Help cutting pipe threads on a lathe

[Joe Miranda]

Usually when I program our lathe to cut pipe threads I go and get a pipe tap and measure it and program the lathe according to those dimensions. Then I try the corresponding male or female part and tweak the "x" offset until the two parts fit together to the Machinery Handbook's "handtight" engagement spec. While this method works - it is not very scientific. I have read and re-read the Machinery Handbooks description of pipe threads and I admit that I am not smarter than a fifth grader. So, if any of you guys understand this formula, would you mind elightening me so that I can make the mathemetacial computation based on an actual formula rather than measuring a tap?
Thanks,
Joe

 

[johnoder]

Here is a handy chart. It does give root dia. at end of pipe, and also outside dia. at end of pipe. This info is in the upper left hand box third and fourth up from bottom of that box.

Warning - I do not know if this was checked by whoever made it.

http://img.photobucket.com/albums/v337/johnoder/PipeData.jpg

 

[Joe Miranda]

Here is a handy chart. It does give root dia. at end of pipe, and also outside dia. at end of pipe. This info is in the upper left hand box third and fourth up from bottom of that box.

Warning - I do not know if this was checked by whoever made it.

http://img.photobucket.com/albums/v337/johnoder/PipeData.jpg

John,
Thanks for the info. It looks like all the info I would need but it would sure be nice to have a mathematical formula to figure this out. Thanks for taking the time to answer this post.
Joe

 

[Forrest Addy]

It's a real trick to make "accurate" pipe threads. You almost have to have a gage handy. They like all threads are measured over three wires but because the threads are tapered there's another component - the axial location of the three wire measuremet on the thread. A last revinement has to do with the inclination of the micrometer anvil following the taper and the cosine error that imposes on what should on a measurement that should ideally be made in a radial plane. Too many high falutin complications.

When ever I made taper pipe threads in the absence of proper gages I used new commercial pipe fittings and tried for 1 1/2 to 2 1/2 threads of hand engagement (this varies somewhat with the pipe size). This crude method requires an irrational hope and faith that fitting I used as a gage was somewhere in the middle of the tolerance band.

The best solution for batch production of pipe threads is to have proper gages on hand (Maybe you could borroe or rent them from another shop).That way the customer can't come back with a hassle you don't have a good answer for.

You see pipe thread gages on eBay from time to time.

 

[Keith Krome]

Just curious, but how is the theoretical perfect thread (now referred to as complete thread*) generated, measurement wise?

Is it governed by a formula?

*seen here:
http://www.aasfasteners.com/glossary_of_screw_threads_gaging.htm

 

[Joe Miranda]

pipe threads

"It's a real trick to make "accurate" pipe threads."

Hi Forrest. Always good to hear from you. Yes I'm familiar with measuring threads over wires and thread gages. As a matter of fact, I have this particular thread gage. But my question is, in order to cut pipe threads on a lathe, you have to know the starting and ending diameter. Well, in the Machinery Handbook, it gives us the maximum diameter. OK -that's great. I now know what diameter to turn my shaft. Here's the rub though, how do I scientifically and mathematically calculate the smaller diameter at the beginning of the thread without going and either measuring a tap like I said in my original post or measuring a pipe nipple or coupling like you said you used to do in lieu of proper gaging? I know that the angle of a pipe thread is 3/4" per foot which works out to 1 deg. 47 min. per side. That's all fine and dandy, but it still doesn't answer my question. The Machinery Handbook gives the pitch diameter and the height of the thread and the truncation of the crests and roots for NPT's, but try as I might, no combination of adding, subtracting, multiplying or dividing has yielded me a formula that, when it has been worked out, comes anywhere's near to the diameters of the pipe tap when I measure it. So, once again, I humbly admit to this venerable audience of esteemed professionals and craftsmen, I AM NOT SMARTER THAN A FIFTH GRADER.

Joe

 

[Forrest Addy]

This highly experienced and baffingly intelligent expert has cut threads the mathematical way any number of times but 9.8 out of 10 times he sets the tool, the thread pitch, touches off, sets his zero and cuts taking passes until it liooks like a thread. Then he deburs, measures, and gages taking additional cuts until the thead is in tolerance. Then he notes his settings and goes on to the next thread. No hocus pocus with double depth etc. Go for the end result.

Sorry. What can I say? I'm an analog guy.

Unified and Metric threads are narrowly specified and all the thread parts have been established by their respective standards and all the manufacturing data may be calculated from nominal via mathematical formula. However it takes time to crank through all the numbers and no shop management wishes to pay you to calculate thread data out to a warm fuzzy conclusion. Thus, I "meatball" the thread cutting process making as good an end product as one who spent hours over calculatons.

That said the derivation of standard V thread proportions is a subject worth studying in depth. It leads to a greater understanding of details like tip and root flats, diameter, pitch, and length components when determining thread tolerance and allowance for the various thread classes, and so-on.

For that I reccommend what used to be "Handbook H-28 Screw Thread Standards for Federal Services" originally published by NBS but now it's maintained by NIST. The 1959 issue had the most complete formula set. All Unified screw thread standards are set forth in this standard. Machinery's Handbook covers this meaterial in an abbreviated form that varies in quality depending on edition. H-28 by the way comes in three parts. The basic info is on part I, a monstrous array of tables is in part II, and gaging info and acme material is in part III - I think. Better get all three to be sure of complete info.

Many of the gaging manufacturers of the era had appendices in their catalogs and handbooks. You might take a look at Van Keuren Handbook #36 Vol II at the pipe thread material commencing on page 117. There's a lot of data there but not too much that's helpful.

Also try to find a text on mechanical metrology dating from the 50's. I can't ecall the name or author but it was published by Prentice Hall. The book I'm thinking of had extensive discussion on the checking of tapered threads, gaging, and maybe some material on calculation.

