Gear Cutting has been a dark art that I knew little of until recently.

The restoration of a Myford ML7 has plunged me into this art in a big
way. When I bought the project I knew that all the headstock gears
were stuffed. Just looking at them didnt intimidate me.

I'm starting to wonder whether the mystery sets the prices of gears.
It seems to me that if you were tooled up to make them they wouldnt be
that difficult.

Myford's still sell the gears that I need to restore the lathe so I
priced the gears I need. Amazing really, the a lathe made half a
century ago still has factory spares available.
What I needed was basically everything from the headstock to the
first gear on the banjo. They all fit in a very small pile in the palm
of one hand. By the time the parts get through the distributor to me
that little pile in the hand has a cost of over $aus750.

When I first read the price quote sent to me I thought "you're joking"
the other words I said to myself cant be written here by civilised
people.
an entire set of involute gear cutters isnt much more than that!

I bought and read up on gear cutting. Ivan Law's Book "Gears and Gear
Cutting" from the model engineer people made it seem so easy.
So I thought that in secret I'd have a bash and see if actually doing
it would teach me along the way. I made a gear in 2024 aluminium just
by bolting the blank beside the existing gear and guessing the cut
positions. the gear worked but I had some fat and some thin teeth.
I just positioned the cutter on edge and used the tool post like a
shaper. with 3 thou cuts the load on the lathe wasnt extreme.

I realised from this experiment that accurate positioning of the cut
was the deciding factor in a usable gear. this meant that some
mechanical detent was needed to position the blank.

Today I had another go starting from scratch

Involute cutter shapes sound esoteric but really they are almost a
circular arc on each face. trouble is that they are the wrong way
around to make a tool for.
The long and short of it is that I picked the least worn gear for use
as a template and freehand ground a HSS tool blank until I was happy
that the shape was close. In Ivans book there is a formula for working
out the diameter of the curve but I wasnt sure of the numbers I
needed. While I was mulling this over I spotted a washer on the floor
of the workshop, a little electrical one. Held up to the light the
washer proved to be the size I needed, it was an almost exact copy of
the involute shape of the gear I was copying. that made the freehand
grinding quite easy. I ground in the curve then tested it against the
washer then looked at the symmetry of the tool. It probably took ten
minutes of tweaking until I was happy with the cutter shape.

This time around I decided to work in cast iron (meehanite) so that if
I was successful I'd have one of the little stack of gears I needed.

I turned up an arbor to hold the existing gear, then a spacer, then
the cast iron blank. I made the blank two gears wide. Following Ivans
suggestion in the book I arranged a simple mechanical detent by using
a tom of scrap steel standing on the lathe bed. this could engage one
of the teeth of the master. I have a aching left hand from holding the
chuck into engagement while I planed out the cuts.

Using the tool on its side in the tool post, shimmed to centre height,
and moving the saddle back and forth to plane out the between teeth
gap is fairly slow work. but it does work.
late this afternoon I parted the finished gear into two and cleaned up
the faces. my gears mesh together smoothly and they mesh with a
smaller and a larger gear without problem.
by jove I think I've done it!
what is it? ten gears to go I think.

After tea I went out and searched through my junk for all the gears I
could find. I needed a 16 tooth gear or a multiple of that. I managed
to locate a 32 tooth gear out of an old photocopier and used every
second tooth for indexing.
One of the guys I work with now has the little 18mm dia 16 tooth gear
he needs to make the thread follower for his Myford Super 7 that he
has wanted for absolutely ages.

Gear cutting, approached the model engineer way, is a piece of cake.

one thing I did learn is that using cutting oil on cast iron almost
halves the effort to plane the tool across the cut. first time I have
ever seen cutting oil make that much difference in the cutting.

Stealth Pilot

On Sun, 11 Nov 2007 23:28:16 +0900, Stealth Pilot
wrote:

Gear Cutting has been a dark art that I knew little of until recently.

The restoration of a Myford ML7 has plunged me into this art in a big
way. When I bought the project I knew that all the headstock gears
were stuffed. Just looking at them didnt intimidate me.

I'm starting to wonder whether the mystery sets the prices of gears.
It seems to me that if you were tooled up to make them they wouldnt be
that difficult.
snip
============
Thanks for the update and real world experience.


