Cutting very, very small metal- possible?

Hi there, y’all. I’m new around here. Just took the tour and did orientation tonight with Ashley and touched base with Matt on some questions already, who directed me here for some more input.

I’m looking to make some parts for wristwatches, which means cutting very, very small pieces of metal. We’re talking things like 0.5mm-1mm wide pieces of sub-0.5mm steel/titanium. Is it possible to use Tormach to make cuts that small or precise? And if not, is that something the incoming fiber laser would be able to do?

I realize that a lot of folks probably come to ASMBLY because they’re working on projects that are too big to handle at home. My problem is pretty much the exact opposite. :slight_smile:

Grateful for any info the braintrust can provide! I expect there are still a lot of things I could do in the space to help move my project along, but prototyping some of those very small parts would be a huge win.

Can you machine things like that on the Tormach? Yes. Will it be a fun an enjoyable experience? Probably not. You’ll really struggle to find a work holding solution that solves your needs, and at those small sizes, I think you’ll struggle to get the RPM you need out of the spindle. That said, it should be capable of repeating the tolerances you’d want.

You’ll definitely struggle to get the tool speed you need for work that fine on the Tormach. The CNC routers would be better in that regard, but workholding would definitely be tricky.

How are watch parts like that usually made? It’s a problem I’ve not though much about before.

Thanks for the feedback, y’all. It sounds like it’s going to be a challenge, but maybe possible.

Typically, mass produced watch hands are made by stamping them out of sheet metal (maybe 0.2mm thick). I’m trying to find a lightweight way to prototype some designs instead of having to go through an outside supplier.

I know of at least one watchmaker who laser cuts his own parts, but I don’t know how beefy a laser he has to be able to cut steel that precisely without damaging the metal.

Jeweler’s Lathes and lots of hand work.

The book you want for this is “Practical Benchwork for Horologists” by Louis Levin.

For workholding, you tend to glue/bolt the blanks to a carrier.

Also: Watchmakers Faceplate See Clickspring: https://www.youtube.com/watch?v=FHE4zMfEZ2Q

That’s 100% correct. The real big boys and girls in the industry do a TON of this stuff by hand or on old school watchmaking tools. And their watches tend to be priced in the low six figures accordingly. :grin:

I’m not looking to swim in that pond, but I also don’t want to just purchase cheap bulk parts from overseas. I’d prefer to build hands and dials that are built specifically for the watches I’m making.

@Jon, you mentioned that the CNC routers would be better- assuming I’m cutting the small parts from a large enough blank that workholding can be solved, does ASMBLY allow metal on the CNCs? I thought I saw something about no liquid cooling on the woodcutting CNCs, but maybe such thin metal wouldn’t require liquid cooling…

You can generally get away with air cooling on nonferrous metals. I know folks have done aluminum on the CNC routers. Is it safe to assume you’d be working mainly in brass?

If you’re using steel, air probably wouldn’t do the job.

With brass I could do dials, for sure. Cutting or engraving. I costs me $18 to buy a blank brass dial out of Switzerland today, so that would make testing and prototyping designs a lot cheaper already. Soldering on the dial feet to fit the movement would be another gap, but not a $15+ per unit gap.

If I was able to do that and then prototype some hand designs in aluminum or brass before sending out designs to be produced in steel, titanium, or gold, that would be another big win.

We think the Lagunas can handle small cuts and small tolerances down to tenths of millimeters then? Sounds like I ought to sign up for an intro to CNCs class…

You can try, but I doubt the machine is sufficiently rigid for that.

Unless you’re doing production, I strongly suspect that the amount of time you will take to do setups that can hold millimeter sized metal bits will outweight the time to simply use a jeweler’s saw and do it by hand.

And, if you’re doing production, you probably want to make a die and stamp things out.

Noted! Thanks for the insight on that. Is it likely to be the same story for the fiber laser that’s coming in, or is it possible that might be better suited to small work with tight tolerances than the Laguna?

I know for a fact that there’s a guy in France who laser cuts his parts: As seen here

But I don’t know what exact setup he has.

Laser is a better choice than most cutting mechanisms as you don’t need the workholding to resist cutting forces. However, looking at that video, he’s not really using the lasers for “cutting through”. It looks more like he is very gradually ablating the surface (presumably to avoid heat buildup) of the material and the repeatability–doesn’t look great.

Given what you are doing–very small planar pieces–is there no etching process that could make these parts? Photolithography is going to be FAR superior than just about any other process for this kind of stuff.

