| Threading part I-II-III
(Jul 13, 2001) |
Threading Questions (Feb
10, 2004) |
| Threading (Jul 16, 2001) |
Correct Thread Depths
without Calculation (Feb 12, 2004) |
| Need a few tips (Feb 19,
2002) |
More Threading - Tapered Pipe
Threads (Feb 15, 2004) |
|
Thread Cutting problem on a 9A (May 20, 2002) |
Threading Primer anyone?
(Mar 8, 2004) |
| Threads Feeds (Sep 28,
2002) |
Threading question (Apr
2, 2004) |
| Threading (Nov 3, 2002) |
First Threads (Apr 27,
2004) |
| Screw Thread Transposing Charts
(Dec 15, 2002) |
Threads on a long rod (14 feet)?
(Oct 25, 2004) |
| Threading? (Feb 9, 2003) |
Threading your bar stock
(Oct 26, 2004) |
| Screw cutting at 25 degrees
(Feb 13, 2003) |
Newbie trying to thread on a
heavy 10 (Nov 17, 2004) |
| Threading question (Mar
28, 2003) |
480 TPI (Dec 6, 2004) |
| Threading on 9 A (Apr
24, 2003) |
Threading with tailstock die
holder (Dec 11, 2004) |
| Now that's what I call a fine
thread (May 7, 2003) |
Cutting 27 tpi? (Dec 17,
2004) |
| Thread size (May 22,
2003) |
Thread pick-up troubles (Jan 19, 2005) |
| Threading question (Oct
3, 2003) |
Internal Threading Bars (Jan 24, 2005) |
| Cutting threads without a thread
dial (Oct 20, 2003) |
1
1/2-8 Threads (Feb 28, 2005) |
| Rough External Threads
(Dec 23, 2003) |
Screwcutting with compound set
parallel to bed (Mar 9, 2005) |
| Threading drill rod?
(Dec 27, 2003) |
Looking for a source for a
threading tool (Mar 18, 2005) |
| Can you cut finer than
160 TPI on a 9" C? (Jan 25, 2004) |
|
| |
| Threading part I-II-III |
| HOW TO CHASE
THREADS ON ANY LATHE 1.SET-UP- Use a fishtail to grind your 60
degree tool bit. A slight top grind so the tip is higher then the
back. Chuck the piece up and use a center to support the end in the
tail stock. Next, position the compound in line with the cross-feed
so the handles line up. Now, kick the compound over so your tool
moves TO THE LEFT. Look at the degree markings on the side. You want
29 1/2 degrees from your starting point. I say it this way because
each brand of lathe can be marked out differently. Tighten the 2
set-screws holding the compound in place. With the bit in its
holder, square it to the work-piece using your fishtail threading
gauge. It is very important that the ground POINT be square to the
work. And just slightly below center. Do this by taking your 6"
scale and place it gently between the work and the work-piece using
slight pressure. This .025 thick scale will tilt towards the work if
the tool is high; and tilt towards the operator if it is below the
centerline. It is a very accurate way of setting any tool-bit in any
lathe. Once you've got that squared away select your feed and
speeds. Use a slow spindle speed. Double-check your work. Lets talk
a moment about the chasing dial. Lathes differ in how the dial is
set up. Many of the machines I have run will read: even threads any
line, odd threads any number, multi lead threads...that is a whole 'nother
world. Now you can be pretty safe in threading by going back to the
SAME spot each time. In fact you can thread without a dial, just
leave the half-nuts engaged. At the end of each cut, crank the tool
out of the cut and reverse the lathe. Now it is a good idea to
practice your indexing. Each lathe has a different 'feel'. The
'sweet-spot' for each dial must be found. ANTICIPATE the index. You
will understand this when you work with it. Even the most worn-out
dials can still do good work. OK it is time to set your dials. Start
with the compound. Run the compound out, then run it back in until
your tool-post is roughly centered over the center of the
cross-feed. Set your dial to '0'.Now bring the cross-feed in until
it barely touches the work. Set this at '0'. This will be your
home-base. In fact, wipe the cross-feed casting and pencil a mark on
it to come back to. END OF PART I Ron (1079) |
| THREADING ON A LATHE PART II RUNNING
THE JOB- Your tool is set, gearing is set, spindle-speed set,
appropriate sacrifice to the local gods done. Now it's time to run
this turkey. Turn on the machine and take .005 on the tool-bit. You
can engage the dial by moving the carriage until the nuts engage
THEN turning the lathe on. At this point you are committed. DO NOT
DISENGAGE THE GEAR-TRAIN FOR ANY REASON! That is, the head-stock and
the thread gears should never be disengaged from one another. If you
want to see the first pass clearly, paint on some layout-fluid on
the work first. At the end of the cut pull the tool out at the same
time you open the half-nuts. This can take a bit of time to learn,
so practice this BEFORE DOING THE ACTUAL WORK. Almost any
single-point thread has an 'escape ditch' at the end of each cut. Go
back to your cross-feed 0 and take another .005 on the tool. Use
some oil. Either black sulphur thread-cutting oil or 90Wt gear or
Hyploid oil. As you progress into the cut, take spring cuts for
every 2 in-feed cuts. This means go back over the work without
advancing the tool. This will save you the embarrassment of having
the tool dive into the work ruining it. When it looks like your
getting near the finish-size, you may want to try the nut on the
shaft your threading. When the last few passes are in sight advance
the tool in using the CROSS-FEED DIAL ITSELF. Being sure to note
where it comes back to. This will clear up both sides of the thread.
IN GENERAL, that's about it. In a third article, I will cover some
finer points. Like what do I do when the whole thing screws up, I
broke the damn tool, and re-setting the tool in the thread. regards,
Ron (1080) |
| Before I get into
today's subject I'd like to address tool forming for a minute. In
most manuals much is said about the angle of the tool bit faces in
relation to one another. In my years of being a machinist doing both
proto-type and production work, I have used a very simple set-up for
all my tool-bits. All I did was enlarge the tool-rest on my grinder
and set it at a 4 degree angle. With the possible exception of a
parting tool or square-thread, I leave it there. By replacing the
stage that came with the bench grinder with a 3" piece of 2X2 angle
iron, you increase the ease of tool sharpening. Tool rests on most
grinders are only good for ruining your tool, your fingers, and your
day! Slot the new stage and make a handle for it that doesn't
require you to dive for that adjustable you keep in your hip pocket.
I always start with the top. Four degrees is plenty of cutting
angle. You only want to grind enough to get a clear dished land. Tip
the tool so that the tip is the highest point of your bit. Next form
the front. If your going to use it for turning a shaft TOWARDS the
chuck, angle it slightly in that direction. When you get a full face
STOP. Now for left-side clearance. The stage is still set for that 4
degrees. Lay the left side of the tool,(the leading edge) against
the wheel. Just use firm pressure dipping it in water to keep your
digits cool. When it has a clean face stop. Now, take a small stone,
India stones are nice, diamond is better but expensive. Hold the
stone by the ends in one hand and the tool bit in the other.
Carefully stone a small radii at the point. I use a back and forth
movement while slowly turning the tool.This actually gives your tool
a longer life. NEVER leave a sharp point on a tool when working
metal. It is the first thing to break. One reason I like this set-up
is that I can rotate my tool against the wheel forming a small
radii. Then I go back and hand stone it. This is very handy when
your doing radius work for a shoulder. WOW! Didn't think I'd get
this far a-field! I tend to 'run my gator' when I get on a favorite
subject. But, this is a VERY short course in cutter-grinding. Now
let's get back to work. INTERMISSION, TAKE A LEAK, PAT THE WIFE AND
GET SOME COFFEE INTERNAL THREADING PART I 1.Start with your
tool-bit. It is nothing more then a boring-bar that has a threading
tip. V-threading is always done with a 60 degree point. The leading
edge of most threading tools tilt away from the work as a compound
angle. This double angle is what does the cutting. 2. Once your tool
is ground tilt the compound 29 1/2 degrees to the RIGHT. I other
words, in the opposite direction as your external threading
operation. When you begin to cut the thread, you will be cranking
the compound OUT instead of IN. Always feed the compound in the
direction of the thread. When we get into left-hand threading, you
will see this is still true. 3. As before, set your bit with a
fishtail so that the V-form of the tool is 90 degrees to the work.
If you grind your tool using a small protractor set to 30 degrees,
it will help to keep your tool form square to the shank. This will
help later when you are setting up the job. 4. Once this is all done
ditch the job. Use a scale and find the back end of the cut. Mark
the point on the bed with a magic marker. Now turn on the machine
and use your cutting tool to under-cut the end of the thread. This
is your 'foxhole'. Because it is near impossible to stop the cut and
back out the tool all at once before you crash into the wall, you
ditch it. About 3 threads long should do it. 5. Now for an indicator
of some kind. Start by laying a very visible mark on the threading
tool. Next you need to set up something a bit more reliable on the
bed. A lathe-stop may only be trouble. A 1" travel indicator is
ideal. Cheap ones for under $20 work well. If I had the extra
shekels I'd go for a 2". These things are invaluable for lathe work.
With them you can do VERY accurate shoulder and groove work. Barring
that grab a magnet and a white strip of paper. Clean off the excess
oil with a rag and place the magnet on the strip of paper where your
ditch is. Why do this? It is a highly visible line to stop at that
should you miss your ditch, won't do damage to the lathe...just your
tool. As in your external threading it's a good idea to practice
your moves BEFORE you do the cut. I do. As you come to the end of
the cut, you only want to release the half-nuts, nothing more. Until
you get the hang of it, this will be a bit nerve-wrecking. But, like
a stone, it will pass. 6.Time for pucker-power. The gears are set,
you have put a mark on the cross-slide as a rough 0 mark. Your dials
have all been zeroed as before. Lay on some oil and courage and get
cutting. Remember, since your cutting with a bar instead of a bit
take no more then .005 at a time. Until you get good at it, do a
spring-cut every other cut. Use back-gear. Remember you will be
feeding the tool in by turning it to the LEFT on your compound. If
you are using a dial, set it at .01 NOT 0! You want to see it coming
so you can anticipate the end of cut. When you arrive in your ditch,
the cross-feed should be dialed IN NOT OUT! 7. It is always a good
idea to start the run by champhering the end of the piece with your
tool. It looks professional and makes starting your male thread
easier. Use oil, it's cheap. When your close to finishing, use the
cross-feed to clean up both sides of the thread. IN GENERAL- Most
guys I know tend to groan when they hear the subject of
single-pointing internal threads. It is not all that bad. Doing
things like using the largest diameter boring-bar you can stuff in
the hole and sharp tools makes it much easier. A 1" hole should be
threaded with a 5/8" bar. Just remember, you can use as big a bar as
the total width of bar, cutter AND lee-way for getting out the ditch
as you want. I don't like using thin flimsy boring-bars. They are
nothing but trouble. One final point. Single-pointing internal
threads will create a small amount of 'bell-mouthing'. This is a
condition that results from the tool going back over the end of the
work repeatedly. There is not much you can do about it. good luck!
Ron (1101) |
| Threading |
| Great post Ron.
Might I make a suggestion for internal threads? For years now, I
have ground my tool bits backwards, so they cut on the back side,
and run the machine in reverse. This lets you cut from your "ditch"
to the open end. Solves two problems. No more bell mouthing as the
tool starts to cut under load. And the big plus, you aren't courting
disaster by cutting toward a blind shoulder. I haven't crashed a
high dollar threading tool in 15 years! (1102) |
| Evening, Although I
have not done internal threading this way, I found that my model C
did not give me good results when cutting away from the spindle on a
left hand thread. I wound up turning the tool upside down and
running the spindle backwards. Since that worked out well I did not
investigate this problem but obviously lash in the machine was
causing the tool not to track consistently in this direction.
