Trying To Grind Jaws

Happycamper

Active User
Registered
Joined
Mar 20, 2013
Messages
153
I have a Fuerda 6" 3 jaw Chuck that is off.... meaning using a precision test bar, it is app .001 at the Chuck and .005 out of round ten inches from the Chuck. I have refaced the back plate and have been trying to grind the jaws but I can't get the jaws to true up. I am using a steel round to tighten the main jaws on to while I run the grinder in and out on the outside jaws. I have leveled the lathe in X and y axis. Any one have any suggestions?
 
Level has no bearing on chuck runout. You need to preload the jaws from the outside, something that can be tricky. There are some clever ways to do it, but they require machining certain tooling aids. If this is a 2 piece jaw chuck (bolt on top jaws and fixed master jaws0, you can use a piece of plate and drill 3 holes @ 120° at the proper place and using longer than standard bolts, afix it to the top of the removable jaws, then tighten the chuck against that. It's a lot of work, but mimics the forces usually seen when chucking a piece of material. That's the key. Think of how the forces play out when you have a piece of material in the chuck.

Another thing that people do is simply put a short piece of bar at the very back of the jaws, and tighten the chuck on it. Of course, this will cause the jaws to close more on the front, because there is no support from any material. Then they grind as far back as they can without crashing into the short bar. When they finish grinding the jaws, they remove that short bar and just leave the chuck loose and grind the short area where the bar was larger than the front area where they ground with the bar in the chuck. Naturally, and especially on an older chuck with wear on the jaw slots, this will cause a bit of bell mouth on the chuck. So it isn't really a great way to do it. If you can outguess that taper and grind accordingly, you can end up with usable jaws though.

Also beware that on on older chuck, if you do grind the jaws, and they come out running true, wear on the scroll may cause the runout to be even worse in positions other than where you ground the jaws. Like different sizes of test bar. May be fine on a 1" but horrible on a 3/4", or just the opposite.

Just a couple of things to think about. There are a few threads here showing how some of our members managed to improve their chucks. A search should bring them up, or some of those members will chime in on this thread.
 
N.O.B. (nature of the beast). Three jaw chucks are not designed to give better tolerance than what you are getting. Most are worse. Use them where you can, and use another work holding method for precision work centering. All that being said, if you chuck your stock in a three jaw chuck and then turn it down in place, it will take the chuck out of the equation, at least for the stock protruding from the chuck. That does not always give you what you want. Due to the scroll design, three jaw chucks are also not particularly good at repeatability. The only other real options are a independent three jaw chuck (built like a four jaw) or a "set true" or similar chuck where you can dial in the entire chuck to center the work. Regardless, you will get different results with each tightening of the jaws and with different jaw opening sizes. It will not be completely predictable or repeatable.
 
Tony and Bob have provided sound advice. If your chuck is off 0.001" at the jaws, and 0.005" 10" further out - that sounds pretty good. If you must trim the jaws, it needs to be done with the jaws loaded in the clamping direction.

You did the right thing by working on some of the basic issues associated with tuning up a chuck. There are a number of other steps involved in chuck tuning and I'd suggest you work through those first (steps posted on this board and others as well).

Of course, you have cleaned the chuck internals well (that is a given) and checked for burrs on all mounting surfaces.

Start by making sure that the back plate mounts well on the spindle. It must repeat. What sort of mount do you have. If you have a D1 (which is what I have so I will comment on it), then mark the orientation so it goes on the same every time. Then tune the taper of the back plate (if the taper in the back plate is a touch small, then it is handy to have a second lathe near by - polish a little out with alox cloth ). The definition of the "correct fit" is somewhat of a personal preference. I believe it should just stick on the nose taper and there must be ZERO gap on the face of the spindle. I worked the taper in the back of the chucks just a tiny bit and found a significant improvement.

