The Breitling Watch Source Forums

I've noticed that my Cosmonaute, which isn't a chronometer actually keeps better time than either my SOSF or Blackbird. All three are within specs, but the Cossie is about 15 seconds fast per week whereas the others are about 25- 30 seconds fast per week. I noticed this on my vintage Breitlings as well. My Venus 178 Chronomat was more accurate than my Cal 12 TransOcean. Are manuals generally more accurate than autos? if so, why?

Accuracy of x vs. y is not really anything to do with function, complication, manufacturer, etc, it's to do with how well the watch is regulated to your use. As an example, extensive use of a chronograph will affect the rate so the watch should be regulated differently to a watch where the chrono is rarely used.

In terms of manual vs. automatic there is one major difference that a good watchmaker should allow for in the regulation and that's the power reserve.

Power reserve in a watch is a spiral spring and while the mechanism is designed to provide consistent power transfer there is more torque through the running train when the reserve is full (tight mainspring) than when the reserve is low (relatively slack mainspring).

The rate is controlled by the oscillations of the balance and the torque affects this in two ways. If there is too much torque (full mainspring) then there can be excessive power transmitted to the balance at the extremes of it's oscillation causing it to 'bounce' off of the pin and rotate back faster (increased rate, watch runs faster). If there is too little torque (low mainspring) then there may be insufficient energy to get the balance to the end of it's oscillation so it never reaches the pin. That lowers the amplitude which increases the frequency and again the watch runs fast (contrary to popular belief that a low power reserve causes the watch to run slow).

In an automatic the power reserve tends to average on the high side - taken off at night for say 8 hours so the reserve drops from 42 hours (in a Breitling ETA based movement) to around 34 hours and then climbs back to 42 during wear and then remains at 42 hours until the watch is agai removed. So on a daily wearer the average reserve is going to be around 38 - 40 hours.

In a manual, general behaviour is to wind daily and the power reserve has a linear decline regardless of whether it is worn or no from 42 hours to 18 hours and then immediately back to 42 so the average reserve is around 30 resulting in a lower average torque and potentially lower average amplitude. That may be more or less accurate depending on how the watch is regulated, but if the watch is regulated for 'average' conditions it may well be more accurate.

This is oversimplifying things of course, the torque and / or amplitude are also affected by a myriad of other factors - condition of hte mainspring, condition of the oils, temperature, humidity, position, isochronism, etc, etc, etc.

Thanks Roff,
That's really helpful. You mentioned that...

"If there is too much torque (full mainspring) then there can be excessive power transmitted to the balance at the extremes of it's oscillation causing it to 'bounce' off of the pin and rotate back faster (increased rate, watch runs faster)."


That might explain why i used to hear my old Venus 178 Chronomat "ping" as it ticked. I always thought it did sound like something being hit.

You may also find that the watchmaker who serviced that watch used a mainspring that was too long or too strong which would drive excess torque through the running train and exacerbate the problem.

A mainspring that is too strong may drive the balance to overbank, but it would have to be quite a bit stronger. Overbanking is usually a result of a balance assembly problem.

If the mainspring is too long the movement will not have sufficient power reserve. There is too much spring in the barrel and you will not get the number of full revolutions you need before the spring runs down.

I once restored a WW1 German airplane clock ( a large watch) that would only run for eight hours. The mainspring had broken and the watchmaker who repaired it heated the spring and bent it back on the outside to form a loop to catch on the barrel hook. The loop took up so much room in the barrel that there was hardly any free space left. I put in a new mainspring of the proper size and, voila!, it ran for 30+ hours.

_________________
1809 Cosmonaute
1903 Navi
Navi Olympus
Super Avenger

If I use the chronograph almost daily, this is affecting the accuracy of the watch?

_________________
End of Discussion.

Of course.

With the chrono running there is additional friction in the running train which affects the rate.

I have one automatic that when the mainspring get wound down the watch loses time quite dramatically. I have another watch where the opposite occurs - it gains time. Not sure what the difference is between the two watches. Both use an eta movement.

The one that loses time has other issues.

A watch should gain with a low mainspring unless there is something else happening.

Depends on how wide the regulator gate is.

As long as the hairspring contacts both sides of the gate as power is decreased, the balance motion will also decrease because the force imparted by the pallet fork will be decreased as well. The amount of time for the balance to make a complete cycle will decrease, effectively making the watch gain time.

Conversely, the effective length of the hairspring will be longer if it is not contacting the gate. It usually occurs because the hairspring is not centered with the balance at rest and in beat. Under that circumstance, the watch will lose time when power drops to the point that the hairspring fails to contact the gate.

There are few hard and fast rules in this business except the rule that states "It's going to cost more to repair it if you tried to fix it first."

_________________
1809 Cosmonaute
1903 Navi
Navi Olympus
Super Avenger

That makes sense Marty, now why didn't I think of that......

I've been at this a bit longer!

_________________
1809 Cosmonaute
1903 Navi
Navi Olympus
Super Avenger