Threading away from the spindle has definite advantages, but this
may be a problem to look out for if you try it. Regards, Tom (1104) |
| Skip has touched on
an idea I read about in a book written by an engineer that worked on
the SR-71 program. In that chapter he recommends turning the
parting-tool upside-down in order to avoid hogging in and breaking
the tool. As a matter of fact he also urged the reader to run all
work clock-wise using the tools upside-down. This has the advantage
of better finishes and no hogging. Since the tool tends to spring in
the opposite direction of force, it is pushed away from the stock. I
always use this method when parting-off. It works very well on small
lathes. HOWEVER, (there always seems to be one of those around!)
many of the bench or precision lathes have RIGHT-HAND THREADS. It is
possible to have your chucked-up work visit the Space Shuttle! The
problem of course can be fixed by figuring a way to pin the chuck to
the spindle. Just HOW to do this without ruining the threads I'm
still thinking about. In the mean-time cutting left-hand threads are
done using the same basic procedures as discussed in the first unit
on threading. The exception being that you cut these threads by
reversing the lead screw so that it rotates opposite to the work,
with the cutting tool moving from left to right. As before in
internal threading, a ditch for starting is necessary. One of the
guys asked me about cutting a left-hand worm for another lathe. I
have to leave for a dig at a quartz mine so I'll answer this later
regards, Ron (1105) |
| Tom and Ron have
both touched on problems with my method of internal threading. I
admit to never cutting an internal thread on my model C. I suspect
the trouble Tom had was excessive end play in the spindle. The model
C doesn't have a very stout thrust washer at the rear. It is,
however, very easy to adjust. You should keep spindle play at an
absolute minimum even when doing normal turning. Excessive play will
result in rough finishes. I would NEVER try to cut off on a machine
with a threaded spindle with the machine in reverse, even if you had
fitted the chuck with a set screw. Doing so WILL eventually result
in damage to things you never knew were damageable! Threading
however, exerts far less cutting force and should be safe if you
install your chuck tightly. The only time I had my chuck unscrew was
with a heavy work piece mounted and the machine reversed before the
spindle stopped. The damage was incredible and instantaneous! (1106) |
| Skip, You got me
curious about that spindle end play so I checked it tonight-it's
about .0015, almost too tight. I suspect the play is in the halfnut
(had it apart, doesn't look the best) as I never have had problems
with other operations. As far as running cutoff tools upside down,
it's been done that way on turret lathes and screw machines almost
forever and does have its advantages. I ran across somebody on the
web who offers a T-slotted cross slide that allows for a rear tool
post as in turret lathe practice. This has the advantages of
mounting the tool upside down while running the spindle forward.
Tom
(1108) |
| In case someone
asks, the slotted cross slide is offered by Andy Lofquist at Metal
Lathe Accessories
http://www.sc-c.com/metallathe/ (1109) |
| Tom, my referral to
not using a tool upside down was meant to say "don't run a threaded
spindle machine in reverse when taking a heavy cut". Sorry if I
confused you. (1112) |
| No confusion, Skip.
I was referring to Ron's post wherein he made mention of the
advantages of running cutoff tools upside down. The rear tool post
just provides a safe way to do that on the South Bend. Tom
(1126) |
| Need a few tips
|
| I've
been trying to turn a single point thread on my 10K and the pitch I
need is 14 TPI. Now, I made sure I was on the "20" gear and set the
left tumbler to the "A" position. The right tumbler all the way to
the right (on the chart these settings line up with "14 TPI"). The
thread that is cut is way too fine though, How can I be sure I have
the right gear in place on the adjustable carrier? I only got the
one on the lathe so I assumed it was original but this dilemma has me
wondering.(3314) |
| The setup sounds
correct. How fine is fine? You gotta make sure you're using the
half-nuts not the clutch for the power long feed. Also, (at least on
a 9") the 20 tooth gear goes on the outermost of the reverse
tumbler. The 40 tooth should be stored on the QC gearbox input shaft
above the 56(?) tooth gear. Attached is a picture of the gear train
from my 9" model A. Any of that help? Paul R. (3315) |
| Paul, First off I
was using the clutch (duh!). I fixed that and tried again. I posted
the photos in the photo section under "single point problem". Now
I'm getting 10 TPI when it's setup for 14. I'm closer, but no cigar,
and I really like cigars! :-) Bill
(3318) |
| Good pictures.
Here's the deal. You've got the two gears on the QC input reversed.
The outer gear (just riding along) should be the 40T, and the one
that's engaged with the 80T idler should be the 56(?) tooth gear.
Swap 'em. (40/56) * 14 = 10. Paul R. (3320) |
| Paul, I swapped that
gear around and everything worked out great! I posted a pic it that
same directory of the successful cut. Bill (3323) |
| Great, Bill, and I
will never divulge how I knew about the power-long feed vs. the half
nuts. Paul R. (3325) |
| Thread Cutting
problem on a 9A |
| After making the first cut and
checking it for accuracy I continue with successive passes using the
thread dial for the proper starting point. For no apparent reason
the tool will advance a half thread and start at the wrong point. I
have checked the thread dial adjustment and everything is fine
there. If you watch it you can see the saddle jump ahead when you
close the half-nut at the point it goes bad and fowls up the thread. Clio (4261) |
| The thread
dial has some backlash, as do the threading gears. And the leadscrew
probably has some end play. This all means that the thread dial
reading will only be meaningful when the leadscrew is engaged and
has driven the carriage (in the correct direction) for a substantial
fraction of a screw turn - which may be several spindle rotations
depending on the thread pitch. Or did I miss understand - did you
mean it jumps at a certain position on the leadscrew every time?
That would suggest a damaged screw and probably nuts too.
Chris (4262) |
| Basically you
have it right. The leadscrew is turning and the threading dial is
turning. When the mark comes up and I close the half-nut some times
the carriage will jump ahead slightly and that is when the tool has
advanced a 1/2 thread and ruined the work. GDS (4264) |
| Jerry, I
think Chris is right. It sounds like the lead screw is moving back
and forth in the gear box. When I did my C to A conversion, I had to
use my old lead screw. The one I got with the gear box was for a
shorter lathe. To fit the lead screw to the "new" gear box I trial
fitted it and then cut down the flat where the gear seats until the
lead screw had very little end play. You can do the same on your
lathe. Just check how far the lead screw moves when you shove back
and forth on the cross slide and add spacers or face off the gear
seat to match. Glen (4265) |
| Jerry, Your
description sounds as though you will find a problem somewhere in
the feed system. As a starting point, I would check everything in
the gear-train starting at the spindle and working my way to the QC
box, then through the lead screw and the apron. Look for loose nuts
and gears, sheared keys, etc. (Have you changed any gears lately?).
I once had a job where one piece was required and it took 3 attempts
because of an un-detected slip of one gear on its shaft. Raymond
(4266) |
| Jerry, I
purchased an old 13" SB with QCGB (c. 1922) which was not wired so I
could not check for screw-cutting, power-feed, etc. When I finally
got a motor and got it wired up and then tried to use the power-feed
on a scrap piece of metal, I found that the carriage would move
along normal for about 1/4 revolution and then, when it should have
been cutting metal, it would rotate 'wildly' with no advance in
lead-screw positioning. My friend Al, an excellent tool and die
maker, helped me investigate and fix our problem. The clutch knob
shaft (star-wheel shaped) was broken at the back end. It was
prevented from falling off but, when pressure (in this case the
engagement of the Automatic Feed Friction Clutch) was applied, the
shaft would rotate but because it was broken, no advancement in feed
was obtained. Although not exactly your problem, you might want to
inspect for broken shafts, etc. BTW, Al made a new shaft out of a
tough metal and there have been no further problems in this regard. Barry (4273) |
| Threads Feeds |
| Do some Southbend
lathes list feed rates in TPI? I ask a seller what the feed rates of
his lathe were and he tells me 4-224 TPI . I say that is the threads
and he says yes threads and feeds. Did SouthBend ever make such a
machine? It's a 13" machine made in 1959. What would typical feed
rates be on a SouthBend of that size and age. Sangtipcoke (6481) |
| They are both
listed in the same matrix bounded by the two gear levers and there
various positions. There are two numbers in each matrix position.
One for threading (TPI) and one for feed rate. Two levers on the
apron are mutually exclusive, one for threading and one for feeding,
longitudinally or crossfeed if the model supports crossfeed. The
seller is correct in the thread range he quotes. As for feed rates
they range fast to slow in direct correlation with the coarseness of
the threads. Ask him to tell you the range of the floating point
numbers. TPI are represented as integers and a couple like 7 1/2. On
the 9" the range from 4 up to I think 7 1/2 require a different feed
screw gear. From 8 on up the same gear is used. Jim (6483) |
| Hi Alan; On
my top lever gearbox, there is a legend that fine feed is 4 times
the tpi. The range on my old lathe is 2 to 112 TPI, for a finest
feed of about 2.2 thou per revolution. Assuming 224 TPI on your
gearbox, this works out to a finest feed of 0.001116 per revolution,
or about 1.1 thou advance per spindle revolution. While my lathe is
a '29 SB13, I think the ratio remains true when they went to double
tumbler gearboxes, as the apron remained quite similar. Threading is
done via direct engagement of the halfnuts, while fine feed is via a
worm and wheel driving a gear train in the apron, giving the 1:4
reduction. Change gear lathes using half nuts for feed (Workshop C
series) typically include some fairly large transposing gear ratios
to get the fine feed down to a nice level. Stan (6486) |
| Alan It looks
like the mid-1950's lathes larger than 10" used a ratio of 1:3, not
1:4 as Stan's older lathe did. My 1958 catalog has a print of the
gearbox index chart for the 13" and larger lathes, and the ratios
are very close to 3. Thus a thread setting of 80 TPI (equals a
thread pitch of 1/80 or .0125") shows a feed rate of .0042"/Rev.
when power feeds are used rather than the half-nuts, in this case a
ratio of 1:2.97. Several other settings produce ratios slightly
above 3. This comes from rounding the exact movement per spindle
revolution to 4 decimal places. In the newer lathes the index chart
shows both the TPI and the feed rate separately, so you don't need
to do the above math each time. By the way, the power cross feeds
are smaller for the same setting, by another factor of about 2.5.
Thus in the above example, a setting of 80 TPI would provide a power
cross feed of about .0017"/Rev. This isn't shown on the chart, you
need to divide. Frank (6488) |
| Threading |
| I'm pretty new to
this forum, but the few days I've been following have given me some
hope with the problem cutting threads with my 9". Don't know what
model to call it as its an A C combined. From what I know, it
started out as a C with a quick change added. A "few" other unknown
worn parts were supposedly changed. The problem is when cutting
threads, after 2nd or 3rd thread it starts gaining and the nut or
whatever will only travel these couple threads and stop. I've tried 6
or 8 comb and 13tpi is only one that will come close to cutting
right. I have a SB parts book and apparently all gears are right.
