#3 Making Thin Brass for Thin Wheels

The ETA/Unitas 6498 has a wheel thickness of 0.27mm-0.30mm (the reference movement seems to have some variation between wheels) which necessitates the need for very thin brass. CZ120 brass sheet does not come in sheets anywhere nearly thin enough. However a reliable solution to this problem was found.

An experiment was carried out to see if a very thin disc could be parted from 8mm CZ121 brass round bar (brass rod). This was performed on the Sherline lathe as a good parting tool was readily available for this.

A few issues had to be overcome. Firstly, on a three jaw chuck, despite it being very tight, the pressure of the parting tool pulled the work out of the chuck gradually, causing misalignment. A four jaw independent chuck solved this. Secondly, the center height of the parting tool must be absolutely aligned to the center. Too high and it rubs on the work and cuts poorly causing lots of heat, too low and the parting tool gets pulled under the work. The parting tool must also be extremely sharp. Finally, due to some flex in the parting tool and the thinness of the brass disc, the parting tool does push on and warp the disc being parted, for which the solution is given below. A final note, I found this to work best without any cutting lubrication.

The final parted discs were dome shaped due to flex in the parting tool as shown below.

The knub in the center of the domed disc was removed with a side cutter and a diamond stone was used to remove any remaining material. Then the disc was gently peened flat with a hammer on an anvil. Finally I gave it a quick cleanup (not to final quality) with lapping paper. In the image below, three parted discs are shown, two with the dome shape and one which has been finished according to this paragraph. Final disc shown on the right below.

The disc was checked and remains at 8mm in diameter after the process. As can be seen below, the disc is perfectly round and flat with a depth of 0.27mm. Success!

Due to the thinness of the part, I would pre-drill the center prior to parting (which would also avoid the need to remove the knub). When making the final wheel, I will also try cutting the teeth prior to parting to see if this is at all feasible without damaging the teeth.

If you would like to join the discussion on this post or leave a comment, please click here.

#2 Taking Measurements (6498 Calibre Movement)

As was explained in the previous article, while the movement being made will be unique, it will also be compatible with wheel and pinion sizes from the Unitas/ETA 6498-1. It was hard to find any existing documentation on the gear-train sizes for the 6498, so I decided to take the measurements myself.

In particular, I needed to measure the major-diameter (tooth-end to tooth-end) of each wheel and pinion, as well as count the teeth/leaves. From this the all important 'module' measurement and pitch diameter can be derived.

In order to take these measurements I used an AmScope 3.5X-90X Trinocular Stereo BOOM Microscope+10MP Camera +4-Zone LED Light which was purchased for SMD soldering work (nothing to do with watchmaking). However, it turned out that through the included AmScope software, it could measure wheels and pinions to a high degree of repeatable accuracy. Bear in mind this does require accurate calibration each time the zoom level is adjusted.

6498-1 Wheel and Pinion Measurements

The Major Diameter was measured for each wheel and pinion in the 6498-1 movement and number of Teeth noted. The appropriate Module Constant was also used to calculate the final Module - in accordance with the recommended method from PP Thornton. With this information, all other dimensions can be calculated as needed. The dimensions from the 6498-1 movement are shown in the table below:

Name	Type	Major Diameter (mm)	Teeth	Module Constant	Module
Setting Wheel	Wheel	3.85	16	2.76	0.205
Intermediate Setting Wheel	Wheel	3.85	16	2.76	0.205
Minute Wheel	Wheel	8.3	36	2.76	0.214
Minute Wheel	Pinion	2.51	10	1.61	0.216
Driver Cannon pinion	Wheel	3.05	12	1.61	0.224
Ratchet Wheel	Wheel	13.76	56	2.76	0.234
Crown Wheel	Wheel	8.86	35	2.76	0.235
Centre Wheel	Wheel	12.48	80	2.76	0.151
Centre Wheel	Pinion	2.54	12	1.61	0.187
Third Wheel	Wheel	8.63	60	2.76	0.138
Third Wheel	Pinion	1.71	10	1.61	0.147
Second Wheel	Wheel	8.42	70	2.76	0.116
Second Wheel	Pinion	1.26	8	1.71	0.130
Escape Wheel	Wheel	6.34	15
Escape Wheel	Pinion	0.97	7	1.71	0.111
Barrel	Gear	17.3	88	2.76	0.191
Balance Wheel	Bal	14.05
Hour Wheel	Gear	8.73	40	2.76	0.204

It should be noted that the 'Module' in the table is the absolute module. Cutters are only available in certain modules, so I will take either the nearest cutter module size if the number is very close, or else I will round down to the nearest available size.

