Making my own watch

[Paul Loatman]

Hello,

I recently joined this forum because it's full of people interested in the same thing i am, that being hand-made watches.

I've been wanting to make my own watch for years now, but just recently have i been able to learn the skills necessary to do this, i'm at a watchmaking school in Quincy Illinois, i've learned a lot of the skills with using a watchmakers lathe and i've been reading about gear ratios and shaping cycloidal gears and making gear cutters on my own time. I'm still not fully capable of making my own wheels and pinions right now so this watch i'm designing will be made using an ETA 6497.

It's going to use all of the original wheels and pinions along with the original regulating system and barrel, i'll be making the carriage for the tourbillon and modifying each part that needs to be changed to fit into my design, although there won't be much to do, the biggest thing i have to do is make the carriage of course, and also i have to either modify the existing fourth wheel so that it's stationary on a bridge or make an entirely new fourth wheel, the pinion is also the same, i may be able to make new pivots for it or just modify the existing pivots to fit under the carriage. This is really early in the planning and design stage so i haven't gotten everything figured out yet, i just wanted to let someone know that i was doing this.

Here's the watch as i have it laid out right now;



The carriage will go over the barrel, center wheel, and cannon pinion minus the cannon, the added height of those three things and the pillar plate will be approximately the height of the entire movement, the third and fourth wheels are both under the carriage and away from the barrel so they shouldn't take up more width than the barrel itself, the dial train is above the barrel and away from the carriage, aside from the cannon pinion which will be cut down to only the height of the pinion itself, so those won't add anymore width either. Another cannon pinion will be added under the hour wheel to actually tell the time, the cannon pinion that's on the center wheel will simply transfer power but not hold a hand.

I haven't completely worked out the set and winding mechanism but that shouldn't be too difficult if i simply recycle the existing system, there's really no reason why i can't.

Also, one last thing, the diameter of the pillar plate is about half a mm less than the original 6497 pillar plate, and the approximate width of the movement will be about 10-11mm but since the dial will essentially be lower than the carriage i won't really need to add that much more room for the crystal and case back.

Thanks for reading, and i'll update as i get further into this project.

[Ophiuchus]

Good luck to you Paul!

This is going to seem redundant to say, but it needs to be said.

Anyone with enough time, proper training, and real, unwavering raw determination can make their own watch. It's been done more than a few times, by many people independently. Only a few have done it to the caliber of becoming well known, from the resulting high level of their work, but people have done it. It's not an impossible task, and this is key- don't let anyone tell you otherwise. I have met many skeptics, who laugh, but ignore these people. Even if they are the head of a watch school, or what have you, ignore them. They have no imagination, nor eyes in their heads, to see that others have already done it. Think of people like Gene Clark, who toiled in relative obscurity compared to famous independents now like Halter, Daniels, Smith, etc.

If you know where to look, in this day and age, with the help of the internet, Mr. Daniel's texts, proper training, time, and you're bright enough, you CAN do it.

And until you make that first watch, don't let anyone convince you otherwise. You're only a madman to them until you prove them wrong.

I look forward to the result.

[Paul Loatman]

Good luck to you Paul!

This is going to seem redundant to say, but it needs to be said.

Anyone with enough time, proper training, and real, unwavering raw determination can make their own watch. It's been done more than a few times, by many people independently. Only a few have done it to the caliber of becoming well known, from the resulting high level of their work, but people have done it. It's not an impossible task, and this is key- don't let anyone tell you otherwise. I have met many skeptics, who laugh, but ignore these people. Even if they are the head of a watch school, or what have you, ignore them. They have no imagination, nor eyes in their heads, to see that others have already done it. Think of people like Gene Clark, who toiled in relative obscurity compared to famous independents now like Halter, Daniels, Smith, etc.

If you know where to look, in this day and age, with the help of the internet, Mr. Daniel's texts, proper training, time, and you're bright enough, you CAN do it.

And until you make that first watch, don't let anyone convince you otherwise. You're only a madman to them until you prove them wrong.

I look forward to the result.

Thanks! It's nice to hear some encouraging words, i'm sure no one at my school would frown upon what i'm doing, they're all great people. I DO know the sort of people you are talking about though.