The bed s/n is 99834, but no idea on what the QC is. As you can see
this is a pretty old lathe and it shows in places. The lead screw
out of the QC has about 1/4 to 3/8 turn slack/slop, could this be
some of the prob? Seems that i can hold the hand wheel on apron and
keep this slop out, still gains on me. Was there a gear spec change
somewhere through the years and maybe I've got to much age difference
between added QC and orig bed, apron, headstock, etc? I've gotten used
to this ole boy and can easily turn out the type work i do even with
the wear, beside I like the dang thing! bill
(6943) |
| I thought
I had a similar problem but it turned out that the compound was
walking allowing the threads to not cut as deep as where I started.
Nut would bind before the male threads got to the back side of a nut
when I tried to put a nut on the newly threaded shaft. Lew (6944) |
| Bill What
gear pattern setup do you have on the end. what stud gear are you
running? By the way, on the A conversion, did they also use a power
crossfeed apron and carriage? Clint (6947) |
| You aren't
describing how you are set up on threading. Set your compound to 29
1/2 degrees. Make sure the bit is centered. I take it you have a
threading dial on your lathe. The marks there are for reference
engagement. As a rule even threads on any mark. Odd threads (13 tpi)
on even marks and 1/2 threads on only one mark (choose a mark say a
number is best and keep the engagement only on this mark). For the
first pass you can take about a .020 cut, then a .010 cut after that
only .005 and .001-.002 on the last couple of passes. Use thread
cutting oil. You need to set the zero on the cross feed dial at the
start. Then make you advancements with the compound slide feed.
Don't know if this is the problem or if with the machine as you
indicate. Tom (6956) |
| Guess I was a
little vague on describing thread cutting prob. Am following SBs
procedure for cutting threads as laid out in their little booklet
"How to run a lathe". (great little book!) And, yes, I do have a
threading guide. I also check with thread gauge and there is where
the gain/separation is really apparent after 2-3 turns. Gears on end
are as shown in SBs parts booklet. 20 tooth stud and 42 tooth
special for the double tumbler gear box. As for the apron/crossfeed,
I don't know if changed from A model with QC, but suspect not, as
"few" other parts changed were implied to be small. Were there diffs
in A C aprons? bill
(6963) |
| Bill, Not
quite sure what you are asking, but just as a general refresher for
all, here's the way the gears are supposed to be from the end of
spindle to the Q/C gear box (see attached picture). From the reverse
tumbler off the spindle we have a 20T (normally) or 40T (optionally)
stud gear on the top of the stud gear. From that we go to the 80T
idler gear. Finally we have the 56T gear-box input gear. The un-used
stud gear (typically 40T) is stored on top of the 56T Q/C gear.
Paul R. (6977) |
| I thought the
42 tooth screw gear was a special, to be used only for cutting 27
tpi [like for 1/8-27 pipe threads] Standard screw gear for a
double-tumbler QC box on a 9 or 10K is 54 tooth. (6980) |
| Sorry guys,
I'm having to many senior moments lately. The gear setup is right,
20-80-56. My computer is a long way from the shop, so I was reading
straight out of SB book instead of looking at my actual gears. Also,
if I would have paused for a moment I would have realized about the
apron having clutch and power feed (mine does) with a QC box. Thanks
for the nice pic Paul, that sure will take the mystery out of what
its supposed to look like. (looks just like mine) Maybe I will
finally get all the info gathered up and we can get an answer. Seems
you guys are headed the way I've been leaning for the problem, a
wrong gear somewhere! Like maybe just one tooth? I sure haven't found
it, hope maybe someone will come up with the answer. bill (6983) |
| Bill, What do
you mean by "it starts gaining"? Do you mean the threads are not
equally spaced? If this is the case, then there is something in the
drive train that is slipping or else you haven't taken up all of the
leadscrew backlash before the cutting tool hits the work. This is
the reason I asked if you were using the half nuts and not the
clutch. It may sound silly but I remember a thread (pun intended,
sorry) where someone was threading with the clutch and guess
what, it slips when the load gets high. It also occurs to me that
the tool could be slipping in its holder or even the chuck if it
wasn't properly seated on the spindle. Glen (6984) |
| Screw Thread
Transposing Charts |
| I'm looking for a
transposing chart to cut inch threads from a South Bend lathe
equipped with a metric QC box. Can anybody help? For those
interested in such issues, I'm attaching the transposing chart to
cut metric threads from an inch QC box. Anthony (8002) |
| I don't have
the actual chart from the machine but I do have a picture in the
1939 catalog showing the chart with gear combinations for ENGLISH
SCREW THREADS using metric pitch lead screw and English transposing
gears, for the 9 inch models. Email me offlist if this will help you
- I can shoot a photo and send it fairly easily - it is only about
1.5 in by 2.5 in. in the book so may not be suitable for a really
large blowup. Can scan in higher resolution and send later but that
will take a day or so. (8007) |
| I've received
several scans of SB charts for lathes with 8 TPI lead screws, both
with and without QC boxes, to cut metric threads. I've also received
a chart for SB changegear lathes with a 3 mm lead screw to cut inch
threads. What I'm looking for is a chart for SB, Boxford, Hercus, or
similar lathes equipped with a metric lead screw and QC box to cut
inch threads. Even though it wasn't what I was looking for, I want
to thank those who sent me the "wrong" charts, your efforts to
assist *are* appreciated. Anthony
(8015) |
| If I
understand you correctly, we already have it:
http://groups.yahoo.com/group/southbendlathe/files/Techinfo/9inchmetricgearing.jpg which includes the chart for
cutting English threads with the metric leadscrew and 127/135
transposing gear and the opposite setup with 100/127 gear. The
conversion math is simple: (mm thread setting on gearbox) *
(127/135) * 25.4 = TPI. I think I wrote that right, but please
somebody double-check me.
Lurch (8023) |
| The formula at least needs an inversion,
since metric threads are pitch in mm (i.e. mm/thread), while English
are threads per inch. So the form has to be: mm thread setting on
gearbox = (constant * 24.5 mm/inch)/(threads/inch) Having said that
I can't make the ratio 127/135 (or 135/127) work out nicely to an
inch thread for standard metric pitches, so this needs some more
work. In any case the solution will need a range of stud gears, for
the same reason that a range of stud gears are required going the
other direction. Frank
(8028) |
| These are
for inch Model B C and for metric Model B C. And I have the chart to
transpose inch Model A to metric. What I'm looking for is the chart
to transpose metric Model A to inch threads. the conversion math is
simple: (mm thread setting on gearbox) * (127/135) * 25.4 = TPI. I
think I wrote that right, but please somebody double-check me.
Actually you got that wrong, it's /25.4 instead of *25.4 and it
equals pitch. You have to invert to get TPI. So let's try it: 3 mm x
(127/135) /25.4 = .11111111 = 9 TPI (not bad). 2.5 mm x (127/135)
/25.4 = .09259259 = 10.8 TPI (not very useful). 2 mm x (127/135)
/25.4 = .074074 = 13.5 TPI (well 1 mm would give 27 TPI). Let's try
for 24 TPI: 1.1 mm x (127/135) /25.4 = .0407407 = 24.5454545 TPI ?!?
1.2 mm x (127/135) /25.4 = .04444444444 = 22.5 (not much help). How
about 20 TPI: 1.3 mm x (127/135) /25.4 = .0481481 = 20.76923 TPI 1.4
mm x (127/135) /25.4 = .05185185 = 19.285714 TPI So, you see, by
accident you will run into a *very few* examples that will work, but
basically you normally have to select a particular pitch, then,
according to a chart, change either the screw gear (the input gear
on the box) or the stud gear to get a precise and predictable inch
pitch. While I haven't seen the metric A to inch pitch chart I
assure you it exists. I've been told it's *very* hard to find. So
far that's proving to be true. As I said in an earlier message, I
appreciate the efforts that various people have made to assist me in
this matter, even though on this point the efforts haven't paid off.
It may be that in the long run I'll have to work out the chart
myself which is complicated by the fact that, when in direct drive
(8 TPI for inch or 3 mm for metric) the drive ratio through the
gearbox is *not* 1:1. I was hoping to get a copy of "the official
chart" because some people get nervous when you provide them a
"non-official" solution. Anthony
(8033) |
| You're right.
It's apparently not an 8mm pitch leadscrew at all. This from Tony
Griffiths' website: "Metric threading Model A lathes with a 6 mm
pitch leadscrew and a screw cutting gearbox had a standard ex-factory
drive train consisting of: 20t, 45t, 50t and an 80t idler. To
convert this gearbox to cut English threads the following gears are
needed: 18, 22, 26, 38, 54, 64, 88, 135/127 compound and a 48/24
compound." Oh, boy, is the math on this one going to be a !@#$^ with
the possibility of two compound gears in the train at once. (8035) |
| I have a
metric Boxford which is the UK version of the SB. I have a large
plate on the base with lots of gears ratios for cutting imperial
threads from very fine to real coarse pitches using the 127/135
compound. I'll copy the details and post if people would find it
useful. David
(8040) |
| That would be
very useful. I think we've established one could drive themselves
quite daffy trying to guess.
Lurch (8042) |
| Frank writes: Having
said that I can't make the ratio 127/135 (or 135/127) work out
nicely to an inch thread for standard metric pitches, so this needs
some more work. It's a mystery to me too, but if you take the metric
Model B C chart and do the multiplication on several pitches it
actually comes out precisely correct. Anthony
(8052) |
| David writes: I have a
metric Boxford which is the UK version of the SB. I have a large
plate on the base with lots of gears ratios for cutting imperial
threads from very fine to real coarse pitches using the 127/135
compound. I'll copy the details and post if people would find it
useful. Yes Please!!! While you're at it, does your lathe have a
6mm, 3mm or other pitch on the lead screw? I noticed Tony's SB web
site states the following: Metric threading Model A lathes with a 6
mm pitch leadscrew and a screwcutting gearbox had a standard
ex-factory drive train consisting of: 20t, 45t, 50t and an 80t
idler. To convert this gearbox to cut English threads the following
gears are needed: 18, 22, 26, 38, 54, 64, 88, 135/127 compound and a
48/24 compound. Can you confirm these tooth counts on your Boxford?
Thank you so much for your offer to assist. Anthony (8054) |
| Lurch and
Anthony I think that starting with the lead screw pitch for a lathe
with a QC gearbox only confuses things. Without any gear changes, a
lathe with metric leadscrew and QC gearbox choices will cut the
following TPI: TPI = (25.4 mm/in)/(pitch setting in mm on the QC
box) This is just turning mm into inches, and inches/thread into
threads per inch. Those aren't real interesting TPI values for most
of the standard metric pitches (like .75 mm, .8mm, 1 mm, etc) Adding
a 135 driven gear coupled to a 127 driving gear in the gear chain
gets us: TPI = (25.4 * 135/127)/(pitch setting in mm) = 27/(pitch
setting in mm) This almost has to be the way the 135/127 gear is
used, since 127 is conveniently 5 * 25.4. So 1 mm pitch gives us 27
TPI, and none of the other reasonably standard pitches give us
anything useful. If we use a stud gear other than 40, we get: TPI =
27*(40/stud-gear)/(pitch setting in mm) This still doesn't help a
lot, since 27 = 3*3*3, so almost nothing cancels. By any chance did
the metric leadscrew lathe come with something other than a 40 (or
sometimes 20) tooth stud gear? It sure looks like things would be
easier with a 140/127 transposing set (yielding TPI = 28/pitch, with
the standard stud gear, and lots of useful combinations). We may be
forced to wait and see how South Bend actually did it. Frank
(8059) |
| The standard
setup on South Bend and Boxford metric Model A lathes used either a
20 or 50 on the stud and a 45 on the gearbox. (80 idler but anything
that fills the space will do.) I believe the above also applies to
Hercus and other South Bend clones but I haven't verified that.
