fortunately my landlord is a very
gracious retired Mechanical Engineer, and doesn't have a problem with
the steadily growing machine shop in my apartment!

    If you're thinking about buying one of these things, just expect to
treat it like a "kit," as if all they send you is the rough castings for
the parts.  Be prepared to spend a lot of time on it before making chips
with anything approaching precision or reliability.  If you take this
attitude you won't be disappointed with it, and you can have fun making
the improvements.  In the end the lathe will be able to do a lot more
than with a Unimat or a Sherline, and you'll spend less time on it than
building a Gingery lathe.