Q about CNC machines back in the 90s

Posts
773
Likes
1,373
I'm doing some "research" (loosely applied) on movement finishing back in the 90s. Specifically, for Seiko but it can also apply to Swiss/other brands.

I'm aware that with the current state, CNC machines are used for anglage and such. But, I'm wondering if that was the case back in the 90s. I would assume not, but am definitely not in the industry.

I'm just wondering if anyone would have thoughts on this?

Thanks
 
Posts
3,535
Likes
8,863
I'm doing some "research" (loosely applied) on movement finishing back in the 90s. Specifically, for Seiko but it can also apply to Swiss/other brands.

I'm aware that with the current state, CNC machines are used for anglage and such. But, I'm wondering if that was the case back in the 90s. I would assume not, but am definitely not in the industry.

I'm just wondering if anyone would have thoughts on this?

Only time I've worked directly with CNC was in the 1980s from the point of view of getting data from CAD system to CNC controllers, but I do not see any difficulty in doing that sort of work in those days.

Last time I saw a CNC machine in action was at Last Engineering in Ely Cambridgeshire where they were doing IIRC the gearbox casings for the Jaguar F1 team. Five feet long and needed machining all in one piece. OK, different scale but the same tech -- and the same accuracy required.
 
Posts
29,674
Likes
76,836
CNC would have been in widespread use by then I suspect. Specifically for anglage? Not sure but it wasn’t a rare technology even in the early 80’s when I first used it.
 
Posts
2,842
Likes
4,537
I bought my CNC machine in the mid 90s it was an 80s model Mill/Lathe combo. My friend sold it to me to make a point. "That you can not simply press a button and parts come out." He also had a large one and was wondering how to put cranks on it. Ironically, I took the mill spindle of and made a permanent loan of it to the local maker space, which had a CNC but no spindle. I then put crank handles on the lathe.

What was litterally the cutting edge tech in the 1990s was EDM. (electronic discharge machining) also known as spark erosion. My freind subscribed to Home/Shop machinist which had plans. I build one of these, but it was scary. Never could find an electolyte that was practical. Choices were, Kerosene(paraffin), DI water (what my friends machine used) and some sort of special impossible to find oil. When we went to the Basel fair as part of the Horological tour, my friend bought a new swiss EDM for his tool and die shop. He gave me the operator manual for the old Japanese one which I think I still have. Not sure what happened to my machine, I took it to the Maker space to set-up, But I think someone tossed it during a junk cleanup day. I was going to use it to replicate some split second chrono and repeater parts.

Derick Pratt was using EDM at the time and wrote articles in the British Horological Journal about it.

The problem with these machines, is that one has to write a program, which in turn has to be debugged. The same issues arise with 3D printing, (why there are so many online 'parts repositories.' It also does not help that the big players like solidworks, do not make the software available to Hobbyist. Some of the software like Ashlar-vellum the Cobalt-Xenon solid modeling is highly specialized. After seeing this used in one of the Swiss factories, I found a copy locally in the silicon valley. In order to use it I have to run it on a 1990s or early 2000 era mac. Last I checked the upgrade cost more than I paid for the program. More that the company wanted to rent the program with a three month or one year license.

I have dozens of project to do with these machines, and have not done so over the last 20 years. I now have access to a laser cutter, but it does not do metal. Metal was done with waterjet, Edm or plasma. In practice, the output is as good as the X/Y lead screws on the machine bed.

Industry has it's own way of doing things. Usually with apprentice/trade programs, and with a brute force approach. Often the parts are drawn on one system. then the operator re enters all the data from the drawings into the G-code. Mostly the work comes down to building jigs and fixtures.

Swiss watchmaking, has been about automation, They tend to stay up on the leading edges of tech. I saw some pretty neat 3d metal printing machines on a tour of spaceX. would not surprise me if they were Swiss.

-j
 
Posts
2,842
Likes
4,537
Addenda: I forgot, For my Electronics engineering certificate in 1982. I re wrote a G-code intepreter to work from one computer time share system (HP200F) to run on a Pr1me system. (That no one remembers) I also did my final paper on restoring a high speed paper tape punch. The tech was old then (late 1970s early 1980s.)

-j
 
Posts
17,942
Likes
37,522
We used to get the chads out of the punch machines and fill some annoying person's desk drawer with them on days when work was boring.
 
Posts
29,674
Likes
76,836
What was litterally the cutting edge tech in the 1990s was EDM. (electronic discharge machining) also known as spark erosion.

These machines have been around since the 70's as CNC controlled machines. I was working with machine shops who had these machines in the mid-80's (both wire EDM and Sinker style machines).
 
Posts
29,674
Likes
76,836
We used to get the chads out of the punch machines and fill some annoying person's desk drawer with them on days when work was boring.

Not quite as bad as the guys that would fill your toolbox with grease using the automatic grease gun...
 
Posts
3,535
Likes
8,863
a Pr1me system. (That no one remembers)

I remember Primes. First used one in 1975 when I joined a project to produce a design system for piping and steel-work which was developed on a Prime 300. That was a big success and the project heavyweights formed their own company to make systems for electrical and mechanical engineers. That also was written on Primes and when I re-joined them Prime had world-wide distribution rights every where except western Europe and my first task was to train up their people. Then I could take my small team to sit quietly and look at the emerging generation of Motorola 68000 workstations. Which is where

Some of the software like Ashlar-vellum the Cobalt-Xenon solid modeling is highly specialized.

Ashlar-Vellum comes in, kinda. Vellum was the second CAD product from Martin Newell, brother of one of my bosses Dick Newell. Martin was working in 1982 for Cadlinc who actually made all the early Sun Microsystems computers and he & I spent a week hand-waving and making sketches on paper about how we might be able to use the coming WIMP hardware, which didn't then quite exist in usable form. There were quite a few similarities between Martin's Pegasus system and my user interface for our ported mechanical CAD sold as Drafting/3000 😉 😉