I've got this Starrett center gauge no. C391. A very common
in toolboxes everywhere. So, can anyone explain the use of
gauge in particular the double depth of thread and also the
scale that apparently is related to various pitches of
an associated decimal measurement. The 60deg v-notches are
self explanatory but the scales? The starrett web site is of
they just say the scales are 14, 20, 24 and 32nd's of an
The "double depth" is to tell you how deep to make your final
cut when threading on the lathe. It assumes that the lathe has a
cross-feed which measures diameter removed instead of radius.
That is the actual table of "double depth" data. The scales on
either side of the words "double depth" (plus other words) are for a
quick sanity check after you have made your first scratch cut -- for the
common threads on the edges of the scale. 32, 24, 20 and 14 TPI. you
lay the scale edge for your thread (if it is one of those four) against
the scratch cut and make sure that all of the scale lines match the
scratch line. Only then are you sure that you have the right settings
in the quick-change box, or the right gear train built up in the banjo.
A simple quick check which keeps you from spoiling a potentially
expensive workpiece with a lot of time invested in the other operations.
For checking 14, 20, 24 and 32 TPI scratch cuts.
And the V notches are both for checking the grinding of the
threading tool (if you are using a shop ground one or are grinding it
yourself, and for making sure that you have the tool at the proper angle
to the workpiece.
I *think* that there is more detail in _the Starrett Book for
Student Machinists_, but I don't have a copy to check. Or maybe the
_Tools and Rules for Precision Measuring_. Both are listed in the back
of the catalog, just before you hit the reference tables.
Actually, it's so you can calculate undercut diameters, or to determine the
configuration of a tool for a thread that would have to be machined to a
shoulder, allowing for enough depth of tool without excess width. That
provides the fullest possible thread near the shoulder (when no undercut is
allowed). Minor diameter of threads isn't a reliable way to gauge depth,
or pitch diameter. Wires come to mind, or thread gages.
Thanks. I hadn't really looked at the C391 that closely.
The double depth table matches the numbers for root / minor diameter
on my threading table spreadsheet which were calculated as 2 * 0.6495/
Pitch, assuming a tool tip width of 0.125/Pitch, both from Machinery's
I usually leave the tip 'slightly rounded' and feed the compound in
0.75/Pitch from initial contact, which should be correct for the
proper tip width but is oversize with the finer tip, then finish to
fit or wire measurement.
I think I could approximate the tip width without regrinding the bit
each time by cutting in 0.75/P, then moving the bit sideways by
shifting the gear train a notch. I avoid the issue of thread root
weakness by buying load-rated forged eye bolts for high tension loads
like the recent I-beam trolley hangers.
to determine the
machined to a
Thanks. I hadn't really looked at the C391 that closely.
The double depth table matches the numbers for root / minor
on my threading table spreadsheet which were calculated as 2
Pitch, assuming a tool tip width of 0.125/Pitch, both from
I usually leave the tip 'slightly rounded' and feed the
0.75/Pitch from initial contact, which should be correct for
proper tip width but is oversize with the finer tip, then
fit or wire measurement.
I think I could approximate the tip width without regrinding
each time by cutting in 0.75/P, then moving the bit sideways
shifting the gear train a notch. I avoid the issue of thread
weakness by buying load-rated forged eye bolts for high
like the recent I-beam trolley hangers.
So what I'm reading is the information given on this gauge
is pretty much a
waste of time. Why give double depth of thread when the
rest only affects the radius. It would be far more handy if
gave the actual compound rest movement required for a given
And the four most common threads (14, 20, 24, 32) ? Thread
gauges are far more useful than the scales given. My
curiosity has been
When you cut an ending groove with a different tool and dial readings,
it's deep enough when your calipers read (OD - DoubleDepth). If I have
several threads to do I'll leave the dials zeroed for threading and
cut the grooves by measurement. A groove that's slightly too deep is
far less important than getting the threads right.
The actual compound movement varies with the tip width if you measure
in from initial contact. If you ground a sharp point you'd have to cut
in the full Sharp Vee depth to get the pitch diameter correct. As I
mentioned, the two numbers I found useful are the double depth they
gave and the compound feed which is specific to the way I personally
Those four pitch rulers work for fractional and multiple pitches as
well. You can check 7, 10, 12, 14, 16, 20, 24, 28, and 32.
There's nothing wrong with using the double depth as a guide, but to use it
for sizing threads makes no sense. The tolerance for the pitch diameter is
generally smaller than the tolerance for the minor, and that has relevance
only when the tool is properly ground for the given pitch. The flat that is
required on the tool tip gives cause for greater error than might be
imagined. It's all too easy to go too deep with the formula, especially
when you must take into account the starting pass, which, again, you're at
the mercy of the major diameter. The tip radius you speak of keeps you out
of trouble, but may well be out of tolerance for the geometry of a given
Theory for threads does not translate well into operation without gauging.
People that rely on compound movement for sizing threads are commonly known
for making scrap.
Please bear in mind, I'm speaking from the perspective of one that had to
satisfy independent, critical inspection. I also worked in places where
was no inspection, ever, and no scrap produced. Anything turned out was
assembled and shipped.
[ ... ]
a) *Some* compound feed dials read in diameter, not radius.
b) Way back when (when the thread gauge was designed) lathes
did not always have a compound anyway.
