[the4thseal]

I would like to build something like this. is there any information out there besides this video? thoughts?

https://www.youtube.com/watch?v=sFrVdoOhu1Q


this is a must watch

[jsheerin]

That's really cool - thanks for posting. You could look at 'Precision Machine Design' by Slocum for more ideas on stuff like that. He talks about air bearings and a lot of other concepts that lathe demonstrates.

[RCaffin]

Look up single point diamond turning for optics. I think they go significantly better than 1 micron.
Yes, air bearings are used on those machines.
I had better add: very big bikkies are required.

cheers
Roger

[hanermo]

Its not all that hard .. I have done it.

Actual mechnical resolution is better than 1 micron.
The first 1 micron movement I got back in 2005 or so .. with brushed, old, "trademill" dc servos.

At the moment, I am using AC brushless servos on x and z and C (spindle drive).
And a suitable hardware controller, the CSMIO-IP-S from Cslabs.

Theoretical step size is 0.2 microns, with belt drive to ballscrew on x and z, both at 1:2.
Very good, very rigid, belt drive.
HTD 8 mm, 30 mm wide, tight belts.

Big servos (so the belts can be tight), 220V, 750W, 10.000 counts, 3000 rpm, x and z.
2,5 kW on spindle, at 1:3, with HTD 8/30.
Ballscrew of 32 mm on z.

All above is a fair bit of work, maybe 300 hours.
About 200 kg in mounts, new saddle carrier (behind lathe), servo motor mount on top of spindle (so heat rises does not expand the spindle).

The CSMIO system, with about 150 IO lines, 24 V DC, costs about 2000 € in bits and pieces,
ENC encoder module, MPGs, wiring, DC 24 V PSUs (2), cables (4 of) to the servos.

Servos cost about 700€ each, plus 1500€ for the spindle drive.
Belt drives about 500€.
Ballscrew 2000 € (for 3, with 400W AC brushless servo for b axis (TS), not yet installed).
5400 for above.
2000 controller.
1000 in steel, paint, etc.
8400 in parts.

I will likely change the arrangement to run on linear bearings, 35 mm hiwin, on z and x and b.
Maybe 800 € extra, for 3 axis.
12x24" size lathe, 500 kg.

This will allow smaller minimum movement with less stiction.
I expect about 0.4 microns, and this will also need a new ballscrew on x (current is old, rolled, and "loose", bought in 2002 when I was just starting).

Now looking for a ground screw, in C0/C1/C3 quality, with 100+ mm of movement, in 20-25 mm D, for x.

[jips]

Its not all that hard .. I have done it.

Actual mechnical resolution is better than 1 micron.
The first 1 micron movement I got back in 2005 or so .. with brushed, old, "trademill" dc servos.

At the moment, I am using AC brushless servos on x and z and C (spindle drive).
And a suitable hardware controller, the CSMIO-IP-S from Cslabs.

Theoretical step size is 0.2 microns, with belt drive to ballscrew on x and z, both at 1:2.
Very good, very rigid, belt drive.
HTD 8 mm, 30 mm wide, tight belts.

Big servos (so the belts can be tight), 220V, 750W, 10.000 counts, 3000 rpm, x and z.
2,5 kW on spindle, at 1:3, with HTD 8/30.
Ballscrew of 32 mm on z.

All above is a fair bit of work, maybe 300 hours.
About 200 kg in mounts, new saddle carrier (behind lathe), servo motor mount on top of spindle (so heat rises does not expand the spindle).

The CSMIO system, with about 150 IO lines, 24 V DC, costs about 2000 € in bits and pieces,
ENC encoder module, MPGs, wiring, DC 24 V PSUs (2), cables (4 of) to the servos.

Servos cost about 700€ each, plus 1500€ for the spindle drive.
Belt drives about 500€.
Ballscrew 2000 € (for 3, with 400W AC brushless servo for b axis (TS), not yet installed).
5400 for above.
2000 controller.
1000 in steel, paint, etc.
8400 in parts.