All the hype of the big-name swiss companies of their tourbillon watches made me wonder if it really was as easy as i was making it out to be, i spent half the time just making sure my layout would actually work, and i still don't see any reason why it wouldn't, it's all up in the air until i actually make the watch though.

[ASWatch]

Kudos for the ambitious plan, Paul! Be sure to post progress reports when you get to work on it. I hope to make my own movement, too, someday, but I plan to start out with a simple time-only piece. I'll have a go at the tourbillon after that.

At least one question arose from your drawing: Will you be having your hour and minute hands on separate axes? Because that's the way it's drawn at the moment, based on the designated location of the Cannon Pinion. Or were you going to take a second cannon pinion and stick it on a post under the hour wheel? Also, the current plan re: motion works would make for some tricky setting and/or winding arrangement unless you go for a key set/wind, which in turn would force certain requirements for the case design. Or unless you put the crown somewhere other than the 3-o'clock position.

Furthermore, what's the diameter of the movement? That carriage looks huge in relation to the mainplate and barrel, surely it needn't be quite that big to accommodate the 6497 balance? Unless of course you're using the 'straight' lever instead of finding a 'side' lever, in which case the carriage diameter is understandable. Will you be making a new arbor for the escape wheel (or at least modifying the old), as the 6497 original I suspect is a bit overlong for use in this sort of construction?[/i]

[Paul Loatman]

Kudos for the ambitious plan, Paul! Be sure to post progress reports when you get to work on it. I hope to make my own movement, too, someday, but I plan to start out with a simple time-only piece. I'll have a go at the tourbillon after that.

At least one question arose from your drawing: Will you be having your hour and minute hands on separate axes? Because that's the way it's drawn at the moment, based on the designated location of the Cannon Pinion. Or were you going to take a second cannon pinion and stick it on a post under the hour wheel? Also, the current plan re: motion works would make for some tricky setting and/or winding arrangement unless you go for a key set/wind, which in turn would force certain requirements for the case design. Or unless you put the crown somewhere other than the 3-o'clock position.

Furthermore, what's the diameter of the movement? That carriage looks huge in relation to the mainplate and barrel, surely it needn't be quite that big to accommodate the 6497 balance? Unless of course you're using the 'straight' lever instead of finding a 'side' lever, in which case the carriage diameter is understandable. Will you be making a new arbor for the escape wheel (or at least modifying the old), as the 6497 original I suspect is a bit overlong for use in this sort of construction?[/i]

The hour and minutes will both be on the dial at the top, if you read through my whole post i explain this, i'll be cutting the cannon pinion that's on the center wheel down to the point where it's simply the pinion and then add another cannon pinion under the hour wheel on it's own post.

The set and winding will be very similar to the normal 6497, i'll most likely have to make more than 1 crown wheel going to the ratchet wheel for it to reach, but there should be plenty of room for the whole thing, it's hard to see what's going on in the picture since it's only two-dimensional but there should definitely be enough room to put the crown at the 3 o'clock position.

Also, yes the tourbillon will have the same exact pallet fork that's in the normal 6497, i was thinking of making a completely new english lever, but for this particular one i want a straight line carriage where the balance isn't in the center of rotation but instead has it's own orbit. The carriage will be shaped so that it tells the seconds instead of having a hand. I was at first thinking of putting a large second hand on the carriage, but then i didn't want people to think i was copying Jaquet Droz or anything.

[ASWatch]

Sorry for the misunderstandings, Paul. I seem to have missed some key points there in my haste. Please accept my apologies.

I think one of my bigger points of confusion there was the fact that your drawing was of the dial side, not the bridge side, and so I assumed the 'normal' 3-o'clock crown position to be on the left hand side in the picture, which of course it isn't.

And I agree, a carousel construction is a sensible way to make use of existing parts rather than having to make or otherwise source a pallet fork.

Best of luck in your endeavour, and I look forward to seeing and reading about your progress!

[Paul Loatman]

Sorry for the misunderstandings, Paul. I seem to have missed some key points there in my haste. Please accept my apologies.

I think one of my bigger points of confusion there was the fact that your drawing was of the dial side, not the bridge side, and so I assumed the 'normal' 3-o'clock crown position to be on the left hand side in the picture, which of course it isn't.

