Frodsham 84 Strand -- modified movement?

[jplotkin]

Hi All,

Here is a lovely Frodsham movement now housed in a custom-made display case -- an aluminum case held between two glass planes. The movement is key-wind and key-set, and it was described as fusee (although I haven't de-cased to verify this). Movement dates to ~1855.

But I am unfamiliar with this movement design -- has it been modified by the watchmaker who made the custom case?

The pillar plate has some of the ratchet wheels and the click on the back -- is that normal for a fusee movement from this era?

The section where the key winds shows an annulus that has been ground into the pillar plate, removing the gilding and cutting through the "Cha Frodsham" engraving as well as the movement serial number, apparently. This also seems to be a modern modification -- is that correct?

I'd love to figure out if/how/why the original movement was modified (for casing?)

Thank you.

 

[gmorse]

Hi jplotkin,

But I am unfamiliar with this movement design -- has it been modified by the watchmaker who made the custom case?

The pillar plate has some of the ratchet wheels and the click on the back -- is that normal for a fusee movement from this era?

The section where the key winds shows an annulus that has been ground into the pillar plate, removing the gilding and cutting through the "Cha Frodsham" engraving as well as the movement serial number, apparently. This also seems to be a modern modification -- is that correct?

I'd love to figure out if/how/why the original movement was modified (for casing?)

I notice in your third picture that the key appears to be inserted into the bottom of the metal block; is this the winding position, or else why is there a hole there? Because it seems to be in the wrong position to connect with any of the keyless work, unless there's some clever work going on inside the block.

I think this may be a key-wound movement that's been converted to keyless at some point and then converted back again so that it can be set in that metal block. The keyless work, including the two idler wheels, one of which is also the ratchet wheel with its click, would almost always be on the pillar plate under the dial, not on the top plate. There certainly appears to be a barrel arbor just to the left of the click which suggests a fusee.

Some more pictures when you remove it from the block would be interesting.

Regards,

Graham

 

[howtorepairpendulumclocks]

You might want to contact Frodsham's in England. They are still in business, very friendly and knowledgeable.

 

[howtorepairpendulumclocks]

... presume it was in precious metal case that was scrapped. Many were, even in my time.

 

[jplotkin]

Thanks, Graham.

The third picture does not include the key at all. What you see is a brass cylinder protruding out of the bottom of the aluminum square case. This brass cylinder is fixed to the aluminum case and it keeps the case steady when placed into a corresponding hole in the wooden stand. This brass cylinder is not related to winding or setting or use of the key. The brass cylinder is marked F. L. -- presumably the modern watchmaker who modified the movement and made the custom case.

The key is double-sided: one side for setting time from the dial-side, and one for winding from the top plate. The key winds by attaching directly to the arbor that protrudes out of the top plate (see annotated photograph). I suspect this is the fusee arbor, not the barrel arbor (right?).

I will post more photos when I de-case it. But I have confirmed it surely is fusee -- I see a small bit of chain, presumably around the barrel, near the side of the movement, located at the "LON" in LONDON.

But isn't it already clear that the movement has been heavily modified, because the top plate has wheels and a click on it, and the wheels are recessed in grooves that removed the gilding. So this must have been a modification to the original movement was made. Right?

Your theory that it was key-wound, then converted to keyless, then again converted back is interesting.

 

[howtorepairpendulumclocks]

Has it been converted from fusee to going barrel? The winding ratchet for fusee is between the fusee cone and the great wheel, not on the back plate? Whoever has done the work has done a lovely job.

 

[Dr. Jon]

This is a lovely movement and it is freesprung!

It appears to be still a fusee. The mainspring arbor is visible to left of the click on the winding wheels. I suspect the key wind arbor is where it was before the attempted conversion.

If converted, the case would have required the addition of a pendant, The origibal case could ahve been modifed and later sold for gold or the original case recycled for e keywind movement. Isuspect this was an experimant Frodshan did and they may not have cased the movement.

I do not understand why the two gears also have a ratchet, unless they modified the fusee assembly by adding a gear to engage the new ones and moved the ratchet there.

Question: When you wind with the key, do the gears on top turn?

 

[jplotkin]

Ok...here's a critical clue: the click and winding wheels on the top plate are apparently non-functional and purely decorative!