At what compound angle? Even just sticking with 60 degree
threads (since this is a 60 degree center gauge, not a 55 degree for
Whitworth threads, or a 28 degree for Acme threads)? It is common to set
the compound at 29 degrees or 29.5 degrees (both a little below half the
thread angle) -- and each of those requires a slight adjustment of the
Back when it was designed those probably were the most common
Certainly -- but not everyone has thread pitch gauges in an
early shop -- but everyone was expected to have the center gauge.
Not for precise sizing -- no. But it can tell you when you are
getting close enough to *start* measuring. (Granted -- the appearance
of the thread crests can do that too -- once you know what to look for.
Or -- when you are cutting an original thread on a workpiece and
expect to make a mating thread to fit it -- not to any official spec,
but to make a working device -- such as when I was making a few
cylindrical waveguide antennas (four total) with white Lexan rings
threaded to match -- to hold on the metal plate at one end, and the thin
plastic one to keep rain out at the other end.
For a project like that -- there is no point to cutting a unique
thread to a precisely calculated and measured spec -- since you can't
buy parts to screw into it anyway.
Or one-off projects? :-)
Yep -- that can be a problem too.
Let's see. I made some bushings threaded inside and out, but the
outside was to fit a tapped hole, and the inside to fit the leveling
feet. Made the nose piece to fit an L-00 spindle, big male threads,
adapter for a broom handle thread, female 3/4x5 acme. I don't ever
recall making both parts, male and female. DoN did, his example from
earlier in the thread.
I think I'd make first whichever part was easier (smaller) to use to
check the other part.
Depends on the relative sizes. If the male thread is on the
larger part, I might make the female first, so I could bring it to the
male to test as I approached size. In the case which I described
somewhere up-thread, I had put male threads on both inches of an
aluminum cylinder of about 4" OD, and then turned white Delrin rings and
internally threaded them to fit the ends of the cylinder. Both ends (of
four cylinders) were threaded to the same depth, so either end could be
used to test the rings. The only difference in the cylinders was slight
changes in length to tune for the frequency of the wireless
transmitter/receiver on different channels, and a corresponding shift of
the location of the connector and radiating antenna which went in the
If the cylinders had been heavier, I would have made the rings
first (or perhaps a master ring of steel) and then turned the threads
on the ends to fit, and turned the actual Delrin rings to fit a male
Depends on the part, Pete. Fact is, I rarely made mating parts----except in
building tools. Consider that if you're doing defense work, you often make
a single item, with no idea where it goes, or how it fits. That's the
reason for establishing standards, and upholding them, even if they aren't
obvious for given circumstances. If party A orders a part that specifies a
given thread and class fit, you can guarantee the fit by upholding the
All too many treat threads with contempt. As long as it works for them
it's good-----but that doesn't always fit in the real world, and it's not
any harder to do it right than it is to do it wrong.
I try to educate in such a way that choosing the proper path becomes
routine. Sizing pitch diameter by dial advancement is one of the things
best not learned. It is not an acceptable procedure.
Good point, Harold. At least as long as it's not to fit a mop handle,
in which case the appropriate standard looks like 3/4" x 5 TPI,
I'll go back and re-read your post, then save it. I have no intention
of ever making a part for the gov., but who knows, I might make
something for my own defense. So far, I've not made a thread where I
didn't have the mating thread to compare, but I like learning other
methods anyway. The mating thread might not be accessible as a gage
if it's too big or unwieldy.
IIRC the male thread first, fitted it to 3-wire measurement, then used
it as a gauge for the internal threads. Usually I try to design so the
nut can be bought or tapped and the male part is all custom, such as
the axle example I gave above. Much depends on how easy it is to
measure or try the fit. A shaft held between centers can be removed,
checked and replaced while a part in a chuck has to stay there until
done, unless it fits a small 5C chuck.
Came by a couple of 1/2" LH thread nuts, so decided to make shaft
adaptors for motors with 1/2" shafts since all of the available ones
are RH thread. I got to the point of cutting the threads, did my first
pass and checked the scratch against the mating male thread - didn't
quite agree; the original thread was 1/2 - 24, so now I have a funny
spot every 6th crest.
Can't tell you how many times I've done that. Nice thing is, the "funny
spot" virtually disappears if you make a habit of making your first pass a
shallow one. A thou or two is plenty.
In spite of my years of experience, or maybe because of them, dunno, I
always take a shadow cut when chasing threads. Just deep enough to allow a
match with the pitch gage. It's too damned easy to miss a setting,
especially on my lathe, where I've messed with the ratio, having cut all the
feeds and threads in half. I like light feeds for finishing and found my
Graziano too fast in that regard. I've lived with the changed ratio since
1967----but still screw up occasionally. Maybe a function of getting old,
"Harold and Susan Vordos" wrote in message >
threads, did my first
thread - didn't
have a funny
is, the "funny
your first pass a
them, dunno, I
enough to allow a
having cut all the
finishing and found my
changed ratio since
of getting old,
A couple years ago I made a thread toy called a puzzle nut.
It was cut
at 9 tpi double start. Since this is an odd number thread
and it was
being doneon a SB model A 9 inch I used the numbers on the
dial for one groove and the unnumbered marks for the other
Did both the male and the female this way. The nut was cut
and red loctited to the ends of the male section. It worked
looked like 18 tpi but the nut moved fast from end to end.
I decided to try a triple! For that I started with a longer
rod with a
dowel pin in the side so it would rest against a chuck jaw.
groove at 9 tpi as before but for the next groove I moved
the rod pin
to the next jaw and then the third jaw to make a triple
This one was really fast moving! Sure acted strange rolling
back and forth as the thread looked to be fine at 27 tpi but
hauled ass from end to end! That was fun......;>)