The HF mini lathe hasn't been used since I got the slightly bigger much
older, and much better HF lathe. I wanted to use it the other day though
because it has gear selectable forward and reverse carriage feed. The I
remembered I stripped those little nylon gears. I am sure I can get a hold
of HF customer service and get a part number for those gears if its not in
my manual for the machine, but I was wondering if somebody knew of a source
for metal replacements? I know Little Machine Shop was the threading gear
set in metal, but I didn't see the rest of the gears for the machine.
My next thought was maybe to try and make some metal gears, but I have never
done that before. Any suggestions on the approach for that? Best alloy for
I don't know in that particular case, but sometimes it's best to leave
the "weak link" weak, so it can protect all the other links, and buy a
few spares of it so you can get back running after you break it again.
Worth considering, anyway.
You'll need a dividing head, and involute gear cutter for the
diametrical pitch and pressure angle and the approximate number of
teeth in your gear. eBay is a good place to shop.
With these two tools, its a piece of cake, cut a tooth to depth,
rotate head, cut next tooth.
You might read "Gear and gear cutting" by Ivan Law to get a good grasp
of the subject. You'll need machinery handbook to calculate your
depths and blank and rotation values, etc.
To replace a plastic gear, I'd use AL. Otherwise, I've made gears out
of 4140 and then hardened. I've had one running in my backhoe for ten
This is a good basic skill to have. You're sure to be confused at
first, but then it will seem easy. Just like learning to ride a
I don't have a dividing head, but I did recently fit my little rotary table
with a lathe chuck, that is reasonable centered and parallel. Is that "good
enough", or do you think a head with dividing plates is a must.
Involute gear cutter? Is that one of those "wheel" type cutters I often see
in videos on a horizontal mill being used for gear cutting?
I could certainly cut aluminum. You wouldn't worry about galling in this
application? I don't really think of aluminum usually when I think of an
interacting mechanical part. As a static or linking part sure, but aluminum
to steel with mechanical interaction and friction?
LOL. On the mini lathe they would outlast the entire rest of the machine.
I have seen some gears cut with a slitting saw, but I wasn't super
comfortable with the process. What do you think? Probably go pretty quick
if I went with aluminum.
Its not at all a bad thought, and I probably won't destroy them all that
often anyway. I don't recall exactly what I did, but it seems I did
something stupid to destroy them in the first place. Hey. Better to learn
on a mini lathe than a real piece of machinery right?
That's a good principle, but replacement gears aren't necessarily easy
Better to have a sacificial key or taper pin I think. It could maybe be
incorporated into the new steel gears.
Someone used to joke about the $100 gear that stripped to protect the 20
cent key, and the wiring harness that burnt to protect the fuse.
The gears on the 6" Sears/AA lathe are white metal. On mine the ways
were seriously worn from use but the gears are still fine.
This aluminum gear has held up well to heavy loading:
ground the cutter bit with straight sides and rounded the teeth
afterwards with a file. Law's book gives a more accurate way to shape
This is my fixture to shape a cutter bit more closely to the tooth
gap, in this case a 30 degree involute spline on a hydraulic pump
shaft. The brass-tipped screw centers the pump shaft tooth space on
the slot the bit slides in.
I ground the bit freehand after marking the face beside the contact
areas with a fine-tipped felt pen. For the close fitting I pulled
strips of plastic-film sandpaper from the hobby store between the
splined shaft and the bit.
Bluing didn't stain the smooth surface very well. I think smoke from a
candle flame would work better.
LMS had ALL the gears plus extra pitches last I looked. And like the
man said, you need a weak link in the chain, otherwise things could
get more expensive than just a stripped gear. They are standard
metric module gears, guys were using ones scavanged from laser
printers and the like for oddball threads.
You must not have looked very hard:
They also have a spare parts kit with some of the other gears, a belt
and some fuses.
Yes -- even easier with a horizontal spindle mill with lever
feed, but not bad with a vertical spindle mill.
However -- the first trick (aside from getting a dividing head)
is finding the proper gear tooth cutter for *your* gear. Given that it
is an HF machine, the gear will probably be metric, which here in the US
makes both the proper gear tooth cutters, *and* the tools for measuring
the gear pitch. IIRC, the pitch is specified in "module" for metric
gears, while in either diametrical or circular pitch for inch gears.
Of course, it is possible to grind special lathe bits to the
shape needed, and mount them in a rotating shaft to take the place of
the gear tooth cutter -- but it will be slower since it has only one
tooth, while the cutters probably have eight or ten teeth.
That sounds like a good choice -- with one caveat: *If* it is going to mesh with another aluminum gear, you will
have galling (transfer of metal from one gear to the other) at the
pressure points. You want unlike metals (e.g. stainless steel meshing
with aluminum, or brass, or whatever. Stainless and Aluminum are
probably the worst for galling.