When I wind it using the key on the arbor protruding through top plate, the exposed keyless wheels do not advance, and the ratchet does not click. So maybe they are just hold-overs from an earlier setup as keyless (as Graham guessed), and they have been placed there for purely decorative purposes?

Perhaps this was an experiment done by Frodsham himself, or maybe by the modern watchmaker who did the conversion into this aluminum case?

Unfortunately, the only serial number we have is from the dial (which may not have been the original one on this movement) -- because the serial engraving on the movement is either absent or it was removed during the placement of the (non-functional) winding wheels.

 

[howtorepairpendulumclocks]

The fusee arbor has a square which presumably held a wheel that engaged with the two (now obsolete) wheels. The fusee and this system are inherently incompatible as the fusee cone turns in both directions... it couldn't work with that click there.

 

[gmorse]

Hi jplotkin,

But isn't it already clear that the movement has been heavily modified, because the top plate has wheels and a click on it, and the wheels are recessed in grooves that removed the gilding. So this must have been a modification to the original movement was made. Right?

Yes, the movement would not have had any of those wheels fitted originally, and I very much doubt that Frodshams had anything to do with their fitting, because if they had, for one thing the newly cut recesses for the wheels would have been gilt to match the rest of the top plate, and for another, I can't see them disfiguring one of their good watches in this way. Whoever did the work to fit the movement in this display case probably left the redundant wheels in place to reduce the ugly appearance of those recesses.

Regards,

Graham

 

[gmorse]

Hi jplotkin,

If you look at the redundant wheel which would have engaged with the (now missing) wheel on the fusee square, you can see that its recess isn't quite circular, which suggests that it could move in and out of engagement, like a rocking bar system.

Regards,

Graham

 

[howtorepairpendulumclocks]

That's a good point and would make sense. It would be cool to fully disassemble for a "forensic" look at all the parts and as above, ping Frodsham's a line, they just might have a view.

 

[jplotkin]

Hi jplotkin,

If you look at the redundant wheel which would have engaged with the (now missing) wheel on the fusee square, you can see that its recess isn't quite circular, which suggests that it could move in and out of engagement, like a rocking bar system.

Regards,

Graham

Yes, I noticed that too. But there does not seem to be enough a room in the recess to fit a full rocking bar. It's mysterious.

I am loath to disassemble for forensic purposes, despite my curiosity about the history. I think the movement is rather lovely, despite the modifications, and I like the custom case.

One sad note: the movement arrived by USPS in a completely water-logged package that had been damaged during shipping. And the movement was wet when I opened it. So I'm not sure it will operate properly. This might argue for disassembly and cleaning, but fusee's are beyond my own ability...

 

[gmorse]

Hi ,

And the movement was wet when I opened it. So I'm not sure it will operate properly. This might argue for disassembly and cleaning, but fusee's are beyond my own ability...

Well, if you dried it out thoroughly with some warm air as soon as possible, preferably out of the case, you might be lucky, but I don't recommend running it at all until it's been checked over.

Regards,

Graham

 

[SKennedy]

FIrst, as I don't think anyone else has pointed it out, this appears to have a reverse fusee since the fusee and barrel are back to front compared to their normal positions in a movement. The screw by the m of Frodsham is probably holding the reverse fusee stopwork arm in place.

I think it was converted to an early form of keyless winding though we can't see how the winding is transmitted to the visible wheels. There would have been a wheel sitting in that groove that was attached to the fusee square. To engage winding you probably had to push a pin in the side of the original case which would have shifted the wheel with the ratchet across to engage with the (now missing) fusee wheel. The reason for this would have been so there was disengagement between the fusee wheel and the winding system, otherwise the winding wheels would be driven backwards as the watch ran.

 

[jplotkin]

Thanks, Seth. Keen eye to note the screw by "m" and infer its likely purpose.

I will try to dry/clean the movement. Although it's far from a pristine movement, I enjoy the history of back-and-forth modifications that it bears out, and so well worthwhile preserving.

 

[SKennedy]

A further thought. Since the chain will run from the outside of the barrel towards the centre wheel side of the fusee then there'd be nothing in the way directly below that upper winding wheel. So it might be fitted on an arbor that runs through to the dial side of the pillar plate and the other end has a larger winding wheel attached to it which meshes with a winding stem bevel/crown wheel. It would be interesting to find out!

