PDA

View Full Version : Single point threading: Not understanding the 59.5 degrees



MotorradMike
07-26-2014, 07:20 PM
I understand that advancing the compound will cut the one side rather than have the cross-slide push in straight and cut both sides likely resulting in chatter.

I don't understand why we need to deviate from 60 degrees.

Any insight welcome. I'm about to cut a thread that's very important to me to get right.


Thanks guys.

justanengineer
07-26-2014, 07:27 PM
The half degree less of compound angle is to give clearance so that advancing the compound cuts with the left side of the cutter but doesnt rub with the right. Some folks will say the half degree doesnt matter, for me it seems to cause chatter if its missing.

Toolguy
07-26-2014, 07:30 PM
Because you want to still be cutting a little bit on the trailing side so the cut isn't wider than the tool. I like to set mine to 28 degrees off straight. The threading tool is 30 degrees each way off center for a total included angle of 60.

KiddZimaHater
07-26-2014, 07:31 PM
Check out this thread:
29.5 Degree Threading-angle (http://bbs.homeshopmachinist.net/threads/53433-Threading-angle)

Ohio Mike
07-26-2014, 07:41 PM
I like justanengineer's explanation. The thread profile is determined by the proper grind on the tool and making sure its on center. From there you can do some funky stuff and still get a usable thread. The idea of the 29.5 deg angle was cutter relief. Otherwise your threading tool is also a parting tool. Lots of chatter possible.

Old Hat
07-26-2014, 07:45 PM
Because you want to still be cutting a little bit on the trailing side so the cut isn't wider than the tool. I like to set mine to 28 degrees off straight. The threading tool is 30 degrees each way off center for a total included angle of 60.

For most alloys this is good advice.
When cutting gummy or stringy alloys I go 30 degrees on the head till the last couple fitting passes
and then go str8 in on the cross slide, shaving both thread faces........

.... But I should point out I thread up-side down and out into space.
I start in the undercut. Gotta get yer mind around it, at first.
A hard-headed old German Toolmaker sold me on this.

I've never looked in the rear-view mirror yet.
===================
PS.....
try this on some scrap .. .. .. nothing to loose
but your assurance that the Sun comes up in the East!

You can also do some turning in a steady=rest ``` with no top.
The tool-bit faces the ways. You can't watch the chip,
but you can put a drip can over the bit
and get better coverage than you could hope for.
On plain bearing spindles the spindle has no lift
and you can acheive mindblowing surface finishes
chatter and vibration free.
===============
Warning! I don't use threaded chuck lathes!
Obviously this could end baddddly!
Unless you've provided an Anti~unscrew mod to your chuck.***

``` solid rest not roller rest
a roller-rest is a miss-invention.

*** Ideas published in the American Machinist Hints & Reciepts books
that used to be published every 5 years.

Old Hat
07-26-2014, 08:18 PM
Turning bit faceing down.

http://i126.photobucket.com/albums/p92/swadge/Copyof000_2076.jpg (http://s126.photobucket.com/user/swadge/media/Copyof000_2076.jpg.html)

http://i126.photobucket.com/albums/p92/swadge/I%20did%20it%20My%20way/000_2011-1.jpg (http://s126.photobucket.com/user/swadge/media/I%20did%20it%20My%20way/000_2011-1.jpg.html)

PStechPaul
07-26-2014, 08:21 PM
I have only cut LH square threads so I am interested in this as well. The square thread is very much like parting, but I found that I could cut a narrower thread and then shift the tool slightly to obtain full width and cut on the tip and edge. The LH thread also meant that I was cutting left-to-right, so there was a bit more thread relief, and also (for certain arguably valid reasons) I completed the entire operation with the half-nut engaged, and I used a manual crank for better control (not having a low back gear).

I found this step-by-step tutorial that seems to explain it pretty well:
http://firearmsdesigner.com/?p=52

As I understand it, you want to cut on just one surface of the tool, and for a 60 degree cutter profile set exactly at 30 degrees, the compound will cause the trailing edge of the tool to drag along the right side of the thread as it advances. With a slightly sharper angle, the leading edge will still be 30 degrees, because that is where it should be set, but the advance will be set at a slightly sharper angle of 29.5 degrees. This is supposed to assure that the cut will advance more toward the headstock.

However, it appears to me that any angle of less than 30 degrees will cause some cutting on the right-side edge, although much less than that on the left. And the reason for this makes sense. If you use a larger angle, say, 31 degrees, the leading edge will be at 30 degrees, but the trailing edge will move further from the original cut, and will in fact form a more obtuse angle and a total thread angle of 61 degrees. Also, this surface will consist of a series of discontinuous cuts, and not a good thread form.

So, the idea is to use the 60 degree cutting tool profile to its full advantage and cut both surfaces, although the leading cut will be much deeper, and the trailing cut will be just a "skim" cut for a smooth finish. This seems different from what I read above.

And the other very important factor for the cutting tool geometry is proper helix angle. The leading angle must be a little greater than the helix angle, while the trailing edge may be vertical, since the helix always angles away from vertical.

And, for square threads at least, and probably for 60 degree, the top of the cutting tool should be aligned horizontal and not perpendicular to the helix angle as I have seen illustrated. It may not make any difference for a point, but for square or Acme threads, anything other than parallel to the work axis will cause a gouge rather than a concentric root.

This is my own opinion, but based on my understanding of the geometry involved and limited experience. Please verify if I am correct, or not. ;)

dp
07-26-2014, 08:40 PM
Paul nails it with this:
However, it appears to me that any angle of less than 30 degrees will cause some cutting on the right-side edge, although much less than that on the left. And the reason for this makes sense. If you use a larger angle, say, 31 degrees, the leading edge will be at 30 degrees, but the trailing edge will move further from the original cut, and will in fact form a more obtuse angle and a total thread angle of 61 degrees. Also, this surface will consist of a series of discontinuous cuts, and not a good thread form.

