Dogwood's Watchmaking Journey

[dogwood]

I had a fun little project cross my bench this morning. @Alex_oz had an IWC Ingenieur from an unknown factory that suddenly stopped ticking. I told him I’d take a look but told him that since it’s powered by a 7750 clone movement, that unless the problem was super obvious I probably couldn’t help. I don’t like working on the 7750 at all. In the past I’ve found that servicing the 7750 was like trying to pacify a sac of angry raccoons on meth.

Fortunately when the watch arrived and I opened it up, the problem was immediately obvious: the little Sheppard’s crook spring that controls the position of the chronograph’s tilting clutch had come free and was jammed in the balance.



I was worried that the errant spring might have damaged the hairspring, but I was able to pull it free without much trouble and immediately the balance started swinging.



Now the problem was that I needed to get the spring back to where it belonged. Fortunately this only involved removing one bridge. Here’s a pic of a different 7750 movement that I pulled out of my parts drawer to use as a reference on how / where the spring needed to be reinstalled.



After reinstalling the spring, I adjusted the eccentric screw (red arrow) to ensure that the chronograph tilting clutch mechanism spacing (green arrow) was optimal (you don’t want the teeth to engage too much, they should tough just enough to allow power to transfer).



This was a fun little project. I still don’t like working on the 7750. But maybe it’s only as bad as a sac of badgers on coke.

 

[dogwood]

I recently had the pleasure of servicing the new Dandong 3285 (CF3285 or DD3285). This is basically a VS3235 with different date / motion works on the dial side. Something that I found interesting is the implementation of the hour jumping mechanism. I thought I'd share a few pics since I haven't seen many of these posted.

This is the hour hand jumping spring. The mechanism is somewhat disassembled to show the spring. When assembled, the pinion (red arrow) sits on top of the spring, and the four lobes on the pinion slide past the jumping spring (blue arrow). Conveniently, for assembly, there are two little cutouts (orange arrow) to allow you to compress the spring with tweezers.



Here's a short video of the assembled mechanism functioning from the crown.

 

[dogwood]

I've had a lull working on watches recently, and took the opportunity to do some upgrade to the tooling on my bench. I've never had confidence in my watch press, so I ordered a beast of an upgrade off AliExpress.



As expected the assembly was pretty bad and the parts were covered in metal chips from the manufacturing process.



I needed to disassemble the tool, and thoroughly clean all the parts before re-assembling and greasing it (pretty much exactly what you'd do with a watch movement that arrived from China).



The tool did clean up very nicely. And as a further upgrade, the tool included a really nice set of aluminum / nylon press dies.



This new press is an absolute beast. For comparison, the first press I ever bought weighed 300g. The one I've been using for the last year is about 1.4kg. And this press is well over 5kg.



The next project was to try and fix the heating element in my Elma Solvex parts cleaning machine. The heater stopped working a few weeks ago. This isn't a terrible thing since it's currently summer in Canada, so there's not much risk of condensation on the drying parts if I'm just using room temperature air. But I'd ordered a replacement heating element and figured I'd try to do the swap.



The machine is relatively easy to disassemble (at least the bottom half). And I was able to remove the blower and heating element assembly.



And thankfully, the German engineers made the wiring harness simple enough that even a retired physicist like me could figure out how it worked.



Unfortunately, replacing the heating element didn't fix the problem. I think the issue might actually be with the relay that controls the heater. Fixing that might be above my pay grade. I've contacted Elma support, so I'll keep y'all updated on progress here.

However, while I had the machine apart, I noticed that the intake air filter wasn't very good on the machine. The filter is basically a 2mm thick piece of foam.



While the air in my workshop is very clean (I run a HEPA filter to keep the air as free from dust as possible), I'd like to have better filtration on the air that's going to be blasted at my freshly cleaned watch parts. And, since I don't have a functional heating element at the moment, I don't really need to worry about maintaining whatever the design spec air flow is through the dryer.

The air intake is just these slots in the housing at the back of the machine:



I thought about designing some kind of 3D printed contraption that would fit over the existing foam filter. But that seemed like overkill, and I simply repurposed some pandemic surplus filter media (aka a surgical mask) and some masking tape to create an effective dust filter over the air intake.



This solution obviously won't work when the heater is fixed, since the heater was spec'd for a certain air flow, and my improvised air filter is reducing that air flow. Nonetheless, while the weather is warm (and humidity is low), I think this is a workable solution.

