How can I do this 3 lead thread?

I do lots of 4 lead threads, but I now need to do a 3 lead 36 TPI thread on the Atlas 6" lathe. I set up the gearing for 12 TPI,
but I'm having trouble figuring out how to do the different starts. The gearing calls for the 32 gear at the spindle through an idler to a 20/40 pair (or 24/48, 32/64) to a 48 on the leadscrew. The leadscrew is 16 TPI.
It can't be done with the thread dial alone, so I'm trying to figure out if I could slip some gears for the second and third leads. If the gear on the spindle was divisible by 3, there'd be no problem.
Anyone have any thoughts?
-Bruno
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Set your cross slide at 0 deg, machine the first thread as normal., then crank the cross slide .0277" for the next one, and another .0277 for the last one.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Dave Lyon wrote:

I've never actually cut a multiple lead thread but I've heard of this method. Also if you can turn the part between centers, use a faceplate with holes (or whatever) 120 degrees apart to drive the dog which rotates the part for each new cut. I've also seen lathe chucks with 360 degrees marked around the perimeter. I'm guessing that you'd just loosen the chuck somehow, turn it the required distance and re-tighten in the new place.
My question is how deep to cut the threads? Can you still use thread wires? Do you have to cut all three threads once then measure?
Randy
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Turn between centers using the 3 jaw chuck to hold a scrap of bar stock turned to a 60 deg. angle for a center. Put the dog against each jaw in turn to maintain spacing. I often use the 3 jaw with a turned center in it for turning between centers. I have a piece of stock marked so it goes in the chuck the same way and I don't halve to take a cut on it unless I'm doing close work. I can take a cut on the center quicker than I can change the chuck to a drive plate and back again. Tom

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

I don't understand what you mean by this. The degree settings are on the compound (not cross slide) and that's set to 29 degrees to cut the thread. Are you suggesting I cut the thread at 0 degrees going straight in?
But using your logic (I think), I just ran an experiment. I set the compound to 30 degress and cut a lead as normal. I then reset the tool by bringing the cross slide in by .0481" ( 1/36 * sqrt(3) ) and the compound out by .0555" ( 1/36 * 2 ) for the next pass and again for the third pass. Under magnification and with a thread gage, it looks just right.
THANKS! -Bruno
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Sorry, I did mean compound. That's exactly what I meant. With small threads, going straight in usually doesn't produce too many problems.

OK, I had to trig it out, but I got the same numbers you did. That works too.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Spindle/Idler * Idler/Lead
32/24 * 48/48 is one suggested gearing.
This is the same as 48/24 * 32/48
and now you have a spindle gear which is a multiple of 3.
Will this or some other equivilant fit?
Bruno wrote:

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
The spindle stud gear is fixed (Atlas 618).
A trigonometric solution has been found. Far easier than slipping gears.

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Another route that works is to set the compound at 30 degrees. Set your compound and cross slide to zero for just touching the thread surface. Cut your first thread. The back out the cross slide. Advance your compound by twice the thread width (sin30=.5). In your case 2/12=0.1666. Set your cross slide so its .866 (tan30=.866) times this value further out , in your case .1666/12= .01389. You're now just touching the bar one thread further ahead. This works for any number of thread leads.
--
<()> An apple a day keeps the doctor away.
Two apples a day gets the doctor's OK.
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Karl, this is basically what I just did, but used 36, not 12, as the thread width and it worked. Is that what you meant?
-Bruno
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Ya, and I also meant cosine not tangent. Listen to what I mean, not what I say.
Karl
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