This is the primary reason I set my compound at 29.5. The other reasons are just good ideas - this one is good physics.

David Powell
07-26-2014, 08:42 PM
I understand that advancing the compound will cut the one side rather than have the cross-slide push in straight and cut both sides likely resulting in chatter.

I don't understand why we need to deviate from 60 degrees.

Any insight welcome. I'm about to cut a thread that's very important to me to get right.


Thanks guys.

I have been threading fairly successfully, both for work and play, for fifty years. I have used most methods. I keep returning to setting the topslide ( called compound over here) parallel with the axis of the lathe, grinding the tool with a flat or near flat top, grinding plenty of clearance at the sides to make sure there is clearance for the helix angle, then when threading RH threads I advance the compound one thous towards the chuck for every three I move the cross slide inwards. For left hand I advance towards the tailstock. Most of my single point threading has been for 1" dia or below. For two or three start threads it is essential that each finishing cut is taken with the same advance on the topslide, regardless of how I have indexed the piece. Sometimes I take a second or third cut just with the topslide feed before returning to feeding with the cross slide. It is simple, involves littlel or no memory or math, and, especially on older machines with fixed dials has less chance of difficulty in depthing. Mine are not the best threads I have ever seen, but they are far from the worst, and I can set up and be threading very quickly. You pay your money and take your choice. regards David Powell.

Nicholas
07-26-2014, 08:55 PM
https://www.youtube.com/watch?v=WqTliEGrhL4&feature=youtu.be

sarge41
07-26-2014, 09:04 PM
MotorradMike: The practice of using the 29.5 degree (instead of 59.5) is a leftover from the old days when a tool was set to cut on one side only to take a load off of the tool. The practice of cutting on 29.5 instead of 30 deg. was because many old lathes were a bit loose. If you cut only on 30 deg., you might wind up with a step on the right flank. Using 29.5 give a little bit of cleanup on the right flank. Kinda longwinded, but this was the way it was explained to me many years ago.

Sarge

Doozer
07-26-2014, 09:43 PM
Check out this thread:
29.5 Degree Threading-angle (http://bbs.homeshopmachinist.net/threads/53433-Threading-angle)


Oh yeah, I had one of my better rants in that one. :)

-Doozer

KJ1I
07-26-2014, 10:12 PM
Since on one else mentioned it (without having to look elsewhere), the cutter used for threading has a tip angle of 60 degrees. The 29.5* (or 28 or 30) is the angle of the compound. Watching a video of the cutter setup makes things a lot more clear than words on paper (or screen in this case).

PStechPaul
07-26-2014, 10:20 PM
The YouTube video explains it well, but I have two small nits to pick.

One is the jump from threading using the compound feed angle, but not explaining that for each 0.010" the depth of cut moves by 0.1 * cos(30) or 0.0866".

And the other is his statement that the cross-feed calibration is based on diameter, while on my lathe and many others, it is based on actual movement (or radius) and thus 0.001" means 0.002" diameter. But radius is what counts for thread depth.

Also, a good way to determine when the thread is fully cut is to coat the surface of the work (pre-turned to the correct diameter, of course), with some layout dye or other visible coloration, and then cutting until there is either a thin band of color, or it just disappears. A looser fit can be obtained by feeding in just a bit further.

danlb
07-26-2014, 10:49 PM
Also, a good way to determine when the thread is fully cut is to coat the surface of the work (pre-turned to the correct diameter, of course), with some layout dye or other visible coloration, and then cutting until there is either a thin band of color, or it just disappears. A looser fit can be obtained by feeding in just a bit further.

Sorry, but I have to disagree strongly. Cutting until the color is almost gone or until it disappears will give you a poor thread. If you look at the definition of an ISO or UN thread, you will see that the crest of the thread (that you colored) is supposed to be a specific width in relation to the pitch.

If you keep going in until the dye is gone you will end up with sharp crests and a smaller diameter than the spec calls for.


The right way to do it is to turn the shaft till it matches the minor diameter (minus a few thousandths) and then cut the thread to the correct depth per the tables. If you are using a full form threading tool you go a little less deep than when using a tool with a sharp V tip. That is also described in the charts and tables.

Thread cutting is a very simple process with some very simple math to follow. It's well defined in the books, and even well described on Wikipedia. Wiki was correct last time I checked it, that might have changed since then. :)


Dan

Old Hat
07-26-2014, 10:58 PM
It's well defined in the books, and even well described on Wikipedia. Wiki was correct last time I checked it, that might have changed since then. :)


Dan

I don't do Wiki, I do some math, and I do LOTS of Thread Wires!
There's no better way, if you want perfection.
Even a go-no-go isn't as good.

http://littlemachineshop.com/Products/Images/480/480.1962.jpg

PStechPaul
07-26-2014, 11:46 PM
I have not actually cut threads on the lathe, so my idea may not be correct, but I have seen that method described somewhere. I agree that sharp threads are usually not desired, but according to the specification the rounded edge or "flat" is quite narrow:

http://upload.wikimedia.org/wikipedia/commons/thumb/4/4b/ISO_and_UTS_Thread_Dimensions.svg/800px-ISO_and_UTS_Thread_Dimensions.svg.png
http://en.wikipedia.org/wiki/Unified_Thread_Standard

The major diameter of a fractional size thread is the actual dimension (say, 0.2500" for 1/4") and the flat or round edge is 1/8P, so for a typical 1/4"-20 thread it will be 0.00625".

The problem I see with thread cutting is the "fillet" in the thread, which is shown to be 1/4 the theoretical thread depth H. So for the 1/4"-20 thread, H is 0.866*P, or 0.0433", 1/4 that is 0.0108", and the width would be 0.0108/0.866 = 0.0125" wide. Thus I think the cutter should have a nose with a radius of 0.0062".