Edit: I think I've just figured out what that click is doing. It would prevent the winding from operating in the wrong direction as it would lock the wheels if the winding crown was turned backwards.

Edit #2: It is quite possible that the movement was made by Kullberg as they were the most keen on reverse fusee.

 

[John Matthews]

I suspect the escapement may be an inverted double roller. A number of examples of keywound, freesprung and with reverse fusee, signed Frodsham are known - I posted one here. Also, in a presentation case, but not modified.

John

 

[Dr. Jon]

The Kullberg Frodshams are listed in Mercer's book on the Frodshams. This may be another line since it is signed "Arnold Frodsham".

 

[John Matthews]

The Kullberg Frodshams are listed in Mercer's book on the Frodshams

This list is for later watches, these have serial numbers that commence with '0'. The earliest is #02033 and is dated 1862. The watches that have Arnold & Chas Frodsham on the dial & Chas Frodsham 84 Strand on the back plate, as here, do not have the '0' and are dated 1854 for #2838 and 1860 for #8484. The date range is not sequential, the date range commencing from 1845 (#6445 * #6551). The sequence shows that caution needs to be exercised when trying to determine date of 'finishing' from serial numbers alone. The date given in the first post ~1855 appears to me to be very reasonable.

Charles Frodsham purchased 84 Strand, and Arnold's business shortly after the latter's death in 1843, which is entirely consistent with this sequence of serial numbers and the dates provided by their hallmarked cases.

The reversed fusee was used by Mudge in the 1770s. It had also been used, separately, by Arnold & Barraud. It is believed that Frodsham followed their lead and introduced it as part of his new caliber in 1850. Regarding Kullberg and the reversed fusee, Mercer's says ...

​

"It is interesting to note that Kullberg came to London for the first time in 1851, when he wished to see the Crystal Palace Exhibition and it would appear that Kullberg adopted the reversed fusee (which is a characteristic of his movements), having one assumes, considering Frodsham's principles to be excellent."​

​

John

 

[SKennedy]

Thank you John. Its always mystified me why more makers, particularly at the better quality end of things, didn't adopt reverse fusees.

Here's a couple of photos of slightly later (early 1860s in this instance) Nicole keyless work. It is not quite the same set up as seems to be on this Frodsham but I reckon there are similarities. The exisiting layout of the Frodsham presumably required something different making.

The main winding wheel is attached directly to the barrel arbor (this is a going barrel watch) and a click is incorporated. The intermediate wheel for handsetting is fitted to the third of the levers and brought into action when the case pin set button is pushed. The spring and shape of the lever at the top left causes a bit of a click-detent feel to this action.

 

[SKennedy]

Edit: I think I've just figured out what that click is doing. It would prevent the winding from operating in the wrong direction as it would lock the wheels if the winding crown was turned backwards.

I was thinking about this while laying awake last night, as you do. I realised that if the fusee is wound in an anticlockwise direction as is normal then that wheel with the click would have to turn clockwise when winding which is the direction the click would prevent. So, back to square one. Not sure what the click is for. Perhaps its to do with the engaging/disengaging movement. Trying to rotate that wheel clockwise from its rest position would cause it to 'climb' the click and pivot on the rocking bar towards the fusee but during the act of winding it would have be clear of the click.

 

[howtorepairpendulumclocks]

Just chipping' in... if it clears things up... 'reverse' or 'reversed' fuse doesn't mean the fusee is wound in the normal direction of winding. The fusee can be designed to wind either way depending where it lies in the train. It is wound one way and runs or unwinds the opposite way. The 'reverse' or 'reversed' relates too the path of the chain or line. Most clocks have a reversed fusee where the line or chain passes between the fusee cone and the barrel forming an 'S' shape. It is to do with load on the bearings. Hope this clears that element up if it was misleading.

 

[John Matthews]

This may help - from Mercer's Frodsham ...



