What's new
What's new

Threading a barrel, accuracy

02dag

Plastic
Joined
Jul 27, 2012
Location
Fjellhamar, Norway
I have a barrel which I want to thread with 16 TPI.

I have a lathe that has a 2mm leadscrew. The closest I can get to 16tpi is 15.875 TPI

Is this accurate enough or will the nut get stuck at some point?

I have already tried once, but didnt manage to screw the nut on entirely. Not sure if thats caused by the threads though.


I am using a normal 60degree cutting tool. And I am practicing on something else than the barrel.
 
I would want to get closer than that. I have lathes with inch lead screws that use 127 tooth gears for converting to metric threads. I would assume going the other way would use the same conversion gear.
The make and model of your lathe might help with getting a closer number or finding the correct conversion gear set.
 
Just keep cutting deeper, the nut will fit but the results are not satisfactory. You will end up with a poor fit, not a very good option.
 
That's a .0629 pitch. 16 TPI is .0625 pitch. You would be losing almost .001" every two turns. How sloppy can your thread fit be? How long is the threaded joint?
 
That's a .0629 pitch. 16 TPI is .0625 pitch. You would be losing almost .001" every two turns. How sloppy can your thread fit be? How long is the threaded joint?

When you put it that way it does not sound so bad. I read a book where someone described doing these mismatched threads. Assemble them with high pressure grease and a some of the threads get "moved" into place.
 
When you put it that way it does not sound so bad. I read a book where someone described doing these mismatched threads. Assemble them with high pressure grease and a some of the threads get "moved" into place.


Guess it's a matter of opinion. To me this just seems a good way to jack up both the barrel AND the receiver.

JMHO

-Ron
 
Guess it's a matter of opinion. To me this just seems a good way to jack up both the barrel AND the receiver.

JMHO

-Ron

It has been done in the past, I have bought 2 metric transposing sets because I did not want to be a hack. Some have called me too picky. 300sniper pointed out it is only .005 off in ten turns. With a couple thousands clearance the barrel threads could move a little if assemble with a good lube. I think most receivers are harder than barrel steel. The OP has not been back to see if there is a better solution with his lathe.
 
I have a Chinese mini lathe type wm210v. They are all in mm, but I bought an extra gearset for imperial threading. I have done some calculations and can do it more precise if I can get a gear with 47 teeth as an example. I can also buy a 16tpi lead screw probably.

This is for an ar15 barrel with barrel extension nut. The length of the barrel where the nut is located is approx .625 long. The threading is a little shorter since there is a small area near the barrel nut shoulder without threads. I am using an extension from Brownells.

The spindle gear is 56 teeth. Lead screw is 2 mm and is on the EF axle. This is what I have to work with

Screen Shot 10-17-20 at 04.58 PM.jpg
 
It has been done in the past, I have bought 2 metric transposing sets because I did not want to be a hack. Some have called me too picky. 300sniper pointed out it is only .005 off in ten turns. With a couple thousands clearance the barrel threads could move a little if assemble with a good lube. I think most receivers are harder than barrel steel. The OP has not been back to see if there is a better solution with his lathe.


I will admit that I'm a little OCD about stuff like that but I think of it as I do any tool, etc. - the threads just won't be working "as designed". In ten turns the "first" thread would be off by .005, the third off by .004, the fifth by .003, etc., etc. so you are applying different amounts of force to different threads. Will Lock-tite even set up properly if there is no force being applied to the thread?

It's things like these that make me want to do it "correctly".

JMHO

-Ron
 
Lathe cut threads will have small variations in effective diameter all along the thread amounting to more than the small difference quoted ,effectively varying the pitch...what you lose on the swings,you pick up on the roundabout .....nothings perfect ,unless the threads are ground on a ultra precision machine.
 
Just looking at it...the first thread may be ok but the rest won't be engaged in the right way so load initially will be on the "first thread" then transfer to the others and to do this it will have to move.
So Just send it out to a person who can cut the thread properly i think will be best.

on edit
so i agree with ron ...
 
On thinwall structures ,as the tenon and receiver ring are,the thread will self adjust for contact by flexing /distorting the whole assy......so the concept of a 001 cumulative error in the pitch is only theoretical ,and doesnt happen when the fittings are tightened to 150ft/lb or so........We have also assumed the lathe leadscrew is in as new condition ,and wear in the lead screw doesnt apply its own variable into the thread.
 
Thank you all for informative answers. I will use the links for gear calculations which @mickri provided. I have already had a quick look at the one for my lathe, and it seems that I can get to 1.587 or 1.588 mm which in tpi is 15.994 or 16.005 which should be close enough.

