How to true my lathe chuck?

[BBchevy396]

I have a 30" x 240" Lathe, Low hours, with a 18" SCA three jaw chuck, It runs out about .003" . I tried grinding the jaws with a grinder mounted in the toolpost, a bearing race clamped in the chuck , and turning about 30rpm. I got it closer than the .006, it was when I got it, but cant seem to get any better then .003" Am I doing this correctly? Any thoughts? I cant afford to replace the chuck for $5000.+. I think it is the masters because i Also measured the run-out with the top jaws off. I dont want to replace them either $1000+, they look good, so does the scroll. Any one proficient at this? Thanks, Jim

[mxtras]

Could it be the headstock bearings? Maybe it's not the chuck, Jim.

Have you checked for slop in the bearings?

Scott

[HuFlungDung]

3 jaw chucks can be temperamental. Mind you, .003 is not bad considering that is only a real misposition of only .0015. Is it perfectly repeatable at .003 runout no matter what diameter you are chucking on, or no matter which way you rotate your part in the jaws?

I'm not clear on your grinding method, as you should actually stress the jaws in the radial-out direction when you grind the inside. Otherwise, you will get a bit of jaw tilt which will manifest itself as an unstable workpiece, with greater and greater runout, the further you get from the chuck. You can detect if the jaw taper is significant by clamping an accurate piece of shafting in the chuck, then dial it in or note the high spot, and hit the shaft with a deadblow hammer, and notice how easy it is to obtain signficant deflection of the alignment.

I have ground lathe jaws by drilling a shallow hole in the top of each jaw. Then, make up a ring with pins to fit the holes and clamp the ring, with the jaws pressing inwards. Any tilt in the jaws and jaw slideways will then be compensated and you can grind a straight cylindrical pass inside.

Rather than grinding the heck out of the jaws, I think I might consider buying or machining up an adjust-true style backplate for it. That is, with the push-pull setscrews to move the chuck radially, then lock it in place with the normal backplate bolts.

[DareBee]

Wow .003 TIR on that chuck is great!
It seems to me that if you used proper procedure to grind the jaws then something is wrong (wore?) somewhere.
I piece of paper under the low jaw should give you perfection.
PS my new 4 jaw that size was $1g and I can indicate to less than .0005 TIR (not what you wanted to hear but just a suggestion if .003 isnt good enough)

[Mcgyver]

like the others have said .003 is top notch for a three jaw. Personally I don't care much whether its .003 or .006 - If it needs to be exact, use a collet or 4 jaw, if not whether its .003 or .006 usually doesn't matter. Remember though everything turned with one set up will be perfecty concentric - ie the lack concentricity is only between the turned surfaces and the surfaces gripped by the jaws. If you plan your work so that the turned and bored surfaces are done at one set up it is a convenient chuck to use.

[BBchevy396]

Obviously, I didn't grind the jaws exactly right,. I ended up with a slight amount of taper that is screwing things up. I dont have a problem with most projects, but I repair alot of Boat shafts., approx. 1 1/2" to 4"Dia , 6' to 20' long . Weld buildups, machining, and straightening. The problem manifests itself when turning down to original dia., or on shaft center repairs.... Some idiot screws a puller into the center, without protecting the shaft, and destroys it.... or worse.... uses the brute force, sheer ignorance, bigger Hammer method of removing couplers and propellors , gotta have a center! well, most of the time, it is nice!
I'll try to source a boring ring, and try it again! Thanks for the input!

[BBchevy396]

Forgot to mention, I've got a 30" & 15" 4 jaw, and dont have a problem with them, but gotta like a accurate 3 jaw, especially if removing & rechucking the work repetedly!

[DareBee]

Originally Posted by BBchevy396 Forgot to mention, I've got a 30" & 15" 4 jaw, and dont have a problem with them, but gotta like a accurate 3 jaw, especially if removing & rechucking the work repetedly!

Definately agree with that.

[Mcgyver]

Originally Posted by HuFlungDung I'm not clear on your grinding method, as you should actually stress the jaws in the radial-out direction when you grind the inside. Otherwise, you will get a bit of jaw tilt which will manifest itself as an unstable workpiece,

while imo the tir is not that important, holding the work an angle is! while i've done it, grinding jaws is not something i do daily, so take this as half question half statement....HuFlung I was thought that you had to stress them radial-in because of backlash and how the jaw gets loaded against the body (ie where's the fulcrum cuz there has to be some clearance). I've use a turned ring on the inside at the very back and then undercut afterwards

BB have you indicated to see how much of a taper there is? This could also be dirt/chip on the mounting or even under the back plate, what happens when you indicated the chuck body?

[HuFlungDung]

Ya, my statement was not very clear. As you did, chucking a ring on the inside, that is correct. The jaws have to be loaded the same way for grinding as they do in use.

[One of Many]

I have ground many jaws and the first thing to note is that it will only be accurate at that position to which is was ground. The scroll and master jaws will have far to many variables in their mating surfaces to pull off much better than .003 in a different position on a worn chuck. Some abused chucks can be much worse. That is where boring soft jaws is the best solution for each common size for max gripping strength.

As long as the top jaws are longer than the master jaw where the grip surface is. I chuck up on an 1/8" thick disk on the master jaw surface to leave the top jaw grip surface free to grind. It is the same basic procedure used for boring soft jaws too. The disk should put the jaws chucking diameter at something within the most used grip range.

If the master and top jaws are of the same grip surface or will be ground that way to prevent a step, a ring can be made to bolt through on the outer steps of the jaw. The idea being to keep the jaws forced outward as if it is gripping. Keeping the grinding wheel dressed will be crucial so the full surface is on the same diameter and let it spark out on finish.

There are several methods to doing this procedure. Here is another description.

DC

[fyffe555]

Not a machinist and so I don't know if this is correct(!) but I recall being taught at Uni. methods of improving machining accuracy. For three jaw lathes the routine was to take a gauge bar of the same diameter of the piece you needed to work and chuck it up and measure runout. The bar was of known tolerance and roundness and usually was a bar originally used on a surface table tolerance measuring set. If the runout was within a few thou after chucking in different rotaed positions the next part was to chuck up the guage bars counterpart, a big and heavy cast iron ring with an internal diameter of the same tolerances as the guage bar. When chucked on the top jaw with the ring against the chuck the jaws would be at the required work peice diameter. IE target 1" work piece, gauge bar 1", ring 1" plus the depth of the jaws. Then we went through a few thou deep cleaning up with a cross slide mounted grinder passing until the sparks stopped.

EDIT: forgot a bit. When the guage bar was in the chuck we ground the *outer* faces of the jaw to mate with the ring, before chucking the ring to the newly ground jaw.. \EDIT

Our next shop assignment was to make a gauge bar and matching ring set for another size which was a whole other story.

As you can imagine at a university the lathes where soundly abused and new chuck jaws where a common supply.