French or Swiss Automaton repeat spring.

onewatchnut

NAWCC Fellow
NAWCC Member
May 8, 2004
50
6
8
Chesapeake, VA
Country
Region
I have a verge fusee automaton quarter hour repeater with a broken repeat spring. It appears to be either French or Swiss. The repeat mainspring is broken. It does not appear to be original. Apparently the original spring was replaced with part of a small pocket watch mainspring. it had to have been shortened to accommodate the size of the barrel.

I need to know how long the spring needs to be. The barrel can be adjusted to preload the spring. The wheel attached to the arbor makes less than one revolution for a 12 O'clock strike, so I would not think the spring needs to be calculated for a normal going barrel length.

i have attached pictures of the arbor and repeat wheel, and a picture of the barrel installed in the plate. you can see the click for the preload adjustment on the outside of the barrel.

The repeat regulator is a train of wheels terminating in a tilting pinion. Someone also adjusted the tilting pinion enough to break off the pivot of the associated wheel creating another problem. I want to get the repeat working before I tackle the regulator.

The arbor diameter is 2.48 mm and the barrel is 8.28 mm. The calulator calls for a 158 mm spring for a going barrel. Can anyone offer the approximate length of mainspring to try?

Thanks in advance,

Marty

Without barrel_A.jpg With barrel_A.jpg
 

onewatchnut

NAWCC Fellow
NAWCC Member
May 8, 2004
50
6
8
Chesapeake, VA
Country
Region
Just a WAG but I suggest three time the outer diameter so you get about 3 coils.
Thanks. It appears to be a good guess.

Digging through Richard Watkins treatise "The Repeater", I found the repeater spring is three to four coils. Of course, the thickness of the spring is not mentioned.

As I recall, shortening a spring makes it stronger but I don't know if the function is linear. I might start with a handful of springs with different thickness and cut them to the same coiled length. I should obtain some useful information from the experiment.

After all, it's only time and money.
 

gmorse

NAWCC Member
Jan 7, 2011
14,569
3,690
113
Breamore, Hampshire, UK
Country
Region
Hi onewatchnut,
As I recall, shortening a spring makes it stronger but I don't know if the function is linear. I might start with a handful of springs with different thickness and cut them to the same coiled length. I should obtain some useful information from the experiment.
The strength of a spring is inversely proportional to its length, proportional to its width and proportional to the cube of its thickness. To save yourself a lot of time and expense, have a look at David Boettcher's web page on the subject, which includes some useful calculators and will allow you to plug in various values to see what effect they have.

Regards,

Graham
 

onewatchnut

NAWCC Fellow
NAWCC Member
May 8, 2004
50
6
8
Chesapeake, VA
Country
Region
Hi onewatchnut,


The strength of a spring is inversely proportional to its length, proportional to its width and proportional to the cube of its thickness. To save yourself a lot of time and expense, have a look at David Boettcher's web page on the subject, which includes some useful calculators and will allow you to plug in various values to see what effect they have.

Regards,

Graham
Thank you Graham. I took a rapid scan of the referenced site and there appears to be a wealth of information there. I will take an in depth look tomorrow.
 

Chris Radek

NAWCC Member
Apr 13, 2014
1,216
659
113
Lincoln, NE, USA
timeguy.com
Country
Just a WAG but I suggest three time the outer diameter so you get about 3 coils.
I agree, but three coils will be more like 9x the diameter, so I'd try 70 or 80mm length.

I don't think you'll be able to narrow down the right strength (thickness) until the regulator pinion is fixed. I feel like in this application I have used a fairly weak spring, like something meant for a 10 ligne manual wind watch.
 

onewatchnut

NAWCC Fellow
NAWCC Member
May 8, 2004
50
6
8
Chesapeake, VA
Country
Region
I agree the regulator has to function to determine the final strength of the spring. My main focus is getting the repeat to run so I can see if there are latent problems that are not apparent by examination. As long as it runs through the process at a reasonable speed, I can move on to the train wheel repair.

If there was a pointer on the regulator arbor, it is long gone. That missing clue probably led to someone thinking it is a winding arbor. It only takes a turn, or less, to damage the train wheel.

Thanks for the help.

Marty
 

dshumans

NAWCC Member
Sep 17, 2009
477
112
43
Marty,
You may already know, but it is very likely possible to put a new pinion on the original repeater train wheel. I have done that numerous times. There are many books and articles on repivoting that are helpful. It's just: soften the metal of the pivot, drill a hole, make a plug for the hole, insert it and cut the plug down to the pivot size.
Doug
 

onewatchnut

NAWCC Fellow
NAWCC Member
May 8, 2004
50
6
8
Chesapeake, VA
Country
Region
Marty,
You may already know, but it is very likely possible to put a new pinion on the original repeater train wheel. I have done that numerous times. There are many books and articles on repivoting that are helpful. It's just: soften the metal of the pivot, drill a hole, make a plug for the hole, insert it and cut the plug down to the pivot size.
Doug
Thanks Doug,
I use carbide drills to drill the hole in the arbor so I don't have to soften the metal. They are a little pricey, but they save a lot of time. Unfortunately, I'm not quite there yet.

i have the hour strike working, a bit fast without the regulating train, using a wristwatch mainspring. I have not cut the length yet.

In the attached picture, you can see the gathering pallet (top center) is hook shaped. It is preventing the spring from unwinding completely and allowing slack in the chain. I have no idea if the pallet is original to the watch, or if someone bent it into that shape. I do not work on a lot of repeaters, but I don't remember seeing a hooked pallet. All the ones I remember have been straight. I have not cased the watch so I don't know if the slack chain will allow the stem to go far enough to strike twelve hours.