Apparently South Bend metric Model As used a 6 mm lead screw,
Boxford, Hercus, and probably other SB clones used a 3 mm lead
screw. Anthony
(8085) |
| Threading? |
| My model "C" is up and running.
I have turned a few test pieces and checked them. Everything seems
to be in good working order. I turned down a piece of stock to.492"
(matching a 1/2" bolt I had laying around), put on the 32 and 52
gears, per the chart, and started in. First pass was good. My
question is, how do you return to where you started to make the next
pass? My lathe has no threading dial ( if I am correct in assuming
that is what I am missing), so how do you index with one , if you
have one? What do I need to do threading correctly? Rich
(9174) |
| Rich,
Without a threading dial, you have to leave the half-nuts engaged.
What you do is to make your pass; withdraw the threading tool and
stop the motor; reverse the motor and let the tool back-up to the
beginning of the cut; then set the tool for the next pass. While
doing all this, the half-nuts stay engaged all the time. As you can
see, this is a slower process and is harder on your half-nuts. Also,
it makes threading to a shoulder a real challenge! I would get a
threading dial when you can. It will save a lot (including your
sanity). Webb
(9177) |
| Rich If
you d like to make a threading dial there s an article in the files
section with plans. Very handy item to have but tend to be pricey;
even on eBay. Lew(9179) |
| Rich You can
also stop the spindle by releasing the belt tension. This typically
causes the spindle to stop faster (less inertia). Once the spindle
has stopped you can retract the cutting tool and reverse the motor
at your leisure. Keep track of how far in you got by the horizontal
feed dial or compound feed dial (set at 29 degrees), so you can feed
a deterministic further amount on the next cut. Using the belt
tension release you can in fact thread to a shoulder (the thread
dial doesn't make this any easier). Frank
(9180) |
| Rich, if
you do not care what type of threading dial you have ,I bought one
off e-bay for $5 fits a 6" atlas which also has an 8 TPI leadscrew. I
was not going to pay big bucks just to have an original southbend
part as my lathe never came with one anyways. Scott (9181) |
| I assume you
are referring to a "real Atlas Atlas" not a Craftsman or an AA109
which are frequently called Atlases. I could definitely use a
threading gauge for my SB 9" C and have several 6" Craftsman lathes
but the tpi on them is not the same as the SB. (9186) |
| Rich, It
could have been from a larger atlas, but it is definitely from an
atlas, if you want a picture just let me know. Scott(9193) |
| Frank writes: Using
the belt tension release you can in fact thread to a shoulder (the
thread dial doesn't make this any easier). Actually, with some
provisos, it does. If it's acceptable to have a run out groove, and
if you can reliably open the half nuts at precisely the same point
of traverse, when you open the half nuts without retracting the
cutting tool you will simply cut out a relief groove at the end of
travel, the groove having the same form as the thread but without a
helical advance which is what turns the groove into a thread. To
look at it from another perspective, the run out groove would be a
screw thread of zero pitch. Can this be done? Of course!! Hardinge,
Monarch, Pratt Whitney, and many other lathe makers depended on a
variation of this concept. Hendey did it a century ago. In their
cases they disengaged a clutch in the screw cutting gear train to
stop the advance of the cutting tool. I know this appears to go
against everything you've been told (never disengage the gear train)
but they used a forward-neutral-reverse-mechanism in the screw
cutting train that ensured picking up any pitch thread, no matter if
it was in the same system as the leadscrew (let's suppose inch) or
any other (let's suppose metric). If you can reliably pick up the
thread you can afford to "break" the gear train or open the half
nuts. The above concept is superior to opening the half nuts because
it allows picking up any thread, independent of its type, but as
long as you stay within type (inch-inch, metric-metric, etc.)
opening the half nuts at a precise point in the traverse will serve
as well. All you're really interested in is stopping the carriage
traverse at a precise point and being able to pick up the thread for
the next cutting pass. The threading dial, which was the starting
point of this monologue, enables you to pick up the thread after
opening the half nuts. If you want more information on the subject
read "Screwcutting in the Lathe" by Martin Cleeve. And
there have been articles on knock out mechanisms for the half nuts
in ME and some other home machining magazines. Anthony
(9202) |
| Scott writes: if you do not
care what type of threading dial you have, I bought one off e-bay
for $5 fits a 6" atlas which also has an 8 TPI leadscrew. Incorrect,
the 6" Atlas (as well as the 6" AA) use a 16 TPI leadscrew. The gear
on the 16 TPI threading dial has half the pitch required for the
South Bend leadscrew. Anthony
(9203) |
| Screw cutting
at 25 degrees |
| I've
noticed that the majority of users use 27 1/2 or 29 degree setting
for the top slide when screw cutting. After reading George Thomas's
excellent book 'The Model Engineers Workshop Manual' I followed his
recommendations of using a 25 degree setting. He noticed that
sometimes the trailing flank was scored. The reasons were found to
be that on some materials the trailing edge was scored by the
springy chip being deflected into the trailing flank, as there is no
further cut on that flank the scoring remains. By changing the angle
to 25 degrees in addition to the main cut there is also a shaving
cut of the trailing flank. As an additional bonus instead of the depthing having a multiplier of 1.127D ( approx 1 1/8) the new
multiplier is 1.1. Thus a depthing that say requires a 0.08 actual
now requires a movement of 0.088 on the top-slide. Bernard
(9282) |
| Bernard;
Was this for Whitworth (55 degree) threads? The 29 degree setting
does make a shaving cut on the trailing flank for 60 degree thread
forms, but would be completely clear of the trailing flank in the
case of a 55 degree thread form. Stan (9285) |
| Good point
Stan. It would vary from thread types as per the angle of the
thread. Divide the thread angle by 2 then subtract a 1/2 a degree or
1 degree. So for the 60 degree American thread is 60/2 = 30. 30-1/2
= 29.5. I personally don't think this is the problem. I've cut
threads with a straight in feed. You have to do this on multi-lead
threads. At least how I set it up. The threads looked ok. Since he's
cutting cold rolled (probably 1018) its not as much of a factor.
Still, he should set the compound correctly, just to minimize the
possibilities. I think the main problem is with the cutting tool.
Frank, I got your address. I might check tonight to see if I have a
threading tool to spare. What size of shank do you use? I might be
able to spare a HSS. Tom (9286) |
| One more
thing, Frank, check the tool over hang. You want this as short as
possible. Especially with threading. Now being a beginner, you
might ask 'what is tool over hang? Well its the length
from the tool post to the tip of the cutting tool. Hope its clear
and do this with all tooling.
Are you using cutting oil? Get some from the local hardware store.
It does help. Tom (9287) |
| Stan,
You're quite correct, this setting is for Whit threads. I keep
forgetting that UNx threads had a different form factor. (9288) |
| Threading
question |
| Lacking a
thread dial for my 9"C, is it possible to reverse the motor and back
the tool out of the thread it has just cut to index for another
pass? I have a reversible motor I can mount to my countershaft that
I thought about using if this is possible. Rich
(9938) |
| Rich,
Reversing the motor to set up for the next cut is the one thing that
ALWAYS works. Glen (9941) |
| That is the
basic procedure in "how to run a lathe". (9943) |
| Rich Yes, but
DO retract the tool from the cut first as the slack in the halfnuts
will shift the tool over on reverse and destroy the good thread.
Also, threading to a shoulder is nearly impossible as the motor and
drive will take a few extra revs to stop before it's ok to reverse.
Working close to the chuck is out of the question. You really ought
to plan on getting the threading dial. Learn to use the topslide
(compound) set to 29 degrees so that the cross-slide is always set to
zero for each cut. The topslide is then only used to set the infeed
cut. Also easy to remember the last setting. When you get good
at threading, stopping a cut at a given point by whipping the tool
out of the cut before stopping will make you shine. RichD (9946) |
| Rich, use
the belt release to stop the spindle. Without the angular momentum
of the 1750 RPM motor, the spindle will stop quite quickly,
especially at a low backgear speed. The spindle/leadscrew
relationship is maintained, so this works fine for threading. This
works without a threading dial (stop the spindle with the belt
release, reverse the motor at your leisure, re-engage the belt). It
is also a good way to stop at a shoulder even of you have a
threading dial. Frank (9949) |
| Rich; No
problem doing it this way, without a threading dial or for certain
weird threads that don't map to the threading dial it's a standard
way to do things. Just be sure the lathe stops all the way before
reversing and your speed is low enough the chuck isn't inclined to
unscrew. Done threads this way quite a few times. Stan (9951) |
| I've done it
that way at that day job many times. On most lathes it's the only
way to cut a metric thread, and usually easier then dealing with the
transposing gears to get the exact pitch. we have an Italian lath at
work and it's set up with an American lead screw and ratios so when
we do metric threads for the Japanese injection mold machines we
have, that's how we do it. Kerry (9967) |
| Threading on 9
A |
| My " How to run a
lathe " does not say. Do you use your back gears to thread?
I've tried everything I could think of and is still not coming out right.
Have read all kinds of books, and none of them say. Could anyone
give me i.g. on this machine to try? Used thread dial, and will not
cut in the same place. Rob
(10413) |
| Rob, A few
things that can mess you up: Have the apron feed selector lever in
the middle position. This disables the power long- and cross-feed
and allows the half-nuts to engage. Use the threading dial if you
have one, else leave the half-nuts engaged and just stop, back out
the cross, and reverse the motor to get back beyond the starting
point. When threading, you are using the HALF- NUTS lever (pull up
to engage), not the power-feed star-knob! Don't ask me how I know
;-). Make sure your lead-screw and half-nuts are in okay condition.
At this point you should be able to chase the same thread. If not
there is slipping between the spindle and the lead-screw. After you
are chasing threads, you should be able to verify the TPI by lightly
cutting the surface and then inspect it with a fish-tail thread
gage. Paul
R. (10415) |
| Yes, use your
back gears. Run real slow. On brass and aluminum (which is mostly
what I use) I use slowest speed and lots of cutting oil on brass and
kerosene on aluminum. Feed with the compound properly set.
Frank (10416) |
| Like Frank
said, use the back gears. This is especially crucial on coarse
threads. The back gears have nothing to do with the rate of feed per
revolution of the spindle, it just makes it easier to start and stop
at the right place when the spindle is turning slower. It is also
helpful to use the belt tensioner to stop as you don't have the
rotational momentum of the motor and pulleys to keep the spindle
turning. Unless I am making a very long thread, I always reverse the
motor to start a new cut. I learned to make threads on a Sherline
with no half nuts or thread dial. I guess I just haven't learned to
trust 'em yet. Glen (10417) |
| Gentlemen
Half the battle with threading and thread dial indicators is
visualizing WTHIGO. Many years ago I was given the following
exercise which helped a lot:- 1) Set up the lathe so the cutting
tool runs away from the chuck. 2) Use a bed stop to fix the start
(chuck) end of the thread. 3) Fix a piece of paper firmly around a
reasonable size mandrel and mount in chuck. (Make sure the paper
can't slip and rotate round the carrier) 4) Replace cutting tool
with a sharp pencil. 5) Put a temporary rotation position mark on
the chuck (use tape, spirt pen, wax stick or whatever you have) 6)
Set-up to cut a thread of the same tpi as the leadscrew and lowest
backgear speed. 7) Experiment taking care always to start
"threading" with the saddle hard up against the stop. Replace paper
and sharpen pencil as required. Marking the chuck and always
starting from the same point on the bed makes it easy to see how the
thread dial "calculates" the relative positions of chuck (rotating)
and tool (longitudinal). The pencil marks make it much easier to see
what happens when the half-nuts are engaged in the wrong position.
