# How do you calculate decimal threads?

I'm making a 5c collet closer, the collet thread afaikt from the internet is 1.238-20. I'll freely admit, I generally just look up threads in machinery's handbook and
That one isn't in there.
I'm thinking that boring my chunk of 1 3/8" 4130 tubing to 1.238 - 1/20 is a possible solution before threading, much like calculating a tap drill. Then using my selection of collets to check for fit is a plan that may work.
I'd really like to know how to calculate this as in I'll build a spreadsheet for future oddball threads.
Thanks,
Wes
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>

We kicked this one around a while ago. I measured my 5C set to make the closer for a drill grinding fixture and found a range of over 0.005", so I cut the thread to fit the largest collet, ~1.245 IIRC. http://picasaweb.google.com/KB1DAL/Tools#5288898039281271986 The 1/2" bushing is to center the shank of an S&D bit.
Cutting in a calculated distance from the initial contact only works if the OD or ID is correct and the tip of the tool is right for that thread pitch. If you round the tip properly you need a separate tool bit for each pitch, PITA.
I don't usually bother. The external threading bit tip is eyeballed to shape for about 8TPI and the 29 degree infeed on my spreadsheet threading chart is thus too small by up to about 0.005", a reasonably safe value. I finish the thread with a tap or die if possible, or cut internal threads to fit after making and measuring the external one.
The formulas on my chart are 29 Infeed = 0.75/Pitch for nominal tip flat width Root diameter = OD - 2*0.6495/Pitch. Tip flat width = 0.125/Pitch, wear limit is 0.25/Pitch They are pretty much from Machinery's Handbook, slightly adjusted so mistakes leave metal.
Root diameter is listed as the -offset which I subtract from the measured OD. 0.6495/Pitch would be the minimum amount to feed in for the ending groove, but the crossfeed dial is zeroed for the threading bit instead of the cutoff tool.
jsw
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Wes,
An American National Thread Form, 20 pitch, 75% thread is calculated: Nominal diameter - (1.29904 x .75 / 20}. Thus you would bore the internal diameter to 1.238 and then single-point cut the threads with a 60 degree threading tool to a depth of 0.2013.
Bob Swinney
I'm making a 5c collet closer, the collet thread afaikt from the internet is 1.238-20. I'll freely admit, I generally just look up threads in machinery's handbook and cut the thread.
That one isn't in there.
I'm thinking that boring my chunk of 1 3/8" 4130 tubing to 1.238 - 1/20 is a possible solution before threading, much like calculating a tap drill. Then using my selection of collets to check for fit is a plan that may work.
I'd really like to know how to calculate this as in I'll build a spreadsheet for future oddball threads.
Thanks,
Wes
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Robert Swinney wrote:

I hope you misplaced a decimal point. :-) ...lew...
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>

For 20 TPI I get an infeed of 0.038" on the angled compound or 0.032" on the crossfeed, assuming the tip flat width is 0.006" and you zero the dials when it contacts the work. Also close the halfnuts at any line on the threading indicator. I haven't thoroughly tested the spreadsheet's accuracy since I don't reshape the tip for every job, but it's close.
jsw
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Lew sez: "I hope you misplaced a decimal point. :-) Naw! You're threading internally. The threaded portion of the bore would increase by 0.2013
Bob Swinney
Robert Swinney wrote:

...lew...
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>

DoN Nichols wrote a little program that John Beidl made usable under Windows. I've been using that for years. http://www.metalworking.com/dropbox/_1999_retired_files/Threadsw.txt The one I use is v5.0.0.0, 2 Mar 99. It's worked on every Windows machine I've owned, currently running Win XP Pro.
A formula that seems to work for internal threads is: ID = OD - .975/TPI
I made a spreadsheet with these values and printed a chart that is taped to my lathe.
It's a guide, not gospel. I've made parts to fit a mating part 1500 miles away (and they did) but with internal threads it's really best to have a try gage at the lathe if at all possible. Perhaps not so true with ultra-rigid CNC machinery, I don't know anything about that.
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
I will agree the only way to do it properly (even on a cnc lathe) is with gauges (go and no-go). This is an odball thread but using the machinery handbook you can figure out the proper dimensions. On page 1733 of the .pdf version of the 17th edition it has a diagram of the unc/unf thread series. You can use the values there to calculate your thread depth and Minor diameter. If you want to make a thread gauge so that you know the threads are %100 correct on pg. 1922 of the same pdf gives the standards for computing the dimensions for a thread gauge both go and no-go sizes.
In all reality, this is not necessary if you are just making a collet closer. You can just figure out the root dimension of the 1.238-20 thread (1.19469 check my math.... ) and then cut your ID about .010-. 015 larger and cut your threads until the collet spins cleanly into the thread.
-Brian

<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>

Scratch that I was thinking pitch diameter not OD..... My brain isnt functioning today but if you can follow the diagram (make sure you look at the internal thread diagram, it's the third one) you can figure out the PD and all the other dimensions you need to know.
-B
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>

Thanks for the program tip.

Oh heck, I'm not using CNC, I'm using my Clausing 6903 engine lathe.
I managed to cut a thread that feels okay but that was after I used the compound to go in .700 deep to start a relief cut for my threading tool to run out into. Well that didn't work out so well. I had a relief .350" from the end of my stock. Darn compound reads half values. I can understand that on a cross feed, but it doesn't make a lot for since on the compound. Oh well, I'm getting to know my lathe by trial and error.
Thanks Don(s)
Wes