New life for an old lathe

I've got a lathe. An old one.

It's a Stark jeweler's lathe, uses #3 (sometimes called 3SS) collets, and it was old when granddad bought it. Probably made 1890 +/- a decade or so.

It's still mainly usable, but examining some collets I see the threads are deformed, and I suspect the drawbar has worn excessively (so only the crests of the threads are under stress).

It's time to make a new drawbar, but my problem is this: the old references say the threaded part of those collets (actually called split chucks in the catalog) is 0.520" 26tpi, and they seem to measure at 13mm 1.0mm pitch, which is 0.512" 25.4 tpi.

So, I could buy a M13 x 1.0 tap, and reform the collets and build a new drawbar to fit, OR I could get a custom-built 0.520" 26 tpi tap and keep it historically accurate (the collets might be worn and it'll be a bit loose regardless). One of these is more historically correct, but WHICH?

In that era, it's likely the collets were made in Europe and with metric tooling, Stark being the US agent. My references might be from aftermarket measurements, and I'm unsure how accurate those numbers are.

Reply to
whit3rd
Loading thread data ...

Wouldn't it be easier to measure the drawbar, or the drawbar and collet together?

If the collets are really 0.520 and they're worn larger, then you're not going to have much meat there to do anything with a 0.512 diameter tap.

Having said all that: if you do start running taps through the collets, it may be best to use a thread-forming tap, rather than a cutting one. If the collets are worn you'd want to preserve as much material as possible.

Reply to
Tim Wescott

The 'drawbar' is a tube, and the threads are on the inside of the drawbar and outside of the collet (similar to 5C collets). I'm sorting through the collet collection to find the least-deformed threads, to do some remeasuring with a measuring microscope now...

The collet threads OD is in the 0.510" to 0.515" range, but if the crests of a 0.520" thread are allowed to be blunt, that COULD be the inch size that my old handbook shows.

Reply to
whit3rd

If you can find an M13 X 1mm bolt (which is not common) you could see if it fits in the drawtube easily. You might also check some of the other fasteners on the lathe to see if they are metric. Are some of the other dimensions of things in integer millimeters or nice fractions of an inch?

Reply to
anorton

Measure the collets and establish a maximum size and machine a new draw bar to match. A thread gauge will tell you if they are 60 degree threads.

Reply to
John B.

The best way to measure the threads is to measure the pitch diameter and the pitch. Use the three wire method to ascertain the pitch diameter and then measure the pitch with a gauge, an optical comparator of some sort, or with a dial travel indicator. Since it seems like the drawbar still works to some degree from your above description you could accurately mark the drawbar and turn it several revolutions and measure the amount the collet moves. Dividing this motion by the number of turns gives you the pitch. If the thread is a

60 degree thread form then Machinery's Handbook will give you all the info needed to either single point new threads in a new drawbar that fit a gauge that you make or have a tap ground that will cut new threads in your new drawbar. If the thread is not a 60 degree form Machinery's Handbook still has the info needed to figure out what thread it really is and what you need to do to make a new drawbar correctly. It doesn't really matter what is historically correct according to catalogs, books, etc. What matters, if you care, is making the drawbar with threads that match your collets. Eric
Reply to
etpm

As a followup to my first reply to this post, if you are unsure of your ability to figure out what thread the collets actually are I would be willing to measure one or more of the collets and give you the resultant data. You would need to ship them to me. I would even ship them back. And even though I like beer you wouldn't need to send any beer tokens with the collets. If you are interested let me know. Eric

Reply to
etpm

Are you *sure* that they are deformed? At least some watchmaker's lathes have buttress form collet threads. (Sort of like this, as close as I can show with ASCII graphics:

/|/|/|/|/|/|/|/|/|/|

so the force to draw the collet tight is on the nearly vertical face of the thread, and the other face does not need to handle much stress.

Look at a collet with a magnifier for the profile. See whether it is buttress profile. (Easiest to see where the threads are interrupted by the keyway). If so, your choices are to either find a

*very* expensive tap with the right thread form, or grind a tool and cut the thread on a lathe with the proper leadscrew gearing (likely not yours, as most of those did not have a leadscrew, let alone a gear driven one.

