What could be the cause of this widely varying rate?

[JohnR725]

4 hours ago, Nucejoe said:

Something ought to change to cause a change in rate.

What changed in this low quality movement, in nine hrs

I'm not 100% clear on something here?

On 11/16/2023 at 12:29 PM, VWatchie said:

Here are the rates I got while letting the watch rest on its back overnight in a room with the same temperature (give or take one or two degrees) night after night. Each rate represents a night’s rest of about 9 hours. +8, +7, +4, +4, +3, +4, +12, +7, +8, +14, +23, +7, +9, +16, +18, +17, +22, +21, +30, +15.

So what we're seeing is the watch was set exactly to time and nine hours later over a variety of days that's how fast the watch was running?

We've already established that we don't know how wound up the watch was. It's always assumed that are automatic watches always fully wound up but it depends upon how much motions somebody has so unless it's actually wound up at the time we don't actually know so that will be a cause of variation.

So what can cause a change of rate? We've already established amplitude what can cause amplitude changes beyond the mainspring? What about the gear train could that be a source of problem or if the watch has a calendar that the calendar changing have anything to do with this where even the gear train related to the calendar? Could any of this cause an amplitude variation?

For instance I a timing machine I've ever noticed how the numbers seem to fluctuate? Now remember here were needing to look at really attempt of a second one second is a good enough for the timekeeping were looking at. Which I do think on the 1900 I think you may build a change it to measure tenths of a second. But in a case have you ever noticed your numbers slowly going up and down? What if you regulate the watch on one of those swings with the swinging up and down is over a five minutes for instance and you regulate of the watch during one of those up and down how would that affect the actual running time of the watch?

So for instance what about this image

What if one of those wheels was slightly out of round or what if the pivot was slightly bent what would that do to our timekeeping? In other words if the gear train was less than perfect and producing a power fluctuations what would that do

so the interesting features of paper tape machines you can look for patterns for instance much harder to see on the LCD screen where the relatively short.

With modern software and some modern timing machines for software to make a computer and a timing machine you can run time plots another way to look at the timing machine output. Makes it much easier to look for issues like this.

So for instance here is a example of a cheaper watch.

Ideally with time plots you're always going to see an up down gear trains will always produce pulsate's and the power but they should be relatively random. You never want to see a pattern on the time plot.

Here's an interesting time plot notice how nice the numbers really look even though graphically they look bad. We can definitely see a pattern but the timing is really really good And because this is a Rolex I believe this watch has no regulator pins

One of the interesting things with mechanical timepieces are they will average their mistakes over time. This is why if you look at Rolex timing specifications once they verify the watch pleases the timing machine then they run the watch for several days and average those numbers together to come up with an average rate per day. So on a instantaneous day your watch may appear to be outside of specifications but there timekeeping specifications is an average over time.

One of the interesting problems here is if I interpret correctly the watches set exactly to time and nine hours later it's look that and we see random timing but normally watch companies average an average over 24 hours would that change anything? Plus they do a little more averaging would that change anything

Here's something Rolex has half wound is 24 hours later fully wound up with courses fully wound up. Notice how in one position its 12 seconds fast how horrible. Except their criteria system so the first criteria is basically the isochronism the difference between a fully wound up and 24 hours later. The second criteria is averaging. Now this is where most watches are going to fall down this example is pleasing a timing machine.

Then what about real-world timekeeping? Notice once again they average their numbers so on a instantaneous particular day at might look bad but the overall average is good.

Oh and of course all of this assumes that whatever you using the time your watch is more accurate than the watch itself.

Oh and while were here discussing precision timekeeping a little light reading in case you think your watch that pleases the timing machine or overnight keeps time maybe you could please a standard? But you will note that the standards and Rolex all base their timing on 24 hours.

 

ISO-3159-2009 chronometer certification standard in English.pdf

Edited November 20, 2023 by JohnR725

[Nucejoe]

39 minutes ago, JohnR725 said:

I'm not 100% clear on something here?

 What can cause a change of rate in a low grade watch, but can not do the same in a high grade one. 

 

[Neverenoughwatches]

49 minutes ago, JohnR725 said:

I'm not 100% clear on something here?

So what we're seeing is the watch was set exactly to time and nine hours later over a variety of days that's how fast the watch was running?

We've already established that we don't know how wound up the watch was. It's always assumed that are automatic watches always fully wound up but it depends upon how much motions somebody has so unless it's actually wound up at the time we don't actually know so that will be a cause of variation.