And I agree, a carousel construction is a sensible way to make use of existing parts rather than having to make or otherwise source a pallet fork.

Best of luck in your endeavour, and I look forward to seeing and reading about your progress!

No need to apologize, I apologize if i sounded negative, it's hard to convey everything properly over the internet, i know the drawing is confusing if you don't know how it's all supposed to be put together, so i welcome any questions.

The reason why i did it from the dial side is because the escapement and balance will be facing in that direction, if i didn't flip the escapement upside down then the carriage would rotate counterclockwise and that would also make the fourth wheel pinion turn counterclockwise from the dial side and that wouldn't work since obviously the fourth wheel should be turning clockwise because it normally holds the seconds hand.

I wish i knew how to use AutoCad better so i could get this all in 3 dimensions, but i have the design in my head and that's good enough for me to make it.

I'm getting really close to actually starting on the watch, i'm just waiting for the materials to be shipped to me, so when i get that i'll post pictures of the plates and bridges as i prototype them.

[Steyr]

Hello Paul,

I am glad finding this new post. Sounds like you're heading the target many have. I'm looking forward reading your progress.
Some questions I had going through your drawing were addressed already. Ho, maybe one : how did you get the gears distances ? profile projector ?

Regarding the CAD programs, you might not need 3D at all, but if you really want to spend some time to get the 3rd one, you could take a loog at other CAD softs. Autocad is definitely the best 2D I've ever used, but 3D parametric (don't know if it can be translated like this) such as Solidworks ; Pro/E ; Inventor... are really much more powerful in this dimension. Worth to spend a bit of time there rather than on Autocad.

Christian.

[Paul Loatman]

Hello Paul,

I am glad finding this new post. Sounds like you're heading the target many have. I'm looking forward reading your progress.
Some questions I had going through your drawing were addressed already. Ho, maybe one : how did you get the gears distances ? profile projector ?

Regarding the CAD programs, you might not need 3D at all, but if you really want to spend some time to get the 3rd one, you could take a loog at other CAD softs. Autocad is definitely the best 2D I've ever used, but 3D parametric (don't know if it can be translated like this) such as Solidworks ; Pro/E ; Inventor... are really much more powerful in this dimension. Worth to spend a bit of time there rather than on Autocad.

Christian.

To get the gears distance i used a caliper to gauge the wheel and pinion sizes and measured the size i got on the caliper with a bench micrometer, i don't have anything like a profile projector as they're really large and expensive machines. You really don't need perfect measurements in designing a watch unless you're making it with computerized machinery anyways, i'm making the watch by hand so no matter how accurate the design is it won't help make the actual watch any more accurate than my hands and eyes can do, the benefit of using an existing movement is that i already have the distances of the pivots measured out so it won't be as difficult as doing it from scratch. Also, of course i will be test fitting everything in the new plates and bridges before i continue further, i need to get the train wheels fitted accurately as i can before i do anything else, i don't expect it to work perfectly on the first plate i use.

About the 3D software, i've never heard of the other ones you mentioned so i'll check them out, thanks for that.

[ASWatch]

If you've got access to a depthing tool, you can use that to accurately trace the pivot distances onto a plate. Easily at least as accurate as ETA does them anyway.

I also came across a light and pretty easy to use 3d CAD package (doesn't do quite as much as Solidworks, Pro/E, Katia and the like, but if it's just used for design work it doesn't have to have all the tool routing etc. capabilities), called VariCAD, that you can download a fully functional (including save/export) 30 day trial version of from their website. One of the key points of interest with this software was the price, though: 500 euros for a full license, instead of ~15-20k for the big boys' toys. I only toyed with it for the trial period, though, didn't have the extra cash to buy it then but will have another look later on. Did get one of my case designs turned into 3d from my 2d drawings, which I've done with qCAD, which is freely available for Linux, although as far as I know not for Windows systems. Wasn't terribly expensive anyway I thought...

[Steyr]

As InkRose said and some guys I've met during my internship : no need to focus on microns when talking about 6497
The depthing tool surely is a good tools, and not that expensive. I've seen homemade ones as well.

