PDA

View Full Version : single point threading unsuccessful, any pointers as to where I went wrong?



Jim Doherty
12-24-2012, 04:43 PM
I watched videos, read threads and thought I had it figured out but no dice.

Pics of the failed attempt.

http://i837.photobucket.com/albums/zz293/jcdoherty/Picture037-1.jpg

I had to cut it off with a hacksaw.

pics of the set up

http://i837.photobucket.com/albums/zz293/jcdoherty/Picture034-1.jpg


http://i837.photobucket.com/albums/zz293/jcdoherty/Picture032-1.jpg

The thread gauge fits like a champ

http://i837.photobucket.com/albums/zz293/jcdoherty/Picture038.jpg


I mic'd the OD and mine is .002 under the one from the hardware store.

I chamfered the leading edge and the hardware store nut screwed on about halfway then it felt like it was binding.

Under magnification 10x a see a ton of tiny tears in the metal so I thought its probably a rough surface and though screwing on the nut with a wrench would clean it up.

I got about one full turn of the wrench and no more, not forward, not backwards, nothing.

I advanced the tool into the work using the compound.

I would appreciate any thoughts as to where I messed up.


Thanks, Jim

Davidhcnc
12-24-2012, 04:48 PM
You need to keep taking cuts until the nut fits.

Rough tears are a function of the material, tool sharpness, lubrication, rigidity, etc. This a seperate issue to the fit.

Tony Ennis
12-24-2012, 04:49 PM
Nut appropriate for the thread? Metric nut and imperial thread? Will another nut thread on to what remains? Oil the thread before putting another nut on.

damengineer
12-24-2012, 04:57 PM
There are others here who are a lot more experienced, but here is my $0.02 mostly questions. What type of material is it. Is the tool sharp. Is the tool above or below centerline, this will be much worse depending on how far out from the chuck you are trying to thread. I have had small stock crawl up the tool bit and tear metal off. Try resharpening the tool and turning the compound on 29deg and advance the cut into the thread about 0.015 at a time... I aways use a light cut for threading as I am just not that good at it, and my old lathe is really worn near the chuck. If possible use a die for anything less than 1 in dia...



If it ain't broke, it ain't mine.

90LX_Notch
12-24-2012, 05:12 PM
I usually finish with one or two spring cuts.

gvasale
12-24-2012, 05:26 PM
if you're going to have it stick out that far, cenetr dill the end & hold it with a dead or live center.

SDL
12-24-2012, 05:29 PM
For a thread to fit not only does the OD have to be correct, but so does the the mean diameter.

There is several ways to get there.

1 Keep taking a bit off until the nut fits.
2 Finish of with a die set to a known matching bolt
3 Measure using thread wires and a mic or with a thread mic.

Have another go, it will work

Steve

danlb
12-24-2012, 05:30 PM
It really sounds like the wrong nut was used. I've had almost identical problems before when using a nut that was almost the same. A 22 TPI where it should have been 24TPI for instance. There is enough slop in the threads that the first few will work, then when enough threads are engaged they tolerance will be used up and it will seize. I'm assuming that was not the case here.

I can't tell from the pictures. Did you use a sharp pointed tool, or did you use one that had the appropriate radius for the specific thread? The major diameter of the screw can be perfect, but if the root of the groove is not deep enough the crest of the nut will rub on the root, eventually galling and sticking.

Dan

Maker
12-24-2012, 05:34 PM
Are you cutting dry?

Try cutting fluid.

mygrizzly1022
12-24-2012, 05:39 PM
Hi all

"I usually finish with one or two spring cuts".... Just to keep me in the loop! What is a spring cut?

Regards...Bert

Dr Stan
12-24-2012, 05:42 PM
if you're going to have it stick out that far, cenetr dill the end & hold it with a dead or live center.

gvasale nailed it. As a rule of thumb your work should stick out no more than 3X the dia without support, typically a dead or live center in the tail stock. Other factors include type of material of the work piece and the cutting tool, wet or dry, tool sharpness, set-up accuracy, etc.

If you do not have a copy of South Bend's "how to run a lathe" order it immediately. Lots of knowledge in that book about threading and other lathe operations. Also buy a copy of Machinery's Handbook. A used one off FleaBay will more than suffice for the HSMer. Get your specs for threads, and lots of other stuff, from MH.

Ohio Mike
12-24-2012, 05:43 PM
I hope you just have that much hanging out for setup. Even then I wouldn't do it. If you have more than your thread length hanging out of the chuck you should really have a center holding the end, lots of spring there. Otherwise you're going to really have to baby it along. If you move your thread gauge up does it still fit nice? My guess is the thread is tapered if not it may just be the rough finish. Another test would be to run a die up it and I think you'll find it cuts more as it goes.

Finish can be impacted by tool geometry, depth of cut, fee, speed, material being cut and the cutting lube used. If you're new threading try to get some 12L14 or even better 1144 which cuts like a dream. Trying to thread plain old cold roll steel isn't a good place to start. Nasty stuff to thread. Also running a file over the thread will knock the sharpness off.

Ohio Mike
12-24-2012, 05:46 PM
Hi all

"I usually finish with one or two spring cuts".... Just to keep me in the loop! What is a spring cut?

Regards...Bert

A repetition of the previous cut without advancing the cutting tool. Its done to negate the "spring" in both the tool and the work piece.

Dr Stan
12-24-2012, 05:47 PM
Hi all

"I usually finish with one or two spring cuts".... Just to keep me in the loop! What is a spring cut?

Regards...Bert



AKA free ride in which a 2nd, 3rd etc pass is taken without advancing the tool. Used to take the spring out of the work piece that is induced by the cutting forces.