However, this fillet may be optional, while the external flat may not be, and the choice of rounding may be optional. I don't know, but just sayin' :)

J Tiers
07-27-2014, 12:05 AM
Any non-CNC lathe is loose....if it does not use ball screws.

So......

with a heavy machine, this is not an issue in general, but with a typical lightweight Atlas, or even a Logan , Southbend, some Sheldons, etc of typical 10" or 12" size......

If you do NOT keep a load on the forward side of the cutter, loading the halfnuts against one flank of the leadscrew, it is possible for the backside of the cut to push the cutter ahead of it. This can happen if the leadscrew and halfnuts are not tight (and most likely they are not). That takes the leadscrew out of the picture, sliding the carriage to some point in the "slop range" of the screw and nut setup. The carriage is not following the leadscrew then, and so you get "drunken" threads, with a wobble in them.

That's the origin of the "load the cutter" idea.

After that basis, the angle is as mentioned, just to keep a "cleanup cut" going on the rear side, without enough force to "take charge" and push the carriage.

A heavy machine with a large carriage weighing as much as a whole Atlas lathe, is not going to easily be pushed like that, so with a big machine you may be able to just cut in with the crosslide and not mess with a compound, if the machine even has one.

Paul Alciatore
07-27-2014, 12:50 AM
First, I must point out that the recommended angle is NOT 59.5 degrees. This is actually getting it backwards.

I believe the recommended angle is 29.5 degrees or if you are measuring it the other way, then 60.5 degrees.

Here is an illustration of what happens:

http://img.photobucket.com/albums/v55/EPAIII/Web%20Post%20Photos/ThreadCutAngle/ThreadingAngleIllustrated_zpsdf679795.jpg (http://smg.photobucket.com/user/EPAIII/media/Web%20Post%20Photos/ThreadCutAngle/ThreadingAngleIllustrated_zpsdf679795.jpg.html)

The top half shows the effect of setting the compound at 60.5 degrees from the lathe axis or at 29.5 degrees from the cross feed axis. In all cases here, the single point, cutting tool has a 60 degree tip angle as viewed from above. In the recommended method you can see that each succeeding cut primarily advances the left edge of the cutting tool into the right flank of the thread being cut and most of the cutting action occurs there and a proper thread form is produced. But the right edge of the cutting tool is also advanced just a bit into the left flank of the thread so that flank is also properly formed. You can see the final result on the top right.

On different lathes, the angular scale for setting the compound angle is marked differently. On some the zero marks are on the left and right so they coincide with the main axis of the lathe. On these lathes the recommended angular setting is 60.5 degrees, NOT 59.5 degrees. On other lathes this angular scale has it's zero marked at the end of the cross slide that points to the operator and the left and right marks are 90 degrees. On these lathes the correct angular setting is 29.5 degrees as you are measuring from the cross slide axis and only half of the 60 degree thread angle is used. I have shown these two angles at the top right.

On the bottom of the drawing I show the effect of setting the compound at 59.5 degrees as the OP has suggested. As you can see, the successive cuts at this angle allow the right hand side of the cutting tool to pull away from the side of the thread and this will leave the stepped thread flank that I show in the bottom right. This is not desirable.

This is the usual reason stated for using a 29.5 degree angle setting of the compound.

This should be sufficient reason for using the 29.5/60.5 degree setting. But if you read the thread referred to by KiddZimaHater you can see some more comments that I made at that time about deeper reasons in this situation. Here's that link again. Those comments are still valid.

http://bbs.homeshopmachinist.net/threads/53433-Threading-angle

If all these .5 angles are confusing, here is what I actually do in the shop. I set the compound pointing straight out toward my belly button, then move it 30 degrees toward the tailstock AND then move it back toward my belly button about 1 degree. Thirty degrees is a lot easier to remember and one degree works just as well as 1/2 degree. And it does not matter how the compound scale is marked.

Grab compound
Point at me
Right 30
Left 1
Lock it and thread

Old Hat
07-27-2014, 01:04 AM
Any non-CNC lathe is loose....if it does not use ball screws.

So......
.

I'm going to make a few adjustments on your assertion.
I understand your point, to a degree it might be valid . . . .
untill you factor in (Use).
Use based on familliarity and understanding.

When familliarity is lacking, substitute validation.
In that case which may be as often as half the time or more.

A machinist or a toolmaker must by necessity, make several cuts
before a finish cut. Might be ruffing and semi. If completely unfamilliar
with a machine it might be ALL cuts (passes) of equall values.
This to provide a mark, or a point, to work against, untill the performance
of said machine is showing it's nature so to speak.

A good Man will use his sharpest tool, (his mind) to construct a method of application
that will ballance out all of the machine's quirks and inadequicies.

OR'
He will reject the machine for that task and move on to another.
I see that many of you have pulled up much good information.
Hobby or Profession, the clock is ticking, unless you are payed by the gov't
or some foundation with a cause or two.

I daily watched a man of about 44 master everything that the shop threw at him.
His weapon? a 1936 Bullard Spiral Drive with only a 5 place Heidenhain DRO
to guide his efforts.
Information is good, it's important!
Mastery of the situation and all it's vary~ables and puzzels reigns supreme.
Even over information.

Hopefuldave
07-27-2014, 08:49 AM
I have not actually cut threads on the lathe, so my idea may not be correct, but I have seen that method described somewhere. I agree that sharp threads are usually not desired, but according to the specification the rounded edge or "flat" is quite narrow:

The major diameter of a fractional size thread is the actual dimension (say, 0.2500" for 1/4") and the flat or round edge is 1/8P, so for a typical 1/4"-20 thread it will be 0.00625".

The problem I see with thread cutting is the "fillet" in the thread, which is shown to be 1/4 the theoretical thread depth H. So for the 1/4"-20 thread, H is 0.866*P, or 0.0433", 1/4 that is 0.0108", and the width would be 0.0108/0.866 = 0.0125" wide. Thus I think the cutter should have a nose with a radius of 0.0062".