Photolithography looks awesome- it solves for the issue of scalability. Cutting everything out by hand is fine for prototyping, but it would be nice to be able to do a production run as well.
It does require a pretty gnarly setup, though. A lot of what I’m seeing is guys fabricating a bunch of machines to make the process possible at home, and that sort of work would be an entirely separate rabbit hole of learning and trial and error just to get that process online. Not a deal breaker, but it would set me back at least a few months more than just being able to plug in some gcode and hit run.

That video that I linked out to is a guy who makes a few hundred watches a year, including one I own. The quality on those hands is incredible, so the repeatability is there for cutting steel and titanium that small with lasers. Just not sure of the equipment requirements.

I’ll do some more digging on etching though. Not something I know anything about, but since it looks like you can etch all the way through, that might be the way to go.

Yeah, photolith etching is messy. But it’s really accurate. It’s how we make PCB boards as well as integrated circuits.

Is it? Have you checked the repeatability of the hands? What kind of hand tuning was required afterwards?

Do NOT underestimate what can be and is done by hand–especially with a good jig. My favorite soldering tech once removed 100 2mmx2mm LGA packages, flipped them over, tinned exactly 3 pads by hand, flipped them back over and hand-soldered them back down. It took her 3 days.

Humans get really good at things when they practice.

There’s always going to be a hand finishing component to this. Even the photochemical process isn’t going to produce finished pieces. Just cut to the right dimensions. That’s the part I need to be repeatable. And doing that by hand with a jeweler’s saw and file is going to have a much longer onramp than a computer assisted process.

If I were a trained watchmaker in the Daniels school of doing everything by hand, including making your own tools from hand, I wouldn’t be going down this road. :grinning:

But this convo gives me a few different rabbit holes to chase down, which is helpful.

I would be very curious to see photos of the process once you sort out what strategy you’re taking. Watchmaking isn’t something I’ve thought much about honestly, but this conversation has me really curious to learn more.

1 Like

I’m happy to share anything I’m able to figure out!

There’s sort of three groups of independent watchmakers:

  • The type who buy a bunch of cheap parts, slap them together, and launch a Kickstarter campaign for them (see any watch under $500 on Kickstarter)
  • The type who use some off the shelf parts (typically movements) and use custom designed parts to add their own design touches to the watch (Sartory Billard, who I linked above, is a good example of this)
  • The actual, bona fide watchmakers who design and build their own movements with hundreds of individual parts, often made by hand, and design entirely novel watches (the brand MB&F is on the more avante garde end of this spectrum, Roger Smith on the traditional side)

I’m really aiming to land in the middle, and I’m trying to keep a lot of my prototyping and design as close to home as I can so I’m not sending designs out and waiting for them to ship from overseas. I’d love to be able to think of an idea and pop down to ASMBLY to make it.

While working on figuring that out, I’ll still end up in the CNC and laser classes, because I’d love to be able to make small trays and holders for parts and tools that it would take me a lot longer to build in my own woodshop at home.

Both the lasers and CNC are amazing tools, and I absolutely recommend learning them, if not directly for your watchmaking endeavors, they are just good things to have in your arsenal of approaches. They might be what you use to make jigs, or other supporting efforts for your watches.

I know it has been discussed in this thread, but it’s probably really good to also get yourself on the path to learning the Tormach. Again, it might or might not directly help you now, but for those who do work in metal, it seems pretty foundational. I think the path to learning the Tormach will be 1) metal shop safety, 2) intro to milling, 3) fusion 360CAM, and then 4) you do a one on one with @EricP for Tormach. (Eric or Ethan, please correct if I’ve gotten this wrong)

3D printing might also support your efforts. I think a fair number of folks learn on our machines and then specialize for their own tools depending on specific interests.

(edited to add complete path for Tormach)

The Tormach has the longest tech tree for anything here at Asmbly. The classes you need are Metal Shop Safety, Intro to Milling, Fusion 360 CAM, then a one-on-one Tormach class with Eric.

Love it. Sounds like it’ll take some time to get there, but time I’ve got.

I can 100% see the lasers and CNCs be helpful in parallel parts of what I’m trying to do, and I’ve got existing woodworking projects that will benefit from the higher end tools at ASMBLY versus my homeowner grade stuff at home.

Really appreciating all the feedback and ideas on this from everybody. Even if it turns out that none of the tools at ASMBLY can specifically do the small part cutting part of everything, this thread is still giving me good options for that and more ideas for what I can do with the available tools. Super helpful. :pray:

1 Like

machining is possible and I think it’s always good to learn processes, for development currently until you get up to speed on stuff, your fastest and most economical option is probably sendcutsend.com laser cut service.