J Tiers
12-24-2012, 06:23 PM
The thread LOOKS fine..... I'm not so sure I would call it a "failed attempt". Looks technically executed quite decently as far as the cutting and result is concerned. Measurement obviously is the final arbitrator.

The errors left are diameter, possibly a taper issue, or thread pitch, possibly an issue with the sharp V where the tip gets a burr on it that can jam up. Flank angle looks symmetrical.

It would not have much taper from sticking out that far.... there would be taper, but it's so far that the taper should be small angle-wise (and it should have been almost impossible to do the cut).

presumably it was NOT sticking out. I'm still not convinced taper is an explanation, a shallow taper takes some distance to bind.

A straight diameter issue should have refused to take the nut even that far.

A thread pitch mismatch is a good chance..... right diameter, but slightly wrong pitch. Since the gage fits, then assuming the gage is OK, the screw thread is fine for pitch. That leaves the nut.

I suspect a mis-match of thread and nut. Maybe nut is metric? I see the gage has "SAE" on it, so it is presumably what you wanted pitch-wise.

Above is barring some sort of swarf getting caught in the threads (possibly attached "fuzz"),

KiddZimaHater
12-24-2012, 07:05 PM
Tip #574 : Always chase your single-point threading jobs with a die. :p

Peter.
12-24-2012, 07:24 PM
Is that stainless you're threading? Not easy the put a thread on if it's 304/316 I have found.

I would check the angle of the tool. Since I bought a graduated loupe I found that my '60 degrees' was more like 65 even though it looked spot on in the fishtail gauge.

tdmidget
12-24-2012, 07:27 PM
Well for starters, that is NOT a thread gage. It's a pitch gage.
The first thing I would check is the pitch on the nut, assuming that it started easily. The threads are not supposed to be sharply pointed, they have a radius at the crest. This should be knocked down with a file before checking. It can cause this problem.
Thread gages:

http://precisiontoolsandgaging.com/Assets/images/limitgage%20thread%20plug.jpg

a

tdmidget
12-24-2012, 07:29 PM
Tip #574 : Always chase your single-point threading jobs with a die. :p




Damn good advice.

cameron
12-24-2012, 07:38 PM
A couple of posters seem to have assumed that if there was taper due to deflection the fit would be tighter towards the chuck. That's the opposite of what happens.

Lew Hartswick
12-24-2012, 09:14 PM
A couple of posters seem to have assumed that if there was taper due to deflection the fit would be tighter towards the chuck. That's the opposite of what happens.
That was first thought, the end will be bigger than the part toward the headstock due to "springing" away from the cutter. I think just several light
passes over the crests with a flat file will do the trick.
Oh and then brush with a fine bristle wire (brass if possible) brush while
spinning.
...lew...

Oldbrock
12-24-2012, 09:26 PM
Your threading looks pretty good, did you run a tap through the nut first to check it was correct? From what I can see of your setup it looks right providing the tool is center height and there is enough front clearance. Run a lathe file over the thread before the last spring pass then make sure the thread is clean before trying your (gauge) nut. I have at this game for over 65 years and I have yet to use a die after single pointing a thread. I can cut a much nicer thread than any button die can. Peter

Ohio Mike
12-24-2012, 10:44 PM
A couple of posters seem to have assumed that if there was taper due to deflection the fit would be tighter towards the chuck. That's the opposite of what happens.

Yes, I think I was to quick to post. The taper would be opposite due to deflection.

chip's
12-24-2012, 11:03 PM
I thought the threads looked quite nice for a first time. Putting it between centers would help deflection but I don't think that is the problem. Chase em' again a few times and the nut should fit OK. Unless the nut is the wrong pitch or has some other problem. Keep trying there is much knowledge here and people are happy to help, good luck, keep at it.

mike4
12-24-2012, 11:13 PM
As Oldbrock said there could be something wrong with the nut , never assume that it is ok just because it came from a store shelf.
Michael

J Tiers
12-24-2012, 11:18 PM
The sharp V is not to current spec, and as I pointed out, it can have burrs etc that jam up. If the nut is not compatible, as many won't be, it could also jam up when the sharp V hits the major diameter.

To fix that "for real" you need to use a 'topping" insert cutter. Hitting with a file can be a "backwoods" solution also, and I'd give it a shot to see.

But. as I said, it's pretty nice for a first attempt, frankly......


A couple of posters seem to have assumed that if there was taper due to deflection the fit would be tighter towards the chuck. That's the opposite of what happens.

That is true, in general.....

But, there is a tendency for the FRONT end (leading end) of the part to be cut deeper, due to issues starting the cut, with the cut stabilizing in a deflected position after a bit of movement. It's somewhat related to bell-mouthing of a boring cut.

So there is a possibility that there is a taper on the leading end of the cut that we don't see because it's presently jammed inside the nut on the cut-off end of the part.

I think that taper could be steeper than a regular deflection taper, which would be in the opposite direction as you point out. But the rest of the cut is so nice that I doubt it could have been done sticking out that far, at least without a follower rest. So the leading edge taper issue may not be a factor.


Well for starters, that is NOT a thread gage. It's a pitch gage.


Sure it is... it's a "Thread pitch gage"......

What you show are various gages for checking the combination of pitch diameter, form, and pitch at once (but not identifying the actual problem).

Both are forms of thread gage, since they are used on threads, and not on tires or drills, etc.......

michigan doug
12-24-2012, 11:51 PM
It only gets easier.

What kind of steel?

How far was it sticking out?

Go make another one.

