Rose & Son London - Early Verge Fusee Watch (1st watch)

[Sooth]

Hi everyone, I've been a clock collector for many years and recently got interested in acquiring an early key wind fusee watch. I don't really have a lot of experience with watches, but I do have a lot of patience and I'm in no rush with this particular little project.

I found this one online for what I hope was a decent price, however, I'm not too sure exactly what I have. The only other search result I found for Rose & Son was a discussion here about Swiss fakes. I believe this watch likely falls in that same category. I wanted to ask your thoughts on the watch. I'm not too concerned if it's a London-made piece or not, and I'm already fairly convinced that it isn't.
See other thread here: https://mb.nawcc.org/threads/rose-son-verge-fusee-anyone-heard-of-this-maker.94579/

From what I can tell, it looks like the outer case is not original to the main watch since the hallmarks are different, and it's not quite an exact fit (but it's very close). I made a very lazy attempt to date the hallmarks and it looks like the outer case might be in the range of 1800-1810 based on the F. The hallmarks on the inner case are very worn, so I'm completely unsure. The name on the dial does match the engraved signature on the rear plate of the movement. I quite liked the fancy shaped pillars. I thought these were square, but they're actually pentagonal, which seems like a lot more effort to do.

The ruby setting on the back cock is rather interesting. It appears to have facets on the upper (outer) part of the jewel which seems strange to me as I'm used to seeing watch jewels as just simple doughnut rings or cabochon style end stones. I thought it might be a later addition, but the screw locations (when looking at the back) appear to have been incorporated into the design. As I'm new to watches I'm not sure if I should take the screws off and clean the stone or leave the entire assembly together.

As for repairs:

The hair spring is broken (but present), and having removed the balance, I'm a bit worried that replacing the spring might be quite a challenge for me. I'm not sure exactly how precise the replacement spring would need to be as far as length and number of coils. I've seen a number of these early watches with very sparse number of coils and rather coarse looking hairsprings. I'll likely ask more questions about this in a separate thread. I may have some loose ones in boxes of scrap parts from my grandfather's things (he worked on watches).

For the crystal, the watch needs one approximately 450mm and "high dome" style. I've seen some early fusee watches with and without a ground flat circle in the glass, and I'd really prefer one without the flat, however the replacement ones I can find online only seem to offer the type with the flat ground top. Does anyone know a source for a regular high dome crystal? It's too bad it doesn't use a standard crystal since I have about 150 spare ones in all sorts of sizes.

The missing hands should not be an issue as I can find or make some to suit. The potence bits are kind of an overly complicated mess of brackets and screws. I have not tackled that part yet. I can't really understand why there are so many pieces involved. The mainspring appears to be ok, but I see some rust from the barrel opening, so it may or may not be safe to use. I do also have hundreds of spares, but no idea of the size needed yet. The fusee chain appears to be in perfectly fine condition, but there is some light rust, and it's fairly dirty and a bit stiff. I'm fairly confident that it will clean up nicely.

There's a bunch of light surface rust here and there, but all the bits and pieces appear to be in fantastic shape (nothing broken, no ugly file marks or apparent repairs, no excessive scratches etc). The only repairs I see are that there were attempts to close the balance cock screw hole to tighten it, and some slight "hole closing" efforts made on some pivot holes (unless this is original).

 

[Sooth]

Here are some additional images.

 

[Sooth]

The dial was quite dirty but cleaned up beautifully. Zero hairline cracks which is amazing considering the age and that it had no glass for some time. Attached are some close-ups of the hallmarks. The numbers not quite visible are: 6 then 1 4 0 2 I have no idea what these are meant to be for.

 

[gmorse]

Hi Sooth,

Welcome to European pocket watches!

You have a lot of questions and to your credit you've answered many of them correctly already. It is a 'fake', made not in London but somewhere in Switzerland. There's an interesting video on the subject by Dr. Rebecca Struthers.

Some of the hallmarks do appear to be genuine and show that the outer case was assayed in London in 1801/2 and was made by Russell Pontifex at 24 Primrose Street, Bishopgate Street, London. The inner case, (old name 'box'), appears to have faux marks, and in view of this you're quite right to suspect that they don't belong together, although it wasn't that uncommon to have the two cases made by different people.

