I just finished a cast iron adapter plate that mounts a plain-back 5C collet chuck to the 2"-8 tpi spindle of my lathe. I can easily dial in the mounted chuck to .0005" runout, the limit of my measuring capability.
When I remove and reinstall the chuck/adapter plate assembly and check the runout, it varies from .0005" to .004". It only takes a couple of minutes to mount the indicator, loosen the bolts and recenter the chuck, but I was hoping for a little more repeatability.
Jim Stewart fired this volley in news:pM2dnWvDEJ_ExMPUnZ2dnUVZ snipped-for-privacy@omsoft.com:
I think you need a different lathe for each chuck. Yeah... you're probably asking a bit much, but you can check a few things.
"Just-any-ol'" faceplate may not have a matching centering taper on its back to fit your spindle-nose. Some spindles don't have them at all.
On my Reed, part of the job of fitting a new faceplate was to cut a short, steep taper in the back of the hole, so it would JUST fit flush on the spindle shoulder as it engaged the matching male taper at the headstock end of the threaded area of the spindle nose. It was a twitchy operation (at the time) because I didn't figure out the math to determine how much to cut at taper to move the plate back xx ten- thousanths. Like another discussion going on here, I had to use plasti-gauge to measure the plate-to-shoulder gap, because no other tool I had would reach and measure there at the same time.
As clunky as that old lathe is, the faceplate mounts true to about a thousanth every time. That's still not half-a-thou, but close enough for almost everything I do with it. My 3-jaw chuck isn't that accurate.
You may have made the same error I did by expecting a snug thread fit to align a chuck. I found out that the thread was supposed to be fairly loose and the position should be established by the spigot and its recess.
I'm not so sure Don ... If he machined the recess on the back of the plate so it seats on the register face , it should be pretty close . I can remove and reseat my 4 jaw with a piece of stock dialed in and have it close enough that my dial indicator can't mearure the difference.
I've machine more 1"-10 and 1 1/2"-8 backing plates over the years than I care to remember...
Here's how I did it, and the results were acceptable at that time. Forgot whether it was .0005" or .0007" repeatability, but it always was less than .001", even on the little Atlas 6" lathe I had for 25 years.
1) Chuck the new 7 unmachined plate and turn the front face (outboard face) flat.
2) Bolt to face or sturdy driving plate. I prefer this because the work may shift in a light chuck when cleaning/boring a cored hole.
3) Face boss and bore hole to tap drill size.
4) Bore locating bore to depth, and with dia. equal to the locating dia. on the spindle, + .0003" to + .0006" or so. Personally I keep this a close as possible to less than .0005", but it depends on your skill as machinist and ability to measure accurately. Turn a 1/32" chamfer on the outside edge of the locating bore.
5) Now machine the thread to match the spindle. I made an acceptance thread gauge to verify the fit-up to the spindle without trying it, but you can unscrew the assembly and try it. If necessary it is easily returned for re-machining to open up the thread. I found the thread gauge a great time saver. As mentioned earlier the thread should be an easy fit on the spindle.
6) Because the threading operation will leave a marking of some depth in the locating bore, deburr this using emery cloth.
7) Clean the bore with a tooth brush and oil lightly, and you are good to go.
8) After removal and prior to re-installation always inspect the threads on spindle and bore, and clean thoroughly, oiling lightly.
The reason for this order of machining is that if the locating bore is finish machined after the threading operation the interrupted cut would, in all likelihood, result in a bore not concentric to the thread.
Another thing, I found commercially finished threaded backing plates have these locating dia. machined waaay oversized.
The method as described above has given predictable results.
I was taught that the chuck will center on the thread , and square on the register face . The actual mechanics aren't that important to me , as long as it works . I recently cast and machined an adapter plate with a "spindle" to mount my chucks on a rotary table . As long as I can center my work on the table , I'll be happy . Haven't had a chance yet to mount the table and check runout , the wife's had me runnin' like a rabbit this holiday season .
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.