.

What diameter and material was this bar.?

Nick

I would argue a 3 jaw is probably the cause of the runout, you've done pretty well as it is to eliminate that much runout.

Think about it, the jaws are never going to register an absolutely true position on a self centre chuck. The only way to do this is to use a 4 jaw or between centres turning.

Moving the tailstock from left to right will only eliminate any relative bias on that axis. On a self centre fixing, the tailstock can only be used to offer support for an unsupported part, to a decent degree of accuracy, but cannot offer total accuracy.

What makes the conundrum worse is that if you centre drilled the part in the 3 jaw, that will also reflect any corresponding chuck runout when tested on the bar's outer unmachined surface.

Silver steel is a better material than mild to test onto because it's often purchase precision ground.

Michael W

Edited By Michael-w on 14/01/2017 00:42:56

Hello Allan,

If the lathe is set up properly you should be able to turn accurately. Runout is a problem only when you move the work piece in the chuck. I am not familiar with Harold Hall's methods but I am sure they will be a good guide to set up your lathe. Regards, Michael.

I agree Nick, nothing to do with runout.

With the 70mm sticking out I'm assuming that the bar enlarges away from the chuck.

0.005mm ain't much

When it's an error that small try running the same cut several times before you do anything else.

Before even doing this personally I would adjust all of the gibs for slight resistance especially the saddle on this lathe. You can get minor differences based on where the feed starts from / the saddle moving slightly when the cut starts etc. It can even twist slowly as the cut progresses. Various things can happen and depending on the lathe the taper it produces could go either way.

Tailstock – assuming the error on the headstock is that small I would simply adjust it over a turned length of 150mm or via a test bar. It's a very small adjustment to make though. A DTI can help with that so the amount of movement can be seen. A test bar is a good idea really because it allows the relative height of the tailstock to be checked by running a dti across it at several places. That can also give some indication of a bowed bed. Tailstcok centres are usually a tiny bit high. Some might strain the bed to correct it. Personally I would probably re ream the morse socket providing correcting the error doesn't enlarge it too much. But I have only done that on modern bed designs with vertical V guides. That makes suggestions on how to do it difficult.

I had one of these lathes when they 1st started coming over. I oiled the bed up, removed the lead screws and set the gibs up for slight resistance. It produced a nice finish but I was very unhappy when I started mic'ing work on it. The headstock was way out and the tailstock about 1mm too high, quill sloping as well. The headstock could be shimmed up. I eventually sold it to some one who just wanted to shorten bolts. It was an "improved" model over the original Chinese one- hand scraped tailstock and centre distance increased from 18 to 20" by chopping off the taillstock nose. Couldn't turn up to the end of work with a centre in then. Noticing that is when I though it's got to go.

John

–

I though that too Neil but little points can improve numbers like those and the man might be used to better. Nearly if not all dovetail beds have short saddles so good adjustments in that area are more or less essential. Thing give as well and 0.0025mm isn't much at all.

Actually if the one I had wasn't a pigs ear I would probably still have it. I'm glad they have started making them a lot better than they were initially.

Super 7 ? – might be out of the frying pan into the fire.

John

–

.

Sorry, Michael … I'm confused … Did Colchester really make the Cromwell lathes ?

**LINK**

http://www.lathes.co.uk/cromwell/index.html

MichaelG.

.

I see that you edited whilst I was posting my question.

Edited By Michael Gilligan on 15/01/2017 14:24:58