The clues to its continental origins are mostly subtle, but you mention the shape of the pillars; these are certainly pentagonal and they're not found in English work but are typical of these fakes. Square pillars on English watches fell out of fashion in the 1770s and this clearly isn't that old. The complex potence arrangements allow for adjustment of depth and drops without dismantling the movement, unlike English practice which was much simpler. Other clues are the motion work wheels which are relatively small and the pillar plate bar which is an unusual shape. The red endstone on the balance cock table may well be just cosmetic and covering a blind plug as the actual bearing. Steel end-plates instead of jewelling are not usually found in English work.

Balance springs on very early watches only had one or two coils and gradually became longer as steel making and finishing technologies became better. Without finding a suitable donor you'll have to shorten and open out a newer spring to make the balance oscillate at the correct rate.

The state of the rest of the steel-work doesn't look encouraging but you'll have to see how it cleans up after stabilising the rust. These watches, whether fake or genuine, were made long before any concept of standardisation existed, so finding donor parts which fit, even on a movement which appears similar or even identical, is unlikely.

The correct crystal is indeed a plain high-dome without a flat in the centre, known as a 'bulls-eye'. These are a later 19th century fashion.

'Punching-up' pivot holes certainly isn't an original feature, although the holes would have been bushed as part of the manufacturing processes, and in this instance the recesses around the fusee holes are probably original, another clue to the origin.

I've probably left out some stuff, but please come back with any questions. Bear in mind that this could introduce a new obsession to your life!

Regards,

Graham

 

[John Matthews]

As you will have discovered there are many examples of watches of this type signed Rose and Son, London. All the movements I have seen have features from which it may be inferred are not made in the traditional English manner of the late C18th early C19th. Some of the movements are housed in pair cases. Often the outer of which has genuine London hallmarks. While it is almost certain that the movements originated in Switzerland or France, these watches were being sold in London and it is possible that the date of the hallmark might represent final casing prior to sale, or at least the earliest date. No doubt existing cases will have been reused, all in an attempt to reinforce the deception.

For the earlier example on the forum that you provided the link to, the hallmark on that outer case is 1789/90 and not 1809/10 as thought at the time of posting

John

 

[Sooth]

Graham and John, Thanks both so much for all the detailed information. I have worked a bit more on the watch. The fusee chain is now nice and clean (as clean as I hope to get it without specialized equipment) and I have figured out a few other small things:

First: the mainspring is indeed broken.
Second: when you say "the pillar plate bar is an unusual shape" I assume you mean the sub-plate that holds the 3rd wheel and the contrate wheel?
Third: the balance cock doesn't appear to conceal a typical brass or steel plug, the back is totally flat, and looking through the minuscule hole, I can see the red of the stone. See images.
Fourth: having looked at the balance staff, I can see that the hairspring is just held in place by a teeny tiny pin on the side of the central "hub". I can't imagine how pieces this small were made so accurately in the 1700s/1800s.
Fifth: what I thought might be a burr or a defect on the balance wheel rim appears to actually be some kind of banking pin? I wasn't aware this "was a thing" on balance wheels. I am not sure if this registers against a corresponding stop. I'd need to have a closer look at the pieces again. It sticks out of the bottom side. See image.

Regarding the rust: it looks much worse that it actually is. The bulk of it just wipes off with a cotton swab, it really is almost just sitting on top. I don't believe there is much corrosion to any of the essential parts (pinions/pivots).

Note: It would be nice to return this watch to running condition, but it's not necessarily one that I plan to take out and wear for a day. I had planned for it to be more of a show piece and something to admire and run every once in a while. I'm not overly concerned with having it be accurate, but will be happy to try my best to get it as close as possible.

 

[gmorse]

Hi Sooth,

First: the mainspring is indeed broken.
Second: when you say "the pillar plate bar is an unusual shape" I assume you mean the sub-plate that holds the 3rd wheel and the contrate wheel?

Yes, that's what it's called.

Third: the balance cock doesn't appear to conceal a typical brass or steel plug, the back is totally flat, and looking through the minuscule hole, I can see the red of the stone. See images.

That's good, many of these are as I described, just cosmetic.

Fourth: having looked at the balance staff, I can see that the hairspring is just held in place by a teeny tiny pin on the side of the central "hub". I can't imagine how pieces this small were made so accurately in the 1700s/1800s.

