lathe design

I've been looking at lathe design - you can probably guess why, but I shan't tell you, that would be too (un-?) fairbrotherish.

However, I have some questions, and I hope you can help.

Does anyone know of any good sources on lathe design, preferably online, but I can try the library if that's warranted. I'm especially interested in smallish lathes, about 70mm above bed and 200mm between centres, but capable of very high accuracy (~0.001mm).

Is belt drive better, or is a fully geared head a reasonable choice? Remembering also that I am probably going to have to cut any unbuyable gears...

In the headstock of a small lathe with some kind of belt drive, the pulleys are usually arranged so the largest is nearest the chuck. Any special reason?

Is myford type backgearing better than the traditional type with two cog sets at either end of the pulleys? Why?

I have a zillion more, but that's enough for one post. :)

Thanks,

-- Peter

"Sweet dreams are made of Anything that gets you in the seam And I feel like I'm Seventeen again." Eurythmics

Reply to
Peter Fairbrother
Loading thread data ...

You'll need more than sweet dreams to attain ~0.001mm, you'll need fantasies and the attendant treasure trove...

Tom

Reply to
Tom

And a touch like a bloody midwife

-- Regards,

John Stevenson Nottingham, England.

Reply to
John Stevenson

You could also try asking on rec.crafts.metalworking A lot of their members 'claim' to have hand scrapped old clapped out WWII South Bends to hold better than this over 24", not taking the curvature of the earth into account.

I often wonder why we went to the trouble of inventing cylindrical grinders when people with these skills are so abundant

-- Regards,

John Stevenson Nottingham, England.

Reply to
John Stevenson

Remarkably, they also seem to manage these feats (or should that read "these feets"?) while recumbant in their armchairs, with a glass of beer in one hand & the TV remote in the other.

Regards, Tony

Reply to
Tony Jeffree

"scrapped" is probably nearer the truth than you intended!

From this thread, the project book by Harold Hall I'm working to which says I should aim for better than 0.005mm over the length of the cylindrical squares to be made is rather a challenge for a novice then? I began to worry about this tolerance given that my clock gauge only indicates down to

0.02mm, so I doubt I can even measure to 0.04mm with any certainty - (as a novice in a shed)?

Steve

Reply to
Steve

Reply to
Richard Bennett

What's a midwife's touch supposed to be like? I'm guessing that's a common metaphor, but I haven't heard it.

The only midwife I know (who happens to be a guy) is a rather rough-n'-ready type. He is said to be very good, but I have never needed his services. He does get bloody sometimes though, maybe his touch changes then.

Come to think of it, you wouldn't want a midwife who dropped the baby because her hands were slippery with blood, would you? I guess that's what it's about.

On a somewhat battered college Harrison 300 I can machine a short round section to size to considerably better than one 0.01 mm division on my micrometer (say to 0.003 mm), that just takes patience (and a good feed helps). On a small purpose built lathe I'd expect to do better than that without great difficulty.

I don't know about doing it over any length though.

Thanks for the rcm suggestion, I think?

-- Peter Fairbrother

Who would skip and who would plod Or who would lie quite still-ly And who would ride backwards on a giraffe Stopping every so often to laugh by Mike Heron

Reply to
Peter Fairbrother

Bollocks

What the man is aiming for is ONE micron tolerance on a home built lathe. Designed by someone with no knowledge of machine tools and built in a garden shed with a machine mart bench drill, a set of measuring instruments that can't read to one micron and a slice of cold toast [ metric slice that is ]

Let's take a good look at this micron or 0.001 of a millimeter.

Coefficient of thermal expansion of say Aluminium is approx 0.0255 microns per millimeter, per degree C rise.

So lets take it you are machining a 50mm bar, that's1.275 microns per deg C

So unless you can prevent the work from rising less than one degree C you have blown your tolerance.

If you have a draughty shed then forget it.

I do agree with the word interesting though

Also at this point we won't discuss the tolerances of the tool marks.

-- Regards,

John Stevenson Nottingham, England.

Reply to
John Stevenson

Well I'm glad that question has been answered, we can now move on to building the optical comparator required...

Tom

Reply to
Tom

Which could easily be made from a set of NHS bi-focals and a catapult frame for the gimbals

-- Regards,

John Stevenson Nottingham, England.

Reply to
John Stevenson

You could check out single point diamond turning for inspiration. This achieves even greater tolerance and good surface finish for optics. eg:

formatting link
Scrim

Reply to
Scrim

Do the NHS specs come certified to a ~.001mm specification or would one need to make a lens grinding machine to improve the optical magnification and then build a collimating jig? Then of course, there would be the problem of parallax..

Tom

Reply to
Tom

The Lathes.co.uk site has some interesting pieces of passing commentry on lathe design while discussing the features of various lathes.

-- Neil Barnes

Reply to
Neil Barnes

this -001mm seems rather vague, is it the radius or focal length of the lens? It must of course be operated with both eyes open or closed - or could it be in varying sequence. The problem of paralax could be solved by using one lens above the other.........

Anyone got a tongue bandage?

Reply to
Neil Ellwood

Hand "scrapped"? I think you have misintentionally spelt that correctly John.

Regards,Mark.

Reply to
Mark McGrath

Aiming for about that. Two would be okay. Even five would be better than nothing.

Got some - and I'm a fast learner.

Oh dear me no. I don't have a shed. I'm a math geek, not a shedi.

The lathe will be built initially in the workshops at Trowbridge College, where they have Harrison lathes and Bridgeport mills and such (including a Super7), fine for roughing work to 0.1 or even 0.01 mm, depending on size etc.

I'm a part-time student of model engineering there. Incidentally I recommend it if anyone local is interested, the tutor is good (though busy), and you get to play with all kinds of machines.

Then I can use the modern cnc lathes and mills and the very accurate measuring equipment at [..] University to do the finishing work (as it's a bit unofficial I'm not naming anybody).

It will be used in a proper workshop, on the first floor of a modern brick building, with a concrete underfloor and climate control (well, with central heating at least).

BTW, an optical comparator? didn't they use them in the war, for gun parts and suchlike? - video measuring technology, and now 3D laser, is the "in" technique.

Reply to
Peter Fairbrother

In which case ignore everything I have said as it was all practical and not theory.

-- Regards,

John Stevenson Nottingham, England.

Reply to
John Stevenson

me thinks you should be looking at a sliding head machine, very acurate, very expensive, IIRC Toss make them, but be prepaired to part with a serios amount of wedge £100k + for a new one

Reply to
Tim Bird

Tony Jeffree writes................

I don't think they 'do' feet, only the metric stuff!

Mike

Reply to
Mike Whittome

PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.