Finest regards,

doug

PeteM
12-25-2012, 12:09 AM
To add to the above, threads are commonly rolled (not cut) in cheap grades of steel. Typical low carbon steel will tear a bit when cut -- and also gall when a tight fit. That looks to be one of your problems. Sharp tools, a positive rake, cutting oil, a rigid setup etc. will all help; as will a different material.

The tops of your thread look to have a flat to me. If your stock is on nominal diameter or that .002 under, you haven't cut the thread quite deep enough. Chasing the thread is the simplest fix.

tdmidget
12-25-2012, 01:03 AM
http://bbs.homeshopmachinist.net/images/misc/quote_icon.png Originally Posted by tdmidget http://bbs.homeshopmachinist.net/images/buttons/viewpost-right.png (http://bbs.homeshopmachinist.net/showthread.php?p=818239#post818239)
Well for starters, that is NOT a thread gage. It's a pitch gage.
Sure it is... it's a "Thread pitch gage"......

What you show are various gages for checking the combination of pitch diameter, form, and pitch at once (but not identifying the actual problem).

Both are forms of thread gage, since they are used on threads, and not on tires or drills, etc.......

I got to disagree. If it checks pitch, it's a pitch gage. If it checks the complete thread, it's a thread gage. I work for a company that commonly sends an "8 thread tap and die set" to jobs. It has taps and dies for threads from 3/4" to 6", 8 threads per inch. According to that pitch gage they are all the same. Or is it just the pitch?

danlb
12-25-2012, 01:37 AM
I'm not an expert, but the gages shown in post #18 look like they could be go-nogo gages. One end should work and the other should not.

But back to the OP's question;

What type of tool did you use? How deep did you cut? What was the major diameter? What standard bolt were you trying to create?

Using a file to 'knock off the tips of the thread" is not needed. Here's why. When considering the unified thread form the major diameter of the male thread is the flat top of the crest. That means if you turn the rod to that diameter the root (bottom) of the nut's thread will not bottom out. Likewise the nut's crest will be the minor diameter so it is slightly bigger than the roots of the screw's threads.

The form of the single point tool is supposed to have a radius on it that ensures that the male root is deeper than the nut's crest will reach. By coincidence a sharp V tool will also make the root deeper than the matching crest will reach and will be compatible with a unified thread.


Pictures, a depth calculator and some discussion of thread profiles can be found at http://www.tanj.com/cgi-bin/tpi.cgi

Dan

tdmidget
12-25-2012, 01:49 AM
"I'm not an expert, but the gages shown in post #18 look like they could be go-nogo gages. One end should work and the other should not."

All functional gages are go/ no go. That is why they are best in production. The part is good or it isn't. No judgement required or allowed.

danlb
12-25-2012, 02:59 AM
"I'm not an expert, but the gages shown in post #18 look like they could be go-nogo gages. One end should work and the other should not."

All functional gages are go/ no go. That is why they are best in production. The part is good or it isn't. No judgement required or allowed.

The judgment comes when you have to decide what class of fit is involved, right?

The problem with that type of gage is that it's only good for one specific size and one class of fit. It's great for production runs. It's not so good for making something to match odd one off parts. If the part calls for an odd size you might have no choice but to fall back to tools like thread mics, tpi gages and thread wires.

My experience has been that with proper measurements I do not have to test fit a part after single point threading Unified threads. The proper form, cut to the proper depth almost always gives a usable part.

Dan

willmac
12-25-2012, 03:18 AM
Running a file very lightly over the threads after screwcutting is normal practice, especially when dealing with fairly soft gummy steel as this appears to be. The file should only very lightly cut and the objective is to remove the burs that can be raised by the cutting action. This can happen even if you have the right setup with a sharp, correctly ground tool. I like to finish critical threads with a hand held chaser whenever possible. It only takes a minute and appreciably improves the crest and root condition and in some cases the flank finish. For some threads such as Whitworth, a chaser is more important, because the crests should preferably have a radius. You can't make this radius just by single pointing unless you use a full form insert tool, which the OP probably does not have. You can use a die as several have suggested, but I find dies vary quite a bit in accuracy, especially drunkenness, and you can spoil an accurate thread if you are not careful. This is not so important for American, UN... threads obviously.

darryl
12-25-2012, 03:29 AM
Nobody has mentioned this yet- does the relief on the sides of the cutting tool take into account the helix angle of the thread? For a right hand thread, the right side of the cutter can be straight up and down, but the left side has to have enough undercut to clear, plus give a few degrees of relief angle. For a left hand thread it's just the opposite. This is going to play a role in how clean the thread is cut. I'm not sure, but I think that some holders allow you to adjust for that.

I don't know about others, but I like to pass a file over the crests of the threads a few times as you arrive at the final depth. Do this just before taking the spring passes. You will get rid of most of the burring which will make the testing more meaningful. You may have to back the compound out so the tool actually touches the burr. You could also use a three cornered file to help clean up the thread itself. Note that the file is not used to shape the thread at all, just to help deburr. Note also that any use of a file near the chuck is dangerous when the spindle is turning.

outback
12-25-2012, 03:37 AM
Jim;

One thing I noticed is your toolbit has no rake angle. The right clearance angle and rake angle will effect turning characteristics. Cutting oil makes
a different to.

How did you figure your depth of cut on the thread? In my experience when a nut only turns a half turn on the threads you are still way oversized.
When the thread makes 2 or 3 turns on the thread you are still .002 or .003 oversized. I think you were several passes away from turning a thread to size. Thread measuring wires will tell the actual story.

Nothing wrong with the thread OD being about .005 undersize.

You can also finish threads with a 3 corner file. A 3 corner file can remove tiny fragments from the thread.