Unka' George [George McDuffee]
============
Merchants have no country.
The mere spot they stand on
does not constitute so strong an attachment
as that from which they draw their gains.

Thomas Jefferson (1743-1826),
U.S. president. Letter, 17 March 1814.

Stealth Pilot wrote:
Gear Cutting has been a dark art that I knew little of until recently.

The restoration of a Myford ML7 has plunged me into this art in a big
way. When I bought the project I knew that all the headstock gears
were stuffed. Just looking at them didnt intimidate me.

I'm starting to wonder whether the mystery sets the prices of gears.
It seems to me that if you were tooled up to make them they wouldnt be
that difficult.

Myford's still sell the gears that I need to restore the lathe so I
priced the gears I need. Amazing really, the a lathe made half a
century ago still has factory spares available.
What I needed was basically everything from the headstock to the
first gear on the banjo. They all fit in a very small pile in the palm
of one hand. By the time the parts get through the distributor to me
that little pile in the hand has a cost of over $aus750.

When I first read the price quote sent to me I thought "you're joking"
the other words I said to myself cant be written here by civilised
people.
an entire set of involute gear cutters isnt much more than that!

I bought and read up on gear cutting. Ivan Law's Book "Gears and Gear
Cutting" from the model engineer people made it seem so easy.
So I thought that in secret I'd have a bash and see if actually doing
it would teach me along the way. I made a gear in 2024 aluminium just
by bolting the blank beside the existing gear and guessing the cut
positions. the gear worked but I had some fat and some thin teeth.
I just positioned the cutter on edge and used the tool post like a
shaper. with 3 thou cuts the load on the lathe wasnt extreme.

I realised from this experiment that accurate positioning of the cut
was the deciding factor in a usable gear. this meant that some
mechanical detent was needed to position the blank.

Today I had another go starting from scratch

Involute cutter shapes sound esoteric but really they are almost a
circular arc on each face. trouble is that they are the wrong way
around to make a tool for.
The long and short of it is that I picked the least worn gear for use
as a template and freehand ground a HSS tool blank until I was happy
that the shape was close. In Ivans book there is a formula for working
out the diameter of the curve but I wasnt sure of the numbers I
needed. While I was mulling this over I spotted a washer on the floor
of the workshop, a little electrical one. Held up to the light the
washer proved to be the size I needed, it was an almost exact copy of
the involute shape of the gear I was copying. that made the freehand
grinding quite easy. I ground in the curve then tested it against the
washer then looked at the symmetry of the tool. It probably took ten
minutes of tweaking until I was happy with the cutter shape.

This time around I decided to work in cast iron (meehanite) so that if
I was successful I'd have one of the little stack of gears I needed.

I turned up an arbor to hold the existing gear, then a spacer, then
the cast iron blank. I made the blank two gears wide. Following Ivans
suggestion in the book I arranged a simple mechanical detent by using
a tom of scrap steel standing on the lathe bed. this could engage one
of the teeth of the master. I have a aching left hand from holding the
chuck into engagement while I planed out the cuts.

Using the tool on its side in the tool post, shimmed to centre height,
and moving the saddle back and forth to plane out the between teeth
gap is fairly slow work. but it does work.
late this afternoon I parted the finished gear into two and cleaned up
the faces. my gears mesh together smoothly and they mesh with a
smaller and a larger gear without problem.
by jove I think I've done it!
what is it? ten gears to go I think.

After tea I went out and searched through my junk for all the gears I
could find. I needed a 16 tooth gear or a multiple of that. I managed
to locate a 32 tooth gear out of an old photocopier and used every
second tooth for indexing.
One of the guys I work with now has the little 18mm dia 16 tooth gear
he needs to make the thread follower for his Myford Super 7 that he
has wanted for absolutely ages.

Gear cutting, approached the model engineer way, is a piece of cake.

one thing I did learn is that using cutting oil on cast iron almost
halves the effort to plane the tool across the cut. first time I have
ever seen cutting oil make that much difference in the cutting.

Stealth Pilot



You have discovered a secret that keeps SO many guys from trying this
stuff.

Good enough, is bloody well good enough!

Function counts. More than having a perfect, precise, just right
measurement. If it don't work, it ain't worth a pinch of... :-)

Gears are really quite simple to make. On a commercial scale, once you
do the setup, you then bang out as many as you think you will need, then
setup for a different size. Setup is the eater of time, and thus the
cost driver.