Cutting away from the chuck means that you don't have to worry about
dropping the nuts before a smash-up. Once you have got things
straight using a fixed starting point, try starting a bit further
down the bed watching the rotational position of the chuck and the
thread dial position at the half-nut engagement points. It should
soon become clear how the thread dial automatically compensates for
saddle offset allowing the chuck to turn further so that the tool is
still in register with the thread despite starting off further down
the bed. See if you can use different dial graduations and still get
the right engagement point. When the thread tpi matches the
leadscrew there is plenty of choice! Now try it with different
threads (hint:- the relative multiple between leadscrew tpi and
thread selected has an effect on which combinations of graduations
it is safe to use). An hour or so of playing should make clear what
is going on understanding the mechanics is then quite simple. If you
really get stuck (I did!) drop the belts and turn the chuck slowly
by hand which should make it really clear how the dial graduations
select the correct tool to thread registration when the half-nuts
are engaged. Clive (10419) |
| Clive
Absolutely excellent explanation of how to do it for good results.
Now here is one that belongs in the FAQ. JWE (10421) |
| This is
exactly why I most often reverse the motor to start the next cut. I
assure you that I am not stupid but I choose to think hard on those
problems which require my intellect and not those on which the
flipping of a switch will make go away. Glen (10434) |
| Rob, As I
re-read your question, it dealt specifically with the back gear. The
normal gearing is set so the motor spins the jackshaft and then that
in turn spins he lathe spindle which in turn spins the gear train.
with the back gear disengaged, the lathe is running full speed or
high gear. the gear train starts on the back of the spindle and runs
through some different gears. on a Model C or Workshop, you must
manually change the gear ratio to get the thread correct. the label
on the gear cover would indicate what the proper gears to be used
are. The back gear comes into play as regards to lathe speed. I
call it low gear. the motor spins the jackshaft which spins the step
pulley which spins the back-gear which spins the spindle. everything
is in slow motion, but all the thread and feed ratios are unchanged.
Of course you need the stack of gears to know change them. Also as
was pointed out, the half nuts are the ones to use, not the feed
lever. And regards the thread dial, always start at 1 regardless
where you need to put the carriage or thread pitch and you will find
it doesn't matter for the rest of the thread pitches, or leave the
half nuts engaged and feed the carriage back and forth by reversing
the motor. Dave
(10443) |
| Now that's
what I call a fine thread |
| I had
always considered the ultra fine thread tpi settings on my Heavy 10
gearbox (max tpi is 480) a pure side effect of the fine feed
provision and merely marked up for the sake of completeness and
possible satisfaction of some obscure American humorous imperative.
Not anymore though. In the April issue of Laser Focus World Azeer
Enterprises are offering 200 tpi adjustment screws and micrometer
heads. That's 5 thou per rev. And the thread depth is probably less.
How do you cut a thread that fine and for that
matter manage to start the nut without cross-threading the beast.
I've known machinists who would consider that an acceptable finish
for non precision work! I presume that no one has screw cut a
matching nut and stud at 480 tpi. Or have they? Clive
(10869) |
| I think that
first, I would consider rolling said thread, and I would definitely
put in a land to allow the parts to align before the threads engage.
I am impressed with a 0.005 " feed per rev. that means the
thread
width must be less than half of that or 0.0025" and that thread
depth is probably also less then 0.005" one bad thing would be to
have a dull cutter and try to run it a second time to clean it up !
Dave (10878) |
| My SB9 has
always been set up with the following: A 105 tooth turning gear,
engaging a 2:1 reduction gear, engaging a 6:1 reduction gear, driven
off the spindle (24 tooth). This gives a feed of 0.0024" per turn of
the spindle. I use this for feeding not threading. Jim B. (10880) |
| Thread size
|
| What thread type would have been used for the tapped holes in a 1944
10in heavy. Being an Englishman in his 30s I've only ever known
metric.
(11309) |
| Your
best bet is to order the parts manual from Rose at LeBlond.
Immediately after that go to the group site and print off the SB
publication that gives the common fastener size by SB part number
used on the machine. I think most have a common English size
equivalent, it was most helpful to me. At one time the screws and
bolts originally on the machine were hardened and, I think, blued.
Is non-metric hardware even available in most UK hardware stores
these days? Bill (11399) |
| I've
heard of metric threads, but am not sure how they differ. Being an
American manufacture of the pre-metric days, I would be quite sure
they are standard SAE screws. Measure the OD and convert to inches
and then count how many threads there are in one inch. the shopswarf
index lists the different threads so you can look up the sizes. in
the States, we use the outer diameter of the major diameter
(converted to a fraction or an inch) for the first number. 0.5
inches is 1/2 inch. then the number of threads per inch as the
second. if there were 13 threads per inch the bolt would be 1/2-13
then the length of the thread diameter. so if the bolt were 2 inches
long, the head would add another 1/4 or so to the overall length. so
a 2" long bolt would really be about 2.25 or so long. Not sure if
that if any great help, but if you run into American threads often,
look for a thread gauge. probably less than 20 pounds. Looks kinda
like a set of feeler gauges, but with teeth. Dave
(11400) |
| Threading
question |
| How do you consistently stop at the same place for
internal threading? I don't like to machine a 'finish slot' at the
end of the internal thread for reasons of integrity; if the wall
thickness is thin to begin with, I figure having the slot there will
make it weaker. I've been getting by with just idling the threading
bit at the end of the cut, and trying to bring it to a stop at more
or less exactly the same place... but noticed that if I mistimed the
feed disengage by a little bit, the tip of the fragile carbide
threading bit will get broken off as it hits the 'fresh' uncut tube
section. At first I tried watching the cutting bit, as you do on the
outside cuts; but it's too difficult leaning into and over the ways
since my machine is large and thick and the ways are right around
elbow height... now I just set the carriage stop on the ways at the
end of the cut, but don't lock it down; it's only there as an
indicator to tell me when to disengage. I use super low speed to
allow me time to disengage quickly; maybe 25rpm or so. Any other
ideas? Or am I completely off base? Also, I've noticed that when
using the normal tables, for instance the Atlas Manual of Lathe
Operation, Table I, Depth and Double Depth of Thread for National
Form and Vee Form tools... my numbers don't seem to be anywhere near
what the manual indicates. For instance, at 32tpi, the depth of the
compound feed (set at 29 degrees) according to this manual should be
.027" for a Vee Form tool... but when I true up a cylinder and do an
actual cut, I've noticed that I have to feed the compound something
closer to 0.85" to get a complete "V" cut (not a 75% finished NF
semi V)... why such a huge difference?? I've indicated my compound
and it's turning in the proper amount, so my actual figures aren't
off.
(14259) |
| Actually, the
part should be stronger with the slot cut into it. Threads are a big
time stress raiser. If the material is too thin for an under cut,
then its way to thin with threads. The thread depth isn't much less
than the undercut. Also put a 45 degree internal chamfer. It helps
keep burrs on the inside from happening. Tom (14261) |
| You don't
just stop the thread hard. You can do a few things. You can use a
hand crank on the spindle and when about 1/8 turn from the end of
the thread move out of the cut. You could run the lathe in reverse
and cut out into open space. You could use a tap :-) or make a
piloted tap if this is a special application. The run out area isn't
going to weaken the tube much more than the threading will, provided
you make sure you radius the corners to avoid stress risers.
Stopping the V thread abruptly is going to give you the worst
combination of stress risers, with two sharp corners plus a sharp
point where the other two risers meet. You can reinforce the area
where the internal threads end by leaving a bit of material turned
to minor diameter of the threads on the piece that screws in. If it
fits well within the tube, this will move the bending into the part
of the tube that is full thickness. As to depth of cut: Sounds like
you are not accounting for a few things: The depth of cut will be
equal to the indicated depth on the compound times the sin of the
compound angle. Add in the variation from sharp to NF form and you
get pretty close, assuming that "0.85" is a typo, and that this
number should be "0.085" 0.085 inches * .485 = 0.041 inches 0.0412
inches * 0.75 = 0.031 inches. The remaining 4 thou is probably
thread roughness, backlash, tool spring, variables in how you
measure the cut thread, the usual shop gremlins. Stan (14262) |
| I cut inside
threads on the back side of the part with the tool held upside down.
You can see what you are doing without leaning over the lathe and
getting caught in moving parts. Glen (14263) |
| Stan, those
are excellent suggestions. Thanks. I didn't really think of
reversing the lathe, but I have a gear drive and at that speed it
sort of comes to a stop very slowly; I don't like reversing while
it's still moving and although there is a large handle for
disengaging and braking the spindle, I haven't yet tried this in
connection with any threading operation as I tend to not want to
move anything until I'm satisfied the thread's complete. I'm sure
you know the feeling. I'll experiment and see what happens. I do use
a chamfer, but I've tended to use a 60 degree included angle instead
of 45 degrees (makes it 30 actual degrees) as it's convenient...
it's the same angle as on the indexable cutting bits. Is 45
significant for some reason? With respect to my odd numbers, you're
right; it was a typo, supposed to be 0.085" ...but the sine portion
is off. I took that into account, as the Atlas table has a normal
column for referencing the pure cross feed number, and another
column for use with the cross feed when set at the normal 29
degrees, which takes into account the angle. I know what you mean by
slop, but this goes way beyond and I still can't figure it out.
walt
(14264) |
| Or cut them
on the back side with the spindle in reverse. You need a reversed
threading bit to do this. Ed (14265) |
| And
preferably a *NON* threaded spindle nose (i.e.. D-Series, A-Series
or L-series). Scott Logan (14266) |
| I
always make the bottom relief groove, only as deep as the thread.
Set up an indicator on the bed to follow carriage movement and set
to zero at the beginning point in the slot. Yes, you will have to
run in reverse as the tool has to be at the rear side of the cut to
get a RH thread from in to out. Release the belt tension before
starting the motor and ease the spindle into motion for a cut. Or
use a front facing tool and slow speed watching the indicator for
when to kick out the half nuts. Still with a good bottom groove to
stop in. The depth of cut is highly dependant on the width of the
tool tip. Getting this exactly right is not practical, so I always
use a gauge piece to test the "class of fit" I want. RichD (14267) |
| Cutting
threads without a thread dial |
| How
do you cut threads without a thread dial? I've seen this come up a
few times lately. Here's a trick an old-timer taught me years ago. Set
the thread at the q/c box or install change gears for the required
number of threads, set compound at 30 degrees, "zero" the crosslide,
all like you would with any thread. Now move the carriage stop to
the tailstock side of the carriage (if cutting right hand threads).