The balance spring is pinned into a brass collet which is a press fit on the actual hub. It can be levered off with care, but don't damage the spring itself in the process; best not to drink any coffee before this task!. This picture is of a lever balance but the spring attachment is the same.

Fifth: what I thought might be a burr or a defect on the balance wheel rim appears to actually be some kind of banking pin? I wasn't aware this "was a thing" on balance wheels. I am not sure if this registers against a corresponding stop. I'd need to have a closer look at the pieces again. It sticks out of the bottom side. See image.

Yes, that's what it is, and all balance wheels have to have such a mechanism, although there's a variety of ways to achieve it. It should bank against a peg or pin somewhere under the balance cock. The angle between the flags on the balance staff looks relatively small, it's usually between 90˚ and 110˚.

Regards,

Graham

 

[Sooth]

Hmmm, I may have already hit a small snag. I am not sure if the mainspring size I need is commercially available. The original is 3.99mm wide (4mm would likely be fine), thickness is 0.19mm and the length is around 19.5 inches. I do potentially have some old spare/scrap mainsprings that MIGHT work, but they are too wide, so I would need to grind or stone the material down to 4mm width, which honestly will be a huge pain. I'm not too familiar with watch parts supply houses, as I'm still mainly into clocks. Which main supply houses tend to carry the best stock/selection? I know we have Perrins here in Canada, but they don't have an online catalogue, you need to mail-order it. I have heard of some people using Meadows and Passmore in UK, are there others that are recommended.

As a side note, I found a way to clean the rust off the parts with relative ease. Things are slowly moving along.

 

[Girl59]

Hi everyone, I've been a clock collector for many years and recently got interested in acquiring an early key wind fusee watch. I don't really have a lot of experience with watches, but I do have a lot of patience and I'm in no rush with this particular little project.

I found this one online for what I hope was a decent price, however, I'm not too sure exactly what I have. The only other search result I found for Rose & Son was a discussion here about Swiss fakes. I believe this watch likely falls in that same category. I wanted to ask your thoughts on the watch. I'm not too concerned if it's a London-made piece or not, and I'm already fairly convinced that it isn't.
See other thread here: https://mb.nawcc.org/threads/rose-son-verge-fusee-anyone-heard-of-this-maker.94579/

From what I can tell, it looks like the outer case is not original to the main watch since the hallmarks are different, and it's not quite an exact fit (but it's very close). I made a very lazy attempt to date the hallmarks and it looks like the outer case might be in the range of 1800-1810 based on the F. The hallmarks on the inner case are very worn, so I'm completely unsure. The name on the dial does match the engraved signature on the rear plate of the movement. I quite liked the fancy shaped pillars. I thought these were square, but they're actually pentagonal, which seems like a lot more effort to do.

The ruby setting on the back cock is rather interesting. It appears to have facets on the upper (outer) part of the jewel which seems strange to me as I'm used to seeing watch jewels as just simple doughnut rings or cabochon style end stones. I thought it might be a later addition, but the screw locations (when looking at the back) appear to have been incorporated into the design. As I'm new to watches I'm not sure if I should take the screws off and clean the stone or leave the entire assembly together.

As for repairs:

The hair spring is broken (but present), and having removed the balance, I'm a bit worried that replacing the spring might be quite a challenge for me. I'm not sure exactly how precise the replacement spring would need to be as far as length and number of coils. I've seen a number of these early watches with very sparse number of coils and rather coarse looking hairsprings. I'll likely ask more questions about this in a separate thread. I may have some loose ones in boxes of scrap parts from my grandfather's things (he worked on watches).

For the crystal, the watch needs one approximately 450mm and "high dome" style. I've seen some early fusee watches with and without a ground flat circle in the glass, and I'd really prefer one without the flat, however the replacement ones I can find online only seem to offer the type with the flat ground top. Does anyone know a source for a regular high dome crystal? It's too bad it doesn't use a standard crystal since I have about 150 spare ones in all sorts of sizes.

The missing hands should not be an issue as I can find or make some to suit. The potence bits are kind of an overly complicated mess of brackets and screws. I have not tackled that part yet. I can't really understand why there are so many pieces involved. The mainspring appears to be ok, but I see some rust from the barrel opening, so it may or may not be safe to use. I do also have hundreds of spares, but no idea of the size needed yet. The fusee chain appears to be in perfectly fine condition, but there is some light rust, and it's fairly dirty and a bit stiff. I'm fairly confident that it will clean up nicely.