Of course, having a hobber and the correct hobs, goes a long ways, and
those cost someone, eh!

Keep pounding out imperfect, useable gears, and at some point you will
probably go back and redo some, or many of them. In the meantime, you
have gears, that you didn't have before, even if imperfect.

Watch that you do not get so out of whak that your homecut gears do a
number on the ones that are/were with the lathe already. If you are
making the whole set, though.... :-)

Congrats, man! Keep on pluggin' and rest easy, knowing that getting
started was the hard part.

Cheers
Trevor Jones

Stealth Pilot wrote:
Gear Cutting has been a dark art that I knew little of until recently.
Cutting high-speed, high load gears, like an automobile
transaxle, require a lot of precision, and some of the gear
forms are insanely complicated, like hypoid gears.

On the other hand, moderate-speed gears with modest loads, like
a lathe's leadscrew drive, is another kind of gearing entirely.
Very light gears with significant errors will still work
remarkably well. A lathe's back gears or headstock gear train
is a little more difficult, but should still be able to be made
in a small shop.

Low-performance gears used to be cut in small shops for all sort
of jobs, even making simple gears to make a gear train for
cutting more complex gears such as helical gearing. (On a
universal mill, you provided a gear train to couple the dividing
head to the X axis leadscrew to form the helix angle. Every
different helix angle required a different gear ratio.)

I haven't heard of small shops making gears as a routine job
anymore. But, the same tools still can still do this job.

Jon

"Stealth Pilot" wrote in message
...

Gear Cutting has been a dark art that I knew little of until recently.

The restoration of a Myford ML7 has plunged me into this art in a big
way. When I bought the project I knew that all the headstock gears
were stuffed. Just looking at them didnt intimidate me.

I'm starting to wonder whether the mystery sets the prices of gears.
It seems to me that if you were tooled up to make them they wouldnt be
that difficult.

Myford's still sell the gears that I need to restore the lathe so I
priced the gears I need. Amazing really, the a lathe made half a
century ago still has factory spares available.
What I needed was basically everything from the headstock to the
first gear on the banjo. They all fit in a very small pile in the palm
of one hand. By the time the parts get through the distributor to me
that little pile in the hand has a cost of over $aus750.

When I first read the price quote sent to me I thought "you're joking"
the other words I said to myself cant be written here by civilised
people.
an entire set of involute gear cutters isnt much more than that!

I bought and read up on gear cutting. Ivan Law's Book "Gears and Gear
Cutting" from the model engineer people made it seem so easy.
So I thought that in secret I'd have a bash and see if actually doing
it would teach me along the way. I made a gear in 2024 aluminium just
by bolting the blank beside the existing gear and guessing the cut
positions. the gear worked but I had some fat and some thin teeth.
I just positioned the cutter on edge and used the tool post like a
shaper. with 3 thou cuts the load on the lathe wasnt extreme.

I realised from this experiment that accurate positioning of the cut
was the deciding factor in a usable gear. this meant that some
mechanical detent was needed to position the blank.

Today I had another go starting from scratch

Involute cutter shapes sound esoteric but really they are almost a
circular arc on each face. trouble is that they are the wrong way
around to make a tool for.
The long and short of it is that I picked the least worn gear for use
as a template and freehand ground a HSS tool blank until I was happy
that the shape was close. In Ivans book there is a formula for working
out the diameter of the curve but I wasnt sure of the numbers I
needed. While I was mulling this over I spotted a washer on the floor
of the workshop, a little electrical one. Held up to the light the
washer proved to be the size I needed, it was an almost exact copy of
the involute shape of the gear I was copying. that made the freehand
grinding quite easy. I ground in the curve then tested it against the
washer then looked at the symmetry of the tool. It probably took ten
minutes of tweaking until I was happy with the cutter shape.

This time around I decided to work in cast iron (meehanite) so that if
I was successful I'd have one of the little stack of gears I needed.

I turned up an arbor to hold the existing gear, then a spacer, then
the cast iron blank. I made the blank two gears wide. Following Ivans
suggestion in the book I arranged a simple mechanical detent by using
a tom of scrap steel standing on the lathe bed. this could engage one
of the teeth of the master. I have a aching left hand from holding the
chuck into engagement while I planed out the cuts.