====================One way is to use the compound to offset the thread by a given amount, but this prevents you from using the compound [set at 30 degrees] to control the depth of cut of the thread. The offset would be one thread pitch or 1/36 or 0.027777778 with the compound set exactly parallel to the spindle. If you do this don't rely on the compound/top slide graduations, but use an indicator and test bar between centers.
IMNSHO, a better way is to slip a gear.
What gears do you have available?
All the intermediate compound gear combinations you list do is a change of 2:1 between the spindle gear and the lead screw gear. Any combination that will physically fit will be OK. This is important because it may make the gear you require available for the spindle/leadscrew.
To get 12 TPI with a 16 TPI lead screw requires when the spindle makes 12 revolutions the lead screw makes 16 revolutions [to advance 1 inch] thus you need an overall ratio of 12:16 or 3:4.
The problem then is to select a gear set from the ones you have with a spindle or lead screw gear that has a factor of 3:4. You can use the intermediate compound gears to allow the use of spindle/lead screw gears with even (sub) multiples of teeth. A spreadsheet program is a great help here. Assuming your gear set starts at 20 and goes through 100 you would have:
Spindle        Lead screw gear gear for 4:3 ratio 20        26.66666667 21        28 22        29.33333333 23        30.66666667 24        32 25        33.33333333 26        34.66666667 27        36 28        37.33333333 29        38.66666667 30        40 31        41.33333333 32        42.66666667 33        44 34        45.33333333 35        46.66666667 36        48 37        49.33333333 38        50.66666667 39        52 40        53.33333333 41        54.66666667 42        56 43        57.33333333 44        58.66666667 45        60 46        61.33333333 47        62.66666667 48        64 49        65.33333333 50        66.66666667 51        68 52        69.33333333 53        70.66666667 54        72 55        73.33333333 56        74.66666667 57        76 58        77.33333333 59        78.66666667 60        80 61        81.33333333 62        82.66666667 63        84 64        85.33333333 65        86.66666667 66        88 67        89.33333333 68        90.66666667 69        92 70        93.33333333 71        94.66666667 72        96 73        97.33333333 74        98.66666667 75        100
Of course only the ones with whole numbers of teeth are physically possible. Remember you can use intermediate compound gears to divide or multiply the teeth required by 2/3/4 etc. Thus you could use a 75:50 gear with a 30:60 [1:2] compound to get the same affect 75:100 = 3:4 75/50X30/60 = 3:4
If you frequently do multiple start threads, an indexing faceplate can prove very helpful [I am assuming your are threading between centers.]
The older machining books frequently have examples of these shop made tools.
"Machine Shop Methods" by Milne has an example on pages 131-132. He also has a short section on multiple start threads on pages 130-131 which may be helpful.
Milne is available from Lindsey books as is a reprint of change gear calculations.
see: http://lindsaybks.com/bks/milne/index.html http://lindsaybks.com/bks7/sscut/index.html
What are you building -- this sounds interesting.
BTW you may wish to double check the set-up information. It should be 36:48 with no compound gears for 12 TPI with a 16 TPI Lead screw, just series transmission gears which won't affect the ratio. As 48 is 3 X 16 slipping 16 teeth would seem to be the easiest method to get a triple lead.
Unka' George (George McDuffee) .............................. Only in Britain could it be thought a defect to be "too clever by half." The probability is that too many people are too stupid by three-quarters.
John Major (b. 1943), British Conservative politician, prime minister. Quoted in: Observer (London, 7 July 1991).
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

    One way to do it is to orient the compound parallel to the bed (and don't use it for infeed as is usually done), and cut the first thread, then advance the compound 1/36" (0.0278")and cut the second. When that one is complete, advance another 1/36" (to a total of 0.0556") and cut the third pass.
    It means having to do the infeed with the cross slide instead of the compound, but apparently the UK normally cuts threads with direct infeed instead of the angled infeed which the compound gives you.
    Now to see what others have suggested.
    Good Luck,         DoN.
--
Email: < snipped-for-privacy@d-and-d.com> | Voice (all times): (703) 938-4564
(too) near Washington D.C. | http://www.d-and-d.com/dnichols/DoN.html
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
if you go with the 24/48 idler they are divisible by three
DoN. Nichols wrote:

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here. All logos and trade names are the property of their respective owners.