However, this fillet may be optional, while the external flat may not be, and the choice of rounding may be optional. I don't know, but just sayin' :)

Ideally the root of the thread should be a radius, not a flat nor a.vee - a radius isn't as significant a stress-raiser, which is why Whitworth / BSF threaded parts are noticeably stronger than UNC / UNF for the same major diameter...

J Tiers
07-27-2014, 08:54 AM
Old Hat ....

There certainly may be a point in that somewhere. So far I am not sure which statement is intended to be the point...

However....

A loose lathe is not an unusable lathe. This is good since ALL are loose because there has to be at least oil space in any sliding contact, meaning that there is some slop on all the screws. Might only be a few thou, or it might all add up to 50 thou, possibly even more in bad cases and coarse screws.

"Using the mind", one figures out that you "take up backlash" in the same direction always, and then you stay on dimension. If you do not do that, then even a DRO won't help. It only tells you how much the part is now "out" because of the cutter shift you allowed to happen when you didn't take up backlash.

When you thread by keeping the front side of the cutter loaded, you are really just "taking up backlash". Which is one way of "using the mind" (as you suggested) to overcome a problem.

With a heavy lathe, the carriage is heavy, lots of friction, and it may not matter.... The friction is acting like the old trick of snugging the table locks on a mill to reduce the effect of slop and backlash. A smaller threading job won't be able to move the heavy table, and so it will not "take charge".

MotorradMike
07-27-2014, 09:06 AM
The youtube video did it.

My compound is marked 0 degrees when parallel to the chuck axis.
The video said "Don't go beyond 30 degrees" to prevent the stepped poor thread form.
I need to set mine at 60.5 degrees by the marks but I do it with an angle gauge anyway 'cause this little lathe is only marked to 45.

So, depending on how your compound is marked; Don't go beyond 30 - or - Don't go less than 60.

PS: Nice rant Doozer!

danlb
07-27-2014, 10:57 AM
I have not actually cut threads on the lathe, so my idea may not be correct, but I have seen that method described somewhere. I agree that sharp threads are usually not desired, but according to the specification the rounded edge or "flat" is quite narrow:

[ snip ]

The problem I see with thread cutting is the "fillet" in the thread, which is shown to be 1/4 the theoretical thread depth H. So for the 1/4"-20 thread, H is 0.866*P, or 0.0433", 1/4 that is 0.0108", and the width would be 0.0108/0.866 = 0.0125" wide. Thus I think the cutter should have a nose with a radius of 0.0062".

However, this fillet may be optional, while the external flat may not be, and the choice of rounding may be optional. I don't know, but just sayin' :)

It is REALLY helpful to learn the nomenclature while you are learning about the standard. It makes all the little comments make more sense. For instance you can refer to the top of the threads as the "crest" and even google it for more information. You can refer to the bottom of the thread groove as the root, and some of the formulas start to make more sense.


Another thing to note is that there are TWO threads in that illustration ( internal and external) and each has a different width crest. The internal thread's crest is 1/4 of the pitch wide and the external is 1/8 of the pitch.

In my opinion, it's important to understand these distinctions because threading is often one of the last things you do when turning a piece. It's a real shame to spend an hour or two on a part and then overshoot the dimensions when you thread it. The only thing worse is when you make a poor thread and manage to get TWO freshly made parts screwed together before the razor sharp crest folds over and jams permanently.

Dan

Paul Alciatore
07-27-2014, 11:34 AM
On that razor sharp edge thing, I always take a brass wire brush to the thread while it is still turning and before measuring or testing.

Old Hat
07-27-2014, 11:58 AM
Maybe I'll check back tonight . . . . .
Try~ing to stay open minded here.

So many threads on this forum, where some of you go way way way way way
deeper with the well, than it takes to get water.

Just say'n.

danlb
07-27-2014, 12:35 PM
Maybe I'll check back tonight . . . . .
Try~ing to stay open minded here.

So many threads on this forum, where some of you go way way way way way
deeper with the well, than it takes to get water.

Just say'n.

This from the guy who suggested using thread wires? :)

Do thread wires do anything to ensure that you have a proper thread form, or does it just make sure the depth of the thread is correct for the major diameter that you are using?


Dan

Old Hat
07-27-2014, 12:40 PM
I want an electron microscope for Christmas so I can make sure
all my molecules end up in the right place.

I wouldn't want to make a tool where the crown on my thread was fit for a Queen,
when it was supposed to be for a King.

J Tiers
07-27-2014, 01:54 PM
This from the guy who suggested using thread wires? :)

Do thread wires do anything to ensure that you have a proper thread form, or does it just make sure the depth of the thread is correct for the major diameter that you are using?


Dan

Nothing directly to do with thread form.... But if the form is correct, they show if the pitch diameter should be right. (Or IS right, if the wires are the size to contact at the pitch diameter).

if the form is off, the sky is more-or-less the limit. Lots of ways to contact the pitch line correctly but have a bad thread... about as many as there are different angles, at least, then you can get into weird thread shapes.

Old Hat
07-27-2014, 02:18 PM
Here's my point.
Often over a thousand non-participating viewers show up in the count below.

One might be age 17. He's got access to machinery at the shop his buddy's dad owns.
He & his partners in crime are building a dunebuggy or a rat-bike.

They are allready fleXing their minds mucsels and mechanical apptitude.
They might allready understand that in most cases any lathe-tool might
work better if the tip aint an infinitely perfect point.

I'd like them to think they can make a well suited single point thread,
without going to college, or haveing to look up standard ISO-Whitworth class 3/B section 2497.64190
and reading thru pages of thread theory with .00000#'s described with words they have never heard before.

I'll have them worry about getting a good fit on the working faces of the mating threads.
The biZness part.
You guys can scare them off with nomenclature and details involving the open-air spaces
at the root and crown of their friendly little thread.

OMG, they might start thinking about a career any day now.