So what can cause a change of rate? We've already established amplitude what can cause amplitude changes beyond the mainspring? What about the gear train could that be a source of problem or if the watch has a calendar that the calendar changing have anything to do with this where even the gear train related to the calendar? Could any of this cause an amplitude variation?

For instance I a timing machine I've ever noticed how the numbers seem to fluctuate? Now remember here were needing to look at really attempt of a second one second is a good enough for the timekeeping were looking at. Which I do think on the 1900 I think you may build a change it to measure tenths of a second. But in a case have you ever noticed your numbers slowly going up and down? What if you regulate the watch on one of those swings with the swinging up and down is over a five minutes for instance and you regulate of the watch during one of those up and down how would that affect the actual running time of the watch?

So for instance what about this image

What if one of those wheels was slightly out of round or what if the pivot was slightly bent what would that do to our timekeeping? In other words if the gear train was less than perfect and producing a power fluctuations what would that do

so the interesting features of paper tape machines you can look for patterns for instance much harder to see on the LCD screen where the relatively short.

With modern software and some modern timing machines for software to make a computer and a timing machine you can run time plots another way to look at the timing machine output. Makes it much easier to look for issues like this.

So for instance here is a example of a cheaper watch.

Ideally with time plots you're always going to see an up down gear trains will always produce pulsate's and the power but they should be relatively random. You never want to see a pattern on the time plot.

Here's an interesting time plot notice how nice the numbers really look even though graphically they look bad. We can definitely see a pattern but the timing is really really good And because this is a Rolex I believe this watch has no regulator pins

One of the interesting things with mechanical timepieces are they will average their mistakes over time. This is why if you look at Rolex timing specifications once they verify the watch pleases the timing machine then they run the watch for several days and average those numbers together to come up with an average rate per day. So on a instantaneous day your watch may appear to be outside of specifications but there timekeeping specifications is an average over time.

One of the interesting problems here is if I interpret correctly the watches set exactly to time and nine hours later it's look that and we see random timing but normally watch companies average an average over 24 hours would that change anything? Plus they do a little more averaging would that change anything

Here's something Rolex has half wound is 24 hours later fully wound up with courses fully wound up. Notice how in one position its 12 seconds fast how horrible. Except their criteria system so the first criteria is basically the isochronism the difference between a fully wound up and 24 hours later. The second criteria is averaging. Now this is where most watches are going to fall down this example is pleasing a timing machine.

Then what about real-world timekeeping? Notice once again they average their numbers so on a instantaneous particular day at might look bad but the overall average is good.

Oh and of course all of this assumes that whatever you using the time your watch is more accurate than the watch itself.

Oh and while were here discussing precision timekeeping a little light reading in case you think your watch that pleases the timing machine or overnight keeps time maybe you could please a standard? But you will note that the standards and Rolex all base their timing on 24 hours.

 

ISO-3159-2009 chronometer certification standard in English.pdf 290.12 kB · 0 downloads

So there is no short-term consistency in a watch because basically its not mechanically perfect and like everything else in the world " its close but not quite Carling " and will never be perfect. We just have to settle for long term daily averages, maybe over the course of a year the averages get  better still but only if the watch stays in the same working condition .

9 minutes ago, Nucejoe said:

 What can cause a change of rate in a low grade watch, but can not do the same in a high grade one. 

 

Outside of external influences that would depend entirely on the difference of the components of the two grades of watches and the regulation of them, wouldn't it ?  An oscillator that oscillates more evenly, a mainspring that delivers energy at a more consistent rate over time, a balance that is super poised in all positions ?

[JohnR725]

10 hours ago, Nucejoe said:

11 hours ago, JohnR725 said:

I'm not 100% clear on something here?

 What can cause a change of rate in a low grade watch, but can not do the same in a high grade one. 

By the way you quoted that I'm not 100% clear on the time span of how long the watch was being timed for in other words was it reset each night so it's only being run for a nine hour cycle or is it 24 hours cycle that's what I was not clear on

the difference between a high grade and a low grade watch is a lot of things. Although did we define what a high grade watches?

Let's make an assumption that a high grade watch is manufactured to better tolerances finishes better materials. More time and effort was spent making the watch more perfect.

11 hours ago, Neverenoughwatches said:

oscillator that oscillates

Oh thinking about the oscillation what about frequency? Notice how typical watches oscillated 18,000 beats per hour but what about newer watches their going faster would that have an impact on timekeeping?