Using the tool on its side in the tool post, shimmed to centre height,
and moving the saddle back and forth to plane out the between teeth
gap is fairly slow work. but it does work.
late this afternoon I parted the finished gear into two and cleaned up
the faces. my gears mesh together smoothly and they mesh with a
smaller and a larger gear without problem.
by jove I think I've done it!
what is it? ten gears to go I think.

After tea I went out and searched through my junk for all the gears I
could find. I needed a 16 tooth gear or a multiple of that. I managed
to locate a 32 tooth gear out of an old photocopier and used every
second tooth for indexing.
One of the guys I work with now has the little 18mm dia 16 tooth gear
he needs to make the thread follower for his Myford Super 7 that he
has wanted for absolutely ages.

Gear cutting, approached the model engineer way, is a piece of cake.

one thing I did learn is that using cutting oil on cast iron almost
halves the effort to plane the tool across the cut. first time I have
ever seen cutting oil make that much difference in the cutting.

Stealth Pilot




Cheers - good reading. Thanks
r.

"Stealth Pilot" wrote in message
...

Gear Cutting has been a dark art that I knew little of until recently.

The restoration of a Myford ML7 has plunged me into this art in a big
way. When I bought the project I knew that all the headstock gears
were stuffed. Just looking at them didnt intimidate me.

I'm starting to wonder whether the mystery sets the prices of gears.
It seems to me that if you were tooled up to make them they wouldnt be
that difficult.

Myford's still sell the gears that I need to restore the lathe so I
priced the gears I need. Amazing really, the a lathe made half a
century ago still has factory spares available.
What I needed was basically everything from the headstock to the
first gear on the banjo. They all fit in a very small pile in the palm
of one hand. By the time the parts get through the distributor to me
that little pile in the hand has a cost of over $aus750.

When I first read the price quote sent to me I thought "you're joking"
the other words I said to myself cant be written here by civilised
people.
an entire set of involute gear cutters isnt much more than that!

I bought and read up on gear cutting. Ivan Law's Book "Gears and Gear
Cutting" from the model engineer people made it seem so easy.
So I thought that in secret I'd have a bash and see if actually doing
it would teach me along the way. I made a gear in 2024 aluminium just
by bolting the blank beside the existing gear and guessing the cut
positions. the gear worked but I had some fat and some thin teeth.
I just positioned the cutter on edge and used the tool post like a
shaper. with 3 thou cuts the load on the lathe wasnt extreme.

I realised from this experiment that accurate positioning of the cut
was the deciding factor in a usable gear. this meant that some
mechanical detent was needed to position the blank.

Today I had another go starting from scratch

Involute cutter shapes sound esoteric but really they are almost a
circular arc on each face. trouble is that they are the wrong way
around to make a tool for.
The long and short of it is that I picked the least worn gear for use
as a template and freehand ground a HSS tool blank until I was happy
that the shape was close. In Ivans book there is a formula for working
out the diameter of the curve but I wasnt sure of the numbers I
needed. While I was mulling this over I spotted a washer on the floor
of the workshop, a little electrical one. Held up to the light the
washer proved to be the size I needed, it was an almost exact copy of
the involute shape of the gear I was copying. that made the freehand
grinding quite easy. I ground in the curve then tested it against the
washer then looked at the symmetry of the tool. It probably took ten
minutes of tweaking until I was happy with the cutter shape.

This time around I decided to work in cast iron (meehanite) so that if
I was successful I'd have one of the little stack of gears I needed.

I turned up an arbor to hold the existing gear, then a spacer, then
the cast iron blank. I made the blank two gears wide. Following Ivans
suggestion in the book I arranged a simple mechanical detent by using
a tom of scrap steel standing on the lathe bed. this could engage one
of the teeth of the master. I have a aching left hand from holding the
chuck into engagement while I planed out the cuts.

Using the tool on its side in the tool post, shimmed to centre height,
and moving the saddle back and forth to plane out the between teeth
gap is fairly slow work. but it does work.
late this afternoon I parted the finished gear into two and cleaned up
the faces. my gears mesh together smoothly and they mesh with a
smaller and a larger gear without problem.
by jove I think I've done it!
what is it? ten gears to go I think.