I'd like them to know the world will quit turning when it destroys the need
for young Men who can do seat-of-the-pants Metal-Working.

They have their whole lives left to dive into the world of refining things
beyond any reasonable point of existance.

J Tiers
07-27-2014, 05:19 PM
Eh, it's now, to be aware of it, or later and have an afternoon's scrap, plus a PO'd boss later.

if this scares off anyone, the first red-faced screaming foreman will probably have them drop dead.....

danlb
07-27-2014, 05:34 PM
Here's my point.

[ snip ]

I'd like them to think they can make a well suited single point thread,
without going to college, or haveing to look up standard ISO-Whitworth class 3/B section 2497.64190
and reading thru pages of thread theory with .00000#'s described with words they have never heard before.

I'll have them worry about getting a good fit on the working faces of the mating threads.
The biZness part.
You guys can scare them off with nomenclature and details involving the open-air spaces
at the root and crown of their friendly little thread.

[ snip ]



You seem to be saying that doing it the right way is too onerous for a new person. You imply that it's too difficult to understand without a degree in mathematics or engineering.

I disagree.

The whole world of machining is full of terms that you really should know if you want to be able to do a decent job with some degree of assurance that it won't end up scrap. How can you find the right way to face a bar on a lathe if you don't ever bother to learn what facing is? How do you figure out why you can't get a good finish when turning if you never learned the terms for tool geometry like "clearance"?

I see a lot of professional as well as self taught folks who seem to think that anything that uses a formula is some sort of snobbery and that you can guess at it and be close enough. This is especially true of threading.

When it comes to threads, I use the formula. I follow the procedures. I have not had to scrap a part due to a bad thread in many years. I have not had a sloppy thread either. I feel no trepidation when I have to thread something, because I know that it will work.

Old Hat
07-27-2014, 05:42 PM
If I was as far off the beam as your illustration indicates.........
I wouldn't be the guy welding all the screw-ups, I'd be causing the screw-ups.

I haven't had a close second, amoung my co-workers for error free machining
in this century, across 3 employers.
(and three part time gigs)

================
Measure twice. Cut once. Weld. Repeat.
( Welding solves many problems.)
==========================

That's your signature not mine.

MotorradMike
07-27-2014, 05:56 PM
So many threads on this forum, where some of you go way way way way way
deeper with the well, than it takes to get water.

Just say'n.

Go as deep as you want boys.
I almost always stop reading but I appreciate the depth of knowledge you guys have; there's really not a down side to that.
Thanks.

MotorradMike
07-27-2014, 06:01 PM
The thread has been cut and it's pretty good.
Not a super finish but the best I've ever got in O1 steel.

Nice to have confidence through understanding.

Once again, I owe you guys.

danlb
07-27-2014, 07:24 PM
If I was as far off the beam as your illustration indicates.........
I wouldn't be the guy welding all the screw-ups, I'd be causing the screw-ups.

I haven't had a close second, amoung my co-workers for error free machining
in this century, across 3 employers.
(and three part time gigs)
.

I respectfully submit that if you are that good, it means that you have already learned all the terms and have been taught how to do the tasks and operate the machinery properly. If you are that good you are following some repeatable process or procedure, even if it is not documented. That's different from cutting a thread 'till it looks right' without knowing what the right look really is.

On the other hand... If your co-workers are no where as good as you, it may well be that they never bothered to understand why they do what they have been trained to do. It's hard to improve your work when you are working from rote.


Dan
BTW... My signature is meant to be humorous, along the lines of the old woodworker's lament "I've cut it twice and it's still too short!"

Old Hat
07-27-2014, 08:46 PM
This must be where we are quite far appart. I know the text-book now.
But along with many others, who were inducted in a time where CNC was on the horizon
and industry told themselves "we will soon be able to do without those pesky
tool-makers and machinists" and cut apprenticeship levels to the bone......

I had to learn most of it by trial and error. Any of us who could keep up at all
with the few older tradesmen learned on a share as you go basis.
I even got to be quite good on a Hydrotell (Cincinnatti duplicator) and a
Gettys Electro-Trace or somethink like that. I had no Idea what I was doing.....
.... had to peice it together.

I was put in a jig-grinding room with three Moores screaming for 10 hours at a time.
I could tell most of what I was shown couldn't possibly be correct.
Was continually being told I was a rebble, and just do what you're told.
The two hands that new what they were doing retired early out of frustration
when the "Kids" took over from their Dad's, and proceeded to wreck the buisness.

One day a young hand came back from Colorado, and they hired him back.
He turned everything up-side down and told the "Kids" how far out of the water
they were. He was amazed at how close to the target I had gotten on my own.

Machining comes as much from the gut as the brain.
Too much emphasis on the brain-work can scare away the young.

thaiguzzi
07-28-2014, 12:43 AM
This must be where we are quite far appart. I know the text-book now.
But along with many others, who were inducted in a time where CNC was on the horizon
and industry told themselves "we will soon be able to do without those pesky
tool-makers and machinists" and cut apprenticeship levels to the bone......

I had to learn most of it by trial and error. Any of us who could keep up at all
with the few older tradesmen learned on a share as you go basis.
I even got to be quite good on a Hydrotell (Cincinnatti duplicator) and a
Gettys Electro-Trace or somethink like that. I had no Idea what I was doing.....
.... had to peice it together.

I was put in a jig-grinding room with three Moores screaming for 10 hours at a time.
I could tell most of what I was shown couldn't possibly be correct.
Was continually being told I was a rebble, and just do what you're told.
The two hands that new what they were doing retired early out of frustration
when the "Kids" took over from their Dad's, and proceeded to wreck the buisness.

One day a young hand came back from Colorado, and they hired him back.
He turned everything up-side down and told the "Kids" how far out of the water
they were. He was amazed at how close to the target I had gotten on my own.

Machining comes as much from the gut as the brain.