 

[WatchSmith.US]

Looks like Christmas morning at Dogwood’s house! Love the new tools!

My Bergeon rack press required a teardown and stoning of a couple of deformed (dented) teeth on the rack that would catch during a press.

 

[dogwood]

Looks like Christmas morning at Dogwood’s house! Love the new tools!

My Bergeon rack press required a teardown and stoning of a couple of deformed (dented) teeth on the rack that would catch during a press.

The rack teeth on my press are really rough. They almost look like they were cast rather than machined or ground. But that seems ludicrous considering that the part was clearly turned on a lathe to get it to size and cut the top threads for the stopping nut. I did take a file to the rack teeth to remove some of the burrs that were hanging over from the cut threads, but I think I may want to take a stone to the teeth to them a bit smoother. The teeth on the pinion are very smooth, so it’s an odd combo.

Such are the trials and tribulations of AliExpress tools. Much like rep watches, we pay substantially less than for western made tools, but the Chinese versions come with some odd quirks.

 

[dogwood]

Final addition to the bench (for now)… a set of bespoke 3D printed bezel dies for my press. These were designed and manufactured by @Oascom. I’m really looking forward to using them.

 

[Prizon]

awesome!

 

[Oascom]

Final addition to the bench (for now)… a set of bespoke 3D printed bezel dies for my press. These were designed and manufactured by @Oascom. I’m really looking forward to using them.

Enjoy!

 

[dogwood]

I had an interesting day working on one of the new SH3285 movements (aka the "VR3285"). This movement is a hybrid just like the VR3235 -- it's part 3135 design and part 3235. However, unlike the VR3235 which is mostly 3135 design, the SH3285 has many more 3235 style designed systems. For example, the VR3235 has a 3135 style keyless works, but the SH3285 has a 3235 style keyless works. In theory this would be a good thing. The 3235 keyless works is nicer to work on (in my opinion) since it has fewer parts and no thin springs that have a tendency to launch themselves into orbit when you poke them the wrong way.

However, a movement from one of the new Explorer 2s appeared on my bench with a keyless works that wasn't working properly. When I got the movement open, I saw the issue was that the setting lever had broken.



The setting level connects directly to the stem, which is operated by the clumsy fingers of humans, so this part needs to be strong since humans can push and pull on things with quite a lot of force. Looking at the broken part under the microscope, I worry that Shanghai movement factory might have under designed this part, or maybe made it out of an alloy that's not strong enough for the forces it will have to endure.



Fortunately the part appears to be gen-spen since I was able to replace it with a gen 3235 part, and the keyless works appears to function properly (although it doesn't click into each crown position as I'd like).



However, as I was putting the watch back together, I noticed that after I'd set the hands that the movement didn't run when I pushed the crown back in. I hadn't tested if the movement ran before since it arrived on my bench with a broken keyless works, but when I flipped the movement over, I saw the problem. The jewel for the center seconds wheel on the train bridge was missing.



I looked at the hole where the jewel should be and measured it to be 1.10mm in diameter. The hole also looks rough. I don't know if this is normal for these SH movements, but when I tried pressing in a new 1.10mm jewel, the jewel didn't have any staying power.



I decided to broach the hole open to 1.20mm and see if I could press in a larger jewel.



I was able to press in a 1.2mm jewel, but I had to swap it out a few different times because the spare jewels I have are an assorted set, so it took me a while to find one that would fit the ~0.15mm pivot on the center seconds wheel arbor.



However, after pressing in the jewel, and re-installing the bridge and removing it, adjusting the end shake, re-installing it, removing it, adjusting the end shake again, re-installing it again, I still couldn't get the train wheels to spin freely. I suspected that the issue was the end-shake on the escape wheel, so adjusted the escape wheel jewel on the train bridge, and managed to get the wheels to turn, but not very "freely". I'm worried that when I broached the jewel hole, I may have moved the hole's center. Technically I couldn't have moved it by more than 0.10mm since I only broached it from 1.10mm to 1.20mm, but maybe that's enough to cause the train wheel teeth to not mesh well enough to run freely. Regardless, I have more work on this movement before it will run again.

 

[dogwood]

I had the chance to use my new press to reinstall a caseback crystal that had popped out. And I have to say, I love Love LOVE this new press.

