Timegrapher diagnosis for the amateur…

[M'Bob]

What do you all think of the scenario where a new watch, like an Omega with a co-axial movement, was sitting on the shelf for five years before purchase. If the intention is to wear it, should it be serviced prior, or likely good for a few years more?

 

[Joe_A]

Archer

I think a timing machine can tell you when the watch is definitely in need of servicing or repair.

What it can't always tell you is when that it doesn't need servicing or repair.

Al,

If I have understood the above second line correctly, you are suggesting that, there are times when a watch may appear to hit all the right marks on the timegrapher . . . and still be in need of service?

I have only one vintage watch that I have purchased thus far that I did not send out for service within a month or three of having acquired it and in the one case it was because the watch wound and set easily, indicated low beat error and high amplitude and the amplitude changed by position the way I had learned to expect. The difference in performance by position was similar to that of a serviced watch. Inside the watch appeared clean and the screw heads appeared unmolested.

If the watch in question were a high-value watch, I'd have sent it out by now, but so far, it remains un-serviced.

Am I missing the point?

Edit: Just to be clear, I am always happy to learn something and that is why I post here.

 

[YY77]

M'Bob

What do you all think of the scenario where a new watch, like an Omega with a co-axial movement, was sitting on the shelf for five years before purchase. If the intention is to wear it, should it be serviced prior, or likely good for a few years more?

I bought my Seamaster Trilogy in 05 '21 (OB Madrid) with the 8806 co-axial, so about 4 years on the shelf. It's running great and I have a backup if it acts up within the 5 year warranty period.

 

[Canuck]

One thing to bear in mind is that in one cycle of the balance wheel, a timing machine hears things that we humans are unable to hear. And many of those extraneous sounds end up showing on the screen, or detail tape. Perhaps we expect the machine to hear only the roller jewel contacting the pallet horn as it shifts the lever. But these machines hear multitudinous tiny sounds. I have some control of what my Vibrograph hears by adjusting the volume to achieve a clean pattern.

I recently serviced a Hamilton 992B for a brother collector. He indicated that it was out about 90 seconds per month (3 seconds per day). I put it on my timegrapher, and it would have been impossible for me to determine a rate as the machine was showing a delta of up to ten seconds variation, per day, without touching the regulator. And I was unable to achieve an absolutely clean, uniform pattern that we all prefer. The watch has a LOT of miles on it over its 60 odd years. I moved it to the Vibrograph, and got a clean, uniform pattern showing a gain of 3 seconds per day. Perhaps those who own modern solid state digital timing machines might be happier with a Vibrograph (ya, like THAT is going to happen!).

 

[Archer]

M'Bob

What do you all think of the scenario where a new watch, like an Omega with a co-axial movement, was sitting on the shelf for five years before purchase. If the intention is to wear it, should it be serviced prior, or likely good for a few years more?

If it comes with a warranty, then wear it. If something goes wrong in the 5 year warranty period, you will be covered.

If there's no warranty, then it becomes a matter of what are you paying for the watch that might need service sooner than expected - is a discount you can get going to cover that possible service.

 

[Archer]

Joe_A

If I have understood the above second line correctly, you are suggesting that, there are times when a watch may appear to hit all the right marks on the timegrapher . . . and still be in need of service?

Here's the incoming timing checks on a Speedmaster with an 1861 I serviced recently:



Full wind - amplitude is healthy in all positions, beat error is good. The only two things that may indicate something is up, are the fast rates, and the fact that the balance amplitude difference between dial down and dial up (first two readings) is 23 degrees.

Onto a visual inspection...balance jewels on this side look well oiled - note that the balance would be riding on this jewel when in the dial up position, so this would be the 303 degrees position:



Escape wheel jewel has some oil:



Third wheel jewel is pretty much dry:



Closer look:



The third wheel jewel being dry is not unusual. For whatever reason, if there's going to be dry train wheel on these, it's almost always the third wheel jewel. I replace more worn third wheels than any other wheel on this movement.

Coupling wheel oil looks quite congealed:



Chronograph wheel jewel is dry:



Escapement is dry:



So no parts were needed as nothing had worn...yet. I wouldn't call this a catastrophic condition or anything, but it's certainly not all "right" as it was when it arrived in my shop. Just by looking at the timing results, I don't think most would see the condition this was in - IMO a visual inspection in addition to timing checks is much better than timing checks alone.