There's a bunch of light surface rust here and there, but all the bits and pieces appear to be in fantastic shape (nothing broken, no ugly file marks or apparent repairs, no excessive scratches etc). The only repairs I see are that there were attempts to close the balance cock screw hole to tighten it, and some slight "hole closing" efforts made on some pivot holes (unless this is original).

View attachment 680486 View attachment 680487 View attachment 680488 View attachment 680489 View attachment 680490 View attachment 680491

Hi, Sooth. I'm a novice at anything time-related but have developed an interest in pocket watches and am something of an Anglophile. I'm enjoying your photos. Especially love those that show the wheel train, etc., between the plates. Great to get this particular view of the workings. Your watch is interesting and the movement is lovely. Good luck!

 

[Sooth]

Some updates on the rust removal and cleaning. First photo is contrate wheel cleaned, and 3rd wheel before cleaning.

You can see the punch marks around the balance cock screw hole in the last photo. Those scratches on the front plate are much less apparent in real life.

 

[gmorse]

Hi Sooth,

I have heard of some people using Meadows and Passmore in UK, are there others that are recommended.

As a side note, I found a way to clean the rust off the parts with relative ease. Things are slowly moving along.

In the UK, Cousins are probably the largest materials house, you should be able to order watch mainsprings by size there. M&P are shut down for their autumn break at the moment until the end of November, (typically quaint!). If you can't source the exact size, I suggest you look at David Boettcher's website on mainsprings, which includes calculators using barrel dimensions, and can offer you workable alternatives. Height can be a little less than the inside of the barrel and thickness can be less than the existing spring, which probably isn't original and is possibly too thick anyway, although 0.19mm sounds OK for a watch like this. Remember that spring strength is proportional to height but to the cube of thickness. I don't recommend using NOS blue steel springs, they're often set and generally unstable and unreliable. The modern alloy springs may not match the fusee profile so well, but they're far safer for the movement.

As a matter of interest, how did you treat the rust?

Regards,

Graham

 

[Sooth]

Hi Graham,

Do you know of any watch parts retailers within North America, or would you suggest Cousins and paying for international shipping? I may need to do my own googling, or spend more time searching the watch forums here...

It's quite unfortunate that the original(?) mainspring is broken because it actually seems to be in rather good shape. I did try searching in Cousins and there are of course no matches for the spring size I need. The barrel inner diameter is 15.17mm or 0.59in

Using the "quick estimate" from your link above gives me thickness 0.17mm and length of 455mm which is pretty much what the old spring was.

"As a matter of interest, how did you treat the rust?"

I was afraid someone would ask this, but I suppose it was to be expected. I'm a bit hesitant to say, because it will likely be frowned upon and not recommended. I tried a few things. Generally I approach these things in an incremental fashion: start with the weakest solution/option and work my way up. I tried a short soak in vinegar, but the results were not great. I tried gentle scrubbing, but that would realistically only work on the plates or the sturdier parts. I needed something that would work effectively on the fine and delicate wheel work.

There are of course dozens of types of rust removers on the market, as well as iron removers, rust converters, plus less reliable (and potentially risky) home remedy solutions. Most are based on either a strong acid or a strong base (like baking soda). From past experience and personal knowledge, I know that a strong base will generally darken brass (which I want to avoid), and acid should not affect the remaining gold plating on the watch. I did not like the idea of long acid-bath soaks using a weak acid, so I decided to see what else I had on hand. I have a number of polishes and abrasives, strong acids like muriatic, and also CLR. I did not want to introduce any sort of abrasives of polishes as these would be a nightmare to fully remove from pivot holes or wheel teeth, so I looked at the CLR. A quick read of the label says "do not use on brass". Hmmm... why though? So I did a bit more research. According to the company's website, the product may strip-off brass finishes and may cause some pitting in certain grades of brass. I assume they mean lacquer topcoats, or brass plating, but it's not super clear. Since that's kind of vague, I did more online searching. Several people did report having success using CLR on various brass items without any ill effects, so I decided to try it in a very controlled way. I used dampened Q-tips and a small artist's brush. The parts were only in contact with the CLR very briefly, maybe 20 seconds to 1 minute max before they were carefully rinsed off and dried. This is likely not the best or safest way to remove the rust, but it allowed for very gentle cleaning using a liquid chemical rather than something more aggressive.

