Which lathe?

Jewelry Discussion
21

As with all other things, which tool you are going to get depends on
what you need to do with it. Clausings and Hardinges are awesome
machines, but you won’t need that degree of power and accuracy to
machine wax models.

I have a Sherline, and though it definitely has its limitations,
overall I think it is a good value for money, and the accessories
are relatively inexpensive. When buying a “bare” lathe you may end
up spending as much again on associated tooling and accessories.

People do use Sherlines to make titanium jewelry:

http://www.ganoksin.com/gnkurl/1ss

If you Google machining forums, you will find that inexpensive
Chinese lathes get some pretty good reviews as being a good value
for money. Many companies sell basically the same lathe; Grizzly
gets pretty good reviews for customer service, Harbor Freight gets
the worst. There’s some good info here, though slightly outdated:
http://www.ganoksin.com/gnkurl/1st

Rob

22

Any lathe like a Sherline that can hold tolerances like Jim was
referring to (within.001") is more than enough for a jewelry bench.
You cannot maintain those same tolerances with the handtools you are
always carrying on about.

To call Sherline lathes borderline junk shows once again that you
will pick an argument with anyone over anything, just to sooth your
own massive ego, I must assume.

Sam Brown
San Jose, CA

23 
I would say the lathe is the first machine tool to have and then a
mill ifnecessary.

Yep, especially considering that jewelers tend to make rings. Two
links:

Machinist’s forum, like Orchid. Pretty much the mother lode:

For those of you/us with small machines, if you don’t know these
guys you’ll be happy to learn of them:

24

Gary, rather embarrasingly, we have a 1950’s Pultra lathe, but I
have no idea how to use it, so it mostly sits idle, needing a bit of
work before it’s useable. Given some spare time, I’ll teach myself
to use it, but who has spare time? One day, I promise, one day!

Jamie Hal
http://primitive.ganoksin.com

25 
Any lathe like a Sherline that can hold tolerances like Jim was
referring to (within.001") is more than enough for a jewelry
bench. You cannot maintain those same tolerances with the handtools
you are always carrying on about.

Holding tolerances is a loaded statement. My first job was lathe
operator in toolmaking shop. My lathe had spindle eccentricity
visible by eye.

Nevertheless, I customarily worked in with six delta tolerances,
which means 1 micron precision. So you see one has nothing to do with
another. Even a bad lathe would maintain local centricity. As far as
the rest of your comments, they hardly deserve a response.

Leonid Surpin

26 
I can find my way around a lathe, but I would not call myself an
expert. 

Nevertheless, I customarily worked in with six delta tolerances,
which means 1 micron precision.

So which is it Leonid? Either you are not and expert or you are one
of the best machinist in the world in your first job and able to
hold 1 micron tolerances on a lathe with an eccentric headstock.
Frankly I think it is you are just full of it.

James Binnion
James Binnion Metal Arts

27

Hi Leonid:

Nevertheless, I customarily worked in with six delta tolerances,
which means 1 micron precision. So you see one has nothing to do
with another. Even a bad lathe would maintain local centricity.

One micron tolerances? .001 mm? (aka 0.00000393 inches.) Just what,
exactly, required a manual lathe to crank out parts that were
toleranced in microns? How did you go about measuring these parts? A
hand held micrometer won’t get close to that level of accuracy. The
heat of your hand will foul up the reading, nevermind what the heat
of the lathe, or the heat of cutting will do to the size of the
part. I have measured things down into the micron range…using an
electron microscope. Not something that’s going to happen in the
average jeweler’s shop. Just by way of comparison, the German digital
positioning system on my big lathe only reads to ten microns. (when
German machinists see no point in being any more precise…)

Meanwhile, he’s not entirely up in the air on this one: if the
spindle on the lathe were eccentric, but the bearings weren’t
completely trashed, he could probably hold pretty good concentricity
on the part by simply doing the whole thing in one shot, and taking
light cuts. Yes, the part would be eccentric in relation to the rest
of the bar, but concentric relative to itself, which is all that
really matters once you cut it off.

