My dark 20 years history in making

I know next to nothing about watchmaking. Did you cut all those parts by using the lathe as a milling machine by putting a mill in the spindle and mounting the stock on the cross slide? (Thanks for continuing to share this journey with us!)

1 Like

Nope. First I bulit a router , then I built the small portable router with the big one (clip above). If you watch the hot wire CAD picture and virtually remove both the posts for the z Axis but add a gantry instead you have my big router. Then I used the big one to build the lathe.

Most above parts are done with the first big router, a few with the lathe.
Mostly used tool for that parts is a 0.3mm flat tip shaft bit. Some engravers and bigger flat tips for the surface patterns the posts of that tourbillon in the last pic are syringe needle pieces threaded with 0.5mm from both sides. E.g. the gears are made using a 0.3mm shaftmill . The precision plained brass is fixed to a HPL lost stock plate using CA. The part is shaped and afterwards the CA is dissolved over night in Acetone…

thats all…

6 Likes

WOW! That is truly amazing!

1 Like

There is much more “material” left but too much of the same kind might be boring…


This is done in similar way but not from Aircraft grade aluminium but from cold working steel. Conceptual design… too heavy…
Very fancy parts can be done same way, routed half way from stock then refilled with CA (activated with mild tenside as it is an anaerobe adhesive which needs initators to crystallize on)
The dashes have a width of appr 0.08mm the screwhole is 0.5mm still embedded in CA to be dissolved

2 Likes

Getting closer to today … When buying my Ramps board for my 3D Printer I thought the boards were so cheap that I had to take 2 sets… thinkering about whatfore to use the second set I decided to do something beside p

rinting

2 Likes

I think I now should break the timeline as we are close to status quo, - but not to bore you I left out many projects over the time…

3 Likes

Recently: as I have told, I’m currently digging into watchmaking.
The latest challenge are two vintage quarz watches with defective coils of the lavette stepper motors.
Unfortunately those coils are wound with 0.02mm
enamel wire which is not possible to be wound by hand. At least not by n= 1800 .
So I started buliding a coil winder…

!
First part is the wire auto support with someway a constant tension source.
( until now ist just mechanics but the uc boards should arrive by next week)

5 Likes

28mm Axis with the winding stepper motor

Shafts still have to be fitted / trimmed and the locking screw holes to be threaded …


Done…
Functional test with existing electronics and slightly too high current values ( did not want to adjust those)… done…

2 Likes

Meanwhile I setup the harnesses of the light barriers… as I try to have a 5 states wire support indicator using an array of light barriers, I recognized that there is too much light emitted so the phototransistors are not fully high impedance. Perhaps I need a comparator PCB to have clear digital signals.
I also did a simple sketch to get the axis moving but recognized a strange thing I will have to check tomorrow.
Each I/O I need is initialized to be an output and pre loaded with the initial state. The sketch consists of two for loops to apply the stepping signals to the polulu driver shields. When the program starts the feeding motor does several steps into the wrong direction before changing to the right one… tomorrow I will have to see if ths would be hardware bound.

But the stepping feed is pretty acurate for the cheapish setup I run. Step width is between 0.03 and 0.05mm
With quite some elastic play under applied load conditions. But how much load can I get from a 0.02mm Copper wire, othogonal to the moving direction. :grin:

But thats it just for today…

1 Like

Today is just short, created some code lines for a discrete cheap control and tested… next would be to connect the stepper stage to the arduino instead of changing port levels only.

The screnn of the oscilloscope shows the 5 weightened frequencies which will afterwards feed the stepper to provide wire

1 Like

Both boards have a very reduced core loop program not to get the controller too much from telling the steppers.
So display and variable parametrization are just open to be implemented, but will probably be added before the winding process and not during… but until now it seems to work… Test is done with 0.05mm enamel wire but 0.02mm should also work

2 Likes

Finished the first revision of the dispenser software
With slightly decreasing the smooth characteristics of the stepper. Now with the right buttons pair you select the wire diameter which is shown with the resistance per meter.
Second is to adjust the resistance you want to dispense /apply then press enter and the dispenser counts the length, calculates the resistance and as the adjusted value is reached it shows an alarm.

2 Likes

First system test on a MIYOTA stator.

Enamel wire 0.05mm Copper diameter.

7Ohm means 7.x m wire…

Wound in rather 35min.

2 Likes

Finally implemented the teach in function for the winder… now its just the case for the controller

2 Likes

Further testing… had a discussion with a colleague and he gave me a few meters of 0.02mm diameter enamel wire from a mobile repair for testing…

Putting this into the device was pure horror but finally I made it and surprisingly the machine had no problem processing this. I ’ m also searching for 0.01mm enamel wire as AWG 54 magnet wire for benchmarking… but a single KG of 0.01 is sold for roundabout 10000 bucks +VAT in Germany… rather 250 KM of wire… but I just need some hundred meters.

2 Likes

While waiting for reply of a possible Chinese wire source,a I thought about what to do next and ended up with the idea of building a lapping machine to grind and polish watchparts.
The concept bases on a half open planetary gear where the object carrier is moving on a hypocycloide trajectory over my mirrorplates carrying different lapping films,!
20200202_183515|243x500


.

3 Likes

Whats left is a nice box around the electronics and some Brass weight and object holder.

1 Like

Turned the two carrier weight elements and thought to initially process those… its very slow . But the results are rather flat… but still with fine scratches … now I will have to find a compound to polish them afterwards…

1 Like

Trying to grok how that works. Any chance of a video showing the underside of that in operation too? :slight_smile:

Watch the previous clip…assume that the weights have centric holes where the pins are picked in and which move the weights (later with the objects stuck to the underside of the weights over the abrassive structure. Pressed down by their own weight ( which may be a little low for a pressure polish…)

1 Like