I should add a disclaimer: use at your own risk. I would definitely not soak anything in CLR for any periods of time. I don't actually know the chemical composition of this cleaning product, but it's generally designed to remove rust, lime, and calcium. The label says it contains lactic acid, gluconic acid, dodecyl dimethylammonium and 1-butoxy-2-propanol (whatever those are), but it may have other ingredients not listed. It appears to be largely acid-based so I felt relatively safe trying it out.

 

[Sooth]

I've done a significant amount of searching, and I was really not expecting the mainspring to be such an issue to source. I re-measured the original and it is indeed 0.17 thick and not 0.19. Most of the commercial ones are all in the range of 0.22, and almost no suppliers carry any sizes wider than around 3mm. There are a few that do, but it jumps to 4.2 or larger and there is nothing around 3.8 to 4mm.

The closest I found was from Cousins: 4.20 x 0.17 x 560. It would be too wide, as well as too long, so I'd need to remake the ends (it doesn't say what kinds of ends are on the spring) and stone down the height.

I also worked on modifying a spare hairspring. I still have quite a bit of refining and tweaking to do on it. The remains of the original are on the right.

 

[gmorse]

Hi Sooth,

The closest I found was from Cousins: 4.20 x 0.17 x 560. It would be too wide, as well as too long, so I'd need to remake the ends (it doesn't say what kinds of ends are on the spring) and stone down the height.

That spring is listed as a 'non-automatic', so it will have a hole at the inner end and a small tab riveted into the outer end which bears against the hooking point in the barrel wall. That could be a proper hook or a recess machined into the wall. I'm just waiting for your picture of the barrel and broken spring to load, (tediously slow at the moment as others have reported); and finally, I can see that it's a hole end, so if the length is near enough to what David's calculator suggests you won't need to do anything apart from reducing the height. The tab will work with a proper hooking, or if you have to reduce the length significantly then you'll have to make a hole, as you would with a clock spring. You're lucky it isn't an English style hooking, with a rectangular steel hook riveted into the spring which engages with a hole in the barrel wall.

You say that the broken spring is 3.99 mm high, but have you carefully measured the actual inside height of the barrel, (not over the central boss)? The existing spring may be a 'close enough will do' replacement; watches this old will have probably had any number of replacement mainsprings. If you do have to reduce the height, it's easiest to do this with it still in its restraint ring.

Regarding your CLR cleaner, I can't see anything in it which would do too much damage, provided you keep the immersion time fairly short and don't heat it up. It's a mixture of weak acids, solvents, sequestering agents and wetting agents.

Regards,

Graham

 

[gmorse]

Hi Sooth,

I also worked on modifying a spare hairspring. I still have quite a bit of refining and tweaking to do on it. The remains of the original are on the right.

Forgot to address the balance spring. Is the thickness close to the broken one? This is the most critical dimension to match. When you're manipulating a spring, there are several techniques for making the correct curvature. I usually hold the coil I'm working on with a fairly stout pair of tweezers on a perspex sheet and just stroke the spring in the required direction with a thin piece of brass rod that's sharp at the end and mounted in a wooden handle. That way you're less likely to create sharp bends, holding both tools upright reduces the chances of forming a conical spring, and the perspex helps the brass point slide smoothly.

Next, you'll have the entertainment of vibrating the balance with the new spring, and once it's right, pinning it to the collet.

Regards,

Graham

 

[Sooth]

Graham:

The length of the existing spring is about 19.5" (495mm) and the estimator/calculator had given 455mm, so 560 for the stock spring will be a bit long. The existing spring has hole ends at both ends like a regular clock spring. I can easily anneal/punch/modify the end.

I can't easily measure the interior height of the barrel with the types of calipers that I have, but the old spring fits well within the barrel, and having just checked it again, it's basically 3.98mm in height, so we may as well say that it's 4mm. I have attached an image of the outer broken coil of the old spring fitted into the barrel. If you look on the right hand side of the image, you can just barely see the barrel cover lip, then a small clearance, and then the spring. I'd say there's maybe 0.5mm clearance?