Somewhere, sometime, everything in the machinist's trade has been reduced to a standard and published. The trick is where to find it without paying through the nose for a copy from some standards organization. A good techncal library is a great place to start.

 

[Mark McGrath]

Follow this link.On the lower right of the page there is a link for threading software.Download it,you have to register in that it asks for your name and company but nothing else.It`s very easy to use and covers a lot of threads.

http://www.atistellram.com/Threading/default.htm

Mark.

 

[CCC]

This doesn't directly answer your question, but have you looked into pipe thread gages? The company I worked for turned out NPT-threaded parts by the millions per year, and all the machine setup was done with these. They're like a standard go-nogo thread gauge (precision ground threads on a plug or ring), but there is a flat ground on the side of the threads to show nominal depth of engagement. You screw the gage in to lightly snug, and compare the position of the flat to the end of the threads. Much easier setup than the three-wire method.

 

[Toolznthings]

Hello,
If I understand your question. You can trig out the base of the triangle formed by the overall length of the thread since you know the angle off the centerline.
Formula is: B x Tangent of D = Base
B = thd. length , D = Thd angle, Look up tangent of thd angle.
Double the end result and subtract from OD to get small dia at the beginning. Hope this is understandable and what you are looking for. If not, ignore me since many do !

Brian

 

[huntinguy]

It's in the book, or it should be.

http://cgi.ebay.com/ASME-Gaging-for...-Standard_W0QQitemZ320203883185QQcmdZViewItem

This book will tell you anything you want to know. and lots you don't.

(I have no affiliation with or knowledge of this seller, I post it for reference, A Google search will bring up other copies)

Forrest: I think for pipe I would rather use the 4 wire method, that way I get my taper as well.

 

[Forrest Addy]

Huntinguy. A four wire method will land you in the soup when you aqttempt to measure a single lead tapered thread. Any way you do it you will have one wire loose - unless you're planning something really fancy with a sine bar

Joe, The congugate motion of the cutting tool and the spindle describes a cone that your programming will have to take into consideration. Calculate the elements and offsets to generate a minor diameter cone. When making initial tool settings, I suggest you touch off at the right end part corner and run your sequence of threading passes from that initial point until the tool tip follows the calculated minor diameter cone.

Don't get lost on PD and other threading data. When CNC cutting threads, the tool path says it all and the minor diameter represents the final pass of a series of threading passes from the initial touch off.

Or if you are tool setting, establish an offset X,Z pair that's based on the location of the initial point located at the right end and stock diameter. Set the tool and the tooling offsets from that calculated point. Then work out the stick reduction from the initial point until your tool gets to the minor diameter cone. Don't worry about cutting the major diameter. That will come automatically from cutting the thread. It will be burred up but a light dress with a file will remove them and the crests can be dressed with a touch of a slip stone.

 

[huntinguy]

Huntinguy. A four wire method will land you in the soup when you aqttempt to measure a single lead tapered thread. Any way you do it you will have one wire loose - unless you're planning something really fancy with a sine bar QUOTE]

Yep, sine bar and guage blocks. A guy has got to have a little fun Besides, it makes the boss think I am smart

 

[Joe Miranda]

pipe threads

This is hilarious - all of our totalled life's experience and the sum of us can not come up with the formula. No one has answered my question. I give my sincere appreciation to all of you for your time. But the truth of it is NONE OF US ARE SMARTER THAN A FIFTH GRADER.

Joe

 

[Rstewart]

Whats this deal with the "Fifth Grader" ? Sounds kinda Lame!

 

[Joe Miranda]

pipe threads

Tell ya what rstewart - since you don't have the answer to the question - you join the group! How lame is that?!

Joe

 

[huntinguy]

No easy answer

I think the difficulty is that there is no clean answer to the question you are looking for. If you want the angle- then that is in the book. If you want the tread pitch at any given point, that ain't going to happen. The pipe thread is based on root contact, hence the need for a guage to measure taper and then a guage to measure pitch. One controls the engagement and the other controls the seal. Hence from my earlier post, you can make a thread that fits the thread guage but fails becuse the taper/root does not compress correctly.

The B1.20.5-1 spec will give the very best direction to the question you are asking and show the locations and ways of calculating the tread pitch at a relational location.

Forrest could most likely explain it better.

 

[jkilroy]

Overlooking one aspect of pipe threads

One thing being overlooked here is that pipe threads are supposed to go on PIPE! The problem here is that pipe is specified by its ID and the OD can vary quite a bit. Thats why the handbook gives a max diameter.

The specified angle for all NPT threads is 3/4 inch per foot. The thread angle is 60 degrees, truncated roots and crests are flat. THAT IS THE STANDARD!

If you can't make it work with that then you aren't smarter than a fifth grader. Over complication is not a mark of intelligence.

 

[Joe Miranda]

pipe threads

Alright. I got it figured out. Thanks to all who chimed in. If you need a formula let me know. It's funny really - after 30 years in this trade I finally took the time to make myself understand what the Handbook states about pipe threads. BTW jkilroy, yeah thanks - 60 degrees - 3/4 inch taper per foot - all that was already said in the previous posts and it missed the point of the question entirely. Thanks for the effort though.

Joe

 

[jimboggs]

I know I'm asking for it...........

Joe,

Please forgive this question as I am a complete moron when it comes to the machine trade, but I've cut about a jillion pipe threads over the years with my good o'le standard Ridgid pipe dies, sometimes with the aid of a power drive, sometimes by hand. I geuss I'm lost as to why break your aggots to cut these when it's already been figured out for us? Is it cause of high production count? Which would seem to justify the cost of the dies. Or are these one-offs or something that you don't want to invest in the tooling? They can be rented BTW at most regular rental places.

Regards,

Jim