John

 

[jplotkin]

Hi All,

I have de-cased the Frodsham and come bearing photos (and some videos). I was trying to determine two things

(1) Is the fusee normal, reversed, or normal-diagonal (thanks, John Matthews)
(2) Is the original keyless works Nicole-patent type

I was too timid to separate the plates, which would allow full diagnosis; but hopefully these photos will resolve some mysteries. (Note that there is a case screw still attached in these photos, so don't confuse that with any functional part.)

The keyless works do not seem to have the typical layout of Nicole, in part because there is no time-setting wheel on top of the center of movement, and it lacks the angled winding pinion (45 degree angled from the plane of the plates) that I often see in Nicole keyless. Thoughts?

You can see the click for the fusee, and I have photos of the fusee and barrel when nearly completely wound. Can you determine the orientation of the fusee chain relative to the barrel from these photos? Or the orientation of the catch on the barrel?

 

[gmorse]

Hi jplotkin,

Is the fusee normal, reversed, or normal-diagonal (thanks, John Matthews)

You can see the click for the fusee, and I have photos of the fusee and barrel when nearly completely wound. Can you determine the orientation of the fusee chain relative to the barrel from these photos? Or the orientation of the catch on the barrel?

Yes, it's certainly a reversed fusee, as per the middle drawing in the page from Mercer which John posted earlier.

The keyless work looks similar to the later Nicole design, and the shape of the top plate also suggests Nicole. The earlier design used a steel bridge on the top plate to support the transmission wheel.

PS, can't see any videos.

Regards,

Graham

 

[howtorepairpendulumclocks]

just to chip in for clarity (sorry to be an old pedant), the click pivoted between the plates you can se is the maintaining power click. The normal winding click would be hidden between the back of the fuse cone and the great wheel.

 

[John Matthews]

Thanks for taking the plunge and exposing the movement

If you have a photograph of the balance staff showing the roller and a close-up of the lever, could you post them please?

Thanks

John

 

[gmorse]

Hi howtorepairpendulumclocks,

The normal winding click would be hidden between the back of the fuse cone and the great wheel.

Yes, there are normally two clicks attached in the thickness of the steel maintaining power ratchet wheel.

Regards,

Graham

 

[jplotkin]

Hi John:

I have lots of photos while it was exposed, but the movement is already re-cased now.

Here is the best I have of the balance staff and roller (with serial indication on bottom of cock).




Also, here is one still image drawn from a video I made of the lever in action. I'm not sure if this is the part of the lever you were interested in, though:


I did not entirely disassemble and clean this movement, because I would need practice on a cheap fusee.

However, I noticed that the diamond end-stone on the balance has not been cleaned, and there does not appear to be any oil under the ratchet cap stone (although there was oil on third, fourth, center wheels) -- and I'm tempted to address these two issues. Thoughts?

 

[gmorse]

Hi jplotkin,

There should be oil under all endstones, especially the balance and escape wheel. However, removing the endstone settings isn't always easy from the top; the diamond on the balance cock isn't that difficult, (and that is certainly dirty), but the one on the escape wheel is recessed into the plate, may be a close fit, and you shouldn't try to prise it out. If it has to come out it should be pressed out from the back together with the hole jewel underneath. That would only need the small cock to be removed for the top jewel, but the dial would need removing to get at the one in the pillar plate.

The balance has a typical single table roller of the period, with a cylindrical impulse jewel. There are timing washers under some of the temperature and quarter screws, possibly to slow the balance after a replacement balance spring. On the plus side, it doesn't appear to be in compensation for any earlier lightening of the screws, which is not reversible. Balances don't suddenly become lighter all by themselves!

Regards,

Graham

 

[jplotkin]

Thanks, Graham.

I have always only removed cap jewels by pushing from below -- which requires removing cocks, etc. But I'm sorta hesitant, in this case, because the movement is working pretty well already, and it is nicely cased.

It is really possible/advisable to remove (and clean) the diamond capstone without removing the balance cock and wheel -- simply by unscrewing the screws that hold the diamond capstone while the balance cock is in the movement!? (That would avoid lots of risk associated with removing the hairspring stud on this otherwise nicely working movement; but that approach seems to risk excessive downward pressure on the staff...)