Regarding the balance spring: "Is the thickness close to the broken one?" I have not really checked it. It's so infinitesimally small and delicate, I'm mostly going by eye. I would say the new spring is definitely not the same strength, it's a tiny bit wider. I am kind of hoping that I can play around with either the length or number of coils, but I admit I'm pretty much winging it here. I did see a rather good video series online for making hairsprings by hand and how to curl them and adjust them, so I can always start over. Failing that, I did not really see new hairsprings listed anywhere for sale as replacement parts, so I'm largely working with what I have on hand. The spare one I used is either out of an alarm clock or an old pocket watch. I did not really think the hairspring was that crucial for timekeeping, but I'm starting to think it will make a fairly big difference. I've been working on the assumption that I could work this out through trial and error.

"Next, you'll have the entertainment of vibrating the balance with the new spring, and once it's right, pinning it to the collet." I don't understand the first part of this sentence...

 

[gmorse]

Hi Sooth,

I did not really think the hairspring was that crucial for timekeeping, but I'm starting to think it will make a fairly big difference. I've been working on the assumption that I could work this out through trial and error.

That's an understatement! The balance wheel doesn't have a natural frequency of oscillation all by itself, but when the balance spring is added, it suddenly acquires one. That frequency depends on the inertial mass of the balance, most of which is in the rim, and the elasticity of the spring. (Early balance springs were known as 'pendulum springs', and the spring effectively replicates the action of gravity on a pendulum and allows the watch to function in more or less any orientation). Since you shouldn't alter the fabric of the watch more than absolutely necessary, the part being replaced (the spring) must be modified to fit the existing part (the balance wheel). This means that the frequency of the spring and balance together must be measured somehow if either is replaced, and that's what 'vibrating the balance' does. Although the process is iterative, it isn't exactly trial and error and there is a logical method of achieving this.

"Next, you'll have the entertainment of vibrating the balance with the new spring, and once it's right, pinning it to the collet." I don't understand the first part of this sentence...

It involves attaching the spring to the balance temporarily and either hanging the spring from its effective outer end, (where it passes between the regulator pins, not where it's pinned to the stud), in a gallows arrangement, or just holding it in tweezers, so that the wheel is hanging from the spring and the lower pivot is just touching the crystal of a watch with a centre seconds hand. If you then make the balance turn with a puffer or a soft brush, you can count the vibrations it makes in a minute. For initial tests a shorter period of say, 15 seconds will tell you how close you are. However, in order to find how close you are, you do need to know what the frequency should be to make the centre wheel rotate once in an hour, and for that you need to count the train, which I guess you would do with a clock in this situation.

If you believe that you have a spring that works, don't forget that you must leave enough extra length to reach round to the stud, and also that the diameter of the final spring must allow the outer end to pass between the regulator pins and into the stud without distorting the whole spring sideways.

Regards,

Graham

 

[Sooth]

Graham,

Thanks so much for all the help so far. Although your description helps, I am still a bit confused with how the vibrating of the balance is actually done. I've re-read the paragraph about a dozen times. From what I can make out, the whole assembly is hung upside down with the hairspring stretched out into a cone shape above (from the weight of the balance wheel), and the balance wheel is upside down just barely touching a glass or other suitable surface? If that's correct, I still don't quite understand how it can be made to spin forwards and backwards without one rim of the balance just hitting the surface underneath and fouling the action.

I'm starting to feel like I've bit off more than I can chew.

Would there be any hope in attempting to reuse the original hairspring instead? It's hard to tell but I am guessing that it broke near the centre, so there is hopefully not much length missing. Maybe 4-5mm? Minus a bit more if I re-pin it to the centre collet. It's hard to know how much "extra" there was when it was originally pinned to the backplate as it was completely loose around the balance when the watch arrived.

If this helps, the tooth counts are as follows:

Centre wheel: 60/10
Third wheel: 52/6
Contrate wheel: 48/6
Crown wheel: 13/6

 

[Chris Radek]

I think if you can set the balance on the crystal of a watch with the target beat rate, it helps a lot. Gripping the hairspring in a way that holds it horizontal will keep the balance level, it's easier than you think.

I like to see the hand motion of the "reference" watch and hear the tick both. A nudge of the balance or a little blast of air from your puffer will set it oscillating for plenty long enough to see if it's too fast or too slow. If they stay in sync for 10 seconds you are there.