With the stop in contact with the carriage, lock it down. Now engage
the halfnut and make the first pass, disengage the half nut and
pullout at the end of cut. Return the carriage to contact the stop,
reset the crosslide to zero and set depth at the compound for the
next cut. Engage the halfnut again and it will connect to the
leadscrew at the exact location as the first cut, continue this
until the threads are complete. Try this, it really works. Rick
(14510) |
| Hey thanks, dude
when I get my compound set up and a toolpost to where I can try that
I surely will. (14511) |
| Rick, that
only works for threads that are devisable by the leadscrew pitch and
the stop is not necessary, just convenient. Set the top slide to 29
deg. for a better finish. I single point thread thread often, inside
out, L R hand. RichD (14516) |
| Rich The
stop is more than convenient, it sets up a relationship between the halfnut and the leadscrew. If you have a thread dial, try my method
and it will only let you catch the leadscrew on one graduation on
dial 1, 3 ect. This way you don't have to worry about if the thread
is a multiple of the lead screw thread. 29 degrees works great the
tool just rubs on one side and cuts on the other. My method would
work on left hand or right hand, you would have to put the stop on
the appropriate side. Rick (14527) |
| Its
fairly obvious that this particular method is not infallible for all
possible thread pitches but I don't know whether any of the pitches
for which it fails correspond to practical threads that are actually
used for anything. Personally I'd be suspicious of any pitches
divisible by three but, without working it out, that may be pure
"off-the-cuff" prejudice. What will make it infallible is to put
some sort of witness mark, e.g a paint splash or piece of tape, on
the chuck and only engage the half-nuts as the mark comes towards
you past a suitable reference. Were I to do it I'd set things so I
engaged the half-nuts as the mark comes up to the bearing cap bolt.
This manner of working fixes both the chuck rotation angle and tool
point longitudinal position at the moment of half nut engagement.
Effectively doing this is the same as using the old Exe lathe with
its single point dog clutch to engage the lead-screw which, I am
given to understand, is infallible at all pitches. Marking the
position of number 1 jaw on a three jaw SC chuck is a good idea
anyway. If you have to replace work in the chuck its more likely to
go back concentric if you align it with the same jaws and use the
same pinion to tighten up the chuck so a ficidary mark is a great
help. If the chuck is well past the first flush of youth using
different pinions for tightening can make a big difference. I had
one that would co-operate only when number 2 pinion was used. Some
lucky folks have the jaw numbers stamped on the outside face but
mine have all been numbered inside the jaw slots. Clive
(14542) |
| I don't see
what the big deal is. There are plans to make a thread dial in the
files section of this site. The methods sound a little like the
"mousetrap" game my daughter had years ago. I use a thread dial from
a Sheldon lathe I picked up very reasonably. JP (14553) |
| Rough External
Threads |
| Why? I'm following all the directions (I think)! Doc (15900) |
| This is a
case where a picture is worth a thousand words I think, you got a
digital camera? I'm no expert, in fact I never tried to cut threads
yet with my lathes, might be able to help though. (15902) |
| Doc, are
you closing your half nut at the same place on each cut? Do you have
a thread dial on your apron? If so start each cut on a line or
number on the thread dial. If you are not doing this it could be
what is causing your problem. Duane (15903) |
| My first try
at threading was not my best either. Most failures are due to
inattention or accumulated error. Start with a sharp tool. Be sure it
is on or a touch below center. Three degree rake. Be sure your not
going to fast, put her in grandma, (back-gear). Use light cuts with
oil. Each cut should be no more then .005 per pass. Every 3rd pass
is a spring cut. Toward the end of the job dial straight in .005.
This cleans up both sides of the thread at the same time. Makes the
threads look professional. I leave my threads a bit tight and lap
them to an accurate polished fit. This is easily done with a strip
of 600 silicon carbide wrapped around a 3-cornered file. This way
the nut or mating part has a light drag for a fit. Nut'in 2
it! Another way to lap threads to a high finish is to take a nut the
right size and split it. Add some 600 clover-leaf and place a clamp
on the lapping nut. As you go over the threading job put a slight
tension on the nut. It's a cheap shot but it works. As for making
shims, I'd get a sheet of .001 S.S. shim stock, cut it in strips and
glue them together with some kind of wood glue. Put a 5lb. weight on
them for 3 hours. You should be able to peel them back as needed.
Experiment with different glue for best results. Finally, a word
about the Order of the Great Iron. We are preservationists plain and
simple. It is our passion to save these wonderful machines. They are
a living history in a world of virtual nothing. We sweat and swear
and sometimes bleed as we fix and use them. BUT THEY ARE REAL. As
real as the loss we sense each day we go to work and watch the world
go to hell on a Pentium chip. Our fathers are dying a thousand a
day. That's the rate of death among WW II veterans. Machinists of
our kind are not far behind. We have now become nostalgia. Real magic
is not a computer. It is what happens when someone runs these
machines and makes steam engines, watches or telescopes. We are
wizards that dare to dream about how it was. Ron
(15904) |
| Doc: What
material are you using? Ron (15905) |
| I found that
if I was running the spindle too slow I got crappy threads. Had much
better luck running at pretty much the same speed I'd run if I were
doing normal turning. Now Ron has stated to put it in backgear, and
it sounds like he has 'a bit' more experience than me (like in I've
cut threads all of 3 times, whoop-de doo) so take my advice for what
it is - just another thing to try. It also seems to me that after
running a nut back and forth on the piece a few times, the thread
'cleans up' a bit. Keep at it, it is very satisfying when it works.
John
(15907) |
| Make sure the
compound is set at 29ø to the right. Use sharp tool, correctly
ground, or a formed threading tool. Cutting tool at center height.
Feed with the compound, leaving the crossfeed at the same setting
for each cut. Use the back gears - slow speed. (very slow for
stainless) Engage half nuts at the same number on the threading
dial. (Any mark for even threads, numbered mark for odd threads.)
Use a good thread cutting fluid. You may make the first couple cuts
at about .010, but then reduce the amount you feed the compound. You
may be increasing the depth only .001 to .003 for the final cuts.
Make your last two cuts at the same setting for clean-up. Jim
(15908) |
| Doc,
Crossfeed set at 29.5 deg? Advance cut with the compound not cross
feed. Object is to cut one side of the thread not both sides at a
time. Larry (15910) |
| Thanks for
all the tips I will retry and recheck everything. I never bothered
to understand the gradations of the thread dial and just always
engage at a real number not a line regardless of the thread count.
The roughness is the same in steel or 6061 I'm using a 60 degree
point on a insert tool on a Phase I quick change and using back
gears (14 tpi requires the leadscrew run too fast for my reflexes at
a regular speed) (15913) |
| Robert If you
are using the chamfering bit that comes with most tool sets it will
always make rough threads because it is not designed for threading.
The style E tool despite being advertised for threading is not
usable for threads finer than about 8tpi. The specific ER and EL
cutters will produce fine threads every time. JWE (15915) |
| Thanks. If
you use the ER or EL do you still sent the compound to 29.5 ? (15916) |
| Robert Yes,
or 29 or 30 degrees using a machinists protractor. The most
important is the form of the tool that forms/cuts the thread. The E
style tool when measured as a threading tool is a 120 degree tool
not a 60 degree tool and that is why it cuts lousy threads. JWE (15918) |
| The ER
took care of the persistent rough threads. Thanks (15924) |
| Threading
drill rod? |
| Can anyone tell me
how to thread oil hardening drill rod w/1/2" die? I tried it as is,
and I tried heating it to a red heat and slow cooling it but all the
die does is grind the end off. It was a good die when I started but
don't know about it anymore.
(15985) |
| O1 is a
bugger to thread with a die, try putting a chamfer on the rod first
and use a name brand, "USA made" sharp, adjustable threading die.
Some dies are for rethreading or thread chasing. They are generally
non adjustable. Some Asian made dies are made from an indeterminate
grade of steel. J (15987) |
| Thanks JP.
I'll try chamfering the end but I don't hold out much hope. I used a
craftsman die and it has never failed me before. Not adjustable
though.
Next time I'll use some other material!
(15992) |
| I forgot to
mention the end of the rod should be turned square first, not just
cut and the chamfer can be small, 1/2 thread width will do. For
1/2-20 a .025" chamfer will work. The A1 and W1 material is worse,
leaded steel like 12L14 works out nicely, for stainless 304 isn't
too bad to thread. JP (15997) |
| Can you cut
finer than 160 TPI on a 9" C? |
| I'm interested in
making very fine finishing cuts and would like to "trick" the lathe
into moving the carriage at 200, 300, or even 400 TPI. Is this
possible with a model "C"? If so, how? Best regards, Keith
(16819) |
| Keith,
Really, you need to learn about radiused tool tips. The normal fine
feeds should be adequate. If anybody is fussy about tooling marks,
that's me. Anything finer requires grinding and polishing.
RichD (16820) |
| You need to
keep your eyes open for compound gears and the largest "turning"
gear you can find. They do show up on e-bay. I have a 2: 1 and a 7:1
and a 105 tooth turning gear. (These are the old 20 DP however) I
have seen similar ones on e-bay. Starting from a 24 tooth gear (the
spindle) then you get 105/24*2*7 = 61 turns of the spindle for every
turn of the lead screw or 0.002 movement of the tool fro every turn
of the spindle. With 18 DP gears (standard for SB-9 after 1936) you
could put a 16 tooth gear on the reverse gear drive shaft. However
the diameters of the gears are bigger. For the reduction you want
you would need to go to a triple compound. You could probably get
one more reduction if you picked the steps right. You would then
have something like 105/16*2*5*5 = 328. I am just guessing at the
5:1 I am not sure that its available I haven't paid much attention
to the compounds on e-bay. I just felt that you might need some room
to get the third compound gear in the train. A alternative is to use
Boston Gears 20 DP. They sell adapters to make compound gears, see
page 16 of the open gearing catalogue. (Remember the spindle and the
three gears on the reversing gear assembly are 20 DP only the gear
you bolt on the stub shaft of the reversing gear assembly and the
drive and idler gears are 18 DP. With the Boston Gear compound
assemblies you would need to make new bushings to mount them to the
spider. The bushing has a standard ID of 0.4375. Od is 0.625 Could
be bored out a bit but I doubt 0.5625 is achievable. Also you would
need to adapt the drive gear. Your lead screw is 0.5625 the Boston
gear Gears are 0.625. Still if you want all that reduction its worth
consideration. My 9" is older and did use 20 DP. I still use them
for turning. I am trying to adapt to 18 DP for thread cutting. Jim
B (16821) |
| This brings
to mind another question. Has anyone ever cut matching 460 TPI
threads? What is the finest matching threads someone here has cut,
even if just for the heck of it? JP
(16824) |
| JP, I have
single point cut steel .200mm x 120 TPI threads for screws for a
model. I still have them. I case you ask, it was not done on a SBL. RichD (16827) |
| I just went
down an looked at the possibility of a third compound. I was able to
get the 5:1 ( I said 7:1 its actually 5:1) to drive off the reverse
gear assembly. In one position it drives off the button of the 3
gears in another off the middle. I was able to also get the 2:1 on
the top arm of the spider to engage the 105 tooth gear without
locking the 5;1. That leaves room for a third compound, hung off the
bottom spider, but on an extension arm, to give you additional
reduction. You would need to rig up an additional to allow the small
gear of second 5:1 (or whatever) to engage the big gear on the first
5:1 and simultaneously have the big gear engage the 2:1. Also I
think you might need the 20 DP's to fit. Jim B. (16829) |
| The fine
finishing cut is done with HSS and a polished bit. Take the grinding
marks off with a hard stone or soft wheel and polishing compound.