I would suspect metric threads, but am not sure. Check the thread profile, find a friend with a lathe which can cut each choice, and try both for fit. (Probably needs to be two lathes cover both systems accurately, especially for cutting inside threads. But you could cut a stub outside thread and mesh it with the collet to see which is the better fit.

If it is *not* a buttress profile, just check with thread pitch gauges in both systems.

Good Luck, DoN.

Reply to
DoN. Nichols

The flat faces would be on the side that's loaded when tightened, and that's backward. The threads are deformed, all right.

There's a lot of tooling, I was able to find better examples yesterday, and compare with an M8x 1.0 tap; it's not metric. My inch-dimensioned thread gage doesn't have a leaf for

26 tpi (because the standards are 24, 27, 28...), but the suspect thread DOES align well with the 13 tpi gage, so it seems I'm about to buy a custom-made tap.

I like the idea of getting an accurate diameter with thread wires, but that will require me to find the least-deformed of the original (as opposed to aftermarket) parts; it's gonna take an hour and a microscope.

Reply to
whit3rd

You mean backwards from what you see on the collets? The vertical faces *should* be the ones loaded when tightened. It lets you get tighter before the friction gets too high to allow more tightening.

O.K.

Hmm ... find a friend with a lathe which can cut 26 TPI and make a gauge with that? I know that my 12x24" Clausing offers 26 TPI along with 27 TPI. Any lathe which will cut 13 TPI should also be able to cut

26 TPI. I would offer to make one, except that my shop suffered a fire and the insurance contractors are just getting close to pulling everything out of the shop, cleaning, and evaluating for whether it can be made functional again. It is *amazing* how much acidic soot is left behind by a liter of Vactra No. 2 burning in a closed space. :-(

Hmm ... do you have access to an optical comparator? Or someone nearby with one? Here is one like mine on eBay (#370547156549), except a *lot* more money than I paid for mine.

And this one: (#140887203295) is a lot more reasonable, but takes up more room. Nice that it has a DRO. Mine just has mounting for dial indicators. If I had room, *I* would go for that one. It is the same brand as mine.

Anyway -- it can measure the distance from crest to crest, or root to root, and allow viewing and measuring the angle of each face of the thread, just in case it is a Whitworth, or something else. (It casts a shadow on a screen, magnified by 10X, 20X, or 40X (mine is 20X, the one in the second auction os 10X only without getting other lenses). Not only can you measure the angle, but you can also check for distortion on the thread faces so you can be sure whether it has been deformed or not.

And before doing the three-wire measurement, you need to know the included angle of the threads. The charts are typically for the common 60 degree threads. But Machinery's Handbook has formulas so you can calculate the proper wire size and the correction factor to the measurement over the wires for pretty much any thread angle.

And yes, flat topped crests are found on some threads.

Good Luck, DoN.

Reply to
DoN. Nichols

I agree.

Gunner

The methodology of the left has always been:

  1. Lie
  2. Repeat the lie as many times as possible
  3. Have as many people repeat the lie as often as possible
  4. Eventually, the uninformed believe the lie
  5. The lie will then be made into some form oflaw
  6. Then everyone must conform to the lie
Reply to
Gunner

If you take the new tube and the thread data along with a couple dozen donuts to a local machine shop..you may get the tube done for the price of the donuts. Or a couple 6 packs of beer.

It can be done on just about any small cnc lathe for very very little money.

Ever price a custom tap?

Cringe.....

Gunner

The methodology of the left has always been:

  1. Lie
  2. Repeat the lie as many times as possible
  3. Have as many people repeat the lie as often as possible
  4. Eventually, the uninformed believe the lie
  5. The lie will then be made into some form oflaw
  6. Then everyone must conform to the lie
Reply to
Gunner

I also have a no. 3 Stark Lathe my collets are fine and so is the draw bar. Also I have a 26 TPI Tap and die that will work on your draw bar and colle ts if you want to use them. I did for a couple of my collets and it worked great. I was able to buy one from England and they were made in India and t hey are inexpensive and they the 26TPI is common for their Bicycles. Email me at Suppo32 (at)homesc (dot) com and I can loan these to you so that you can straigten up your collets.

Mart> I've got a lathe. An old one.

Reply to
suppo32

PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.