Edit
Turned out that the calculator suggested three different gearsets for 15.994. Two of them would only work theoretically because the gear holder would touch the spindle. The last one worked

Setup is now 56,80-72,66-52,50 which is spindle, AB,DC,E
 
On thinwall structures ,as the tenon and receiver ring are,the thread will self adjust for contact by flexing /distorting the whole assy......so the concept of a 001 cumulative error in the pitch is only theoretical ,and doesnt happen when the fittings are tightened to 150ft/lb or so........We have also assumed the lathe leadscrew is in as new condition ,and wear in the lead screw doesnt apply its own variable into the thread.


I understand what you are saying but the cumulative error is only theoretical in that it would vary +- by some percentage but would still be cumulative unless the variable +- is greater than 100% of the error. If one wants to get really picky, we don't know just how accurate the threads on the receiver are cut and if adding this cumulative error may push it over the limit or possibly make things better.

In this situation you can probably make things work but in general principle I try to eliminate as many known/preventable sources of errors as I can (that OCD thing) as I have enough trouble with things I can't control. :D

As always - JMHO,

-Ron
 
Thank you all for informative answers. I will use the links for gear calculations which @mickri provided. I have already had a quick look at the one for my lathe, and it seems that I can get to 1.587 or 1.588 mm which in tpi is 15.994 or 16.005 which should be close enough.

Edit
Turned out that the calculator suggested three different gearsets for 15.994. Two of them would only work theoretically because the gear holder would touch the spindle. The last one worked

Setup is now 56,80-72,66-52,50 which is spindle, AB,DC,E

That is close enough by any standard.

And for general info, Loc-Tite does not need "pressure" to cure. Only the exclusion of "air" is required. NEVER squeeze all the air out of the bottle of loc-tite before you cap it. It will set up! read the instructions! ;-)
 
That is close enough by any standard.

And for general info, Loc-Tite does not need "pressure" to cure. Only the exclusion of "air" is required. NEVER squeeze all the air out of the bottle of loc-tite before you cap it. It will set up! read the instructions! ;-)

I used to think it was exclusion of oxygen that set Loctite. Talked to a technician at the company and it is certain metal ions that accelerate the hardening. Stainless and aluminum are essentially inactive with steel, iron, copper and brass being active and fast setting. Instead of using another product to accelerate the hardening on inactive alloys you can fixture them then heat to 150 to 220F to accelerate them. In the past I have had threaded assemblies with Loctite start hardening way before made up. Have not tried it but chilling one or both parts of an active assembly may slow it down until the joint is made up.
 
The first 2 or 3 threads take the vast majority of the load in any case. Like 70% or more. And if it's a mil spec breech extension, the threads have probably been cut before hardening, and could be distorted a bit anyway. They're supposed to be masked before carburizing. I dont remember if the spec was thread/mask/heat treat, or mask/heat treat/thread. But all I've seen show color from heat treat.

Your error is pushing the threads to tight on the outside faces, so to speak. So you're loading the first threads in the direction you want, the way they'll be when tightened.

A well known action maker used to thread the receivers with a tapered thread, of slightly different pitch from the standard on the barrel. This made for full engagement of all threads when torqued. They abandoned this when they discovered nobody reads instructions anymore ;-)
 
I have been single pointing threads off and on for about 35 years now as part of my job. Whenever you set up to thread there are minute differences in the results you get. Number one--take the tool post off and dry out any oil under it. You cannot clamp tight enough on an oil film to hold. Number two--clean your leadscrew with plenty of WD40 and a shop towel till the dirt and grime is truly gone. Number three--take some time and set the tool up as closely to the thread gage as you possibly can with the compound set at precisely 29 1/2 degrees. Number 4--clamp a thin aluminum shim to the workpiece with your tool tip and confirm that it is truly pointed absolutely straight up and down. Number four--check your gearbox thread settings again, even it you have cut that thread fifty times. Number 5--watch for metal build up on the tip of your threading tool and keep it stoned off. Number 6--Thread at the fastest speed you can reliably stop and start. Number 7--you can still get screwed so watch absolutely everything else as closely as possible.
 
Thread fitment of receiver/barrel is detrimental if too tight- you don't want tight threads.
What's critical, is the facing of the receiver ring and barrel shoulder to be perfectly perpendicular to the threads.
"Loose" threads, within reason, are perfectly fine. Threads bring the parts together- it's the shoulders that align the parts.
 








 
Back
Top