Is it worth the effort to make another gathering pallet so everything is properly at rest?

Thanks,
Marty
Pallet_A.jpg
 

gmorse

NAWCC Member
Jan 7, 2011
14,569
3,690
113
Breamore, Hampshire, UK
Country
Region
Hi Marty,
I have no idea if the pallet is original to the watch, or if someone bent it into that shape. I do not work on a lot of repeaters, but I don't remember seeing a hooked pallet. All the ones I remember have been straight.
They often are this shape, but if it's been put back on its square in the wrong orientation you'll have this situation. These don't bend! There has to be a little pre-tension as a rule.

Regards,

Graham
 

onewatchnut

NAWCC Fellow
NAWCC Member
May 8, 2004
50
6
8
Chesapeake, VA
Country
Region
They often are this shape, but if it's been put back on its square in the wrong orientation you'll have this situation.
Hi Graham,

Unfortunately, the current configuration is the only one that works. In the picture attached to my post labeled without barrel_a, you can see the hour rack. it is approximately located in the correct position. No other position of the gathering pallet will run the racks correctly.

My understanding is, with the system at rest, the gathering pallet should be in contact with the quarter rack when it is locked by the all-or-nothing piece. There should be slight pressure on the pallet. Otherwise, the pallet would not travel far enough to lock the rack. In that position, there should be no slack in the chain. When the stem is pressed far enough to unlock the rack, the repeat spring is wound by the chain and the rack is free to travel to the snail. Releasing the stem allows the spring to unwind and the hour rack strikes the hours followed by the gathering pallet moving the rack to strike the quarters until the rack stops and locks again.

This is the simple description as I know it. i can easily be wrong since I do not work on many repeaters. But if I am correct, the problem is that the gathering pallet has traveled too far as a result of the hooked shape. Looking at the arbor straight on, it appears a straight pallet, or a hooked one with the square hole rotated a bit anti-clockwise, is called for. Or, the chain is too long. I am reluctant to leave the watch with the slack chain.

i am also open to all suggestions. Many of you out there are far more experienced than I and might have a simple fix. I always learn something I didn't know when I tackle a complicated watch. There is no way to determine how many people have worked on this beauty before; and what their skill level was.So, one has to be a good sleuth as well as a good watchmaker.

Regards,
Marty
 

onewatchnut

NAWCC Fellow
NAWCC Member
May 8, 2004
50
6
8
Chesapeake, VA
Country
Region
Hi Marty,

It's not unknown for these chains to stretch . . .

Regards,

Graham
Hi Graham,

As I was typing the words "Or the chain is too long." the light bulb went off in my head. It is the one thing I did not examine.

Well the chain is out and on the bench. there is evidence of a poor repair, but not enough to indicate add links. My real question is in regard to the attachment to the arbor pulley. Under close examination, the chain is attached by a pin through the holes of the outside chain plates. It appears the center link was removed and the chain pinned through the two plates. Is this the correct way to mount the chain? One could shorten a half link length by pinning the link vice the plates.

Just a thought.

Regards,
Marty
 

gmorse

NAWCC Member
Jan 7, 2011
14,569
3,690
113
Breamore, Hampshire, UK
Country
Region
Hi Marty,

I would expect that there would be a ring link at each end of the chain. Your remark about it having signs of a poor repair is the key to this I think.

Regards,

Graham
 

dshumans

NAWCC Member
Sep 17, 2009
477
112
43
I think you're on to the answer. It's likely the chain. I have worked on a good number of these and they have al had hooked shape gathering pallets. I believe it may need the hook shape to clear the rack tooth just before it reaches the rack and to advance the rack correctly. It looks like if the chain were shorter, it might work OK. If there is enough chain around the hour strike pulley to draw out all 12 hours, then the chain should be shortened at the pendant end. If there is not enough to draw 12 hours, the chain needs more wrap on the hour strike pulley. Either way, the pulley would then draw the chain tight just as the all-or-nothing stop is hit and the pendant is at full height.
 

onewatchnut

NAWCC Fellow
NAWCC Member
May 8, 2004
50
6
8
Chesapeake, VA
Country
Region
I think you're on to the answer. It's likely the chain. I have worked on a good number of these and they have al had hooked shape gathering pallets. I believe it may need the hook shape to clear the rack tooth just before it reaches the rack and to advance the rack correctly. It looks like if the chain were shorter, it might work OK. If there is enough chain around the hour strike pulley to draw out all 12 hours, then the chain should be shortened at the pendant end. If there is not enough to draw 12 hours, the chain needs more wrap on the hour strike pulley. Either way, the pulley would then draw the chain tight just as the all-or-nothing stop is hit and the pendant is at full height.
Thanks Doug,
It is definetly the chain. I ran through the sequence slowly and the hook shape gathering pallet is correct. It clears the rack and gathers it up correctly regardless of the quarter snail position. The half baked attachment to the strike pulley is part of the problem. it needs to be re-done. It will complete the twelve hour strike, but doesn't leave a lot of over-run. I will work on that end of the chain and see where it goes.
 

dshumans

NAWCC Member
Sep 17, 2009
477
112
43
The over-run on the strike pulley is determined by the stop for the all-or-nothing piece. If the rack end engages the all-or-nothing piece, that's all the over-run it will ever do. Once that works and you have fixed the attachment to the strike pulley, just shorten the chain until it pushes the pendant all the way up and stops with a little tension.
 
  • Like
Reactions: gmorse

Forum statistics

Threads
177,715
Messages
1,557,623
Members
53,661
Latest member
sara.fagan73
Encyclopedia Pages
909
Total wiki contributions
3,061
Last edit
E. Howard & Co. by Clint Geller