I will likely change the arrangement to run on linear bearings, 35 mm hiwin, on z and x and b.
Maybe 800 € extra, for 3 axis.
12x24" size lathe, 500 kg.

This will allow smaller minimum movement with less stiction.
I expect about 0.4 microns, and this will also need a new ballscrew on x (current is old, rolled, and "loose", bought in 2002 when I was just starting).

Now looking for a ground screw, in C0/C1/C3 quality, with 100+ mm of movement, in 20-25 mm D, for x.

Did you got some pictures ?

[hanermo]

Sorry .. pics are still delayed.
Commercial concerns, Im afraid..
As you can see from the numbers, these represent significant financial outlays ... and there are some constraints on publishing stuff.

I need to only show finished work with detailed components..
And actual working stuff (video of machine(s) making parts.
Originally these were planned for end of year, and I am about 1 month late on this.

Partly due to not receiving needed components (shipping in christmas time does not work well over here).

[jips]

Understandable , seems really interesting by the way 😉

[RCaffin]

Hi Hanermo

All very nice, but I can see one problem ahead: setting the zeroes to that accuracy!
How do you do that?

Cheers
Roger

[johnohara]

Somewhat off-topic I guess, but Dan Gelbart's entire video series on YouTube is well worth the time.
He's very generous with his knowledge and expertise and I highly recommend it.

~john

[the4thseal]

yes his videos are fantastic. i would love to see how he "made" his air bearings and so on. i a sure it is well thought through.

[RCaffin]

i would love to see how he "made" his air bearings and so on

I have made air bearings. Most of it is not difficult, but drilling the tiny air holes and making the pockets for the air holes is tricky.

Cheers
Roger

[rowbare]

yes his videos are fantastic. i would love to see how he "made" his air bearings and so on. i a sure it is well thought through.

As I recall, Dan stated somewhere, perhaps in the comments that he bought the air bearings on eBay.

bob

[the4thseal]

have you posted the making of air bearings? he said he bought the spindles but made everything else. i would love to see a "building of" on this lathe. the man's insights are breath taking in their refinement.

[jhovel]

Why don't you ask him? He is quite sociable and won't bite. And he is retired - so to speak...

[the4thseal]

Fair enough, Roger could you post you examples on air bearings?

[RCaffin]

Hi Seal

Sorry - that was a long time ago - probably pre-internet and pre-CAD. We were trying to make miniature air bearings for research gear, under 10 mm diameter, but we were running into some laws of physics about the curvature of the bearing surfaces. We had to use video microscopes to drill the air holes, on a manual Aciera, so the elderly machinist could see when the sub-millimeter drill bit started to bend. The bearings worked, but I would suggest that they should be larger to avoid insanity.

Cheers
Roger

[mactec54]

the4thseal

They are not difficult to make at all, the most important part is how clean & dry your Air is, Here is a Video and PDF on air Bearings,there is lots of information on how to

https://www.youtube.com/watch?v=lOTWx69mghM

[RCaffin]

Hi Mac

OK, my knowledge comes from many years ago, and is somewhat eclipsed by the manual you provided. Really seriously impressive stuff.
I note the manual does warn against making very small air bearings: I wish we had known that the start of our experiments!

Lovely stuff.
Cheers
Roger

[RCaffin]

Damn you Mac.
I have now spent the entire afternoon googling air bearings and porous carbon ...
The technology has advanced a bit since we were playing there.

cheers
Roger

[mactec54]

Damn you Mac.
I have now spent the entire afternoon googling air bearings and porous carbon ...
The technology has advanced a bit since we were playing there.

cheers
Roger

There is so much stuff out there on different subjects, you could spend a whole lifetime just searching through the mountains of information, some is good some is bad, at the end of the day it takes lots of time, when I find something good on a subject I save it, this is a great subject to spend some time on