After tea I went out and searched through my junk for all the gears I
could find. I needed a 16 tooth gear or a multiple of that. I managed
to locate a 32 tooth gear out of an old photocopier and used every
second tooth for indexing.
One of the guys I work with now has the little 18mm dia 16 tooth gear
he needs to make the thread follower for his Myford Super 7 that he
has wanted for absolutely ages.

Gear cutting, approached the model engineer way, is a piece of cake.

one thing I did learn is that using cutting oil on cast iron almost
halves the effort to plane the tool across the cut. first time I have
ever seen cutting oil make that much difference in the cutting.

Stealth Pilot



I thoroughly enjoyed this description and wish we had more like it.
Somewhere on the web I once saw some pictures of a change gear made of some
thin metal ribbon which had been crimped between two gears. The ribbon was
formed into a circle with the proper number of teeth, overlapped a bit and
soldered. A wooden disc with a metal bushing was turned to fit into the
metal tooth circle and was epoxied into it. It looked crude but I have no
doubt it would work for a while.

Don Young

Don Young wrote:

The ribbon was
formed into a circle with the proper number of teeth, overlapped a bit and
soldered. A wooden disc with a metal bushing was turned to fit into the
metal tooth circle and was epoxied into it.

Despair is the mother of invention! :-)


Nick
--
The lowcost-DRO:
http://www.yadro.de

On Nov 12, 5:37 am, Nick Mueller wrote:
Don Young wrote:
The ribbon was
formed into a circle with the proper number of teeth, overlapped a bit and
soldered. A wooden disc with a metal bushing was turned to fit into the
metal tooth circle and was epoxied into it.

Despair is the mother of invention! :-)

Nick
--
The lowcost-DRO:
http://www.yadro.de

IIRC winning a bet was the mother of that invention. Isaac Babbitt
told a customer he could make them a replacement gear in (??) minutes
and used that trick to form a mold and cast it.

Jim Wilkins

On Sun, 11 Nov 2007 22:00:08 -0600 in rec.crafts.metalworking, "Don
Young" wrote,
thin metal ribbon which had been crimped between two gears. The ribbon was
formed into a circle with the proper number of teeth, overlapped a bit and
soldered. A wooden disc with a metal bushing was turned to fit into the
metal tooth circle and was epoxied into it.

That does not sound like much of a way to make a gear. It sounds like a
way to make an EDM tool to cut a gear.

On Sun, 11 Nov 2007 17:18:19 -0600, Jon Elson
wrote:

Stealth Pilot wrote:
Gear Cutting has been a dark art that I knew little of until recently.
Cutting high-speed, high load gears, like an automobile
transaxle, require a lot of precision, and some of the gear
forms are insanely complicated, like hypoid gears.

On the other hand, moderate-speed gears with modest loads, like
a lathe's leadscrew drive, is another kind of gearing entirely.
Very light gears with significant errors will still work
remarkably well. A lathe's back gears or headstock gear train
is a little more difficult, but should still be able to be made
in a small shop.

Low-performance gears used to be cut in small shops for all sort
of jobs, even making simple gears to make a gear train for
cutting more complex gears such as helical gearing. (On a
universal mill, you provided a gear train to couple the dividing
head to the X axis leadscrew to form the helix angle. Every
different helix angle required a different gear ratio.)

I haven't heard of small shops making gears as a routine job
anymore. But, the same tools still can still do this job.

Jon


a question or two for the pros

in the catalogues the treatment of involute cutters is now "we only
get them in to special order" and they take the general information
out of circulation on what cutters are regularly used, or even who
makes the ones sold. often you are forced to make your own.

Q1.are there tooth gauges like there are for screwcutting and radius
curves giving a master shape which can be compared directly by placing
the gauge against a tool or tooth? you'd think that workable gauges
would exist. I've never seen any.

Q2.for gears up to, say the size of a cd, what guides exist to best
practise?

I'm just amazed that you get the raw computational theory but little
information on the limits to practical useful gear sizes.

Stealth Pilot

Jim Wilkins wrote:

IIRC winning a bet was the mother of that invention. Isaac Babbitt
told a customer he could make them a replacement gear in (??) minutes
and used that trick to form a mold and cast it.

That is even a much better trick!


Nick
--
The lowcost-DRO:
http://www.yadro.de

Q1.are there tooth gauges like there are for screwcutting and radius
curves giving a master shape which can be compared directly by placing
the gauge against a tool or tooth?