They made very beautiful tools to help you accomplish this task, but how often will you do it really? It's easy to improvise with whatever you have handy like this:

 

[Sooth]

Chris,

A picture truly is worth 1000 words. I don't really have another functional watch, so I'm not sure what I could use to compare it to. I really am a beginner with watches, so I'm not sure what "rate" I'm even looking for. Would I just be aiming for one revolution back and forth per second? Or di I need to know how many beats per minute and check if I can count them for 15 seconds as was suggested? The only watch I could use is a modern quartz one, and it does have a centre seconds hand. Would that work?

 

[Chris Radek]

I think your train count gives 18026.7 bph (beats per hour) as the correct rate. (The center wheel turns once an hour, so the third wheel geared to it with 60/6 turns 10 times an hour, and you can do the rest and check my math, going on down the train. For the crown wheel, remember that each tooth ticks twice per revolution, once on each flag.)

So sure, if you use a quartz watch that ticks once a second, you can count 5 oscillations (2.5 back-and-forths) per tick. If you get exactly 5, that is 18000 bph, which is within about 10 minutes/day of the correct rate. Leave some extra length so you can adjust the pinning once it's in the watch.

If you aren't practiced at counting like this, your phone might have a slow motion camera feature that would help. (Get the balance and the sweep second hand in the video)

 

[Sooth]

My phone, hahahaha, nope! I am one of "those strange people" who doesn't have a cell phone. I can do video on my camera and replay it in slow motion on my PC so hopefully that will work.

 

[gmorse]

Hi Sooth,

Thanks to Chris for the picture, a practical alternative to the traditional, (and very expensive), vibrating tools.

You don't need to count every beat and you don't need a watch with exactly the same frequency. To make it easier you can count on every other beat, (in other words, only count on the clockwise rotations for example), and the quartz watch is just there to time the period in which you're counting and to provide a convenient surface for the pivot to rest on. By the way, your train count produces an unusual beats per hour figure if your figures are correct, (actually 18026.6 recurring), but it isn't unusual for these old watches to have non-integer beats, so don't worry too much about that.

I have sometimes done this without another watch at all, but using just a piece of glass to rest the pivot on, and using Audacity on my laptop to generate a click track at the appropriate frequency.

Would there be any hope in attempting to reuse the original hairspring instead? It's hard to tell but I am guessing that it broke near the centre, so there is hopefully not much length missing. Maybe 4-5mm? Minus a bit more if I re-pin it to the centre collet. It's hard to know how much "extra" there was when it was originally pinned to the backplate as it was completely loose around the balance when the watch arrived.

Losing that much balance spring will make quite a difference if there isn't any 'spare' at the outer end where it pins to the stud. The other consideration is that the old spring may well have some corrosion which could have contributed to the breakage in the first place, so when you come to make the sharper turn for the collet, some more may break off. When you're testing the new spring on the balance, you can attach it to the balance temporarily with a little piece of Rodico or similar.

When you're doing this sort of work, a micrometer is essential and need not be very expensive.

Regards,

Graham

 

[Sooth]

Small update: I'm starting to feel a bit defeated by this watch. I was not imagining that finding a replacement crystal would be difficult, but apparently there are no commercial replacements available (in glass or plastic) and there are little to no existing "old stock" available unless you get real, real lucky. I don't really imagine that I want to take up glass blowing just for this one specific need. What do collectors do in these cases? Are there really no companies that exist who can fabricate these? I've been looking for days and have come up with no useful leads.

The crystal needed would be 447 (44.7mm) and based on the height of the hand shaft, plus the approximate thickness of the glass, it would need to be about 9-10mm in overall height. I'm aware it's meant to be "Bull's Eye" or "High Dome" type.

It is starting to seem like a waste of time and effort to repair the rest of the watch if there's no possibility of sourcing a crystal.

 

[Chris Radek]

Bill White will make you one to whatever spec you like. His work is excellent.

 

[gmorse]

Hi Sooth,

I'm aware it's meant to be "Bull's Eye" or "High Dome" type.

Yes, a plain high dome would be correct but the bullseye is a later fashion from the 19th century.

Regards,

Graham

 

[Sooth]

Chris thanks so much for the lead. I will see if I can get in contact with Bill.