Otherwise, a friend of mine who is going through a similar project was planning to confirm his initial measurements making some in-line movement. Quite easy to make and keeping some opening where the teeth are in contact, helps to check how is the gearing doing. (hope my english is clear enough )

[Paul Loatman]

If you've got access to a depthing tool, you can use that to accurately trace the pivot distances onto a plate. Easily at least as accurate as ETA does them anyway.

I also came across a light and pretty easy to use 3d CAD package (doesn't do quite as much as Solidworks, Pro/E, Katia and the like, but if it's just used for design work it doesn't have to have all the tool routing etc. capabilities), called VariCAD, that you can download a fully functional (including save/export) 30 day trial version of from their website. One of the key points of interest with this software was the price, though: 500 euros for a full license, instead of ~15-20k for the big boys' toys. I only toyed with it for the trial period, though, didn't have the extra cash to buy it then but will have another look later on. Did get one of my case designs turned into 3d from my 2d drawings, which I've done with qCAD, which is freely available for Linux, although as far as I know not for Windows systems. Wasn't terribly expensive anyway I thought...

A depthing tool! I don't know why i didn't think of that, i've never used one, but i know what they are. They're pretty expensive for what they are, i'll probably just make one.

I've been looking around at the CAD software mentioned, but i'm not sure i really need 3D, the onlly reason why i used autocad was simply to see where everything would fit, the rest i can just figure out as i go.

As InkRose said and some guys I've met during my internship : no need to focus on microns when talking about 6497
The depthing tool surely is a good tools, and not that expensive. I've seen homemade ones as well.

Otherwise, a friend of mine who is going through a similar project was planning to confirm his initial measurements making some in-line movement. Quite easy to make and keeping some opening where the teeth are in contact, helps to check how is the gearing doing. (hope my english is clear enough )

Making a test movement to check measurements might take a bit more time than making a depthing tool, so i'll just make the tool, the test movement will only work for that particular watch, a depthing tool would be useful for any watch after.

[ASWatch]

I think what Steyr was suggesting was making a simple in-line mock-up of the gear train: just two bits of brass plate and someting to hold them apart, drill the holes at the distances you've measured, and check that all transfers are deep enough/don't bottom out, so that you won't necessarily need to remake your base plate/bridge if something turns out to have been inaccurate.

Shouldn't take very long to make as things like endshakes needn't be adjusted to any fine degree, as long as the wheels spin between the plates and it feels nice and smooth. Although, as I said, the 6497 is hardly a masterpiece of precision to begin with, more like an exercise in efficient industrial manufacturing, and any measurements you've taken are likely to be close enough anyway.

[Steyr]

Regarding homemade depthing tools :

http://gadgetbuilder.com/Gear_Depther.html

http://www.horlogerie-suisse.com/forum/viewtopic.php?f=2&t=4927&start=30

Christian.

[Paul Loatman]

I think what Steyr was suggesting was making a simple in-line mock-up of the gear train: just two bits of brass plate and someting to hold them apart, drill the holes at the distances you've measured, and check that all transfers are deep enough/don't bottom out, so that you won't necessarily need to remake your base plate/bridge if something turns out to have been inaccurate.

Shouldn't take very long to make as things like endshakes needn't be adjusted to any fine degree, as long as the wheels spin between the plates and it feels nice and smooth. Although, as I said, the 6497 is hardly a masterpiece of precision to begin with, more like an exercise in efficient industrial manufacturing, and any measurements you've taken are likely to be close enough anyway.

I guess that would work, i'm still going to make the depth tool as i'll likely use it often.

I just got a lot of my lathe materials delivered today, but i need a few more things for my setup before i begin. It turns out i can cut wheels and pinions, and it's not even that difficult to be honest, i don't need to buy gear cutters since i can easily make them. I even found a way to make a gear cutting attachment, which also could be used for other things.

Regarding homemade depthing tools :

http://gadgetbuilder.com/Gear_Depther.html

http://www.horlogerie-suisse.com/forum/viewtopic.php?f=2&t=4927&start=30

Christian.

I've seen those, but they're for clocks, the one i'm going to make will have a slightly different layout for the ease of building. I'll show everyone that when i make it as well.