Cheers, Al

 

[ansoa-66]

As a novice as you are, I will let you know what i gathered after speaking with multiple watchmakers.

One thing that seems to be agreed between all of them is that a "service" is not a "refurbishment".
You will usually see some people telling that a watch after service needs to have a 270+ amplitude. That's something that is only achievable if the watch was really good taking care of, or if you change parts (seems like the mainspring is something that interferes a lot on the amplitude).
What I saw, was that if, for example a 60s watch, has an amplitude of around 200, they don't intervene to solve it. When it becomes really low, 120 or so, they then change parts. If this is a correct assessment or no, I guess it really depends on the opinion of each one.

Also, I would like to say something personal that bothers me. I have seen quite a lot of those refurbishments where people simply swap parts with "small" defects. For example, if a part has some rust, instead of trying to fix it and clean it properly, they run towards getting a new one.. That creates a big big big issue as far as a person that is a hobbyist and that collects watches.
I have came across watches that needed parts and it was simply impossible to find them (I have a close relationship with my watchmaker, every time he needs parts, he asks me to get them and I search for them). I think that we will come to a point that parts going to be to scarce that will be impossible to fix watches when they truly need it.

 

[Canuck]

As indicated above, cleaning a watch is done (generally) NOT to remove dirt! So often I hear the statement “it looks clean”! But congealed oil and dry pivots don’t show up until you strip the movement.

 

[Canuck]

My Hamilton grade 952 which (admittedly), could use servicing. Not perhaps the best test case, but it was handy. I show the same watch on my TYMC digital timer, and my Vibrograph B200 (about 60 years old). A comparison I find interesting. Using the digital timer, note the rate variation is all over the yard, the amplitude 170°, and the beat error 1.3 ms. Put it on the Vibrograph, and the pattern shows pretty much on beat. In one picture on the Vibrograph, the rate was about -12 seconds per day. The watch was nearly run down. After a full wind, the rate on the Vibrograph improves to about - 2 seconds per day. Speaking as someone who repairs a lot of watches, the Vibrograph is of more value to me than the digital timer. But the Vibrograph isn’t suited to some high beat watches. So, two different timers telling two different stories! Timing machines are a tool, but not the be all and end all of satisfactory results. They can only take you so far in ending up with a good result.

 

[M'Bob]

Pondering this one: what could it potentially mean if the rate in three positions is within spec, but the delta is way off?

 

[Canuck]

M'Bob

Pondering this one: what could it potentially mean if the rate in three positions is within spec, but the delta is way off?

My timer may show a rate variation of 10 seconds per day without testing the watch in alternate positions, and without touching the watch when it is on the pick up. As I said earlier, timing machines take a two minute “snapshot” of performance, and extrapolate that forward to estimate how it would time over 24-hours. My most useful timing machine in regards to bringing a watch to time is my wrist! The watch I was testing in the example shown hadn’t been serviced for decades. So this was not necessarily the best test case to illustrate a point. It has since been serviced, and without fussing with the regulation, it is within 10 seconds per day. Not bad for a 115 year old watch, in my view. The amplitude has also vastly improved since the test shown above. In short, final regulation is not done by trusting the rate shown by the timer.

 

[M'Bob]

I appreciate that example. What I’d like to get more of a handle on, though, is what could account for somewhat significant rate changes in different positions, in the positive and negative rates, which could give a mathematically good daily rate but a delta that’s way out of specification.

 

[Canuck]

It’s not so much a discussion of what the watch is doing to give those results as it is a matter of how the machine interprets what the watch is doing. I said it before, and i’ll say it one more time, and only one more time! Final regulation is rarely if ever done with the watch on the machine. These machines can be trusted to a point, but only to a point. Please don’t ask again to give you and explanation that makes sense to you.

 

[Archer]

M'Bob

I appreciate that example. What I’d like to get more of a handle on, though, is what could account for somewhat significant rate changes in different positions, in the positive and negative rates, which could give a mathematically good daily rate but a delta that’s way out of specification.

I’ve had watches come in that we’re clearly in need of service, but in my initial timing tests had a Delta over 100 seconds, but they averaged out to near zero, so it happens. If you wore the watch so that it spent time on your wrist in exactly the same way it did on the timing machine, it would run great. But of course that’s pretty much impossible. That’s why reducing the Delta is so important to good timekeeping.

