Alan
E-xcellent poster
Posts: 234
|
Post by Alan on Jan 24, 2009 10:17:49 GMT
Hi
I took the chuck off my S7 and measured the run-out on the mandrel. The movement on the needle on the DTI hardly moves.
However what is everyones opinion on acceptable run-out, I guess Zero is ideal and every one will have their own ideas. I have attached a ER32 Chuck directly onto the nose however the run-out fluctuates between 0 and 0.010 mm but frankly is this good or rubbish if so I will send it back !
Regards
Alan
|
|
|
|
Post by Deleted on Jan 24, 2009 12:02:09 GMT
Alan. Have you tried it with a piece of ground bar in one of the collets? Half a thou total runout is the maximum acceptable (for me anyway). I think that's 0.012mm.......
JB
|
|
44767
Statesman
Posts: 529
|
Post by 44767 on Jan 24, 2009 12:40:01 GMT
Alan, ER collets have a runout tolerance of up to 0.015mm for the standard range. Cheap imports may be up to 0.05 if you're unlucky! The good quality manufacturers will have a range of precision collets guaranteed to have less than 5 microns (0.005mm). Remember that this is only the collets; the chuck will have a runout tolerance of its own and the two will combine.
Generally, I am happy if I chuck something in the 3-jaw and it indicates less than 0.05mm. If I need it better than that I either clock it in a 4-jaw, use a mandrel or use soft jaws bored to suit. My lathe (Weiler) also has collets which draw directly in to the spindle (no morse taper in this spindle). These always hold the work with less than 0.01mm runout. However the job must be the exact size of the collet to do this as only then will the collet bore be parallel. This is similar to the "5C" collets. My lathe did come with a set of imperial collets from 1/16" to 1" in 1/16's and 1mm to 25mm in 1mm increments.
I hope this helps a little.
|
|
|
|
Post by havoc on Jan 24, 2009 12:45:34 GMT
If I chuck something in the 3-jaw then the run-out is 0.5mm. This is only a problem if you want to re-chuck something, but then the 4-jaw is a better idea. I wouldn't worry about 0.01mm, even not about 0.05mm to be honest.
|
|
russell
Statesman
Chain driven
Posts: 762
|
Post by russell on Jan 27, 2009 15:17:58 GMT
Hi Havoc,
I hope you mean 0.05 mm! If it's really 0.5 mm I'd buy a new chuck.
Russell.
|
|
|
|
Post by havoc on Jan 28, 2009 10:00:15 GMT
No, it is 0.5mm. I chucked a piece of ground silversteel and put the clock on it. Doesn't matter which key hole I use either.
It doesn't matter really as long as you don't have to re-chuck a piece (and take a cut on the outside) as the lathe spindle itself is OK.
|
|
|
|
Post by Deleted on Jan 28, 2009 12:18:45 GMT
Hi Havoc. That's a huge error, even for something that may originate in the East! Has it always been like that? If so, maybe the wrong jaws were fitted when new, because they are supposed to be matched to specific scrolls. Do you get the same error with the outside jaws?
JB
|
|
|
|
Post by engineernut on Jan 28, 2009 16:35:37 GMT
No, it is 0.5mm. I chucked a piece of ground silversteel and put the clock on it. Doesn't matter which key hole I use either. It doesn't matter really as long as you don't have to re-chuck a piece (and take a cut on the outside) as the lathe spindle itself is OK. Is the chuck fitted to a back plate or does it have a thread to screw straight on the lathe nose ? If fitted to a back plate I just wondered if the plate was originally machined accurately as if not the chuck might not be seated centrally ? I feel .5mm (nearly 20thou) is far too much. |
|
|
|
Post by havoc on Jan 28, 2009 19:48:38 GMT
AFAIK it has always been like that. I bought it new and checked it not long after I had everything set up. Never checked with the outside jaws, I'll do that once I can get into the shop again.
It is a type with a backplate. I could check the backplate I guess.
I don't find the papers back that came with it but (again) IIRC then it was within spec, but on the limit. But as said, don't trust any chuck to be perfect. I have turned wheels on that lathe with that chuck. But then making a mandrel first. No problem with concentricity.
|
|
russell
Statesman
Chain driven
Posts: 762
|
Post by russell on Jan 29, 2009 12:28:59 GMT
I was amazed by the accuracy when I fitted a TOS chuck to the 918 lathe: 1 thou run out on a 1/2 in Silver steel bar adjacent to the chuck and 4 thou run out 10 inches away from the chuck. No way would I put up with 0.5 mm. Even the chinese chuck I threw out gave about 5 thou adjacent to the jaws.
|
|
|
|
Post by engineernut on Jan 29, 2009 16:10:11 GMT
AFAIK it has always been like that. I bought it new and checked it not long after I had everything set up. Never checked with the outside jaws, I'll do that once I can get into the shop again. It is a type with a backplate. I could check the backplate I guess. I don't find the papers back that came with it but (again) IIRC then it was within spec, but on the limit. But as said, don't trust any chuck to be perfect. I have turned wheels on that lathe with that chuck. But then making a mandrel first. No problem with concentricity. Presumably the chuck is held on the backplate by 3 or 4 bolts? Try slackening then off slightly and see if you can move the chuck sideways. If so just position the chuck centrally and do the bolts up. When I fitted a chuck to my Hobbymat lathe you had to machine a groove on the back plate that fitted tightly inside the rear of the chuck positioning it accurately. On my Myford the chuck is threaded already and screws straight on the lathe nose. I gave that a check this morning, first time for many years, and there was less than half a thou run out with a piece of 1/2" steel held in it. |
|
|
|
Post by ettingtonliam on Jan 29, 2009 18:26:29 GMT
When I got my Littlejohn lathe, the Burnerd chuck that came with it had about 50 thou clearance on the register of the backplate, no idea why, I'd always been taught to make the register a good push fit into the chuck. It would allow a fair amount of adjustment though to correct (or cause) runout.
Richard
|
|
