Craftsman 6" Tailstock Alignment

[ChandlerJPerry]

Hey all! After my chronicle with the Atlas 12", I've been very pleased with my Craftsman 6" Mk II. I've been trying to get the tailstock aligned, and aside from the inherent difficulties of accomplishing that, I've noticed a peculiarity. I've been turning some CRS over about a 10" length, one end in a three jaw chuck and the other supported by a live center, making small adjustments to the tailstock to try and get it aligned. I've noticed that the middle of the part is consistently a larger diameter than either end, giving sort of a "barrel" shape if that makes sense. It's not by much, usually around .010", but far more runout than I would consider acceptable. Is this also a consequence of tailstock alignment or something else? I don't think I'm causing the workpiece or tool to deflect, as I'm only taking .005" per pass.

 

[benmychree]

You have not said what diameter the CRS is, it is likely tool spring that is causing the barrel shape, the piece need not be so long for such testing, and then there would be less deflection. You also could use a very short piece and center drill it and run a center in the tailstock into the center drilled hole and use that to align the tailstock.

 

[ChandlerJPerry]

You have not said what diameter the CRS is, it is likely tool spring that is causing the barrel shape, the piece need not be so long for such testing, and then there would be less deflection. You also could use a very short piece and center drill it and run a center in the tailstock into the center drilled hole and use that to align the tailstock.

That would make sense, I'm new to all this so I had assumed a longer piece would show the tailstock deflection better, but I hadn't considered the inaccuracy it might produce. The CRS is 1/2".

 

[benmychree]

Another thing is the two collar method; visualize a 1/2" rod with two collars machined on it maybe 3/4" diameter with enough rod to chuck and the other collar at the tailstock end with the center drilled hole supported by the center, the collars made perhaps 5/16 or 3/8# wide and perhaps 4" apart, very light cuts are taken on each collar without adjusting the tool, and the difference in the diameter after the cut noted and the tailstock adjusted accordingly until they come to the same size with subsequent cuts.

 

[wa5cab]

OK. Diameter is 1/2". But what is the length? If more than about 4", that is probably your problem. About the only other possible culprit is a bent bed.

Also, use of a chuck to hold the headstock end is not a good idea. For this operation, you should have a dead center at the headstock and a live center or lubricated dead center at the tailstock. Also, cutting more than 1/2" to 1" at each end is not really necessary. And the female centers at each end should be cut with a proper center drill. Not with anything else.

 

[ChandlerJPerry]

OK. Diameter is 1/2". But what is the length? If more than about 4", that is probably your problem. About the only other possible culprit is a bent bed.

Also, use of a chuck to hold the headstock end is not a good idea. For this operation, you should have a dead center at the headstock and a live center or lubricated dead center at the tailstock. Also, cutting more than 1/2" to 1" at each end is not really necessary. And the female centers at each end should be cut with a proper center drill. Not with anything else.

Thank you for the advice. I had assumed a chuck might introduce accuracy issues but unfortunately my lathe did not come with a dog plate or any dogs, so for now I can't mount a center to the headstock. Would a 4 jaw maybe be more permissible? If it's a lost cause I will wait until I can acquire a drive plate and some dogs, otherwise I'm just working with what I have.

 

[wa5cab]

In this case, 3-jaw, 4-jaw, 6-jaw, etc. all introduce the same possible error.

 

[ChandlerJPerry]

On this topic, how can one avoid that deflection when working on a long piece? One of my projects I hope to make soon is a vise handle that will probably be about 12" in length. Would a follow rest keep everything straight? I am still unfortunately without a follow or steady, I'll have to watch eBay for both.

 

[timmeh]

If you are turning the entire length(10"), yes it's work piece deflection.
"Level" lathe first.
Turn a work piece 3:1(approx.) ratio, 3 long x 1 diameter(checks spindle alignment), similar to two collar method.
Then 2 collar method, supported by live/dead centre, over longer distance(held by 3/4 jaw or dead centre will get you close.
Bed wear will then be your only issue(must be taken into consideration when applying first test).

 

[ChandlerJPerry]

Thanks for the advice everyone, I switched over to a piece of 3/8 aluminum about 4" long and was able to get it dialed in after one cut. I nudged the tailstock over .003" with a dial indicator and now it seems to be cutting true at both ends .