 

[Neverenoughwatches]

8 minutes ago, JohnR725 said:

Oh thinking about the oscillation what about frequency? Notice how typical watches oscillated 18,000 beats per hour but what about newer watches their going faster would that have an impact on timekeeping

The teacher coming out in you John, love it when you make us gear up the old grey matter. So on hi-beat movements things are moving much faster, the balance and the escapement twice as fast if its 5hz compared to a  fairly standard watch.  The amount of time the balance is interacting with the escapement will be half the usual, so the escapement is having only half the amount of time to have an effect on time keeping. Like other escapement designs that have less influence by having less points of contact with the balance, hi-beats  have faster points of contact with escapement  ?

[Marc]

3 hours ago, Neverenoughwatches said:

hi-beats  have faster points of contact with escapement  ?

But more of them........

[Neverenoughwatches]

32 minutes ago, Marc said:

But more of them........

Ah ok marc thanks i didn't know the escapement was any different to a swiss lever.

32 minutes ago, Marc said:

But more of them........

What extra points of contact does a hi beat movement have over a low beat movement ? Or do you mean the balance just has twice as many contact events in the  same amount of time as say a 36,000 compared to an 18,000 ?  If the balance is shifting at twice the speed , then centrifuge must play a part as well, the weight of the balance is being drawn sideways rather than downways so there must be less gravity effect,  yes ?

Edited November 21, 2023 by Neverenoughwatches

[nevenbekriev]

Hi beat movements have over lo beat movements only when they are moving. When the movement stays still, the frequency doesn't affect timekeeping at all.

See, the balance is in contact with escapement only when it is within the lift angle. When it is turnetd more than half lift angle left or right from the zero position, it oscillates free. So, the biger amplitude, the more time of free oscillation. But, the frequency has nothing to do with it, as the 'free' time is the same percent of the whole time, if the amplitude is not changed.

I will advice to look at the rate 9 hours after full wind, or time error, measured every morning,  as like it is random quantity.  It actually is, and has it's expectation (average value) and dispresion. And, the good quality and well adjustet movements show small dispersion, the cheap and not well adjusted usually show big dispersion.  That is all

 

[Marc]

3 hours ago, Neverenoughwatches said:

What extra points of contact does a hi beat movement have over a low beat movement

None,

 

3 hours ago, Neverenoughwatches said:

the balance just has twice as many contact events in the  same amount of time as say a 36,000 compared to an 18,000

correct

 

3 hours ago, Neverenoughwatches said:

If the balance is shifting at twice the speed , then centrifuge must play a part as well, the weight of the balance is being drawn sideways rather than downways so there must be less gravity effect

Neither the mass of the balance nor the gravitational force will vary as a function of rotational speed, and as weight (gravity effect) is a function of those two parameters it won't vary either. A simple experiment to demonstrate this is to use one of those spinning gyroscope toys and weigh it stationary and spinning, there will be no change. A spinning gyroscope will resist changes in the orientation of its axis, but it can't defy gravity.

[VWatchie]

Sorry for not being very active in this thread that I started. I've been meaning to get back to it but too many things have come in between. Anyway, not sure what it brings to the discussion but I would just like to remind you that an escape wheel of a hi-beat movement has more teeth (20) than a low-beat movement (15). Please keep the discussion going. It's very interesting 

Escape wheel of a low-beat movement (Vostok cal 2409)

Escape wheel of hi-beat movement (ETA cal. 2824-2)

Edited November 21, 2023 by VWatchie

[Neverenoughwatches]

7 hours ago, Marc said:

None,

 

correct

 

Neither the mass of the balance nor the gravitational force will vary as a function of rotational speed, and as weight (gravity effect) is a function of those two parameters it won't vary either. A simple experiment to demonstrate this is to use one of those spinning gyroscope toys and weigh it stationary and spinning, there will be no change. A spinning gyroscope will resist changes in the orientation of its axis, but it can't defy gravity.

Thanks marc, obviously the weight  cant change or even shift position as its shape remains unchanged. I was thinking about fair rides lol, i must think harder . What about air resistance  ? A cooling fan can create lift, what about a floating balance .

Edited November 22, 2023 by Neverenoughwatches

[RichardHarris123]

1 hour ago, Neverenoughwatches said:

obviously the weight  cant change

Do you mean mass? The weight of the escapement, the whole watch and indeed yourself is constantly changing. An interesting thought, any watch undergoing radioactive decay, radium/ tritium etc will be slightly less massive for eternity. 