Regulating is fine tuning the final rate. It’s normally done on the timing machine, and then verified after the watch is assembled. Personally I perform static tests for 24 hours in each position, plus another test on a test winder (very different machine from a consumer winder). From there I adjust the average rate to where I want it to be, generally slightly positive (people usually want a few seconds per day positive, but not even 1 second a day negative).

Regulating is the easy part, and the last thing done.

Adjusting, which is done to reduce the Delta is the hard part. That’s all done on a timing machine, and causes for variation depend partly on the design of the movement. Things such as the balance being free sprung or using regulating pins will change the potential causes of positional errors.

I’ll be leaving out the issues that would be caused by wear (such as worn balance pivots for example), or other more fundamental repair issues, because these things all need to be repaired before final adjustment even begins. So these below are pure adjustment issues.

By no means is this exhaustive, especially since it’s New Year’s Eve and I’m well under way celebrating...but here are some possible causes:

Balance spring not concentric (includes coils not being evenly spaced).
Balance spring out of flat (raised up on one side, coned, etc.).
Balance spring not centred in the regulating pins.
Regulating pins not spaced correctly.
Regulating pins not parallel.
Poise errors on balance.

These things are not diagnosed completely with the timing machine, as it requires detailed inspection under magnification to determine what the problem is specifically. I listed the poise error last, because you only start adjusting that once you know everything else is 100% perfect.

This isn’t some magical thing that cannot be explained, even though it seems that’s what some people want you to believe. Honestly some of the answers given in this thread border on nonsense. Anyway, it’s pretty basic stuff for any watchmaker, but would be considered advanced work for an amateur to tackle these things.

 

[M'Bob]

Archer

I’ve had watches come in that we’re clearly in need of service, but in my initial timing tests had a Delta over 100 seconds, but they averaged out to near zero, so it happens. If you wore the watch so that it spent time on your wrist in exactly the same way it did on the timing machine, it would run great. But of course that’s pretty much impossible. That’s why reducing the Delta is so important to good timekeeping.

Regulating is fine tuning the final rate. It’s normally done on the timing machine, and then verified after the watch is assembled. Personally I perform static tests for 24 hours in each position, plus another test on a test winder (very different machine from a consumer winder). From there I adjust the average rate to where I want it to be, generally slightly positive (people usually want a few seconds per day positive, but not even 1 second a day negative).

Regulating is the easy part, and the last thing done.

Adjusting, which is done to reduce the Delta is the hard part. That’s all done on a timing machine, and causes for variation depend partly on the design of the movement. Things such as the balance being free sprung or using regulating pins will change the potential causes of positional errors.

I’ll be leaving out the issues that would be caused by wear (such as worn balance pivots for example), or other more fundamental repair issues, because these things all need to be repaired before final adjustment even begins. So these below are pure adjustment issues.

By no means is this exhaustive, especially since it’s New Year’s Eve and I’m well under way celebrating...but here are some possible causes:

Balance spring not concentric (includes coils not being evenly spaced).
Balance spring out of flat (raised up on one side, coned, etc.).
Balance spring not centred in the regulating pins.
Regulating pins not spaced correctly.
Regulating pins not parallel.
Poise errors on balance.

These things are not diagnosed completely with the timing machine, as it requires detailed inspection under magnification to determine what the problem is specifically. I listed the poise error last, because you only start adjusting that once you know everything else is 100% perfect.

This isn’t some magical thing that cannot be explained, even though it seems that’s what some people want you to believe. Honestly some of the answers given in this thread border on nonsense. Anyway, it’s pretty basic stuff for any watchmaker, but would be considered advanced work for an amateur to tackle these things.

Thanks, this is what I was looking for.

From my limited experience tinkering, I’ve noticed (since I typically check in the usual three positions), what I think is a pattern, where, in the dial-up position, it tends to run fast, amd in the two vertical positions, it seems to run in the negative. These are watches that are in need of a service, usually.

If this is indeed a common pattern, and not an artifact, then I’m trying to make sense of it mechanically. I can understand if a watch isn’t properly lubricated or clean, increased friction and slow running in the vertical positions make sense, as there appears to be more friction due to surface contact of the pinions, which are now horizontal, plus the longer lever arm vs when they are in dial-up, and vertical. But what I’m unclear on, is how this same dry/under-lubricated watch runs faster dial-up, rather than slower as well. So, it’s the variation in the fast vs slow, as related to position (and hence the delta outside spec.) that’s getting me confused. Of course, this may all be an observation of too small a sample size on my part, and the fact that I’m only using a bunch of cal. 321’s as my reference point, which may have positional performance quirks that I’m not aware of.