In defense of the Sherlines, they’re nice little machines for what
they are. Yes, they’re mostly aluminum, but for what people use them
for (namely small stuff) they’re enough. The nice thing about them
is that they’re small enough to pick up and put away when you’re
done, and you don’t have to rewire the house when you install them.
They have lots of nice tooling, and can be tweaked to do all sorts of
interesting things.

I don’t have one, but I’ve played with them, and have been impressed
with them. (I’m a serious lathe snob, so I can understand why some
look down on an aluminum bedded lathe, but for a beginner or for
turning wax, they’re great.) There are reasons why the lathe snobs
gravitate to heavy iron. John’s treatise on rigidity being one of
the most important, but not everybody needs to play that game. For
those who don’t understand why I have 5000 pounds of oily iron eating
up one side of my garage, the sherline’s one of the best options out
there for a new micro-lathe.

For normal mortals, 0.001" is plenty good enough for jewelry work.

FWIW,
Brian

28 
Nevertheless, I customarily worked in with six delta tolerances,
which means 1 micron precision.

I have no idea what “six delta” tolerance is (maybe sigma?), but in
an ordinary machine shop, even with extraordinary tools, you can’t
even measure to 1 micron. Difficult even under controlled laboratory
conditions.

Al Balmer
Pine City, NY

29 
Sherline that can hold tolerances like Jim was referring to
(within.001") is more than enough for a jewelry bench.

On a more reasonable note… Somebody in the newspaper was quoted
as saying they were going out to dinner but “they didn’t want to
have elf-food”. Giant plates with one single bite of over-produced
food and such. The only problem have with the really small lathes is
that - they are tiny, with tiny little handles and everything is just
pint sized. Beyond that they are perfectly fine machines for a great
many people. I would say that anything less (read: cheaper) than the
Sherline or Taig is probably not worth buying, no.

The last real job I did on my lathe - last week - was a channel set
princess wedding band. I cut out the hole to size 7 1/4, cut the
thickness to 2.5mm and then cut it off at 3mm wide. I used dividers
to measure, and the handwheels. Then I cut the channel by hand,
tapered down the shank both ways - thinner and narrower - at the
bottom, and after I cast it I popped it up to it’s real size, 7 1/2.
All very low precision, is the point. I have the metrology to measure
it down to a thou’, but it just wasn’t necessary.

The real answer to the question here is just like anything else -
which lat= he? depends on what you want your lathe to do. I want mine
to make jewelry and also tools and parts for the occasional machine.
Shafts, punches, bearing surfaces, rotors, what have you. So, I want
a more substantial machine but I really can’t justify a two ton
behemoth, either. If you want to do spinning then you’ll need
something really serious. On Ebay there’s a lathe with a four foot
diameter chuck…

The job I did BEFORE that ring was a little button pendant - drill a
hole to set a one carat diamond into, turn it to some diameter
around the hole, contour it with a file, and cut it off. 15mm
diameter, maybe. Just about ANY lathe will do that…

30 
I have no idea what "six delta" tolerance is (maybe sigma?), but
in an ordinary machine shop, even with extraordinary tools, you
can't even measure to 1 micron. Difficult even under controlled
laboratory conditions.

The difference between delta and sigma as follows:

Sigma indicated probability of even. It is statistical measure of how
likely even is to occur. In modern corporate practice sigma 6 is
quite popular. It means to restructure corporate processes so
probability of disruption was no more likely than sigma 6 event.

Concept of delta is the same as in calculus. It indicates difference
from standard. Every delta means one decimal place using meter as
unit of measurement. Delta 6 means six decimal places to the right,
or 0.000001m, more commonly 0.001mm or one micron.