View the bit under magnification and you will see the grind marks
clearly. A polished round nose HSS bit will do it. JP (16830) |
| My model C
chart goes down to feeds of .0021 which unless my math is faulty
translates out to 480 threads per inch. That is with at 16 tooth
stud gear, 80 tooth screw gear and both the 54x18 and 72x18
compounds. Also available are 200, 240 and 320 tpi. This is with the
standard set of gears for a C. Note that the B does not have the
necessary gears in its standard set it is missing the 54x18
compound. Actually I have had my C setup in this range for years but
I agree that a rounded tool bit is required for really fine
finishes.
(16834) |
| The threading
chart at the bottom of this page
http://www.lathes.co.uk/southbend9-inch/page6.html
shows .0021 being 160 TPI with the gear setup you just described
(fig. 4). Keith (16841) |
| The 0.021
feed is through the worm gear and friction clutch in the apron of an
A or B model. This doesn't apply to a C model. Glen (16844) |
| Consider a
shaper with a spiral cutting attachment. JP (16859) |
| You are
absolutely correct. I calculated the pitch wrong. (I should not
calculate after a glass of wine. There is no need for the triple
reduction the double does it fine. Jim B. (16861) |
| Can anyone
give me the parameters for the fine finish tool. i.e., angle of the
"V", radius for the front, and the two relief angles behind and
below the cutting point. I am assuming this thing goes in
perpendicular (with the uncut part of the tool parallel to the bed)
at center height on very slow feed (how fast can you get away with
with a .005" finish cut), with good chatter free support. I am a
newbie who has just learned to do fair roughing cuts. Good at honing
tools from woodwork. Steve (16865) |
| Steve,
Depends on the material. For all materials, front and side clearance
angles of around 5 to 7 degrees. For brass top rake of zero degrees.
For steel top rake of 5 to 7 degrees. For aluminum, top rake of 12
to 15 degrees. A good starting point for tip radius would be 1/16
inch. You can increase this up to the point you start to get
chatter. Stan (16867) |
| Stan has
contributed an EXCELLENT write up on tool grinding a few years ago,
the info hasn't changed. it's located in the files section for the
group. Its definitely a good read. I recommend it as a reference. dennis
(16868) |
| Ah shucks :-)
Glad you found it useful, and thanks for the kind words!
Stan (16871) |
| That is a
Model B chart which takes the reduction gearing on the apron into
consideration. The C has a different chart as it only has the half
nuts to drive the carriage. Therefore the extra compound gear that I
mentioned.
(16873) |
| Proper
surface speed is also important to a good finish. There are charts
in HTRAL and the machinery handbook covering surface speeds for
various materials. Finish cuts are typically .001" to .002". For
insuring chatter free operation you should have the most rigid tool
holder and largest tool bit you can use for the job. JP (16877) |
| Those turning
gears are good, I have a 116 tooth turning gear for my B model. It
will put a satin finish on steel even with a regular tool. RC
(16885) |
| Posted today on
E-Bay are the following, 2592963342 1:6 108/18 tooth compound gear
2592958249 116 tooth turning gear. Jim B.
(16964) |
| Threading
Questions |
| Is there a
quick and easy method of determining the thread depth of different
diameters and threads per inch? This is a bit lost to me. I can cut
the thread, but don't know when the thread is at the right depth
(minor diameter). When using a threading dial on the lathe, if I
remove the part I'm threading and decide to go back and thread some
more using the same thread I've already started, can it be done? I'm
not sure if I'm stating these questions properly. Dave
(17135) |
| If you can
get the exact same location as you had you can do it. A very small
position error shows dramatically. That is why the short answer is,
No. A thread gauge is good for this problem. For common threads you
can use a matching nut. Another method is with 3 wires and a mic or
caliper. Wires of known sizes are placed into the threads and the
overall OD is measured. Its similar to the "rods in the dovetail"
and is in the machinery handbook. JP (17136) |
| Dave, There
are different ways to measure threads. The cheapest and curdest is
as mentioned with a common hardware nut. A similar approach is with
a thread gauge of a go-no go type. Gauges are used for Pipe threads
though, but I think you are referring to standard straight threads.
Threading micrometers are made to measure OD threads. For ID threads
I'd probably use a gauge. They do make ID measuring units, which I
haven't used. They are expensive. The micrometer attachments to
measure ID threads, haven't worked for me. As far as OD threading
mics, they make fixed anvils and replaceable anvils. The fix type
are good for a specific thread range, say 0-1 inch diameter and
18-24 tpi. Just as an example. If your needed thread is 3/4-20 (just
for example) then said mic would work. The replaceable type, you
have a mic of say 0-1 inch or 1-2 inch etc. capacity and you change
the anvils to suite your needed threads per inch (TPI). They cost
initially more. You would save in the long run as the anils will
cost less than buying a bunch of threading mics for various thread
sizes. I think there are some sets made in Poland of the
interchangeable mics. Should be of very good quality and reasonable
price. $169 sticks in my mind, but don't quote me on that. You have
to adjust or calibrate the mic each time you change the anvils. If
you are only doing one type of thread, then the fixed type should be
the way to go. Now if you take out the part you were threading, for
whatever reason, you can pick back up the threads. Its really not
very difficult. First setup as you would for threading, i.e. feed
rate for the specific thread, the threading tool set up to cut. Keep
the threading tool on a plane (this is hard to explain) or larger
diameter than the threads. Turn on the lathe and take basically a
false cut. I.E. start at a distance before the threads, engage the
threading lever. Then when the threading bit is near the part, stop
the lathe (turn it off). The lathe will still turn a bit and advance
the threading tool. After the lathe has stopped, use the cross feed
and the compound (set for threading angle ie 29-30 degrees for a60
degree thread), and move it into match the thread groove. Set the
dials on the crossfeed and compound feed to zero. Back out the
cross feed. Take the tool to the beginning to where you will start
your cut and resume to cut threads as usual. I would turn in the
cross feed back to the zero point an take a cut and not try and feed
in more of a cut on the initial pass. Hope this is clear. Tom
(17139) |
| I thought there was some kind of mathematical formula that would
take into account the diameter and the tpi and would give you the
depth of cut. I have a thread gauge that I've used, and also have
used a nut. Dave
(17142) |
| For "full
depth" 60 degree threads you feed into the work a distance 1/2 the
pitch. If the thread is 20 TPI, the pitch would be 1/20 or .050 inch
per thread. The thread depth would be half that or 0.025. Most of
the time threads are not cut "full depth", I think 75% is common,
but the depth still should be proportional to the pitch. Glen (17143) |
| Dave Further
to the previous excellent advice don't forget that the theoretical
thread depths do assume that you have managed to grind the thread
cutting tool to the right profile with the correct width flat on the
end. Most mortals, like moi, end up with near enough rather than just
right. I find that cutting theoretical full depth turns out about
80% depth which usually works out just fine. Don't forget the spring
cuts (successive cuts with the same tool setting). Frequently you
will need two or three extra runs to clear right down to the size
you have set. This often makes the difference between too tight and
too loose! If you are not sure about the thread dial its well worth
doing some mock cuts with paper wrapped round a suitable round bar
and a pencil substituted for the tool. Then you can see what is
happening. Good practice for if you ever do need to pick up a thread
again. Clive
(17145) |
| Dave, Just
the tpi is all you need. (1/tpi)*.5 = cutting depth Take a 1/4-20
thread and look up the tap drill needed, .201" Ok, so 1/20 = .050"
to lathe cut the thread its half of that. Offset your compound to 29
degrees, 61 from the axis of rotation and then its .025"/.875 (sine
of 61 degrees or the cosine of 29 degrees) the depth on the compound
dial is .029" Take a spring cut or add a bit to make up for tool
pressure. JP (17148) |
| Is there a quick and easy method of determining the thread
depth of different diameters and threads per inch? This is a bit
lost to me. I can cut the thread, but don't know when the thread is
at the right depth (minor diameter). When using a threading dial on
the lathe, if I remove the part I'm threading and decide to go back
and thread some more using the same thread I've already started, can
it be done The book that came with my lathe (Atlas/Craftsman) has
tables to show the reading of the cross slide for standard threads,
both the absolute measure and the compound reading when the compound
is set at 29.7 degrees. And yes you can pick up a thread after
removing the stock from your chuck. Get the tool resting in the
thread and work from there. That is one advantage of the old type
lantern tool post as you can slide the post along the T slot in the
angled compound to pick up the thread. John (17151) |
| Dave,
There is. It is small dia=O.D. -(1.299/Threads per inch).
Rick (17152) |
| Correct
formula for depth of threads is: .6495 divided by N (number of
threads per inch)= Single Depth Of Threads Single Depth of Threads X
2= Double Depth Of Threads Single Depth Of Threads X .5773= Single
Feed Depth with Compound set at 30 Degrees I am a machinist by Trade
and we cut all Unified National threads with compound angle set at
30 Degrees and tool ground to 60 degree included angle. Ron
(17170) |
| Ron, So to
cut a regular set of coarse/fine threads, using your formula's could
you tell me what I would do with the formulas to get my total depth
of cut. I don't understand the "single depth of threads, double
depth of threads, and the single feed depth with compound set at 30
degrees. I apologize for my ignorance. I can do the math, just don't
know which formula(s) to use. Dave
(17176) |
| Seems to be
much fear expressed by members about chasing an old, used or
incomplete thread. Here's what you do in a nutshell. There are two
slightly different methods both with the same results. First Method
(This is the one I use most often) 1--Set compound parallel to
workpiece. 2--Set threading tool using thread gauge or "fishtail" as
some people call them perpendicular to workpiece 3--Set Thread
gearing for correct pitch, start lathe with cutting tool some
distance away from actual cutting surface and engage thread lever at
proper spot on thread dial 4--Let machine start to work its way down
thread BUT at a distance away from actually cutting. 5--Stop lathe
WITH thread lever still engaged( Hint--let lathe coast to a stop and
do not stop by trying to reverse spindle as doing that will allow
backlash in screw and nut to come into play) 6--Using a combination
of compound and crosslide bring threading tool into thread groove
until it is touching on all sides (hint--use a piece of white paper
underneath thread and threading tool where you are trying to see if
tool is touching both sides) 7--Zero Crossfeed dial 8--Zero Compound
Dial 9--Withdraw toolbit from thread one complete turn and disengage
Split-Nut 10--Reset tool one turn in to Zero Crossfeed setting .
12--You are now ready to start to chase thread at this setting.
