Normal Chuck runout?

[BrooklynBravest]

finally got a dial indicator and trying to determine whether the Chuck is spinning true or not.

I placed the indicator on the backing plate of a 3 jaw Chuck and rotating the Chuck slowly by hand it appears to have a .001 to .0015 runout to it.

Is this within the acceptable range?

[John Evans]

The OD of the backing plate has no bearing on the chuck run out. The mounting face and pilot of the plate should be under .0005 . a better check is with a truely round pin/bar in the chuck jaws. Remember even with runout at the jaws if you make your part in one setting without removing the part from the chuck all features will true with each other. Very inportant that the chuck to spindle mounting surfaces are spotless with no chips-burrs etc.

[BrooklynBravest]

It has a screw on chuck, and by eye I can see a slightly disassociation if you will of the point where The Chuck meets the spindle. Again I don't know if the OD of the backing plate at the threads is relevant either. I have not checked the spindle on its own yet, I don't own a test indicator just a plunger type.

[Harold_V]

If you can visually identify runout, it will be greater than .003", as the human eye can not discern less. That's not generally an acceptable amount of runout if it relates to the spindle of a metal working machine.

You commented that you have only a plunger type indicator at your disposal. No problem, so long as you can make a setup that permits the use of such a device. In truth, the resulting reading will actually be more reliable, assuming the device has the required sensitivity.

Before you worry about how well a chuck runs, you should determine if the registers of the spindle are running true. There should be a diameter that serves to force concentricity of the chuck with the spindle, as well as a shoulder that guarantees perpendicularity. Both of those should run true. If they do not, there are issues with the spindle or bearings.

Small diameter spindles are known to get bent in a crash. That might ccount for any visual runout you may see.

Harold

[BrooklynBravest]

The spindle has like a 3/8" non threaded area. Would unscrewing the Chuck and checking that spot reveal anything?

[BrooklynBravest]

Maybe this helps?

http://youtu.be/D0KK-MzMCOc

[Harold_V]

The spindle has like a 3/8" non threaded area. Would unscrewing the Chuck and checking that spot reveal anything?

Depends. It may or may not be relied upon for use as a register. If there's a "collar", for lack of better description, just beyond that area, then it is most likely the register, and serves both purposes (concentricity and perpendicularity). A close examination of how the chuck backing plate is constructed should reveal how it registers. The surface involved will have limited clearance with the mating surface, thus overcoming any possible irregularities in thread concentricity.

Harold

[BrooklynBravest]

What exactly is considered a register?

[spro]

"register" is the combination of unthreaded spindle neck and the face at which the backplate mates. The threads of a chuck/ backplate are loose for a reason but the clearance is minimal between the unthreaded bore. The mate of the back surface of the plate to that spindle "face" , further aligns to the register.
Since that face is the final determination, it was advised to check the spindle run- out and face.

[spro]

If that is okay, the plate can "sometimes" be screwed on backwards and some checks made. You would need some DI or adapter to the plunger type. The idea is to see and record the run-out of both sides of the plate. This can show you a variance between the opposite sides of the plate.

[Harold_V]

Spro's description should prove to be more than adequate, but if you have any doubts about what you see on your machine, try posting a couple pictures, closeups--one of the back of your chuck, detailing the bore on the mounting side, and one of the spindle nose, showing the exterior (interior of the spindle, for this purpose, is not important). Between those two pictures one should be able to identify surfaces used as registers.

Harold

[spro]

It is more likely that the neck face of a threaded plate has been faced before. This depends, of course, but the threaded area presents a smaller id to the unthreaded register and can't face correctly unless forced. When that happens, the threads are being distorted. It was a practice once to know when that happened and place a precision ring shim or bore it out further. I look at this from older stuff yet, newer replacements may not have the same depths of thread or register bore.