As far as what is possible and what is not, let me gently suggest
that your experience if far behind the reality. In order to make two
parts to engage in what is know as interference fit, required
tolerances, if my memory serves me right, about 12 microns. Let me
use this opportunity to plug my DVD “Coronet Cluster” which
demonstrates how such fit is accomplished at the bench, with very
primitive means.

Leonid Surpin

31 
One micron tolerances? .001 mm? (aka 0.00000393 inches.) Just what,
exactly, required a manual lathe to crank out parts that were
toleranced in microns? How did you go about measuring these parts?
A hand held micrometer won't get close to that level of accuracy.

First, the highest precision worked in the shop was delta 8 or
0.00001 mm, so one micron was somewhat pedestrian. Microns are
measured with hand micrometers faced with carbide. Of course they are
of more sophisticated design than of the shelf type. And they do need
to be calibrated often. In places where micrometer cannot reach,
indicator heads are used. For example for determining centricity. One
would rotate chuck by hand and watch needle deflection. It is
somewhat tedious to describe actual methodology. Your concerns are
well understood. Each project required different setup and method of
measurement.

As far as why such tolerance were needed, we made tooling used to
produce micro switches used in aero industry. On the west these
function were done by microchips, but Russia was behind that
technology and used mechanical devices. Hard to know for sure, but I
believe that MIG 21 was using our switches. In 1972, we exhibited in
Industrial Expo in Moscow. We made 10x10x10cm cube comprised of 1000
cubes, 10x10x10mm each. Each and every cube was finished to delta 8
tolerances. When these 1000 cubes were assembled in one large one,
all the air was squeezed out and they held together due to
atmospheric pressure. The only was to take it apart was under vacuum
bell.

Leonid Surpin

32

To All,

Here is another 2 cents worth.

I too have a Sherline Lathe and Mill. They meet or exceed every one
of my expectations. For what we are making anyway.

He are a few comments.

For the price you can’t beat it for quality. I had two other more
expensive ones that were no better. They both wore out. The Sherline
still keeps going strong. They have many available attachments that
are beneficial to choose from.

You can attach a digital read out, which in my opinion is an
absolute must. This is available in either inch or metric
measurement. It’s nice to zero out different stages of a project.

What is harder to find are inside expansion collets to hold either a
wax or a ring. I found some large ones at Harbor Freight that I had
tooled to finger sizes. I also had a set of brass ones made. The hard
part here is what size to make the collets. Considering some of the
shrinkage we have to build in from wax to finished piece.

If you consider the amount of lost material from start to finish at
between three and seven percent, tolerance at the micron level are
in a different industry. I still use a ring mandrel to size my rings.
Maybe I’m old fashioned.

Best regards,

Todd Hawkinson
Southeast Technical College
www.southeastmn.edu/jewelry

33 
Just what, exactly, required a manual lathe to crank out parts
that were toleranced in microns? How did you go about measuring
these parts?

All of which is aside from the fact that some parts of a jet turbine
engine are made to those tolerances, and some parts are not. A
Ferrari engine isn’t made to those tolerances. Telescope mirrors are
ground to those tolerances and even less for good ones. There is no
reason on Earth for any piece of jewelry in any way or any fashion
to be made to that spec. Understand that precision money. I won’t
quote Jim’s assessment of the statement, but I’m definately with him
on this one. “That there’s %^**, and this here’s Shinola”

34

A friend gave me a small lathe, Edelstaal Machinex 5. It would have
done what I needed but it needed parts and I could not find them. I
like to buy better than I need so if I decide to sell it I can find
someone wanting it or wanting to upgrade to it. I bought a Smithy
Midas 1220 LTD 3 in 1 lathe, Mill, Drill Press. It is more than I
need at the moment but life changes and I might use it more to it
capabilities in the future. My capabilities would have to grow
considerably first. I make spheres (out of rock) and put a 45 degree
bevel on galvenived cups and pvc cups to make spheres. I make stands
to hold my finished sphere. I have faceted in the past and hope to
get back to it. I can make dops with this machine. I also make and
sell sphere machines. I can use the Smithy for that also.