Toolbit drawing (http://img.photobucket.com/albums/v30/jglass/Shop%20Demonstrations/CAD%20Drawings/toolbitb.jpg)

Keep trying

Jim

beanbag
12-25-2012, 06:00 AM
next time, coat the threads with a layer of dykem or a sharpie marker and see where the nut rubs

luthor
12-25-2012, 06:15 AM
[QUOTE=darryl;818321]Nobody has mentioned this yet- does the relief on the sides of the cutting tool take into account the helix angle of the thread? For a right hand thread, the right side of the cutter can be straight up and down, but the left side has to have enough undercut to clear, plus give a few degrees of relief angle. For a left hand thread it's just the opposite. This is going to play a role in how clean the thread is cut. I'm not sure, but I think that some holders allow you to adjust for that.

This point is often overlooked and even when highlighted, on forums such as this, people seeking help still don't seem to understand the importance of the clearance angle on the leading edge of a screwcutting tool.

Jim Doherty
12-25-2012, 07:46 AM
Thanks for all the thoughtful insights and comments.

I'm sure the nut is 3/8 - 16, I bought a bolt/nut to have something correct to compare to my threads.

Not a clue on the type of metal.

Use a center and spring cuts. Got it, thanks

Cutting dry? Guilty, I'll correct it.

Called a pitch gauge. Got it, thanks

I didn't think of using a file so I just put a loop of emery over the threads for a couple of seconds to take off the burrs. Would that do the same thing or is a file better?

Oops, in my hurry to try I forgot to grind a rake angle on the tool. I'll correct that and try again. Thank you for the toolbit drawing.

I had no clue as how to determine where to stop cutting depth so I used the pitch gauge, visually compared it to the screw from the hardware store and the OD.

I believe I have enough side clearance but I'll double check, thanks.

I'm going to take the advise of the replies here, apply them and try again.

I'll post the hopefully successful results.

I really appreciate all of you taking the time to help me figure this out.

Thanks, Jim

tdmidget
12-25-2012, 08:26 AM
The judgment comes when you have to decide what class of fit is involved, right?

The problem with that type of gage is that it's only good for one specific size and one class of fit. It's great for production runs. It's not so good for making something to match odd one off parts. If the part calls for an odd size you might have no choice but to fall back to tools like thread mics, tpi gages and thread wires.

My experience has been that with proper measurements I do not have to test fit a part after single point threading Unified threads. The proper form, cut to the proper depth almost always gives a usable part.

Dan
When you are making a part to a print there is no judgement allowed. You make it the way the print says. If it doesn't work, that's an engineering problem.

Mcgyver
12-25-2012, 09:16 AM
Oops, in my hurry to try I forgot to grind a rake angle on the tool. I'll correct that and try again. Thank you for the toolbit drawing.


you don't use rake on a form tool....if you do you are no longer cutting the profile you intend. threading tools ground to 60 degrees should have no rake


I watched videos, read threads and thought I had it figured out but no dice.



That may be the biggest problem. This will sound like a smart ass remark but its not intended to be, it's sincere advise help get the most from this craft. Get some basic books on it, I recommend use grade 12 texts as a great start; cheap and they're written 1) for beginners and 2) usually do a thorough job. You'll stumble through all this disparate advice not understand the reason why or which of it is relevant to the exact situation and problem you have and the next time something goes wrong you'll be stuck again.

You are in the unconsciously incompetent stage; you don't know what you don't know....and the only way to move beyond that painful stage is a baseline of knowledge. You don't what's written in threads or shown in utube videos is correct, `thorough or explained properly and there is no obligation on behalf of the creator to thoroughly cover the subject.....not so with a good text .

anyway, don't take this as "not willing to help, go read a book"....that's not at all the case. I just that its hard for you to get the base of knowledge you need from random posts; no one here's going to write the 20 pages you need to read and frankly, its been done.


My experience has been that with proper measurements I do not have to test fit a part after single point threading Unified threads. The proper form, cut to the proper depth almost always gives a usable part.

agreed, i don't understand this 'not measuring a thread' approach. would you make a shaft for a journal with without mic'ing it and machining to the required fit? Yes, its ok for simple stuff to do trial and error with a hardware store nut, but most work I'm single pointing I want better than that, do not want to upset the work (held with a centre) to do a trial or its a different (say Whitworth) thread who's mate is a 200 lb casting, etc etc. Measure your threads for best results :)

Dr Stan
12-25-2012, 10:03 AM
I didn't think of using a file so I just put a loop of emery over the threads for a couple of seconds to take off the burrs. Would that do the same thing or is a file better?

Jim,

If you continue to experience tearing on the flanks of the thread you probably have too little side clearance on the tool. BTW, I too refrain from putting top rake on a form tool as it does change the shape of the tool and consequently the shape of the form, in this case the threads.

To dress up threads I recommend either a thread file, see http://www.travers.com/skulist.asp?r=s&n=||UserSearch1%3Dthread+file&q=block+id+58700+and+class+level3+id+29974 or a small triangular file such as one found in a set of Jewelers files.

What others have said about thread gages is correct, but for 99.999999....% of HSMers and almost the same percentage of job shops threads will simply be cut to fit. If you're making parts to a Mil-Spec or to meet ISO 9000 requirements that's a whole different can of worms.

One more thing. When I taught threading at the vo-tech I had the students use aluminum for their first threading project. Brass would also be good for the first time. A lot of steels are difficult to obtain good surface finishes, especially with threading. However, if you have some 12L14 (AKA Lead Loy) it is very nice to machine & thread.