I also have the wheel and pinion dimensions plus tooth count for the 6498-2. Please leave a comment on the forum if this is of interest to anyone.

6498-1 Measurement Images

Below are images of the measurements I took of the 6498-1 movement using the microscope. It may however be easier to download them from here as the filename corresponds to the part being measured. Note that the measurement text is on the image itself, but is small and may only be visible when the photo is fully expanded.

Validating Measurement Accuracy

Measurement accuracy was verified in several ways. Firstly, some of the wheels were checked using a Mitutoyo 293-821-30 Digital Micrometer. Secondly, the centre distances were also measured using the AmScope and calculations performed to derive the diameters from the centre distance and number of teeth. Finally, dimensions were originally measured for the 6498-2 movement and were remeasured on a separate occasion for the 6498-1 movement (after recalibration of the scope). For all critical dimensions the three methods agreed within 0.07mm and in some cases the disagreement between measures was less than 0.005mm. Given the movement being examined was also used and noticeably worn, the agreement between calculation using centre distances and major diameter was quite remarkable.

Next Steps

Next I will likely use these measurements to recreate a CAD model of the 6498-1 movement. I will post the drawings for this in a future post.

If you would like to join the discussion on this post or leave a comment, please click here.

#1 The Movement Base Design

Many years ago I designed a watch movement using Fusion 360 which had my own arbitrary wheel and pinion sizes (I will upload the designs for this in a future post). Since then life got in the way (moved country, had two children as well as two full-time job changes) and so the manufacture of the movement only got to the early stages.

Having re-evaluated my goals, one of which is to become a competent watchmaker, then I realised a pretty big problem with my first design. It will be very hard to test until most of the movement is built, and when it inevitably doesn't initially perform as well as hoped, it will be quite hard to diagnose exactly where in my new creation the various problems lie.

At the time I probably wouldn't have easily parted with the design which took me many months to complete. But now, 7 years later I'm looking at this with a fresh pair of eyes and I've decided on a new approach.

While the movement layout and many other aspects (such as the keyless works) will be entirely my own, I will ensure the wheels/pinions are compatible in size with another reference movement. This way I can make one wheel/pinion at a time and drop those parts into the existing reference movement as well as swap parts out in my watch to root-cause specific issues. By placing a handmade part into an existing reference movement it's individual performance can accurately be characterised. It's also a more motivating way to work as results can be measured sooner.

Choosing The Reference Movement

The Unitas/ETA 6498 was originally designed as a pocket watch movement, but it has since featured in many wristwatches. For this reason it is on the larger end of wristwatch movements which makes it a good candidate for being handcrafted from scratch.

The original goal was to use the Unitas/ETA 6498-2 which runs at 21,600 BPH. I generally prefer watches with a higher frequency of vibrations, not only because of the potential accuracy improvement, but also I think the second hand just looks better at a higher frequency.

However, after reverse engineering the 6498-2 it turned out to have some pretty small wheel teeth to cut. Effectively, the 6498-2 is a 6498-1 layout with tighter wheel-pinion ratios and a balance frequency. As a result a 6498-2 looks like it will pose some manufacturing challenges. In particular the Second Wheel has an outer diameter of 8.4mm with 120 teeth - giving an approximate module of 0.07. PP Thornton only makes wheel cutters down to a module of 0.09 and I don't fancy manufacturing my own cutter, or cutting the wheel, on that scale for a first-watch project.

The Unitas 6498-1 movement used to take reference wheel and pinion measurements from

The Unitas/ETA 6498-1 on the other hand runs at 18,000 BPH, but has gear train which lends itself more readily to hand manufacture. The smallest module size needed is 0.11 (for the escape pinion) which is covered by readily available cutters from PP Thornton. Hence I have decided to design the movement to be compatible with the 6498-1 wheel and pinion sizes. But this does also give me the option if I'm feeling brave in the future to upgrade my movement to 21,600 BPH if I can find a way to cut wheels at a module of 0.07.

The 6481-1 Escape Wheel Pinion (Module 0.11)

Keeping the Design Unique

As will be shown in a future post, this movement will be particularly unique even though it will borrow aspects from the 6498 movement. At the same time, the rapid feedback and accuracy measurements I can attain by basing aspects of the movement on the 6498 will fulfil my goal of becoming a competent watchmaker in a reasonable timeframe.

Links

• Unitas/ETA 6498-1 Diagram and Service Chart
• Unitas/ETA 6498-2 Diagram and Service Chart
• PP Thornton's available cutter sizes by Module

If you wish to join the discussion on this post or add a comment please click here.