So far, I've made one gear (and planned others). It was a brass
gear to replace a broken plastic one, and I've previously posted the URL
documenting that project, but if you are still interested, here it is:
[ ... ]
You don't want it *too* strong, or some other part of the
machine will die when the load gets too high.
Before I answer that (assuming that your rotary table does not
have dividing plates), I have to ask *you* -- how many teeth should the
If is is something like 30 teeth, no problem. Most rotary tables
have a 90:1 ratio, so three full turns for each tooth and you are fine.
However, if it were say 28 teeth, you need 3.2143 turns per
tooth. that is 4 degrees for the full turn, and 0.8572 degrees left
over. That is 51' 26" added per tooth. Even if you use a spreadsheet
or other program to print out a chart -- how long do you think that you
could go before you made a mistake? Now, if you have a dividing plate
with 14 (or some integer multiple like 28) holes, that works out right.
Set the arms on the plate to allow an extra motion of 3 (for 14), or 6
(for 28) holes. Then reposition the arms after that, so the next
division advances the same number of holes, and keep going until you
reach the starting point again. (And be careful to remember to turn
three full turns before going that extra three or six holes, or you
will still be out of line.) *This* is why people use dividing heads for making gears.
"Involute" is a term for the math defined shape of the gear
tooth, and an involute gear tooth cutter is one which will form close to
the precise shape needed. These shapes are so the teeth roll against
each other, instead of sliding and wearing.
No friction with a proper involute shape -- just rolling
friction. But this is for *one* gear made of aluminum meshing with the
other original plastic gears. If you have to make two or more gears
which must mesh with each other, you want dissimilar metals (unless you
are using well lubricated steel or cast iron gears) to avoid galling.
Aluminum on steel? Fine! Aluminum on brass? Fine. Brass on steel?
Fine. Aluminum on Aluminum? *No*!
Almost anything will outlast the rest of the machine. Those
plastic gears are made to be the "fuse" -- the weak point which breaks
before other more expensive parts break.
A slitting saw? *No*! Not for power transmitting gears. Now,
if you were making a really large gear (say 1" between teeth or larger),
and had the shape laid out with layout die on the flat, you could use a
vertical bandsaw to cut a rough approximation, leaving a little extra
metal, and then using a file to take to to final shape. This is the way
old machinists made large gears. But your gears are not the right size
at all for this.
A slitting saw *might* work for the gears used in desktop
clocks, which don't handle much torque, and which have the flat gear
meshing with a cage of rods one of which goes into a slot at a time.
And even with this, you would probably have to file the ends of the
teeth in the flat to allow the rods to enter properly.
Now -- an involute gear tooth cutter looks a lot like s slitting
saw -- until you examine the shape of the teeth.
Oh yes -- each involute gear tooth cutter has a different shape,
precisely appropriate for a specific number of teeth, and close enough
for a number around that value. The larger the number of teeth, the
wider the range that is close enough. One end of the set is capable of
cutting any tooth count from 135 teeth to a rack gear (infinite number
of teeth). However, when the number of teeth gets small, the range is
also small. The cutters are marked #1 through #8. If you have to cut
two gears of different numbers of teeth, the odds are that you will need
two cutters unless you are quite lucky. These different cutters make up
for the different angles at which the gear teeth mesh.
Well, I probably wouldn't do it quite that way. I'ld do a quick spread
sheet. Cell 1 would have the tooth count. Cell 2 would have the formula
360/T with T being the number of teeth. Every cell after would have the
formula ((360/T)+PC) with PC being the value in the previous cell. Instead
of trying to keep track of the number of turns etc, I would just advance the
table rotary table to the value given in the numeric table generated by the
spreadsheet and tighten the lock screw. Obviously I would only turn the
table in one direction so as to account for backlash in the screw. I
could do it on paper almost as easily. It would just take a little longer.
13 Tooth Count
13 is only selected for example purposes. You could apply this for any
number of teeth. I would have no cumulative error. Just whatever level of
accuracy I was capable of with the table.
Well, I guess I need to look a little further than that link as well. It
doesn't include the 20 tooth reverse gear or 25 tooth forward gear. Neither
does the spare parts kit found here:
This set does, but it is the only one I could find that does.
Its not a bad option if I used the mill more. 2 yards is a bit steep for
basically one gear though.
Ah... here we go:
Thanks Stanley. I was all prepared to be pissy about your comment, but it
lead to a cheap solution without setting up to cut gears. Thank you. I may
cut some gears anyway just for the experience, but...
When I mounted the chuck on the table it was my intention to convert it to
CNC eventually. Both of my little machine controllers have an unused 4th
port. Its would just be a matter of making a mount and adding a motor. The
problem I see is that the drive screw for this rotary table has too much
back lash in it. I have not looked yet to see if there is anything I can do
about it, but I was thinking it might be easier to just make a spindle for a
4th axis. I was thinking I would like something with a much bigger bore
than the average lathe spindle or rotary table. If I keep it under 3000 RPM
there are some pretty big bore bearings out there.