I just looked them up again just to find they are having a big sale
on them!! Oh well it is only money. ;-))

Any way I like my machine. People who know more than I do might not
like it, I do not know. Your input is welcome. It will not hurt my
feelings. Like I said it is only money.

Check it out at http://www.ganoksin.com/gnkurl/1t3.

If any of you live close you can come by and use it.

John
Grand Junction, CO

35 
The difference between delta and sigma as follows: As anyone who
has taken (or taught, as I have) a statistics course knows, 6 sigma
means that 99.99966% of results are within specified tolerances.
That's without going into details of probability curves and
standard deviation. This is easily verified by a simple web search
or reference to a statistics text.

As a student of mathematics, I of course know the common meaning of
delta, but I have never encountered the application of the word you
describe, nor can I find any references on the web. Perhaps you
would be so good as to provide such?

As far as what is possible and what is not, let me gently suggest
that your experience if far behind the reality. In order to make
two parts to engage in what is know as interference fit, required
tolerances, if my memory serves me right, about 12 microns.

My suggestion to you wouldn’t be so gentle :slight_smile: Allowances for
interference fit depend on the size of the parts and the joint
strength required, but measurement to 1 micron is not required. Tell
me, how did you measure your lathe-produced parts? When I worked at
B&L, we would use optical techniques in a controlled environment.
Maybe machinist’s calipers are getting better?

Al Balmer
Pine City, NY

36 
First, the highest precision worked in the shop was delta 8 or
0.00001 mm, so one micron was somewhat pedestrian.

You do know what a micron is? 0.00001mm is ten microns, which I
would consider more “pedestrian” than one.

Al Balmer
Pine City, NY

37 
I too have a Sherline Lathe and Mill. They meet or exceed every
one of my expectations. For what we are making anyway.

I’ll just say here that there are probably millions of hobbyists and
something more than hobbyists using Sherlines and Taigs to do all
sorts of things. Working steam engines are a common project - many
will sit on a standard playing card. Another project that’s almost
become a competition is building working scale models of jet engines.
All in a home workshop… But a gizmo that holds a wax tube on a
flexshaft while you hold a graver in your hands is - well, let’s just
say it’s more primitive than Icare to fiddle with.

38 
You do know what a micron is? 0.00001mm is ten microns, which I
would consider more "pedestrian" than one.

I know that micron is 0.001 mm. Please check your sources.

Leonid Surpin

39 
I have never encountered the application of the word you describe,
nor can I find any references on the web. Perhaps you would be so
good as to provide such?

I am not so sure it can be found on the web. May be it is particular
to Russia. It is indicated on blue prints as a single triangle
followed by a number.

Measuring in microns is not so difficult as you make it. I have a
gage calibrated in 10 microns divisions. The physical distance
between marks is 1 mm. It is very easy to interpolate up to 2 microns
even with my gage which is relatively primitive. On actual micron
gages the distance increased to enable readouts in microns. It is no
more complicated than that. I suspect that you are using digital
readout gage, which has to rely on current strength to measure the
distance. So measuring microns simply above their design parameters.
With mechanical gages there are no problems.

Leonid Surpin

40 
Sherline that can hold tolerances like Jim was referring to
(within.001") is more than enough for a jewelry bench.

Surviving for decades of using a UNI Mat SL which makes a Sherline
look large. jewellery work and making some rather large tools.

A much larger 7 X 14 at a good price when the tax advantages
couldn’t ignored. I do like the bigger one although even though I do
keep my fingers out of the 80 pound one, which I can carry.

Now I did help my dad move a 300 pound one, not a lot of fun even
with proper rigging. I do tend to be a whimp from time to time :slight_smile:

jeffD
Demand Designs
Analog/Digital Modelling & Goldsmithing
http://www.gmavt.net/~jdemand