J Tiers
12-25-2012, 10:42 AM
The OP did fine...... rough surface, etc, is likely from material, tooling, and speed, not from technique per se. Some materials just cut poorly, and the best machinists still have trouble with them.




All functional gages are go/ no go. That is why they are best in production. The part is good or it isn't. No judgement required or allowed.

They are still not perfect.....

They may allow a thread with offsetting problems to pass, although most would be caught by the "no go" section. It takes a special sort of problem to pass and still be out of spec. However they are the best thing available for quick checking.

I've got a bunch of them, but inevitably NOT the one I want...... then you have to fall back on the other things, wires, triangles, thread pitch gages, thread mics, etc.

All of those have problems, particularly in determining the real issue with a thread that comes up as "bad".

Thread wires do not test the thread, they just check the effective pitch diameter, the form may be wrong, the crest and root might be wrong.

Triangles are a little better in some ways, as they contact more of the thread, and can find wrong angles/form better.

Thread mics are fooled by crest problems.

Thread pitch gages are rough checks, not precision.

An optical comparator is maybe the best, but for most people who would read this forum, this level of precision just is not an issue.

Abner
12-25-2012, 11:05 AM
Great manuals by Southbend, not that you haven't got all the information you need already.
http://wewilliams.net/SBLibrary.htm

Ian B
12-25-2012, 11:30 AM
Jim,

You say that you don't have a clue of the type of material. This might be the problem. Some materials are simply very difficult to single point thread - high tensile steels being a case in point. Just as a quick trial, swap the material for a piece of hard brass, change nothing else, try again - you might be surprised (and encouraged) by the results.

Ian

Black_Moons
12-25-2012, 12:29 PM
Thread mics are fooled by crest problems.


No, Thread mics use diffrent anvils for diffrent TPI ranges and have gaps for the crest to sit in, as well as rounded tips to avoid resting on the root.

J Tiers
12-25-2012, 12:44 PM
No, Thread mics use diffrent anvils for diffrent TPI ranges and have gaps for the crest to sit in, as well as rounded tips to avoid resting on the root.

Don't count on it......

I have Lufkin that DO have the recess, and Starrett that do NOT.

naturally they have different ranges, but that is also simply because the anvil needs to contact a significant part of the thread. One that would "do" for 8 TPI might not be so good for 40 tpi, and vice-versa.

danlb
12-25-2012, 01:09 PM
I had no clue as how to determine where to stop cutting depth so I used the pitch gauge, visually compared it to the screw from the hardware store and the OD.



Bingo!

If you do not measure the depth of the cut you are making an approximation of the correct shape. It will look correct but being just .001 too big will cause it to jam. This is your problem. It looks like most everything else you did was fine.

The depth is easy to determine. The formula is in all the books and can be calculated on the fly if you know the pitch and major diameter. I cheat and use my little program as posted earlier ( tanj.com)

In your case, it says;
Looking for 16 threads per inch
The major diameter is .375
The minor diameter is .30735

Calculations for the proper cutting depth using a sharp V tool, AKA partial profile.
This will provide a flat crest and sharp V groove.

For a thread of: 16.00 TPI
The pitch is: 0.0625 inch per thread
The thread height is: 0.0541 inch ( pitch * .86603 for UN and UNR)
External thread depth 0.0474 inch (H * .875) (for a screw)
Internal thread depth 0.0406 inch (H * .750) (for a nut)

If you cut the thread to a depth of 0.048 inch you will be able to force it on, but it may never come off again.

Dan

outback
12-25-2012, 02:19 PM
you don't use rake on a form tool....if you do you are no longer cutting the profile you intend. threading tools ground to 60 degrees should have no rake
results :)

Mcgyver, I don't agree with that. You are concerned with losses that most us cannot measure. The benefits of clearance and rake angles far outweigh
the minimal losses in the thread profile. When most of us cut threads and when the nut fits the thread it is considered a finished thread. Finish appearance and class of fit are often important but who checks the thread profile?

I think you are a very bright person. I have been reading your threads for years.

For an easy way to check thread depth I have used a tap drill chart to find the tap drill size (minor dia of thread) for a certain thread. Turn that diameter
on the very end of your workpiece. Coat the diameter with some kind of ink. When the thread cutting tool marks that diameter, your thread will very
close to size. Tap drills charts do not tell the actual minor diameter of a thread because tap drill charts usually figure 75% of full thread depth.

Respectly submitted,
Jim

Oldbrock
12-25-2012, 02:39 PM
I too use top rake on my threading tool, sloping down from the cutting edge when using the 29 degree compound feed. I usually start with about 20 thou first pass and decrese for each pass until I am only adding a thou per pass for the last couple of passes. For steel I like soluble oil and water at about 50 to 1 and can get a super finish on even the most gummy steels. It's slow but I usually thread at 50 rpm because it's my shop, it's my time and I'm an old fart and all the time in the world to play. When I threaded in the oilfield shop eons ago I used carbide and around 300 rpm on 4 1/2" API and H 90 tool joints on 4340 drill collars. Big hollow spindles and Herbert 9 B30s are fun to run when you have seemingly unlimited power and super ridgidity. For home shop with light lathes I prefer HSS tooling and slow speeds for screwcutting. Merry Christmas and a Happy and chippy New Year. Peter

darryl
12-25-2012, 04:47 PM
If the tool is ground properly, and the outer diameter is correct, you should be able to see when you're approaching the minor diameter by the width of the crests. If it looks like the crests are as wide as normal threads, then you are pretty close- close enough to try a nut for fit. Eliminating the burring is important at this point. It becomes intuitive fairly quickly. That's how I go about it, and no I don't do work for NASA. I get the nut to fit, and because I'm picky there's usually little play. My first threading job was a 7/16 or so (available threadable rod material, printer rod) by 20 tpi acme-like thread. It started as a square thread, but after breaking the tiny cutting tool once or twice, it became a bit more acmeish (is that a word :)). Made the nut with the same tool and it's been working on my cross slide for many years now. I learned quickly and had lots of frustration at that time, but now I know what all the variables are, and it's actually a limited number :). Threading is about a medium level activity- not rocket science, but more difficult than making an omelet.

