View Full Version : Lathe Concentricity...It's the little things

12-03-2010, 12:33 PM
This may be old news to some of those here who are experienced, trade machinists. For those folks, just jump to the next thread. This will probably be elementry school stuff for you. But for the real HSM guys here, those who are learning on their own, this may be of some help.

I've been using my 10" Logan Lathe for a number of years now but haven't been all that pleased with the concentricity of the chucks at times. Yesterday I decided to take some time to take a closer look at things. When I placed an indicator on the spindle I found that it was concentric and has about .0005" side movement when a heavy load is applied. Not too bad for a machine built in 1943, in my opinion. When a test bar was placed in the 3 jaw chuck I got .007" runout 3" from the jaws; not acceptable, to me anyway. Inspection of the faceplate where it meets the spindle register proved that there is clearance at that point and I began to suspect that might be the issue.

At the risk of opening a can of works here the story takes a slight turn. I did some research with regard to clearance between the chuck back plate and the spindle register. Over the years this has been a subject of contention with many saying that this should be a close fit to insure concentricity while others say the threads do the work. After reading two articles, one at http://www.lathes.co.uk/latheparts/page7.html and the other at http://www.metalwebnews.com/howto/chuckmnt/chuckmnt.html I came to the conclusion that the fit at the spindle register wasn't my problem. Further research provided the following link http://metalwebnews.org/mr-tools/lathe-chucks.pdf at which time a light bulb went on in my head.

After bluing the back of my chuck back plate where it meets the face of the spindle register and screwing it onto the spindle I could see where it was only touching in one area. Judicious scraping and about 20 or 30 repeats of the process, blue, install, inspect, scrape finally brought me to a 360 degree fit of the back plate to the spindle face. The results were an incredible improvement! My 3 jaw chuck now centers within .003", which is what my South Bend Lathe "How to Run a Lathe" books states a new 3 jaw chuck should run out at. With this knowledge I decided to check my other chucks. My four jaw independent back plate was fine so I didn't have to make any changes there. My Jacobs Headstock Chuck with 1 1/2- 8 threads was another story. It ran out at .011", 3" from the jaws. Terrible! My South Bend reference tells me it should be more accurate than my 3 jaw chuck at .002" runout. I blued the back of the Jacobs chuck, screwed it on the spindle and sure enough, there was a high area. Not knowing how hard the Jacobs chuck might be I decided to attempt scraping anyway. I learned that it isn't hard at all and scrapes easily. Further, as luck would have it, after only one scraping cycle I got full 360 degree contact with the spindle. A check of concentricity with my test bar proved that runout was now .002", the specified amount. What an ah-ha moment!!

I am much happier knowing that my lathe now can produce work within tolerances that were acceptable when it was built and for most HSM work to this day. If I ever get requests from NASA to do their lathe work I may have to upgrade, but for now the improvement was well worth the small effort required.

12-03-2010, 04:00 PM
Good post, and congrats on your success.

So often, threads are devoted to scraping, rebuilding, replacing spindle bearings and all manner of involved "solutions" to machine tool accuracy when attention to the basics will as often as not correct the problem.