[JohnR725]

6 hours ago, VWatchie said:

Please keep the discussion going. It's very interesting 

I was concerned we were hijacking your discussion into something possibly related or possibly not

3 hours ago, Neverenoughwatches said:

What about air resistance 

Air resistance what a strange idea that is plus you didn't clarify something? The clarifications would be what about the balance wheel construction is there anything we could do to make air resistant less? Plus what about air resistance at sea level versus the mountains or how much humidity is in the air?

By the way I really hate scanning things to PDFs tedious and time-consuming but if it's a PDF it makes it easier to find in the future.  First thing the scan is only four pages From Omega this get us started then who would later today possibly tomorrow a section I will book which I did recommend everybody read but apparently you haven't taken the time to read it so I'll extract out some sections of that later this will get you started.

The attached PDF talks about switching from the bimetallic balance wheel to the modern balance wheel no screws basically easier to manufacture but is there anything in here related to this conversation? I've only extracted out a couple of things you should look at the PDF.

Omega technical guide number 12 June 1959 Omega have adopt the screwless balance.pdf

Edited November 22, 2023 by JohnR725

[Nucejoe]

21 hours ago, Neverenoughwatches said:

 36,000 compared to an 18,000 ?  

 Adjustments made in regulating  is done " manually"  thus is limited to the scope of man's ability/ accuracy/ percision, so as you double the frequency from 18000 to 36000 , you increase the effect of man's limited physical ability by two folds, therefor  36000bph is twice accurately regulated. 

I couldn't tell the type of balance wheel in Vostok from the pictures, is it glucydur ?  

Rgds

 

[JohnR725]

On 11/18/2023 at 12:39 PM, JohnR725 said:

https://youtu.be/op68gwVvsCo?si=7-jq6k3mRsgTqUeZ

 CL the exciting

On 11/20/2023 at 9:37 AM, Neverenoughwatches said:

So we are now back to how isochronous is the hairspring and what varying effects and changes of the escapement or anything else happen on any of the given 5 days. Its probably not something we can pinpoint and answer, not unless anyone here designs watch movements.

Finally got a chance to do some scanning.

Omega document on self lubricating mainsprings. As far as I know this is my place were I've ever seen any hints at all as to how the self lubricating mainsprings get their lubrication.

Then I made a reference to a book by way of a video but I wasn't sure if anyone was really paying attention to the video and the problem is he just talks about how wonderful the book is which it is but I thought maybe we could do a little better book review today.

I scanned in the index section so you can see all the wonderful stuff that's in the book.

Then I scanned in the first section on the oscillation system where it explains all about that. As we keep seeing a reference to isochronism on the second PDF page bottom right the reference to isochronism. Notice what it says about the escapement it's a bad thing. Then amplitude the reasons why it decreases and reference to gear train issues which is why we would have timing fluctuations. Watches do a beautiful job of averaging over time I'm not sure if eight hours qualifies as enough time depending upon where the problem is in the watch. But in a case we get a nice thing that explains all of and just stuff about the oscillation system pretty simple reading

then I skipped over the balance wheel section we don't need that. Then people's concern over the hairspring such a simple thing a little spiral a metal how complicated could that be? Oh and then I got distracted in the middle of scanning which is why it's in two separate PDFs but it's all there a little more light reading for you.

So now you have your reading assignment

Omega technical guide number 30 1963 self lubricating mainsprings.pdf Watch adjustment Jendritzki s3.pdf Watch adjustment Jendritzki s2.pdf Watch adjustment Jendritzki s1.pdf Watch adjustment Jendritzki index.pdf

[Neverenoughwatches]

3 hours ago, JohnR725 said:

Finally got a chance to do some scanning.

Omega document on self lubricating mainsprings. As far as I know this is my place were I've ever seen any hints at all as to how the self lubricating mainsprings get their lubrication.

Then I made a reference to a book by way of a video but I wasn't sure if anyone was really paying attention to the video and the problem is he just talks about how wonderful the book is which it is but I thought maybe we could do a little better book review today.

I scanned in the index section so you can see all the wonderful stuff that's in the book.

Then I scanned in the first section on the oscillation system where it explains all about that. As we keep seeing a reference to isochronism on the second PDF page bottom right the reference to isochronism. Notice what it says about the escapement it's a bad thing. Then amplitude the reasons why it decreases and reference to gear train issues which is why we would have timing fluctuations. Watches do a beautiful job of averaging over time I'm not sure if eight hours qualifies as enough time depending upon where the problem is in the watch. But in a case we get a nice thing that explains all of and just stuff about the oscillation system pretty simple reading

then I skipped over the balance wheel section we don't need that. Then people's concern over the hairspring such a simple thing a little spiral a metal how complicated could that be? Oh and then I got distracted in the middle of scanning which is why it's in two separate PDFs but it's all there a little more light reading for you.