Not as such, no. You'll find gear guages made with rack profiles that
you can use to get data from an existing gear by rolling the tooth
form on it and seeing what gets you best engagement, but nothing for
shaped profile. If you know the diametral pitch, angle, and number of
teeth you can generate the profile with a little work, and use that to
make your own template...

--Glenn Lyford

wrote:
Q1.are there tooth gauges like there are for screwcutting and radius
curves giving a master shape which can be compared directly by placing
the gauge against a tool or tooth?


Not as such, no. You'll find gear guages made with rack profiles that
you can use to get data from an existing gear by rolling the tooth
form on it and seeing what gets you best engagement, but nothing for
shaped profile. If you know the diametral pitch, angle, and number of
teeth you can generate the profile with a little work, and use that to
make your own template...

--Glenn Lyford

To add to that, the tooth profile is different for each diameter, if
you go to the trouble of working it out and drawing it accurately.

The answer to this is that the cutters are made in 8 sizes with each
covering a range of tooth counts, for a particular pitch diameter.

In cases where the tooth shape is important (more than your average
class of spur gears anyways) cutters will be made for the specific tooth
shape.

I suspect, in this day and age, the market for such cutters is
becoming more linited, with the availability of EDM and CNC grinding
equipment.

Cheers
Trevor Jones

I suspect, in this day and age, the market for such cutters is
becoming more limited, with the availability of EDM and CNC grinding
equipment.

20 years ago I spent a summer at a manufacturing department that made
gears and splines. The whole time I was there I did not see one
involute cutter. Everything was either hobbed, shaped (fellows gear
shaper style, not say, logan shaper style), generated (spiral bevels),
or broached (particularly internal splines). Most were profile ground
to final shape after case hardening.
A bunch of the grinders would use CNC to redress the profiles at
regular intervals.

As much as I like them for home use, I just don't see involute cutters
as a viable way to make gears in quantity for a modern manufacturer.
I wouldn't be surprised these days if less critical gears and one-offs
were simply CNC milled, no special tooling at all (unless you count a
cad file as tooling)...

--Glenn Lyford

a question or two for the pros

in the catalogues the treatment of involute cutters is now "we only
get them in to special order" and they take the general information
out of circulation on what cutters are regularly used, or even who
makes the ones sold. often you are forced to make your own.

Q1.are there tooth gauges like there are for screwcutting and radius
curves giving a master shape which can be compared directly by placing
the gauge against a tool or tooth? you'd think that workable gauges
would exist. I've never seen any.

Q2.for gears up to, say the size of a cd, what guides exist to best
practise?

I'm just amazed that you get the raw computational theory but little
information on the limits to practical useful gear sizes.

Stealth Pilot


Take a look at this YouTube video, "Machining a Spur Gear"
http://www.youtube.com/watch?v=VHTXaU7GZC0

wrote:

I suspect, in this day and age, the market for such cutters is
becoming more limited, with the availability of EDM and CNC grinding
equipment.


20 years ago I spent a summer at a manufacturing department that made
gears and splines. The whole time I was there I did not see one
involute cutter. Everything was either hobbed, shaped (fellows gear
shaper style, not say, logan shaper style), generated (spiral bevels),
or broached (particularly internal splines). Most were profile ground
to final shape after case hardening.
A bunch of the grinders would use CNC to redress the profiles at
regular intervals.

As much as I like them for home use, I just don't see involute cutters
as a viable way to make gears in quantity for a modern manufacturer.
I wouldn't be surprised these days if less critical gears and one-offs
were simply CNC milled, no special tooling at all (unless you count a
cad file as tooling)...

--Glenn Lyford

Yup!

CNC all the way.

The problem lies in the cost effectiveness.

Jokers like us don't bring on the demend for runs large enough to
justify the setup and programming, though, for the types and quantities
we want.

Setup time and low volume = High cost per tooth in the box.

I bet you are right.

I wonder if they are CNC grinding Hypoid gears to order yet, for the
old car set?

Cheers
Trevor Jones

Trevor Jones wrote:
wrote:

I suspect, in this day and age, the market for such cutters is
becoming more limited, with the availability of EDM and CNC grinding
equipment.