 

[Sooth]

More news! I never thought of checking Timesavers, as I didn't think they carried mainsprings this small:

No. 85031 .157 x .0067 x 22" (4.0 x 0.17 x 560mm) OL $5.75

These are loop end but that is a super easy fix. This also means I won't have to stone down the Cousins UK spring I was thinking of getting down to 4mm. They actually have multiple strengths in the 4mm size in more than one "end" type! They have hole end in 4mm but these are too thick, so the one above will work nicely as it's also a bit longer than I need. See their list here: https://timesavers.com/files/mainsprings.pdf

At the moment it looks like the only tricky bit will be the hairspring. I do have a micrometer that I bought maybe 10 years ago and have never used even once (everything I usually need to measure is easily done with my electronic caliper), so I suppose I will see if I can re-learn how to read the micrometer. I do have about 4 junk pocket watches and dollar watches (missing jewels, beat-up cases, damaged or missing screws, etc) which could also be potential hairspring donors.

I'm still a bit confused over the whole hairspring issue. I know that there are various types and strengths of them, but doesn't metal quality and tempering also make a difference? How could I know the proper number of coils to use if the springs were so variable in these early watches? Does the total length matter, or is it largely the "springiness" overall that is largely the deciding factor? My thought process here is let's say I could exactly replicate the same width and thickness of the original hairspring, but my metal is more of a stiff temper and the original is a softer steel, would these not give completely different amplitudes/rates to the balance? I feel like I'm getting lost in the complexities. For now I largely know the rough size, the approximate diameter and number or coils for it to engage in the rating assembly and the pin location for the end on the backplate.

Or am I going about this wrong. Should I instead fit whatever hairspring I want and vibrate the balance until I have the correct beat, and then open up or reshape the hairspring afterwards, assuming it's largely the length (rating assembly pinch point) that will play a factor and the shape can be changed?

I'll see if I can do some more research. The "good" part is that I suppose I can infinitely tinker with the hairspring until it's correct, or fit a wrong one and have the watch function but not keep time.

 

[gmorse]

Hi Sooth,

Is your micrometer imperial or metric? The latter are far easier to read.

I'm still a bit confused over the whole hairspring issue. I know that there are various types and strengths of them, but doesn't metal quality and tempering also make a difference? How could I know the proper number of coils to use if the springs were so variable in these early watches? Does the total length matter, or is it largely the "springiness" overall that is largely the deciding factor? My thought process here is let's say I could exactly replicate the same width and thickness of the original hairspring, but my metal is more of a stiff temper and the original is a softer steel, would these not give completely different amplitudes/rates to the balance? I feel like I'm getting lost in the complexities. For now I largely know the rough size, the approximate diameter and number or coils for it to engage in the rating assembly and the pin location for the end on the backplate.

The composition of the metal does make a difference, but the original spring would have been steel of some sort, and as long as your donor springs are also steel, as is most probable, it shouldn't make too much difference, and certainly less than the contribution of the three dimensions of the spring. It's possible that the old spring is untempered, but at this period I think most would have been tempered, and in any case, this may not be original at all. The strength of the spring and hence its influence on the rate of oscillation of the combined balance and spring is inversely proportional to its length, proportional to its width and proportional to the cube of its thickness. The regulator changes the effective length of the balance spring in order to alter the rate of the watch; shorter is faster, longer is slower, and the fewer the coils on the spring the larger the effect of moving the regulator by one division.

Or am I going about this wrong. Should I instead fit whatever hairspring I want and vibrate the balance until I have the correct beat, and then open up or reshape the hairspring afterwards, assuming it's largely the length (rating assembly pinch point) that will play a factor and the shape can be changed?

Yes, that's certainly one option to find a suitable spring which vibrates at around the right frequency as the original. You can open it out afterwards to give a similar number of coils to the original and only then reduce its length as necessary, remembering to leave some 'spare' to allow for the distance between the index pins and the stud, plus some extra for adjustment. If you find that you could produce a spring with the right frequency but significantly shorter than the old one, it will probably run with an inadequate amplitude; all three of the critical dimensions have to be close to the original.

Regards,

Graham

 

[Sooth]

I realize that I never posted any updated photos of the watch cleaned and reassembled. I still have not quite sorted the replacement hairspring situation yet, so the watch is shown without the balance installed. The watch does run fine. I have started to work on carving/filing the replacement hands, but realized that I will need some escapement files in order to file the minute hand hole.