 

[gmorse]

Hi jplotkin,

It is really possible/advisable to remove (and clean) the diamond capstone without removing the balance cock and wheel -- simply by unscrewing the screws that hold the diamond capstone while the balance cock is in the movement!? (That would avoid lots of risk associated with removing the hairspring stud on this otherwise nicely working movement; but that approach seems to risk excessive downward pressure on the staff...)

Yes, if you have a delicate touch with the screwdriver, because those screws are tiny and shouldn't be in there too tight. However, if the endstone is dirty it's very likely that the hole jewel is as well, but since you've already removed the balance you could work on the balance cock. Doing the same on the potence is another matter altogether though.

Regards,

Graham

 

[John Matthews]

Thank you for responding with the additional photographs.

I'm not sure if this is the part of the lever you were interested in, though:

I was interested to see a clearer close up of ...



John

 

[gmorse]

Hi John,

I was interested to see a clearer close up of ...

That fork does seem rather too wide for the impulse jewel, but that may just be a matter of scale. If it is as wide as it appears, the action of the escapement will be compromised, and impulse will be lost.

Regards,

Graham

 

[jplotkin]

Unfortunately, John, I don't have any close-ups of the fork (and the balance clock is already re-attached now). Except I have this still image from a video, which isn't really all that close.

I note that the fork has a guard pin sticking up vertically, which would make sense for a single-roller safety function. But I don't see any notch in the single-roller table that would operate with this guard pin-- so perhaps this guard pin is non-functional, and the entire lever is non-original and non-matching for this impulse jewel?

 

[John Matthews]

I believe it is a single roller and I think I can see a very very shallow passing arc, when I enhance one of your photographs.



but the roller and fork combination seems to me to be a very 'tightly' engineered

It will be interesting to hear what Graham thinks.

John

 

[gmorse]

Hi jplotkin,

But I don't see any notch in the single-roller table that would operate with this guard pin-- so perhaps this guard pin is non-functional, and the entire lever is non-original and non-matching for this impulse jewel?

I agree with John that the passing crescent is certainly there, although it's very shallow and easily missed. In some escapements it isn't a crescent, as here, but a flat. If it matches the length of the lever and the rest of the geometry, and performs its function, that would be consistent.

The apparent discrepancy between the diameter of the impulse jewel and the width of the lever fork, (if it does in fact exist), could be accounted for by one or other of the components being replacements. The impulse jewel seems to be a good fit in the roller, so the whole roller could have been replaced, or the lever could be a replacement; either could have been due to some accident or damage. This seems possible, especially in the light of the other modifications which this movement has undergone.

Is it possible for you to make a short audio recording of the watch running?

Regards,

Graham

 

[jplotkin]

Graham,

I've attached a 21-second recording. I saved the recording in several different formats (mp4, mp4-audio only, mp3), with all the files compressed into single .zip file. Hopefully one of these audio formats will be readable for you. Any analysis of the beat would be terrific.

Note: in reality the watch is losing 300 seconds/day, but I cannot easily regulate the free-sprung movement. Also my cheap time-grapher iphone app does not accurately measure the rate, even though it works well for almost all other movement I own. And so maybe the fork/beat really is funky.

Also, maybe relevant: I neglected to check whether or not the impulse jewel is loose, but I don't think it is.

Very interested to see what you infer from the audio!

 

[jplotkin]

Update: here are some good close-ups, with the balance cock in place, John.

There is indeed a small passing in the roller. And you can get an idea for the size of the impulse jewel relative to the fork. It doesn't look like the fork is too wide, necessarily -- but I'm no expert.

Also, note that the guard pin seem to be touching the roller plate in the photos. I can't imagine that this is a good thing. But perhaps this is just because the lever moved after I temporarily stopped the escape wheel, to take the photograph...

 

[roughbarked]

Has it been converted from fusee to going barrel? The winding ratchet for fusee is between the fusee cone and the great wheel, not on the back plate? Whoever has done the work has done a lovely job.

Agree. It is nice work.

 

[gmorse]

Hi jplotkin,

Also, note that the guard pin seem to be touching the roller plate in the photos. I can't imagine that this is a good thing. But perhaps this is just because the lever moved after I temporarily stopped the escape wheel, to take the photograph...

The guard pin is doing its job; with the movement stopped it's highly likely to be contacting the roller. When running, the lever should be held against the banking when it isn't in the passing crescent during unlocking and impulse, this is what draw on the pallets is all about.