Too much emphasis on the brain-work can scare away the young.

+1.
A lot of sense in this post.

ckelloug
07-28-2014, 02:46 PM
I cant tell you guys how much I learned about threading making custom 1/2 12TPI bolts for an antique jointer head. I had to make them because modern bolts are 1/2 13TPI and obviously don't fit right. Some idiot previous owner had of course cut a 1/2 13 bolt really short and tried to engage only the few threads to replace a missing bolt. Not a good idea on a jointer cutter head. . .

Anyway, what I learned is that if you're cutting a relatively fine thread like 24 tpi or something like that, almost any technique, good or bad will work. On a 12 TPI thread, the cut is so deep that you have to learn the 29.5 degree compound advance trick. If you don't, you break several threading tools before discovering why people do it that way. Once I got the hang of the compound advance and the not very satisfactory trick of backing the cutter out and stopping the lathe before crashing the tool, I got good bolts.

Advancing straight in, the cut was too much material and the tool would dig and snap the bit. The other thing I learned is that good PeeDee thread wires are available from McMaster Carr and they are only $32. Given thread measuring wires and some axle grease to hold them on the thread while trying to get the mic on it, threading accurately became pretty easy. As long as you use a fishtail gauge to ensure you have the 60 degree V of the tool perpendicular to the bolt axis, and the tool is ground on spec, it comes out pretty well. Before getting the thread wires, I made a lot of scrap bolts where the depth of cut was wrong and I couldn't tell. With the thread wires, you just measure across them and find out if your minor diameter is right.

RichR
07-28-2014, 03:18 PM
Hi ckelloug

I cant tell you guys how much I learned about threading making custom 1/2 12TPI bolts for an antique jointer head. I had to make them because modern bolts are 1/2 13TPI and obviously don't fit right. Some idiot previous owner had of course cut a 1/2 13 bolt really short and tried to engage only the few threads to replace a missing bolt. Not a good idea on a jointer cutter head. . .
When I was making replacement ball handles for the quill on my milling machine I found that they too needed to be threaded 1/2 12. I believe that Sir John
posted in another thread that 1/2 12 was actually very common, possibly on his side of the pond.

danlb
07-28-2014, 06:28 PM
This must be where we are quite far appart.

[ snip ]
Machining comes as much from the gut as the brain.
Too much emphasis on the brain-work can scare away the young.

You could be correct, but my experience is that letting new folks blunder through while they learn from their mistakes with trial and error will scare away more. Think about how many people you see on this forum who are afraid of using a parting tool because it never works for them. That's a simple task that always works well if you know of and follow a few simple guidelines.

My experience is that any competent person can learn new tasks if they are given a sound foundation. That's what the old apprentice programs were all about. Once you know the terms and the concepts you can improve your performance though intuition and observation.

When it comes to machining, welding, etc I am also self taught. I made a lot of mistakes while learning but I had the resources to discover the root causes of my mistakes. One of the things I do early on is to read the lists of terms, jargon and acronyms that apply to a new endeavor. With that simple starting point I can understand what others are trying to tell me. I can understand what I read and I can apply what I know from past experience to the new tasks.

It's my belief that too little emphasis on the brain work will result in unnecessary frustration and will scare away the young too.


Dan

cameron
07-28-2014, 08:31 PM
[QUOTE=ckelloug;929440 Before getting the thread wires, I made a lot of scrap bolts where the depth of cut was wrong and I couldn't tell. With the thread wires, you just measure across them and find out if your minor diameter is right.[/QUOTE]

This is probably too picky, but the thread wires tell you if the pitch diameter is right . At that point, the minor diameter will be "right" only if the flat ( or the rounding) on the tool is correct.

PStechPaul
07-28-2014, 09:38 PM
The thread wire system effectively measures the pitch diameter of the external thread, which may be most important, but not the true thread form or the root(minor) or major diameters. The major diameter can be measured pretty easily with a caliper or micrometer with jaws wide enough to span two or more threads, or with a go/nogo hole gauge. Here is more information on the thread wire system:
http://www.threadcheck.com/3-wire-thread-measuring-systems-info/technicalinfo/

Here is some discussion of how to measure minor diameter:
http://www.engineersedge.com/engineering-forum/showthread.php/1158-Correct-way-to-check-minor-diameter-on-external-threads

It should be possible to use a caliper with a very sharp edge to read the distance between roots (or valleys, or cuts), and then use the angle (which is not the helix angle) to calculate the actual minor diameter. Another way might be a dial indicator with a sharp point or flat edge that can reach the deepest part, and compare to the OD, or have it calibrated at zero at the center of the piece.

dp
07-28-2014, 10:46 PM
I've yet to have a threading project that required thread wires but surely that day is coming. Because I'm a home shop kinda shop I generally am making repair parts and have the mate nut to the thread I'm turning. When I don't have one I make one. One such case was a hardened shoulder bolt that broke off the strut on my Harley Fat Boy. It is a nut welded to the frame, and getting the broken bits out required some careful brazing below the gas tank. Got it out, turned an internal thread that matched the broken part, and turned a new shoulder bolt that fit perfect. I did the best I could to heat treat it and it was still in place when I foolishly sold that bike. What I learned from that experience is never to sell a perfectly good Harley. Oh - and to make a model of the mating part.

andywander
07-28-2014, 11:59 PM
The half degree less of compound angle is to give clearance so that advancing the compound cuts with the left side of the cutter but doesnt rub with the right. Some folks will say the half degree doesnt matter, for me it seems to cause chatter if its missing.

Actually, it's just the opposite-it is done so that the right side of the tool DOES cut a tiny bit.

PStechPaul
07-29-2014, 12:31 AM
While we are talking about single point threading I'd like to discuss the role of the threading dial and engagement of the half-nut. My experience so far has been only the 3/4-8 LH SQ threads, which is the lower limit of TPI for my lathe, and I have a 9/16"-16 leadscrew. I discovered (the hard way) that there are 64 teeth on the wheel and 8 marks, so each mark corresponds to 8 turns of the screw or 1/2". Since the thread pitch is 1/8", every other point is correct, and the other points between create a two-start thread.