 

[Canuck]

I mentioned the test Hamilton 952 that I discussed earlier in this thread. Since the tests that I showed, I have conditioned the watch. As part of the conditioning, I adjusted the regulator curb pins to allow virtually no freedom for the hairspring to vibrate between them, I made a minor adjustment to the poise of the balance, and since this watch has a bi-metallic, steel and brass balance wheel and blued steel hairspring, I de-magnetized it (it was magnetized). The 952 has a “collar button” hairspring stud, so centring the hairspring is simple. Stop the balance wheel when the hairspring is in a resting attitude, loosen the clamp holding the stud, the hairspring centres itself, then tighten the clamp. Done! Test results have improved by 75%, and it has varied less than 6 seconds in 30 hours, with the watch running face up for 12 hours, and in the pocket for 18 hours. The watch is (as far as I am aware) completely original, and it is 115 years old. The results I have attained are good enough for me!

 

[Archer]

M'Bob

From my limited experience tinkering, I’ve noticed (since I typically check in the usual three positions), what I think is a pattern, where, in the dial-up position, it tends to run fast, amd in the two vertical positions, it seems to run in the negative. These are watches that are in need of a service, usually.

That's not any kind of regular or typical pattern that I've seen. But trying to look for patterns on a watch that needs to be serviced is pointless really. The first and only step in dealing with such a watch is to service it, and get any wear, dirt, lubrication issues sorted, then the adjusting comes after.

 

[M'Bob]

Since we’ve touched on the topic of “service”: I’ve had two watchmakers over the last few decades ask me, when bringing a watch in for service, how often I intend to where it. If I said, “about once a month,” they both said that the watch really doesn’t need a full service, but they would pull the balance and make sure it was well-lubricated. I should mention that these are independents, not working at the same facility.

I’m certain there’s a standard service protocol that watchmakers are taught, and would be complete, but the question: is there a case, such as above, where a partial service, emphasizing the more important issues to attend to, is ever indicated, and make sense? Put a different way: are there areas of a watch that tend to wear more aggressively than others, or do most parts wear at a fairly consistent rate? Because if the former is true, can it ever be justified to attend to the more pressing areas if a watch will only be worn infrequently?

 

[Archer]

M'Bob

Since we’ve touched on the topic of “service”: I’ve had two watchmakers over the last few decades ask me, when bringing a watch in for service, how often I intend to where it. If I said, “about once a month,” they both said that the watch really doesn’t need a full service, but they would pull the balance and make sure it was well-lubricated. I should mention that these are independents, not working at the same facility.

I’m certain there’s a standard service protocol that watchmakers are taught, and would be complete, but the question: is there a case, such as above, where a partial service, emphasizing the more important issues to attend to, is ever indicated, and make sense? Put a different way: are there areas of a watch that tend to wear more aggressively than others, or do most parts wear at a fairly consistent rate? Because if the former is true, can it ever be justified to attend to the more pressing areas if a watch will only be worn infrequently?

I guess this would be some sort of partial service. To be honest, it's not something I would do, but I understand other watchmakers may handle things differently for a number of reasons.

Generally speaking, when one area of the watch is in need of servicing, it's not isolated - the entire movement needs attention. I don't ever get watches in that only need lubrication on the balance jewels. If those are dry, the others are as well, so continuing to run the watch in that condition is not helping it. Sure, you can run it very little and it won't have a big impact, but you could do that without doing anything to it, so whatever they did doesn't make a lot of sense to me.

I don't know what these watchmakers actually did when they "pulled the balance" and "made sure it was lubricated" but generally speaking, just adding oil to any jewel without properly cleaning it first, is as likely to make things worse as it is to make things better. Any debris that's in there now becomes a grinding paste, which is not good for pivots.

 

[M'Bob]

Archer

I don't know what these watchmakers actually did when they "pulled the balance" and "made sure it was lubricated" but generally speaking, just adding oil to any jewel without properly cleaning it first, is as likely to make things worse as it is to make things better. Any debris that's in there now becomes a grinding paste, which is not good for pivots

Thanks. In a watch that comes in for service, that was seldom used, does the congealed oil noted above in one of your pictures above count towards the type of debris that is abrasive, even in the absence of dirt or metallic particulate?