I'm investigating your audio files at the moment, but I still think from your pictures that the fork may be too wide. Are you able to use a slightly lower magnification? This would have the benefit of more depth of field.

Regards,

Graham

 

[gmorse]

Hi jplotkin,

I'm investigating your audio files at the moment, but I still think from your pictures that the fork may be too wide.

I've looked at the waveforms and run a noise reduced version through my timing package, (Delph Electronics eTimer), which shows that the correct rate appears to be 16,800bph, whereas it's actually fluctuating around 16,720bph. The waveform also shows the gap between unlock and impulse; whilst there's always some delay between these two peaks, it appears in this case to be larger than normal, and this is also suggested by the escape tooth trace at bottom right which has alternate teeth unevenly spaced. This does suggest that the lever fork is wider than it should be, causing the delay between unlock and impulse as the fork flips from one side of the pin to the other. The double trace on the paper tape emulation at the top centre shows a beat error, although I'm not sure why it isn't displaying this at top left where it normally does.

This zip file contains some modified .WAV files, screen captures of the waveforms and an image of the timer screen.

Regards,

Graham

 

[John Matthews]

When running, the lever should be held against the banking when it isn't in the passing crescent during unlocking and impulse, this is what draw on the pallets is all about.

Graham - does this mean that in the case of the early single rollers, that were without draw, the guard pin would have been in contact with the roller, as without draw, the lever would not have been held against the banking?

John

 

[gmorse]

Hi John,

...does this mean that in the case of the early single rollers, that were without draw, the guard pin would have been in contact with the roller, as without draw, the lever would not have been held against the banking?

Yes, it could well be, because without draw to hold it in contact with the banking, there was nothing to keep it positively away from the roller. Once this was appreciated, the slight recoil inherent in draw was accepted as a necessary price for the increased consistency of action, and draw became universally applied in levers.

Regards,

Graham

 

[gmorse]

Hi jplotkin,

In view of the fact that several of the balance screws have been fitted with timing washers to slow the balance down, and that on current evidence the lever fork doesn't appear to fit the impulse pin, I think that any adjustments should be postponed until these factors have been clarified.

Regards,

Graham

 

[SKennedy]

A wide shallow fork like that might have been working with a triangular roller jewel. If that were broken then, given the unlikelyhood of finding a spare, I can imagine a later repairer replacing the roller with a more common round jewel single roller. A more modern D shape jewel would actually work better since it could be a larger diameter to fit the fork better with the flat providing the clearance as it enters.

 

[jplotkin]

Graham - does this mean that in the case of the early single rollers, that were without draw, the guard pin would have been in contact with the roller, as without draw, the lever would not have been held against the banking?

John

I have some slow-motion video of the movement in action, from the same angle as photographs above. It appears that the guard pin does not actually contact the roller during operation. (It was only in contact during the still photograph because I stopped the movement by holding escape wheel). I will try to post the video, which is interesting.

I will also try to get photos with more depth of field, although it's an awkward angle and and requires shooting through the hairspring at the moment..

 

[jplotkin]

Hi jplotkin,



I've looked at the waveforms and run a noise reduced version through my timing package, (Delph Electronics eTimer), which shows that the correct rate appears to be 16,800bph, whereas it's actually fluctuating around 16,720bph. The waveform also shows the gap between unlock and impulse; whilst there's always some delay between these two peaks, it appears in this case to be larger than normal, and this is also suggested by the escape tooth trace at bottom right which has alternate teeth unevenly spaced. This does suggest that the lever fork is wider than it should be, causing the delay between unlock and impulse as the fork flips from one side of the pin to the other. The double trace on the paper tape emulation at the top centre shows a beat error, although I'm not sure why it isn't displaying this at top left where it normally does.

This zip file contains some modified .WAV files, screen captures of the waveforms and an image of the timer screen.

Regards,

Graham

Wow, thank you so much, Graham.

What is a good rule of thumb a normal gap size between unlock and impulse, relative to the rest of the wave-form?

 

[jplotkin]

Here is the slow-motion video in zipped .mov format; seems to show guard pin away from roller. Also, the fork is quite still at all times except for during impulse.