Runout Wobble on the Chuck?

[John Hasler]

I agree that turning between centers is the best way for Carlos to produce a test bar. That will give him a for sure concentric piece.

Any time I use a center with a chuck (something I try to avoid) I always drill the center hole and turn with the same setup. Otherwise the tailstock fights the chuck and the part loses, as you point out. I see no reason Carlos should need to do that, thought, even if he isn't equipped to turn between centers. He only needs a short, fat bar.

I don't see how turning in a sprung chuck would give him a lobed part as long as the piece is held rigidly [Edit: and without a center]. I do agree that it's best not to do so.

[earlgo]

Harold said, "Turning between centers is the way to create straight pieces, meaning pieces free of bowing."
This may be true for a reasonable diameter to length ratio which depends on a lot of variables including material, strength, etc. Otherwise someone would not have invented the follower rest.
I wish I had not thrown away my 3 sided turning resulting from bum chuck jaws, it would have been a nice illustration.
This is the way I 'fixed' a backplate that was not concentric. I drilled and inserted 6 pins in the edge of the suspect boss and turned the pins to the correct diameter.

backplate with pins.

It worked and I didn't have to buy another expensive backplate.
--earlgo

[SteveM]

This is the way I 'fixed' a backplate that was not concentric. I drilled and inserted 6 pins in the edge of the suspect boss and turned the pins to the correct diameter.Backplate #1.jpg
It worked and I didn't have to buy another expensive backplate.

Very creative and effective!

Steve

[Harold_V]

I don't see how turning in a sprung chuck would give him a lobed part as long as the piece is held rigidly [Edit: and without a center]. I do agree that it's best not to do so.

Simple. When a chuck is sprung, the amount of support of the bar is limited due to the short area of contact, at the rear of the jaws. The short contact area permits the bar to oscillate in the chuck as it rotates. Tool pressure will cause deflection, with the work piece deflecting less when it's aligned with the jaws. That results in a lobed turn. The longer the extension from the chuck, the more it will be lobed.

A short test piece, in this instance, isn't a great way to go, as one is trying to determine how far the chuck is running out, and why. The longer piece will provide a clearer display, assuming the piece is held true to the chuck. That would define how much error the chuck has, if any.

If the chuck is sprung, testing with a long bar may not provide much information, as the jaws permit the work piece to be gripped well even when the piece is not parallel with the theoretical centerline of the chuck. In a case like this, it can be difficult to determine the actual condition. Soft jaws would eliminate the problem, so the shaft would be gripped in a proper attitude.

H

[John Hasler]

Harold writes:
> When a chuck is sprung, the amount of support of the bar is limited due to the short area of contact, at the rear of the jaws. The short contact area permits the bar to oscillate in the chuck as it rotates.

And therefor it isn't being held rigidly.

I wrote:
I don't see how turning in a sprung chuck would give him a lobed part as long as the piece is *held rigidly*

[armscor 1]

Don't know how a Gudgeon is ground but maybe makes a good test pin and readily available.

[Harold_V]

Harold writes:
> When a chuck is sprung, the amount of support of the bar is limited due to the short area of contact, at the rear of the jaws. The short contact area permits the bar to oscillate in the chuck as it rotates.

And therefor it isn't being held rigidly.

I wrote:
I don't see how turning in a sprung chuck would give him a lobed part as long as the piece is *held rigidly*

My advice to you is to try it and see for yourself. I've had the experience, and I understand why it happened. If it's not obvious to you, you would be best served to have the same failure.

That said, it happens because there's less support of the part between jaws than there is when the cutting pressure is aligned with the jaws. For a short cut, where one does not violate the three times diameter rule, no problem, but for a long shaft (which we're talking about), it is exacerbated by the additional unsupported length. Chucks that grip at the rear tend to be useless, and this is one of the reasons.

I trust you read comments from earlgo? "I wish I had not thrown away my 3 sided turning resulting from bum chuck jaws, it would have been a nice illustration."

H

[GlennW]

You can hear the difference in the cut when comparing a three jaw chuck and a 5C collet when cutting small diameter stock. The 5C has full support of the stock.

[John Hasler]

Harold writes:
>For a short cut, where one does not violate the three times diameter rule, no problem, but for a long shaft (which we're talking about),

I thought we were talking about a short, fat shaft just long enough to check the runout of the chuck.

[Harold_V]

Perhaps you were talking about a short shaft, but I wasn't, and I made that clear when I suggested turning between centers.

H

[Mr Ron]

I think the only true way to determine if the headstock spindle and bearings are true is to mount a long bar between centers using a dog and driving plate. If the turned diameter at both ends of the bar (say 12" long) are identical, then you will know the spindle is not bent and the bearings are still good. You also need to make sure the lathe is leveled (bed not twisted). (Advise from a wannabe machinist).

[Harold_V]

That test won't provide any useful information. Because the tailstock may not be in alignment (and most likely isn't), it won't reveal anything except that it is, or it isn't. A combination of twisted bed, misaligned tailstock and bent spindle can still yield a straight shaft. Given good fortune, it would still be round, however. The shaft, due to the bent spindle, would have a turn that isn't concentric with the center hole at the headstock end (due to the bent spindle). The same results can be achieved if the tailstock center is live and does not run true (the tip is not concentric with the bearings). You'd be able to determine if that's the case by inspecting the resulting shaft by two methods---one of them being to install the shaft in surfcenters, or reversing the shaft in the lathe (with dead center, not a live center, in the tailstock), then spinning it on the centers by hand with a decent DTI used. Any eccentricity would then be obvious.

H