20 years ago I spent a summer at a manufacturing department that made
gears and splines. The whole time I was there I did not see one
involute cutter. Everything was either hobbed, shaped (fellows gear
shaper style, not say, logan shaper style), generated (spiral bevels),
or broached (particularly internal splines). Most were profile ground
to final shape after case hardening.
A bunch of the grinders would use CNC to redress the profiles at
regular intervals.

As much as I like them for home use, I just don't see involute cutters
as a viable way to make gears in quantity for a modern manufacturer.
I wouldn't be surprised these days if less critical gears and one-offs
were simply CNC milled, no special tooling at all (unless you count a
cad file as tooling)...

--Glenn Lyford

Yup!

CNC all the way.

The problem lies in the cost effectiveness.

Jokers like us don't bring on the demend for runs large enough to
justify the setup and programming, though, for the types and
quantities we want.

Setup time and low volume = High cost per tooth in the box.

I bet you are right.

I wonder if they are CNC grinding Hypoid gears to order yet, for the
old car set?

Cheers
Trevor Jones

There are definitely companies out there producing hypoid gear sets for
cars long out of production, I don't know whether they are CNC ground or
not though.

"David Billington" wrote in message
...
Trevor Jones wrote:
wrote:

I suspect, in this day and age, the market for such cutters is
becoming more limited, with the availability of EDM and CNC grinding
equipment.


20 years ago I spent a summer at a manufacturing department that made
gears and splines. The whole time I was there I did not see one
involute cutter. Everything was either hobbed, shaped (fellows gear
shaper style, not say, logan shaper style), generated (spiral bevels),
or broached (particularly internal splines). Most were profile ground
to final shape after case hardening.
A bunch of the grinders would use CNC to redress the profiles at
regular intervals.

As much as I like them for home use, I just don't see involute cutters
as a viable way to make gears in quantity for a modern manufacturer.
I wouldn't be surprised these days if less critical gears and one-offs
were simply CNC milled, no special tooling at all (unless you count a
cad file as tooling)...

--Glenn Lyford

Yup!

CNC all the way.

The problem lies in the cost effectiveness.

Jokers like us don't bring on the demend for runs large enough to
justify the setup and programming, though, for the types and quantities
we want.

Setup time and low volume = High cost per tooth in the box.

I bet you are right.

I wonder if they are CNC grinding Hypoid gears to order yet, for the old
car set?

Cheers
Trevor Jones

There are definitely companies out there producing hypoid gear sets for
cars long out of production, I don't know whether they are CNC ground or
not though.

There are some old Gleason gear generators out there. Somebody may have a
few of them to supply the old-car market. They can make a spiral-bevel or a
hypoid gear set. They were production machines and they're relatively quick.

There also are gear job shops who use new CNC face-hobbers and another type
of CNC gear generator that can make them, who make these gears for the
industrial market. They set up and change over a lot faster.

I find it very unlikely that anyone is making them on mills or a grinder of
any kind. The geometry would require a fairly recent CNC, the equations that
have to be reduced to coordinate positions are daunting, and productivity
would be intolerable.

--
Ed Huntress

On Sun, 11 Nov 2007 23:28:16 +0900, Stealth Pilot
wrote:


Gear Cutting has been a dark art that I knew little of until recently.

I bought and read up on gear cutting. Ivan Law's Book "Gears and Gear
Cutting" from the model engineer people made it seem so easy.
So I thought that in secret I'd have a bash and see if actually doing
it would teach me along the way. I made a gear in 2024 aluminium just
by bolting the blank beside the existing gear and guessing the cut
positions. the gear worked but I had some fat and some thin teeth.
I just positioned the cutter on edge and used the tool post like a
shaper. with 3 thou cuts the load on the lathe wasnt extreme.

I realised from this experiment that accurate positioning of the cut
was the deciding factor in a usable gear. this meant that some
mechanical detent was needed to position the blank.

It is sort of a dark art, I'll grant you that. Reading the
engineering texts on the subject is a good way to get tired and
cross-eyed in a hurry.