Another way you can define the minor diameter to help with threading is to turn the OD to the correct size, possibly minus a few thou, then also turn a short section to the minor diameter. When you're cutting the thread, the moment the tool starts marking the minor diameter, you're there. You might go a little deeper to compensate for the rounding on the end of the cutting tool, but the test fitting gets you to the right point anyway.

Carld
12-25-2012, 05:13 PM
Top rake on a threading tool is not a good idea. Look at any manual and you will see that. Look at any threading insert and you will see it doesn't have any back rake or front rake, it is absolutely flat. When you add any rake you will change the V form of the cut and no longer get a perfect V shape. The cutter also has to be as near centered on the axis of the work as possible as well as 0 rake. Also, the side rake on the leading edge of the cut has to be enough to give clearance on the pitch of the thread. A 40 tpi doesn't need as much side clearance as a 4 tpi.

danlb
12-25-2012, 05:24 PM
The need for side clearance also depends on the size of a bolt. A 20 TPI thread on an 8 inch bolt is almost straight up and down, but put it on an 1/4 inch bolt and it slants quite a bit. The more the slant, the more you need to take it into account.

I don't understand the aversion to using the dials to cut the thread depth accurately. If you can turn a rod to .375 inches, you should have no problems cutting a spiral groove 0.048 deep in it for a 16 tpi thread. True, to do it right you should cut the thread in several steps to minimize the load on the cutting tool, but there are even charts (and formulas!) for that too.

Dan

( fixed the 1/4 vs 8 inch. Thanks Luthor for pointing out that I had it backwards. )

willmac
12-25-2012, 06:12 PM
Top rake is not normally used on screwcutting tools.

As others have said, if you grind a tool to say 60 degrees for a metric or typical US thread, then add top rake, you will not cut a true 60 degree thread. This will also tend to screw up depth measurements. You CAN use top rake, but then you have to grind the sides of the tool to compensate so that you end up with a correct straight sided thread form with the correct angle. This is not straightforward. The question of top rake is further complicated by the fact that you would not normally cut anything other than a very fine thread by going straight in with the cross slide. Best practice is to use the compound set at say 29.5 degrees so that nearly all the cutting action takes place on one tool flank. When you do that, the top rake, if you use it should be with respect to that tool flank.

PixMan
12-25-2012, 06:17 PM
Tip #574 : Always chase your single-point threading jobs with a die. :p

I very rarely have to do that. Ever used the pitch-specific full profile lay-down 16ER threading inserts? Nicest single point threading you'll ever see. This is a 3/4-10UNC thread done in 316SS with one:

http://img.photobucket.com/albums/v466/kenm10759/wheel-adapter/IMG_0398-r.jpg

With the lay-dow threading inserts I have now, including the partial profile ones for threads I don't have in the pitch-specific style, I never have problems with thread form. By using the full profile inserts and Vardex "TT-Gen" software to select the proper under-insert anvil that tillts the insert to match the helix angle of the thread I'm making, I get no rubbing and achieve perfect thread form on the roots, flanks and crest of the thread. Moneywise it can be painful to get set up with the tooling, but by getting a toolholder off eBay for cheap and buying a pair of inserts from KBC Tool or MSC for each thread pitch as it comes along I've managed to collect the most common ones needed. I also bought a Vardex kit that has all the various anvils for external RH of internal LH threads, and will soon get the opposite hand anvil set for external LH and internal RH threading.

Also, I no longer need to use the fishtail gauge for setting up. I instead just set the shank of the holder to be parallel to the X-axis travel and start cutting. Yes, the carbide grades I use can deal with the slower speeds of a manual operation, and last a LONG time. In fact, I've not dulled or chipped any yet thogh I admit I don't get to use them much. I do occasionally use ground HSS tools for large pitch or uncommon shape threads such as buttress, though it's quite rare.

Back to the original posting, the first things I'd verify would be the gearing and settings on the lathe to assure it's set at the required pitch, and verify the nut to be the correct pitch as well. Note that a UN thread isn't finished at nominal O.D. For instance, for a 1/4-20UN external RH thread I usually use a 20p full profile insert, turn the O.D. to 1/4" and then let the cresting insert take the O.D. down to .246" so I know when it's done.

HTH

rohart
12-25-2012, 08:14 PM
A great deal of the discussion so far is really about how a good threader can get great threads.

If these are the OP's first threads, he should forget about measuring the right place to stop. His threading looks quite good for a first attempt. His mistake was to stop cutting so early.

As you take more off, first the nut won't start. Then it goes halfway on. Then, and you're cutting one thou at a time buy this stage, the nut goes halfway on, and stops because less is taken off where the work is deflected away from the tool. This is where you don't feed the tool in, but you take a cut or two at the same depth. Only after a few cuts at the same depth with the nut refusing to thread on all the way do you advance a little more.

When is does go on all the way, then you test the nut the other way round, especially if you cut the nut yourself.

When the nut goes on, that's when you start being critical about the thread, the rough flanks, the radius of the tool point, should you have brought a 60deg file out before the last cuts, and so on.