Now I want to figure out what thread pitch or TPI would not require such careful reading of the thread dial. The change gears presently mounted provide a 30:60 step-up ratio to the leadscrew, so the TPI choices from the gearbox are 8, 9, 9.5, 10, 11, 11.5, 12, 13, and 14. It seems to me that the TPI which are not integer multiples or factors of the 16 TPI leadscrew result in a case where the synchronization of the workpiece angular position and that of the leadscrew will only be correct at a very specific point. The TPI is equal to 1/2 the number of teeth on the gearbox selector, which are 16, 18, 19, 20, 22, 23, 24, 26, and 28. It seems that the "right spot" will move at 16/8 = 2 turns, 18/8 = 2.25 turns, 19/8 = 2.375 turns, etc, to 28/8 = 3.5 turns. The 24 tooth gear should be exactly 3 turns. I'm not sure I'm stating it correctly, but what I'm saying is that it takes two turns of the 16 TPI screw to get the 0.125" pitch of 8 TPI, and three turns to get 12 TPI. However, since the leadscrew is moving more slowly, it is probably always exactly 2 turns for each correct position.

But it also seems that the carriage must be set at the same point each time to start each threadcutting pass. For the 8 TPI I did, this did not seem to matter, and every other spot would be OK, but for other settings this does not seem so obvious. According to the information on the lathe, any of the 8 positions is OK for even numbered TPI from 8 to 56, while the odd dial points 1, 3, 5, and 7 are OK for odd numbered TPI 9, 11, 13, 16 and 23, and for the half-fractional threads 9.5 and 11.5, only 2 and 6 are OK.

I will need to get some hands-on experience with my lathe to understand how this works, and maybe I'm complicating it too much. The problem with my lathe is that the 64 tooth gear and unclear markings make it very difficult to engage the half-nut reliably, at least on the 8 TPI setting. I have a feeling that the situation may improve for the higher TPI change gear ranges, and it seems that for 16 TPI (same as leadscrew) there will be a 1:1 relationship of the work to the screw and any position for engagement would be OK. This might also apply for the higher TPI up to 28. I would expect this range of 16-28 TPI would cover most medium size work from 1/4" (20 or 28 TPI) to 3/8" (16 or 20 TPI) and 1/2"-20 NFT, as well as the lathe's 9/16-16.

For the finest thread settings, with 60:30 reduction gears, I think the thread dial may be even less critical, so that all the threads from 32 to 56 TPI can be cut without regard to the dial setting, but I would have to confirm this by experiment or by some rather involved mental gymnastics that I'm not up to performing at this time. :confused:

Perhaps this deserves a thread of its own, if it spawns much additional discussion. I just have a hard time "rapping my brane" around this concept. :D

RichR
07-29-2014, 01:09 AM
But it also seems that the carriage must be set at the same point each time to start each threadcutting pass.
Maybe you meant the threading dial not the carriage in that statement? The carriage can be anywhere, the threading dial is used to guarantee proper
synchronization.

PStechPaul
07-29-2014, 01:55 AM
What I mean is that, with the half-nut disengaged at the end of a cut, I can move the carriage using the handwheel to any place along the ways, but generally close to where I want to start the next cut. The thread dial will still be in the same position it was when the last thread cutting operation was stopped. Then, it will rotate until the correct mark lines up with the indicator, and I re-engage the half-nut to continue threading. From what I've read, the position at which the half-nut is engaged or released only needs to be close to and beyond the limits of the desired thread.

I have verified that this is the case for 8 TPI and the 16 TPI leadscrew, with a simple 2:1 ratio. But it is harder for me to visualize how it works for other fractional ratios.

dp
07-29-2014, 02:07 AM
Old Tiffie did a good workup on use of the thread dial here and is worth looking up. I'll see if I can dredge it up. I miss his presence here but don't fault him for dropping out. I hope he's well.

Edit:

Lordy I can be dense at times. I remember being sufficiently impressed with his post that I blogged it here at my site.

http://metalworkingathome.com/?p=11

Old Hat
07-29-2014, 02:15 AM
I have verified that this is the case for 8 TPI and the 16 TPI leadscrew, with a simple 2:1 ratio. But it is harder for me to visualize how it works for other fractional ratios.

When in dout, leave the half-nut engaged untill you are done.
Wheel out, reverse lathe past your start point.

Bar the spindle to the start area.
Wheel in, cut .
Repeat.
A worn & sloppy lathe requires this method to overcome poor repeatabillity.

All this is just more reason to thread away from the chuck into space.
Try it, you'll never go back.

PStechPaul
07-29-2014, 04:23 AM
dp - thanks for that reference. It's now become about as clear as milk, while before it was clear as mud. I think once I set up the lathe for more normal threading (not coarse LH threads), it will clear up considerably, like adding vinegar to the milk, allowing the curds to float to the top of my brain, so I can see through the whey.

His example of a 16 tooth gear on an 8 TPI screw would be analogous to a 32 tooth gear for my 16 TPI screw, and I think I could have worked with that. But mine has 64 teeth, and thus the effective divisions are much closer together. I was able to hit it right twice in succession, but the next try was off just enough to "screw" it up. Fortunately it was early into the jb and I was taking light cuts, and I could tolerate a slightly smaller OD. But from then on I did as Old Hat suggested and I kept the half-nut engaged, as is necessary anyway for metric threads.

Cutting LH threads already has the carriage moving toward the tailstock and what would be empty space were it not for the live center, but there was still plenty of free space on both ends. So I might consider running the lathe in reverse with the cutter upside down on the "far side" so I can keep the same action, which avoids the "pucker factor" when cutting close to a shoulder.