Today I had another go starting from scratch

Involute cutter shapes sound esoteric but really they are almost a
circular arc on each face. trouble is that they are the wrong way
around to make a tool for.
The long and short of it is that I picked the least worn gear for use
as a template and freehand ground a HSS tool blank until I was happy
that the shape was close. In Ivans book there is a formula for working
out the diameter of the curve but I wasnt sure of the numbers I
needed. While I was mulling this over I spotted a washer on the floor
of the workshop, a little electrical one. Held up to the light the
washer proved to be the size I needed, it was an almost exact copy of
the involute shape of the gear I was copying. that made the freehand
grinding quite easy. I ground in the curve then tested it against the
washer then looked at the symmetry of the tool. It probably took ten
minutes of tweaking until I was happy with the cutter shape.

Sounds like a lot of stuff I do. If I didn't have a grinder, I'd be
out of luck most days.

This time around I decided to work in cast iron (meehanite) so that if
I was successful I'd have one of the little stack of gears I needed.

I turned up an arbor to hold the existing gear, then a spacer, then
the cast iron blank. I made the blank two gears wide. Following Ivans
suggestion in the book I arranged a simple mechanical detent by using
a tom of scrap steel standing on the lathe bed. this could engage one
of the teeth of the master. I have a aching left hand from holding the
chuck into engagement while I planed out the cuts.

Here's where you're beginning to lose me. It may just be a question
of economy, but it sounds like you're going about this the hard way.
I can understand that, but there is a much easier method if you can
invest a little money in tooling.

Basically, all you need is an involute tooth cutter or two that looks
like a little circular saw blade on an arbor (picture here, no
affiliation with the site: http://www.capital-tool.com/cutters.html)
and a dividing head. The dividing head gets mounted on the cross
slide, and the tooth cutter goes in the headstock. Gears mount on an
arbor protruding from the diving head, and you count the number of
clicks as you turn the head to evenly space the teeth according to a
formula (here's a website explaining this:
http://www.atmsite.org/contrib/JSAPP...e/divhead.html)

You get the cutter spinning, then feed the gear into the cutter with
the cross slide, retract, crank the dividing head to the next setting,
and so on until you finish the gear. Not the easiest job in the
world, but not the hardest, either.

My understanding of gear cutting from what I've read on the subject is
that you really want that involute profile to be just right. As the
two gears mesh and push one another, the net effect should be that
there is always smooth contact between the two gears- it's a matter of
rolling, not pushing. If the profile is off, it may still work, but
the teeth will wear out quickly, and/or it can be really noisy.

All the same, a dividing head can be an awfully expensive bit of
equipment, (Grizzly has one for $419.95, and they tend to be a little
less expensive than most.) and the cutters run about $100 as well, so
I'm definately keeping your idea for future reference. Sounds like a
lot more work, but for a handful of gears, it's probably more
worthwhile than shelling out the cash for the head and cutter.

On the other hand, if you intend to make more than a couple of gears,
it's probably worth getting the special equipment to make sure they
hold up in the long run.

On Tue, 13 Nov 2007 18:51:17 +0900, Stealth Pilot
wrote:

in the catalogues the treatment of involute cutters is now "we only
get them in to special order" and they take the general information
out of circulation on what cutters are regularly used, or even who
makes the ones sold. often you are forced to make your own.

Q1.are there tooth gauges like there are for screwcutting and radius
curves giving a master shape which can be compared directly by placing
the gauge against a tool or tooth? you'd think that workable gauges
would exist. I've never seen any.

I've never seen one, but it's not a bad idea. If you, or anyone else
on the group wants to make up a cad drawing of such a thing, I could
crank out a couple of dozen of the things out of thin (16 ga? 14 ga?)
stainless on the laser cutter for a modest sum. In a perfect world,
it'd be one gauge that can measure all common tooth sizes, with an
etched notation for each one.

I could draw it up as well, but at that point, there's a lot of
tedious drafting to get there, and it becomes a much less attractive
proposition- and there is the fact that I am not an accomplished gear
cutter, and may not end up with gauge that will work as well as it
could if I had more experience with the job. No point in wasting time
and material on something that only sort of works the way it is
intended.

If anyone wants to look into this, contact me off-list at
prometheusatcharter.net . I figure I could do them for $5 a gauge,
with minimal shipping costs- especially if they're small enough to fit
in an envelope. Whoever does the drafting can have them all at that
price and mark them up accordingly, or just get a couple for free, and
I'll sell the remainder at the $5 price.

Q2.for gears up to, say the size of a cd, what guides exist to best
practise?

I'm just amazed that you get the raw computational theory but little
information on the limits to practical useful gear sizes.

Stealth Pilot