You can cut a thread with a hacksaw blade. It'll just be a poor thread. Anything pointed will give you a thread that lets a commercial nut screw on. So starting with a good 60deg toolk like the OP will get you a pretty good thread.

And as for thread depth, the radius of the tip is critical - not because of the thread profile that will result, but because of how it influences what you call your start position. If your tool is sharper than the standard, when you advance it to touch the work the flanks will be further from the work than they should be. So to cut to where the flanks are correct you'll appear to have to go deeper than standard. The thread will be weaker too, but that's beside the point. You'll have set your start point too early, so you can't go by the published or calculated thread depth. My threading tools tend to be too pointed, so I'm always having to cut deeper than the book says.

luthor
12-25-2012, 09:49 PM
[QUOTE=danlb;818472]The need for side clearance also depends on the size of a bolt. A 20 TPI thread on a 1/4 inch bolt is almost straight up and down, but put it on an 8 inch bolt and it slants quite a bit. The more the slant, the more you need to take it into account.

I know what you are trying to say here Dan but you have it the wrong way around, the 1/4 bolt would have more lead angle than the the 8" one.

oldtiffie
12-25-2012, 10:09 PM
Top rake on a threading tool is not a good idea. Look at any manual and you will see that. Look at any threading insert and you will see it doesn't have any back rake or front rake, it is absolutely flat. When you add any rake you will change the V form of the cut and no longer get a perfect V shape. The cutter also has to be as near centered on the axis of the work as possible as well as 0 rake. Also, the side rake on the leading edge of the cut has to be enough to give clearance on the pitch of the thread. A 40 tpi doesn't need as much side clearance as a 4 tpi.

Not so sure about that:

http://i200.photobucket.com/albums/aa294/oldtiffie/Lathe_misc/I-Fanger3.jpg

http://i200.photobucket.com/albums/aa294/oldtiffie/Lathe_misc/I-Fanger2.jpg

The top and side rake are ground on the edge/corner of a pedestal drill - by hand - to the guage supplied which compensates for what may be an error otherwise to get the correct thread form.

It is not difficult mathmatically to allow for top and side rake to get the correct form - same as tilting the threading tool to match the helix angle on threads with a helix angle about or greater than 5 degrees.

Thread grinding wheels are off-set to the helix angle and if I recall correctly, the thread angle on the tilted wheel is mostly but not always about thread angle + 1 degree. Thread grinding is just a variation of thread cutting on a lathe.

J Tiers
12-26-2012, 12:22 AM
No matter WHAT the back rake, IF the cutter edge is radial when the thread is finished, the angle will be correct. You can also do the math and figure a corrected angle that cuts the corret form.

Those things are not necessarily easy, and the first only really works with one thread pitch, so it's far easier to make it flat with 60 deg angle and feed radially. But it isn't required.

Carld
12-26-2012, 12:28 AM
As willmac said, it's possible to put top rake on the thread form cutter but the side angles must be computed and ground so the cut will maintain 60 deg. So the question is, how many home machinists can do that or for that matter how many professional machinists can or will do that. The other aspect is, how practical it that.

The thing is, beautiful and perfect threads are cut everyday all over the world with a 0 rake 60 deg cutter so why would you want to complicate the job? The fact that something can be done a different way doesn't make it the most practical or efficient way. On the other hand, it the material requires back rake because of it's machinability then you may have to resort to using back rake.

big job
12-26-2012, 07:01 AM
I agree with Carld . 99.9% of our threading we do a releif cut. Maybe this will
help. First we know the maj. and minor. We know the pitch. Most of the time
the stock (out of the stock pile) will be turned to the major. THAT should tell
you if you are dealing with junk. If machinable we make a releif cut with
parting tool to the root, the termination point, also the starting point with
left hand threads. No top rake, no honning just good ole sharp 60* grind.
Then getting close to the releif cut, try a test nut if not, spring cuts till you get
a perfect fit. Soften lightly with file done deal.. Also yrs ago rainy Sundays
I made a box of male/female all sizes pitches all up to 3inches left and right,
those are my test gages and I stamped them all. We do a share of threading
and all my 3 lathes have thread stops= cant live without them. sam

big job
12-26-2012, 07:19 AM
Another good example problem i just thought of. Last month I
had to machine a pulley and long fine thread both ends cant
recall- but one end 3/4 & other 11/16 for a crankshaft Ford F350
460 cid (snow plow pump). The idea we will locktight the small
threaded end in the crank and the 3/4 will sandwich the whole
assembly. Absoultly horrible looked like a threaded broom
wooden handle. So thats a case of junk steel like pig iron,
but thats all I could find. Perhaps it just could be his problem.
So more downtime and my pickup dont run on air to get
machinable stock>>>good thing, by the hour plus materials.
I made a second reply cause I dont like long ones sam

J Tiers
12-26-2012, 08:20 AM
Don't forget that rake, or positive cutting angle, at least, is available other ways, NOT just with back rake.

You can, for instance, groove the middle of the 0 deg bit so that there is a positive angle inwards toward the center of the cutter. Similar to some cutoff tools. That will not affect the 60 deg angle, although it may complicate sharpening, since sharpening will also slightly lower the edge.

Dr Stan
12-26-2012, 08:52 AM
I had no clue as how to determine where to stop cutting depth so I used the pitch gauge, visually compared it to the screw from the hardware store and the OD.

Here are a couple of "tricks".

1) Use the double depth chart on the fish tail. Look up the TPI and divide the depth by two. Use it as a reference point on the slant depth. Since the slant depth is greater than the straight depth you should be too shallow. Start taking .001 or .002 inch cuts on the slant depth until it fits.