 

[dogwood]

Last night I had an interesting watch on my bench -- one of the new Tungsten weighted DayDates. It does indeed feel heavier than a normal stainless steel rep. But I wanted to figure out where the weight had been added.

Pure tungsten is an incredibly hard material, so it's very unlikely than any parts that require intricate machining (i.e. threaded holes or other complex geometry) are made out of pure tungsten. However, tungsten can be mixed (not alloyed) with copper to make a metal-metal composite material. You'll often hear people say "Tungsten-copper alloy" which is technically incorrect since the two metals don't form a solid solution, a better analogy would be to think of oil and water that's shaken up so there are tiny droplets of both oil and water mixed together and then it gets flash frozen so you get a solid block of fine oil and water particles that are stuck together. But Tungsten-copper is much more easily machined that pure tungsten, the higher the ratio of copper to tungsten the easier it is to machine.

Ratio of Tungsten (W) to Copper (Cu)	Density (g/cm3)
W55 Cu45	12.30
W60 Cu40	12.75
W65 Cu35	13.30
W68 Cu32	13.93
W70 Cu30	14.18
W75 Cu25	14.50
W80 Cu20	15.15
W85 Cu15	15.90
W90 Cu10	16.75

For reference Au750 (i.e. 18K gold) has a density of 15.4g/cm3... so adding a bit of W90-Cu10 (at 16.75 g/cm3) is a good wait to increase the average density of a watch while still using stainless steel (7.5-8.0g/cm3) for the parts of the watch where complex geometries need to be machined into the part.

Interestingly, the case felt a bit thick. I'm not sure what an unweighted / gen DD case should measure, but I measured this weight case to be 13.1mm thick. I think the gen is supposed to be 12.1mm thick, so it feels like the caseback has been extended to fit a 1mm thick weighted slug inside.



I weighed each of the parts of the watch (except the movement which was a standard eta clone 2836). And here are the results. You can clearly see the tungsten slug in the caseback.



The ring is the movement spacer ring... the ring felt "heavy" in my hand, but it does have threaded holes machined into it for the case clamp screws, so I felt it was unlikely to be a high ratio of W-Cu due to the difficulty in cutting threads into high tungsten content W-Cu material.

But weight is just the first step... the next thing I needed to do was to measure the volume of each of the parts. With volume and weight I'd be able to calculate density, and with density, I'd be able to compare to the known densities of various materials to make a guess as to what parts are made of what.

Measuring volume isn't easy without equipment specific for the task. My improvised apparatus was a surface gauge and a whisky glass (the glass in my kitchen with sidewalls that were the most vertical and consistent in thickness.



With this setup I'd be able to drop in the part, and then measure how much the water level increased.



While this works in theory, the whiskey glass is a bit too wide, so the elevation change of the water isn't very large. So, even with a surface gauge that can measure to 0.05mm, I'm not at all confident in my measurements. What I need is a proper graduated cylinder that's tall and narrow.

Regardless, here are the results. I'm not sure that they're even worth presenting due to the large uncertainty of the water heigh change. But I thought I'd present my methods even if the results aren't valuable just in case anybody else has a better setup and can replicate this experiment and get better results.

 

[PsychoTee16]

You must be an Engineer!

 

[MrButterfield]

Could you measure the volume displacement of the water by weight?

1. Fill glass to rim (max amount container can hold)
2. Weigh glass + water
3. Gently place materials in water, causing water to overflow glass
4. Remove materials from water, being careful to let all water drip from materials back into glass
5. Weigh glass + water
6. Compare weights, before to after

Water has known mass per volume, so one should be able to calculate the displaced volume according to mass of the displaced water.

I’m not a scientist or engineer; I just like problems like this, and this would be how I would attempt it.

 

[dogwood]

Could you measure the volume displacement of the water by weight?

1. Fill glass to rim (max amount container can hold)
2. Weigh glass + water
3. Gently place materials in water, causing water to overflow glass
4. Remove materials from water, being careful to let all water drip from materials back into glass
5. Weigh glass + water
6. Compare weights, before to after

Water has known mass per volume, so one should be able to calculate the displaced volume according to mass of the displaced water.

I’m not a scientist or engineer; I just like problems like this, and this would be how I would attempt it.

That’s a really good idea actually. But I wonder how hard it would be to get all the water drops out of the bracelet back into the container.

