Lathe Accuracy Problems

[DerryUKParticipant @derryuk]

<It is seriously annoying as well as a complete waste of the threadwriters time.>

It has happened on other websites to me as well so I am not sure what the problem is. And has already been said if I want to write something 'serious' I now write it locally first then cut and paste.

[Anonymous]

Early on I spent a lot of time chasing a huge error on my lathe at the headstock end, about 4 thou in 4 inches. My lathe is bolted to the manufacturer's stand, presumably by the manufacturer themselves. Fortunately I didn't fiddle with the bolts holding the lathe to the stand, as I'd seen a post on a professional machining forum suggesting that fiddling with said bolts would be a 'really bad idea'. Instead I tried all sorts of clamps and shims on the hold down lugs on the stand itself, to no avail. It was only when I turned the aforementioned casting at the outer end of the bed and got almost zero error that the light began to dawn and I realised that the bed probably wasn't twisted. It turns out that the error was due to the gap piece not being seated properly. A good clean and refit of the gap piece resulted in the smaller errors mentioned in a previous post.

Out of curiosity I've just measured one of the rolls in the bending roll set I made last year for rolling my traction engine wheel rims. The roll is 5.5" long and was turned with one journal in a collet chuck and the other supported by a revolving centre in the tailstock. I used a 2-3" micrometer with a 'tenths' vernier. The error is about 3 tenths over 5 inches.

If the Weiler lathe is bolted to the manufacturer's stand, possibly by the manufacturer, I'd be inclined to leave well alone until all other possibilities have been eliminated.

Regards,

Andrew

[Anonymous]

Posted by Graham Meek on 11/04/2013 22:27:00:

There is a taper on the 8 mm diameter, it is smaller at the threaded end to allow the ball race to be fitted easily and is a push fit at the shouldered end, that taper is 0.001 mm, this is what a well set up lathe is capable of.

Gray,

How on earth do you measure 0.001mm; using 1.001mm and 1.002mm slip gauges I simply can't see the difference on my normal metric micrometer with any kind of consistency?

Regards,

Andrew

[jason udallParticipant @jasonudall57142]

Posted by Andrew Johnston on 12/04/2013 10:42:57:

Posted by Graham Meek on 11/04/2013 22:27:00:

There is a taper on the 8 mm diameter, it is smaller at the threaded end to allow the ball race to be fitted easily and is a push fit at the shouldered end, that taper is 0.001 mm, this is what a well set up lathe is capable of.

Gray,

How on earth do you measure 0.001mm; using 1.001mm and 1.002mm slip gauges I simply can't see the difference on my normal metric micrometer with any kind of consistency?

Regards,

Andrew

Andrew I sort of agree.. I have found many turn the screw until it agrees with the slip……but then I am an electronic engineer and naturaly distrust the lsd…….. ( thats why I like "my" 1/10 micron height guage …) 

Edited By jason udall on 12/04/2013 16:38:26

[jason udallParticipant @jasonudall57142]

An odd thought just struck me

Much has been said about leveling lathes.

Do you need to level surface plates?…

[Ex contributorParticipant @mgnbuk]

Do you need to level surface plates?…

I would suggest yes – but qualify that with it probably depends upon the size of your surface plate & it's mounting arrangement.

At my last employment we had a 6' x 4' Grade "A"cast iron surface plate. That was regularly checked with a precison level before use & it was NAMAS calibrated with a "Talyvel" electronic level connected to a laptop that printed out a 3D repesentation of the errors. After rectification (read "scraped by someone who knew what they were doing&quot the maximum error was around 0.0005" from flat.

I have no experience of the OP's particular Weiler lathe (I did work on a Weiler slant bed CNC lathe many years ago), but wonder if it has an adjustable headstock ? Not all lathes have the headstock aligned using a bed feature (the vees on a Boxford, or the inside face of the rear shear on a Myford) – some have the headstock mount on a flat face & "throw" it with adjusting bolts. This is a standard feature on CNC lathes & allows for re-alignment after the inevitable bumps that happen from time to time, but I have also seen this on (larger) manual lathes. Worth a look, maybe ?

Re-alignment would be achieved with a precision test bar mounted in the spindle taper – after setting the bed level within the maker's specifications using a precision level, of course

[Ian S CParticipant @iansc]

I don't remember it being leveled, but I imagine it was this is the surface table in our newly set up overhaul shop about 1966,I think the table was a war surplus item. Ian S C

Edited By Ian S C on 13/04/2013 12:14:37

[Anonymous]

Gray: Ah, I should have known, you're a better man then me. I have a range of micrometers, including some by Mitutoyo, but most of my outside micrometers are M&W or Starrett. All bought on Ebay, which may of course be the problem, although they have all been checked against slip gauges and screw together rod sets. I sold the 4-5" one that was miles off calibration, amazingly somebody bought it, despite a clear statement that the frame was probably bent.

Jason: a tenth of a micron, blimey that's 100nm. I hope the height gauge is in a temperature controlled room! I'm an electronics engineer too; I don't trust any of the digits, not just the LSD, which is why I don't have any electronic micrometers, calipers or why.

Regards,

Andrew

[HopperParticipant @hopper]

Not all surface plates have to be leveled. The bigger ones of course do. But there are smaller benchtop models about 16 inches square that are made to be moved about and set on benches etc. When I was an apprentice toolmaker, each apprentice in the first year training centre had a surface plate on his bench. They had no bolting down holes and simply sat on the bench, no precision levels involved. The thick cast webbing forming the base of the plate prevented any distortion.

It was used for marking out the dreaded G clamps and toolmakers clamps and machine vices that were then cut out from solid bar by hacksaw and filed flat and parallel within less than .001". So the surface plates were used for bluing to check for flatness as well.

And there are other similar sized surface plates that have handles and/or lifting ring holes so they can be placed on large machinery such as lathes etc to check bedways etc are properly flat when scraping them in.

On the other hand, big surface tables, big enough to wander around on, used for making  car body patterns and jigmaking were all levelled.

So for the small home workshop with a smaller surface plate, probably no need to get too carried away with leveling. A builder's level should do you, if at all. (Providing your benchtop is reasonably level.)

Edited By Hopper on 14/04/2013 11:35:47

[Author]

Posts