At some point I will want to be able to use the thread dial, and I have a few options. One may be if, as I suspect, the finer thread rangers are more forgiving because each increment of the screw will be greater than the thread pitch on the work, at least for even numbered TPI. If so, and if most of my work will be in that realm, I may be able to use what I have (after perhaps making better "tick" marks in the dial and a better "pointer").

Another idea was to make a 32 tooth gear, and I might try that as an exercise in making a worm gear using the method I saw from Brian. Basically I would make a threaded rod to match the teeth in the 9/16"-16 leadscrew (which might be Acme), and then mill some longitudinal slots in it, effectively making a tap. I would make a gear blank from suitable easily machined material, perhaps brass or 12L14, and then make a jig where I would rotate the threaded rod so that it cuts into the gear blank and forms threads. Something like a gear hob or knurling tool.

Another option, which may also be combined with the smaller gear, is to enlarge the dial so that it's easier to read. Possibly repurpose an old rotary tire pressure gauge or a scrap dial indicator or a meat thermometer. Lots of options. Or I could go high-tech and use a motor shaft encoder (which I already have from a scrapped printer as well as a surplus stepper motor), and I could program a PIC to flash an LED or beep (or even actuate a solenoid to engage the half-nut) at the precise position.

Finally, I should in any case take the apron apart and fully clean, inspect, touch up, and lubricate everything. I need to see if the mating threads on the half-nut are in good shape, and it might be good to have a slight bevel on them rather than a sharp edge, to guide it into the leadscrew threads.

Time to make chips! And salsa?! :rolleyes: Maybe I'll take some video and show my technique for operating the lathe for screwing and call it my "half-nut dance"? :)

JCHannum
07-30-2014, 09:21 AM
The thread dial will still be in the same position it was when the last thread cutting operation was stopped. Then, it will rotate until the correct mark lines up with the indicator, and I re-engage the half-nut to continue threading. From what I've read, the position at which the half-nut is engaged or released only needs to be close to and beyond the limits of the desired thread.

I am not sure that I understand what you are saying here. When the half nuts are released, the threading dial will not remain stationary when the carriage is returned to the starting position, but will revolve (unwind) to whatever number is appropriate to the carriage location. The purpose of the threading dial is to assure that the carriage and the leadscrew are in the same position as when the last cut was started. The carriage should be positioned so that the leadscrew will make a couple of revolutions before the tool starts cutting the thread to eliminate any backlash.

As to the angular position of the cross slide, the original purpose of this was to relieve pressure on the form cutter used for threading. It may also help to eliminate chatter in difficult materials. The actual angular setting is variously given as 29*, 29.5* and 30* depending on the source. I think the main intent is to set at an angle larger than 29* and less than 30* as no lathe compound I have ever seen has a protractor graduated accurately enough or adjustment fine enough that it can accurately split a degree, so just set it at whatever point you are comfortable with and don't overthink it. You are using a form tool that, if set properly, is at 90* to the workpiece and that is the thread form it will produce.

I use the Atlas Lathe handbook, in the appendix for threading there is a table which gives the infeed depths for National Form and Vee Form threads with a compound setting of 29* for thread counts from 4TPI to 96TPI. This eliminates the need for thread wires and will produce an accurate thread for any but the fussiest of applications. I have several sets of wires as well as a set of Montgomery thread triangles and have yet to use any of them. Nothing I have made has fallen apart due to improper thread geometry.

I highly recommend the beginning machinist get a copy of the Atlas Lathe handbook, it is still probably the single best introduction to the use of the lathe available and far more informative than the South Bend How to Run a Lathe book.

Danl
07-30-2014, 09:16 PM
Last weekend I visited my brothers machine shop, and was shocked to see him cut internal threads without touching the compound feed at all. He said he learned how to do this from Dennis Turk (Turk Manufacturing, in Hillsboro). My brother emailed me this afternoon because I find this process quite interesting. Here is his response:

"Dan,
Dennis talked me into the new way, total infeed at 0 degrees. It does go against all we have learned, and the reason I did it on the ID was because I've started doing it that way and I was too lazy to turn the compound to the opposite angle, when threading ID normally the handle is to the left instead of the right side, so that the left side of the tool is being advanced. When threading OD the handle is to the right so as you advance the tool it is cutting on the left side only, but ID threading the same setup makes the tool cut on the RH side only. So normal practice is to turn the compound with the handle to the left side so as you advance the BB thread tool OUT the left side of the tool cuts.


Dennis says he never uses the 29 degree method anymore and hasn't for twenty years. I've started 0 deg. infeeds but you have to recall how many you took out with each pass. The old way you just go to zero, then advance the compound, easy to keep track of.


But the Zero degree infeed doesn't add up as it should, total infeed should have been .083", it ended up at .093" or sumpin. In the end I always go by fit anyway, starting with the flat size at the top of the thread. When the flat starts to look like it's the right width I start checking fit."

Interesting. It was a nice appearing thread, and the fit was very good.

Dan L

Old Hat
07-30-2014, 09:31 PM
Sounds very interesting! And Cool!

danlb
07-30-2014, 09:35 PM
His example of a 16 tooth gear on an 8 TPI screw would be analogous to a 32 tooth gear for my 16 TPI screw, and I think I could have worked with that. But mine has 64 teeth, and thus the effective divisions are much closer together. I was able to hit it right twice in succession, but the next try was off just enough to "screw" it up. Fortunately it was early into the jb and I was taking light cuts, and I could tolerate a slightly smaller OD. But from then on I did as Old Hat suggested and I kept the half-nut engaged, as is necessary anyway for metric threads.



There is no reason that you can't stop the lathe between passes. That lets you move the carriage to a point where the right marks line up and then close the halfnuts while the leadscrew is motionless. If you miss the mark just open the halfnuts and try again. When it's at the right spot you can turn the lathe back on.

I blacked out the threading dial marks that were either side of the #1 mark. This makes it very easy to find the #1 mark. The #2 and #7 marks are still there, they are just black on black.


Dan