I was thinking of various geometries of container where there’s some kind of tall thin (like old thermometer) tower where the change in displacement could be read from. But I can’t think of a geometry like that where it’s easy to also insert an object.

 

[j2k]

Mad lad.

 

[MrButterfield]

This shows a variation that combines both of our strategies. It would require getting (or making) some specialized equipment for which I haven’t yet found the proper search term on Amazon. Other sources have it listed as either a “eureka can”, or “overflow vessel”. Combine that with a graduated cylinder of a useful size (widely available on Amazon for cheap).

 

[dogwood]

Continuing the saga of the SH3285 movement from a few posts above… I had a spare vs3235 train bridge, and I thought I’d try using it in the place the SH3285 train bridge that I’d had to press a new jewel into. However, it turns out that the SH3285 movement is ever so slightly not gen spec.

For starters, the crown wheel has a larger diameter than the VS/gen. It also appears as though the position of the crown wheel might be slightly different than the VS/gen movement.



In the process of doing this investigation, I also discovered that my SH3285 escape wheel pivot was bent.



I figured this would be no problem, since I have plenty of parts vr3235 movements on my bench. But it turns out the vr3235 escape wheel isn’t compatible with the SH3285 movement, even though it’s very close. I even tried pressing the jewel of the bridge to make more room. But it just wouldn’t line up correctly.

 

[lollollol]

I sent in my Rolex Oyster Perpetual 39mm with A2824 movement around mid February. I was trying to fix the watch myself and I realized that I probably messed up the keyless and it needed to be reset. I looked around for someone who was able to work on rep watches, and I went with @dogwood due to the positive reviews from other users, plus this thread, where he documents the watches he works on, which I found to be quite interesting. He was able to quickly diagnose the issue and sent me a detailed list of the steps that had to be taken in order to fix the keyless and get it back up and running. I was quoted CAD$50 for the service, while other rep-friendly watch smiths quoted me approximately $70. Since he is on the west coast, it took a few days to get there. There was a bit of a line of watches in front of mine, and started working on it around the beginning of April, and sent me pictures and videos documenting the process, which I greatly appreciated. I got my watch back around the end of April, and was really happy with the communication and the service itself. I understand that he does this on his free time so it took a bit longer, but I did not have any concerns regarding it. He also helped with regulating the watch, as it was running +8 s/d he says he usually charges $50 for regulation, but volunteered to do it for free, which I greatly appreciate.

He is extremely knowledgeable and I will definitely use his services again in the future!

 

[HardyVancity]

I had the pleasure of connecting with @dogwood , who is an exceptional individual with extensive knowledge of watches and their intricate movements. He provided honest advice and is truly a class act.

I have been actively observing this forum for several months, and what led me here was the purchase of my first Super Franken Sub 116600 LN VR3135 movement. I obtained it from a member of this forum who collaborated and built it with a trusted watchmaker recommended within this community.

Given my location on the West coast of Canada, there is a lack of local watch service providers for reps, actually none that I've come across so far. I am incredibly grateful to have crossed paths with @dogwood

Overall, the watch has some issues, but they are all fixable through a service. I want to express my gratitude to Dogwood for his invaluable assistance and help!.

 

[dogwood]

I had the pleasure of connecting with @dogwood , who is an exceptional individual with extensive knowledge of watches and their intricate movements. He provided honest advice and is truly a class act.

I have been actively observing this forum for several months, and what led me here was the purchase of my first Super Franken Sub 116600 LN VR3135 movement. I obtained it from a member of this forum who collaborated and built it with a trusted watchmaker recommended within this community.

Given my location on the West coast of Canada, there is a lack of local watch service providers for reps, actually none that I've come across so far. I am incredibly grateful to have crossed paths with @dogwood

Overall, the watch has some issues, but they are all fixable through a service. I want to express my gratitude to Dogwood for his invaluable assistance and help!.

Thanks for the kind words Hardy.

It’s always great to meet up in person with local Canadian rep-fam members.

Hardy’s watch is a really nice franken, but the movement was serviced 3 years ago and the amplitude has dropped from ~260 in July 2020 (right after the service) to ~200 today (three years later). The movement probably just needs to be cleaned and re-lubricated, and maybe the mainspring needs replacing — normal wear and tear on a mechanical watch.

I’m looking forward to working on your watch.