Hacker News Reader

Comment by jjk166 | Hacker News Reader

A lathe can't actually replicate itself completely. Specifically, a lathe can only make ways smaller than its own cross slide's stroke. It would also be impossible to make a typical lathe bed on a lathe, though you theoretically could design an unconventional lathe bed that is possible to make on a lathe, even if grossly impractical.

The real starting point for machine precision is rubbing 3 granite plates together.

View on HN

You can see a guy following Dave Gingery's instructions to make a lathe bed here [1]

And as you say, a granite surface plate is needed. Of course, Gingery's books only claimed to set up a metalworking shop starting "from scrap" and "simple hand methods" and that "it isn’t long before the developing machines are doing much of the work to produce their own parts" [2]

Of course, to truly make a lathe from scratch, you must first create the universe.

[1] https://www.youtube.com/watch?v=zPGZg45dGXA [2] https://gingerybookstore.com/MetalWorkingShopFromScrapSeries...

Yeah, that's a different book, _Foundations of Mechanical Accuracy_:

https://mitpress.mit.edu/9780262130806/foundations-of-mechan...

Related to the Moore's work, I also enjoyed Engineering reminescences[0] as a historical account how people figured out ways to make accurate things in metal, more than a hundred years ago.

0: https://www.gutenberg.org/ebooks/72043

You might also like English and American Tool Builders by Joseph Wickham Roe, I picked up a copy from a free library and enjoyed it. I should probably skim it again, then set it free into the system for someone else to find.

https://www.gutenberg.org/ebooks/72046

Thanks, seems to fit the theme!

A better link would have been:

https://mooretool.com/about-us/publications/

(the book is long out of print, and used copies are exorbitant, but maybe if enough folks express interest it will get a reprint)

The three plates is the foundationiof accuracy but no tools are needed to create them. you need a lathe to create a lathe - but a lathe can build itself as by the time you need a lathe in construction you already have enough of a lathe built for that next step.

Does anybody actually use the three plate method with granite? It was originally done with cast iron, and I thought cast iron was still the standard material. The plates are covered with dye and rubbed together to find the high points, which are then scraped off, instead of being removed by the rubbing directly.

Granite is a common material for modern surface plates (and a good one because it doesn't rust and doesn't raise burrs if it's chipped), but I believe these are still made using cast iron reference plates.

From what I understand the three plate method is when you are going from 0 to flat as the errors are averaged out.

Doing the "covered with dye and rubbed together to find the high points, which are then scraped off" thing is only if you already have a flat reference surface as you wouldn't have a way to know if the thing you're trying to make flat is really flat.

The real question is how do you get the first flat reference surface when all you have are a few somewhat flat things?

> The real question is how do you get the first flat reference surface when all you have are a few somewhat flat things?

my understanding is the threeplate method allows you to build the reference plate in the first place.

Right. You start with any two plates and make them flat with respect to each other. One will be convex and one concave of course, then you take one of those and your third plate and make those two flat with each other, the switch out again using the third plate and the one not swaped out. Keep repeating until all three are flat with respect with each other - only true flat will have all three flat and the repititions keep getting closer.

of course if you have a known flat surface you can save effort by making the new plate flate to the known flat.

> of course if you have a known flat surface you can save effort by making the new plate flate to the known flat.

All the effort!

one of the good things about granite flat references, is that they last for ages, so you can get them reasonably cheaply second hand, if you can find a second hand machine shop specialist.

Even if your second hand plate is no longer flat enough it is still close enough so there is very little effort needed to bring it back to flat. (machine shops often check them on a regular basis and bring them back to flat as needed)

Also for those who aren't familiar, there's also "hand scraping" for flatness which is more common and used for things like refreshing the "ways" (ie precision linear bearings) on a lathe or other precision machine tool.

This is done like the "dye / rub / scrape" method described above, which I believe is still used as it's superior to grinding for these applications.

See video below for the process: https://www.youtube.com/watch?v=T7w84CrBEE8

Hand scraping is done, but for ultimate flat you need to lap three plates not hand scrape. Hand scrapping is not as flat - but the average is close enough and the imperfections are needed anyway for oil so hand scraping is used for bearing surfaces.

New Scientist published a reminiscence of somebody in the relatively modern era doing the 3 plate dance. I wish I could find it online. They said it was tedious work.

Maudsleys 3 plates are in the London science museum along with Whitworths screw, and some of Marc Brunels stuff. Same room as the meccano differential analyser and the harmonic calculator for tide charts and Babbage bits.

Edit: found it - https://archive.is/iyCzB

It's a straight edge, not a bed, but this is a fun watch: https://www.youtube.com/watch?v=pq47yXFmj24

Much the same process, in one less dimension. And it looks time consuming but I bet immensely satisfying.

Oh yeah smart guy? Well how do you make the granite?

Lathes can certainly make cylinders, and a tube-based lathe bed is not a stretch.

A lathe can't replicate its own assembly, of course. It can't seat the spindle in the constraint bearings, for instance.

A CNC (without the word lathe) can make most of itself, and possible all. Nope: certainly all, if two of its dimensions fit within its work volume.

But it can't make cylinders as long as its own guide ways.

most lathes have a hole in the chuck to feed the work through. so if the material is ground down by hand to a diameter small enough to fit in the hole then to be turned and then removed and flipped over it's possible.

Lathes can make cylinders, but not of unlimited length in one setup so they lose some accuracy making cylinders longer than their carriage travel. And their beds are by necessity longer than their carriage travel, since the carriage rides along the bed and isn't infinitely thin. They also can't make things like motor stator laminations, and you definitely need a motor for a replica of a motorized lathe. So lathes can't replicate themselves exactly.

Milling machines are also just lathes with a different orientation, an extra travel axis, and a motor optimized for higher speeds & lower torques, it's possible (and reasonably common) to use a mill like a lathe or a lathe like a mill in many cases. So "only machine" part is also a stretch.

if the purpose is to bootstrap, you could also use something like a leather belt drive off a central shaft, which would require different power sources but ones that a higher percentage of could be made with a lathe

> Milling machines are also just lathes

Mills are more limited than lathe - they don't have leadscrews, which are a necessity in the "build yourself" phase. You cannot make arbitrary threads with a milling machine. Thread milling gives some capability in this arena, but that's a CNC process.

One can make arbitrary threads (vertically) on a CNC using a single point cutter (limited by the height of the Z-axis/thickness of the stock which one can fit a tool over/into):

https://community.carbide3d.com/t/thread-milling-in-metal-on...

A manual lathe will often have a gearbox which allows cutting threads on it:

https://www.youtube.com/watch?v=E_KF3n3oo08

make me a truly flat surface

It would be hard to make a human into a truly flat surface. I suppose if you have big enough granite blocks...

Reply on news.ycombinator.com