2) Use a magic marker or layout dye on the OD before you start threading. When the marker or dye is almost gone you're very close to depth. BTW, I do this to help see my touch off for setting zero on the cross slide.

Mcgyver
12-26-2012, 09:33 AM
Mcgyver, I don't agree with that.

thats what make the world interesting :) I agree in a lot of cases workable results could come from either approach....in the other thread I outline my why's as to preferring zero rake.

btw, thanks for the nice compliment!

Carld
12-26-2012, 10:12 AM
big bob, you and I cut threads about the same way. For each thread I cut I have a nut to use as a test thread that I have run a tap through and if the thread was tight and some metal cut out of it I save it for a test nut. I have made a test nut from round stock and drilled and tapped to the size I need. It's a poor man's go-nogo gauge. I have a paper chart with single and double depth of threads on it and I cut just short of the double depth and then finish up with fine cuts and testing with the nut until the thread is the way I want it. I also use a fine file to dress the OD of a thread and then use one of the square thread chaser files to clean the thread and may even clean the thread with a wire wheel on the bench grinder. I get a nice finish on most metals.

I now use a stand up insert threading holder for most my threads but I have a selection of HSS and HSS cobalt cutters from 1/4" to 3/4" I use at times. Some metals just don't thread nice but most the time a nice finish can be had with patience.

Jim Doherty
12-26-2012, 01:49 PM
Woo Hoo! I read some more of the responses and tried again with the same tool grind and a bit of oil. Success!

It appears my problem was not cutting deep enough. I cut it to the same depth as before then kept creeping up on it until the nut screwed on cleanly.

http://i837.photobucket.com/albums/zz293/jcdoherty/Picture.jpg


Thank you all for the advice and help.

Now I'm gonna be looking for an excuse to cut threads.

Jim

Davidhcnc
12-26-2012, 02:23 PM
Now that you have that cracked you need to do some internal threads, square threads, square internal, acme.........

aboard_epsilon
12-26-2012, 05:43 PM
Youre probably bumping it at the end of the threading operation ..you should have a relief cut at the end of the thread ..of the root diameter

bumping it will cause the tool to move and loose sinq..beginners should always have a relief ..

You should also have a lead in of the root diameter..of about 3/4 of an inch ...with it supported by the tailstock

When it scratches the lead -in you're at the right depth.

Sometimes people cut most of the thread then use a die to clean up after.

You should also use neat cutting oil or specialist threading grease or oil .

60 degree files can be used as a last operation ..stroke it with the file whilst its turning.

You can also wire brush the threads .

you need also :-

To make sure you're on centre .

and that the tailstock is trammed horizontally and vertically in to compensate for wear in the ways .

always use the lock on the tailstock barrel ..or your barrel will wobble about on an old lathe.

Make sure that there is no play in the lathes gears ..some gears may have a shear pin that is locked on to the gears ..play in this gear will screw the threads up .

Always reverse right to the end of the lead in..this will give the gears chance to take up the backlash

Take spring cuts for final stages .


all the best.and happy christmas ....markj

rohart
12-26-2012, 06:14 PM
[QUOTE=
Make sure that there is no play in the lathes gears ..some gears may have a shear pin that is locked on to the gears ..play in this gear will screw the threads up .

....markj[/QUOTE]

I disagree.

With my lathe there are about 10 gear to gear contacts when threading, three in the headstock, three in the change gears and another four in the gearbox, and then there's the play in the shear pin.

To estimate the play, I set up my lathe to cut a 1mm pitch, and locked the saddle down. Then I rocked the chuck forwards and backwards gently by hand without forcing it. I got about half a turn of the spindle before I felt the play get taken up.

I can cut nice threads. I always feed straight in - I haven't tried the 29 deg feed in yet.

I suggest that what's happening when cutting a RH external thread with the lathe turning normally and the saddle traversing towards the headstock is that the net forces on the tool are towards the tailstock, and that all the play is taken up, and that this state never changes throughout the turning process. The amount of play is irrelevant. More play would still all get taken up.

I guess that I could find a feed in angle - probably minus 20 degrees or so, ie feed in from left to right - that would put enough forward force on the tool to mean the forces were balanced. Then play would matter, because you would have no control of where the saddle was in its 'play window'.

aboard_epsilon
12-26-2012, 06:24 PM
That one you just quoted comes from my latest experience with my 1024

it had a undersize shear pin fitted ...and would sometimes it would take up the backlash in this ..sometimes not...because the whole gear can be set up to be loose or tight ..or medium

never overlook anything ..just passing on anything im aware of.

all the best.markj

rohart
12-26-2012, 06:34 PM
But my point is twofold mark.

1) That even with a reasonably tight lathe that cuts nice threads you have play,

2) That the play is automatically taken up by the geometry of the cutting forces anyway.

If your experience was of a saddle that would wander in its 'play window', then you had cutting forces that allowed, or even encouraged, that effect.

As long as the cutting forces add to saddle friction you should be OK.

I suggested cutting a RH external thread normally, but feeding in at 20 degrees backwards, as a way to balance the forces and get into trouble.

I don't intend to try to analyse it at this time of night on boxing day, but with RH vs LH, internal vs external, tool at front vs tool at back and right way up vs upside down cutting, there may be many ways to run into this trouble quite inadvertantly.

When you experienced your wander, were you threading totally normally, or were you using one of the more unusual geometries ?

aboard_epsilon
12-26-2012, 06:52 PM
i had the tension on the gear that held the loose shear pin ..at medium ..so sometimes it would take up the back lash ..sometimes not .

saddle wonder ..when i had that it was caused by oversize felts getting crushed ..that's something else ive experienced...it can ride on new felts instead of the ways .

all the best.markj