Depth is increased by turning in Crossfeed dial a few thous at a
time until thread is to complete depth or damaged thread has been
cleaned up. With this compound setting you will cut on all sides of
tool not one side as when compound is set to 30 Degrees. Second
Method Basically follow all steps the same as in first method except
compound is now set at 30 Degrees. 1--Follow steps 1 through 10
Depth is now increased by moving in Compound Dial a few thous at a
time until thread is to depth or damaged thread has been cleaned up
PIECE OF CAKE! Ron
(17192) |
| Ron, I
don't understand the third statement (single depth of threads X
.5773 etc.). Please explain. I am also a machinist by trade ( 22
yrs. so far ), and I'm always up for learning something new. Jim (17194) |
| That is the
amount you feed compound in at 30 degree angle. You would then feed
compound in that many thousands of an inch. DO NOT 1/2 that figure as
you would thinking that you are taking for example .010 per side for
a total of .020. This is Single Depth Of Thread. If you double
single depth of thread you obviously have Double Depth Of Thread a
figure you will need when doing an Internal Thread to calculate
proper bore size. Since my computer skills are so poor I will ask my
better half if she will set up a spread sheet with all this info and
the calculated figures for threads from 2 to 40 TPI. I will have to
do some major whining to get this accomplished. I will post to both
this and the Atlas-Craftsman Groups Files if I can figure out how to
do that also. Ron (17197) |
| Dave: Single
Depth of Thread is the amount you feed toolbit in measured with
compound set parallel with workpiece. Most inexperienced people do
not thread this way. They set compound at 30 Degrees. I have cut
most of my threads at the parallel setup. Double depth would really
be the Minor Diameter of the thread. The other formula with the
.5773 multiplier is the amount you would move compound in when
threading at the 30 Degree Angle. There is a trig factor involved
here so that is where the .5773 comes from. When threading just use
the Single Depth of Thread figure or Single Depth of Thread Compound
at 30 Degrees. Use The Double Depth of Thread figure to calculate
bore for internal thread. Hope this makes sense. Been doing this all
my adult life but it is hard to put on paper. Ron
(17198) |
| Dave, Do you
have a fishtail gauge, center gauge? Mine has a double depth chart
on it. JP
(17204) |
| Ron-- Be
assured that I understand all of what you're saying except the .5773
multiplier. The distance the compound travels along the 30 degree
angle will be a larger number than if run in parallel with the
crossfeed. If the thread depth is multiplied by .5773, you will end
up with a smaller number than the thread depth. If thread depth is
.032 as in 20 pitch, then the compound travel is .032 / .875 or
.037. .875 is a trig function of 29 degrees and is easy to remember
as it is the decimal for 7/8ths. I thread with the compound at 29
degrees for threads courser than 12 pitch. For finer than 12, I do
the same as you and feed in with the crossfeed, if the material
isn't to tough, as some stainless is. I cut some 3", 4 pitch stub
acme, 20"s long the other day and there's no way our old machines
will stand crossfeed straight in for that. We also cut quite a bit
of "mystery metal" in repairing threads in oil field and mining
equipment. A work hardened 4140 heat treated drill string tool joint
thread can be quite a chore. These are usually 4 or 5 pitch and cut
on a taper. Jim (17206) |
| Ron, I was
trying to figure out where you came up with the .5773 number and the
only thing I can find is in the use of the 3 wire pitch diameter
measurement. Take the pitch length times .5773 and that is the
diameter of wire that will rest in the thread and touch the pitch
diameter. It is the optimum size wire to use in the measurement. Its
also the cosine of 54.7 degrees. What am I missing here, I can't
figure out where that number comes from for cutting an SAE thread?
JP (17211) |
| You lost
sight of the whole formula though. Tool movement at 30 Degrees is
Double Depth X .57773 which will be a larger # than Single Depth.
.6495 = S.D. X 2=D.D X.57773= FEED DEPTH at 30 Degrees TPI Ron
(17212) |
| You are
correct-- I did not read "double depth" in your original post. The
light came on in my head after I went to bed last night about how
you stated the formula. Jim (17215) |
| So you are
adding 15% for the compound movement, got it, thanks. The
explanation got a bit confusing. JP (17219) |
| SD=
0.6495/TPI or SD=Pitch X 0.6495. REMEMBER that this only works if
the tool bit used is ground with the proper width of flat on its
tip. This can be found with the formula Width of Tip = Pitch x (1/8)
for Ex. 16 TPI equals a pitch of 1/16 or 0.0625; thus Width of tip =
.0625 x (1/8) = .0078" When the compound is set to 30 degrees your
tool bit must travel along a longer path. It is easy to calculate a
constant for this. Assuming again an American Standard Thread with
Pitch = 1" we can start with our SD of 0.6495". To find our new
distance use simple trig. ; We have a known side (0.6495) and a know
angle (30 degrees). If we sketch a right triangle our known distance
is the side adjacent to our angle and our unknown is the hypotenuse
of the triangle. We can use the following formula: cos of angle = adj/hyp
; or hyp = adj/cos of angle thus hyp = 0.6495/cos 30 = 0.750 Again
since we used a pitch of 1" to find this we can use this as a
constant for finding the amount the COMPOUND must travel for a given
number of TPI. Distance = 0.750 / TPI or Distance = 0.750 x Pitch.
Ron's constant of 0.5773 is a ratio of of the above calculated
constant (.750) to the double depth of an American Standard thread
with a pitch = 1". Constant = 0.750/1.299 = .5774 EX. American
Standard Thread with 16 tpi. What is SD for Compound parallel?; what
distance does compound slide when set at 30 degrees? 1) Compound
parallel SD = 0.6495 /tpi = 0.6495 /16 = 0.0406" 2) Compound at 30
degrees Compound Travel = 0.750 / tpi = 0.750 / 16 = 0.0469" or with
Ron's formula DD=SD x 2 = (0.6495/16) x 2 = 0.0812" Compound Travel
= DD x 0.5774 = 0.0812 x 0.5774 = 0.0469" both work but I like using
the first better as DD does not have to be calculated and the
constant 0.750 is easy to remember. You can also calculate constants
for finding SD when using a sharp V tool for American Standard
Threads (i.e. not ground to correct profile with flat tip). If any
one is interested just let me know. I apologize for the length of
this post, but I really got into this. Russ
(17228) |
| Russ:
Excellent explanation!!! You put to paper nicely what I could not.
Like I said, I wish you could all come over and we could thread
together and you would be surprised at how easy it really is. BTW:
In previous post when I said I do most of my threading with
cross-slide Parallel with workpiece and feed in with cross-slide I left
out one other little tip for those that might like to try threading
this way. As you near finishing Single Depth of thread take the
compound and move a couple of thous forward so tool only cuts on
front side of toolbit and test for fit. If tight then take tool back
towards tailstock a couple of thous so tool cuts only on back. Test
for fit again. If tight then repeat procedure again after taking
another thou or so in on crossfeed depth. This method will also work
to help take chatter and or burrs or "rags" out of a thread which
you will sometimes get when threading with my method. Ron
(17229) |
| That's all very
interesting Russ. However, I feel I should point out that the
American Standard Thread Form is archaic. It was replaced many many
decades ago by the Unified Thread Standard. There are important
differences between the two in regards to thread depth. Ed (17266) |
| Ed, you're
absolutely right. Being an avid collector and reader of older
reference material I sometimes find myself using information that is
somewhat dated. However, the mathematical concepts remain valid and
it was with regards to the math I was writing. An examination of
newer thread profile data such as that found in Machinery's Handbook
26th ed. will yield the material necessary to find constants that
are "up to date". As far as thread depth goes, it appears that the
Unified thread form is slightly shallower. Instead of a constant of
0.6495 being used it might be more appropriate to use 0.5413 found
by multiplying 0.86603 by 0.6250 (revised depth compared to a
standard V-thread). Other factors such as class of fit, etc. may
need to be taken into account. Thread cutting tool profiles would
have to be adjusted accordingly as per modern standards. According
to Machinery's Handbook, threads cut to the American National
Standards that were previously discussed are fully interchangeable
with the Unified standards. I apologize for the confusion. Although
I'm relatively young, much of the material that I have educated
myself with has a lot of years on it. Then again some of my
machinery dates from the 19th century, so the two (my library and my
machinery) seem two go hand in hand. Russ
(17301) |
| I agree that 0.86603 times the pitch is the depth for a
sharp V thread. And I agree that 0.6495 times the pitch is the depth
for a American National Thread form. But my text book ("Machine Tool
Practices" 4th edition) says 0.6134 not 0.5413 times the pitch for
external Unified Thread form depth. That is based on a flat at the
top and bottom with a width of 1/8 times the pitch. It gives 0.5413
times the pitch as the depth for an internal Unified thread. So we
are close! Ed
(17308) |
| ED, Machinery's
Handbook 26th edition in table 1 "American Standard Unified Inch
Screw Thread Form Data" lists both internal and external UN thread
forms in the same column. It gives thread depth as 0.54127 X pitch
for both. So...., I looked the same table up in Machinery's handbook
21st. ed. and here the data is listed separately for internal and
external threads. Internal is 0.54127 x pitch and external is
0.61343 x pitch ... the same as you're reference lists.
Correspondingly, the illustrations for thread profiles are also
slightly different. 21st. ed. uses standards revised in 1974 and the
26th. ed uses standards revised in 1989. As I understand it most of
these changes have been made to correct problems with maintaining
tolerances in some production activities. Threads manufactured to
the newer standards are also suppose to be interchangeable with
those made using the older standards. I do not know how true that
is, but I've never had any problems with fits when using data from
older references. Of course I am not doing production runs either!
Russ
(17325) |
| That's
interesting Russ. My text book is dated 1991 so it should of had the
1989 standards. I have a copy of the 25th edition of Machinery
Handbook (1996) and it also shows a single column with 0.54127 x
pitch. Oh well, I guess I'll stick with that. Also, keep in mind
that they are *increasing* the minor diameter since they are
decreasing the depth from 0.61343 x P to 0.54127 x P. Well... wait a
minute. They may have kept the minor diameter the same and decreased
the major diameter. Can't tell since I do not have the earlier
standard. Perhaps you can tell. Ed (17344) |
| I am
finding out there is a lot I do not know about thread standards!
Comparing different texts it looks like they are increasing minor
diameter. Reviewing MH 21st.ed. again revealed that the constant
0.6134 appears to have been derived by including an allowance for
the optional rounded root. This amounted to 0.08333H, or in our case
an addition of 0.07217, being added to the basic constant 0.54127...
which is of course equal to 0.61344 The radius is listed as .14434 x
pitch. Sides were to be straight beyond the apex of the radius. The
width at this point corresponds to 0.25 x pitch, which is the width
of the root for the basic profile of the UN thread. It would seem
that in order to use the constant 0.6134 a tool bit with proper
radius would need to be used. If a tool with a point .25 x pitch
were used at thread depths found with this constant, the flanks of
the thread would be wider apart at the pitch line than the standard
calls for. Also, the resulting flat at the peak of the thread would
be narrower than the standard .125 x pitch. In MH 26th ed. the same
profile diagrams show some changes. First the radius of the optional
rounded root has changed. It is now 0.108 x pitch. The constant used
for depth is derived from .625 x H or in our case = 0.5413. No
allowance for a rounded root is added "except" for the diagram for
the actual UNR thread profile. Even this number has been changed.
Using 0.5413 with a tool bit ground with a flat of 0.25 x pitch
should give the proper depth and resulting profile. BTW the old
standards that were originally discussed used a flat of .125 x pitch
for both root and peak. I also noticed that in the thread tables
such as Table 4a in MH 26th. ed., the numbers given for minor
diameters are based on the UNR form and include the additional depth
for the tool radius. I guess what formula or chart you use depends
to a great extent on what kind of threading tool you use and what
its profile is. I am guessing that older thread tool grinding gauges
may not give the correct geometry if using the latest threading
data. I just realized we haven't even touched on thread Classes and
fits! It might be easier to buy my threaded products at the hardware
store! Russ
(17360) |
| Thanks for
the information Russ. It is complicated. I use an Aloris tool post
with their HSS cutting tool. When it came it had a sharp V on it and
I stoned it down until it "looked" right. I'll have to check it. To
be "right" I would have to use a different cutting tool with a
different width flat for each pitch! I'll just stone it for a
mid-range pitch. There is an easy adjustment for the helix angle
which I use. Also, I'll use the latest depth specifications as a
guide. Usually, like most people, I cut a thread until the part I'm
mating to fits properly. Ed (17384) |
| Correct Thread
Depths without Calculation |
| There is an easy
method of getting the dept |