Once the back plate mounts well on the spindle, then you skim the mounting surface (where the chuck bolts on). It is up to you if you make the register to the chuck snug (to locate the chuck body), or leave it slightly loose and bump the chuck body around before the final tightening (sort of like a "set true"). Then check it all out - that should be good (after all, as Bob pointed out it is a 3 jaw chuck, it won't be perfect). Be sure to try the chuck on all the possible orientations available on the back plate (no guarantee that the way it came to you is the way it was set up in the first place - perhaps one position is better than the others.

If you still are not satisfied, true the back mounting face of the chuck. To do this, use another chuck (say a big 4 jaw) and skim the largest bar with will fit into the body of the 3 jaw - do not remove the bar from the 4 jaw. Make the bar long enough to go through, drill a center hole and support the bar with the tailstock. Then grip the bar with the 3 jaw, positioned with the mounting surface out and check for axial run out. Take a skim cut off the mounting surface of the back of the 3 jaw.

Grinding the jaws is sort of a drastic last measure. The typical 3 jaw chuck has 4 gripping surfaces on the jaws (the ID and the 3 outer steps), and the face of each of the steps are supposed to be in the same plane (so you can quickly reference your work by pushing a piece against that step). If you grind one of those references to a new position, then the others are going to be OTL (out to lunch). To re-reference the clamping and reference surfaces of a chuck is a lot of work (not for the feint of heart).

A 6" 3 jaw is not an expensive chuck. I suggest you work through the basics (most of which you'd have to do on a new chuck anyway), then call it "good enough".

Let us know how you make out,

David
 
Thanks for the replies. What is frustrating is chucking up a long piece of stock, center drilling it then moving it out to be supported by the tail stock. Five thou in ten inches means ten thou in 20 inches which means a bend in the stock when you engage the tail stock. This is a LOO mount Chuck and is about a year old. I have checked the run out on the LOO spindle which is dead nuts on. I skimmed the back plate and it is dead nuts on. I preload the main jaws using a 1" socket which the ID is a little more than that so a 1" grinding wheel fits. I mark the surfaces to be ground with a marksalot and note that all are being ground. I understand the inaccuracy of three jaw chucks but You would think that the ground surfaces would be parallel to the rotation but it comes out slightly canted it seems. Bugs the s**t out of me.

Jerry the unhappy camper.
 
Hi Jerry, The pursuit of precision is endless. You never actually get there. You will get better, but will never get to perfect. I believe it is worth it to keep trying.

What are you building that is long and thin? Sounds like a candidate for working between centers. Of course working a relatively long / thin piece has its' own challenges as it is impossible to eliminate tool push (i.e. using a follower rest, keen narrow tooling, adjusting DOC as you go, and so on).

David
 
I'm not sure they are actually canted. If they are swinging around on the centerline of the spindle they really can't be. Now if they come out tapered, either belled open at the front, or back, they merely grip the stock in that area and the stock hanging out is really unsupported, and not rigid. When you put your indicator on the stock or test bar several inches out, you should be able to "bump" it to where it runs true. If you can't bump it and have it stay where you move it, then the chuck actually is holding by the full jaw length, or a good portion of it. If there then is runout on a piece a few inches out from the chuck that you can't move, definitely there is a jaw problem. But I don't believe that would or could come from grinding.

I'll go ahead and ask the dumb question: You are spinning the chuck under power when you are running the grinder in on the jaws, aren't you? Don't be insulted, please. But it is possible to create the problem you seem to have if you ground the jaws one at a time with the chuck stationary.
 
This is what i use to lock the jaws 3 square gage blocks, then grind the jaws. If that is not good enough make it into a buck chuck so you can cram it into any size dia. you want. I did that to a 6" cushman works great.

101_0728.JPG
 
Kern, please tell me those aren't Jo Blocks. Though, I suppose they could some that are worn out.
 
I personally wouldn't trust the angles ground on the jaws. I doubt they are precision. And yes, please say those aren't good gage blocks at least!
 
Back
Top