So now you have your reading assignment

Omega technical guide number 30 1963 self lubricating mainsprings.pdf 1.08 MB · 2 downloads Watch adjustment Jendritzki s3.pdf 2.43 MB · 2 downloads Watch adjustment Jendritzki s2.pdf 7.13 MB · 1 download Watch adjustment Jendritzki s1.pdf 3.43 MB · 1 download Watch adjustment Jendritzki index.pdf 1.42 MB · 1 download

Thanks for the scanned pages John, this is a book i would really like, but I'm holding out on buying it new for now I'm sure ebay will provide me with an alternative in due course. The section on isochronism and poising some of it has gone over my head and needs a few reads to better grasp more of it. The next section on hairsprings, the first couple of paragraphs has me needing to ask a question about terminal curves.  What we normally call the terminal curve, the last 90 ° section of the spring that sits between the curb pins that is bent away from the main body of coils . If that section is necessary ? Jendritzki diagrams show a hairspring with only a final bend to locate the hairspring to the correct position to fit into the stud. I have come across hairsprings without that long end curve and actually added one as i thought it was missing and unadded to a raw spring. It looks like its not really needed as long as the spring follows the same arc as the regulator. And having a bend past the active point into the main body of the spring could affect the hairspring's oscillation. Its obviously only a small portion depending where the pins sit on that curve. But I'm thinking the active point could also be pulled from that bend as well as from the pins, a straight line so to speak directly from the collet to the curb pins should make the oscillations more even. From point A to point B with no diversion, if I'm taking nonsense feel free to tell me to shut up

On 11/16/2023 at 8:29 PM, VWatchie said:

I have a newly serviced automatic Vostok cal. 2416 with a decent amplitude in all positions, very small beat errors, and a mainspring with a full power reserve after a normal day of use.

Despite having a healthy amplitude, beat error, and power reserve, the rate widely varies. Here are the rates I got while letting the watch rest on its back overnight in a room with the same temperature (give or take one or two degrees) night after night. Each rate represents a night’s rest of about 9 hours. +8, +7, +4, +4, +3, +4, +12, +7, +8, +14, +23, +7, +9, +16, +18, +17, +22, +21, +30, +15.

I know the rate may fluctuate by a few seconds, but in this example, the rates vary between +4 and +30 seconds.

Any ideas?

I have to reply here to you @VWatchie this where it all started. Congratulations on instigating a rather complicated topic, i hold you responsible for my morning's headache

Edited November 28, 2023 by Neverenoughwatches
I felt the person responsible should be highlighted and personally notified 🤣

[JohnR725]

The problem with acquiring new knowledge is new questions come from that. I thought you would've read through the hairsprings by now how complicated could that be? Don't answer that question answer pretty complicated paragraph is why a lot of the stuff really only applies if you're making a watch as we just have to deal with the watch that we been given were not here to reinvent the wheel and modify and change it to be possibly better or destroy it trying to make it better.

If you look at vintage American pocket watches 18 size typical you'll find that there is no terminal curve for the hairspring. They hairspring is also much longer than usual and it's one of the rare times we can purchase a flat hairspring with  collet and no stud. Typically can only purchase over Quill hairsprings premade to exacting specifications. But in those early days you could purchase a flat hairspring and there was no terminal curve and it didn't seem to be a problem with the regulator pins. But if you look at a lot of the modern hairsprings the way they open and close without a terminal curve they hairspring would be bumping into the backside of the regulator pins.

If that is the question I think you're trying to ask is the terminal curve necessary Yes it is clear the regulator pins.

Yes eBay if you're patient sooner or later just about everything shows up and if even more patient sooner or later it might show up at a price you can afford.

[Neverenoughwatches]

Just now, JohnR725 said:

The problem with acquiring new knowledge is new questions come from that. I thought you would've read through the hairsprings by now how complicated could that be? Don't answer that question answer pretty complicated paragraph is why a lot of the stuff really only applies if you're making a watch as we just have to deal with the watch that we been given were not here to reinvent the wheel and modify and change it to be possibly better or destroy it trying to make it better.

If you look at vintage American pocket watches 18 size typical you'll find that there is no terminal curve for the hairspring. They hairspring is also much longer than usual and it's one of the rare times we can purchase a flat hairspring with  collet and no stud. Typically can only purchase over Quill hairsprings premade to exacting specifications. But in those early days you could purchase a flat hairspring and there was no terminal curve and it didn't seem to be a problem with the regulator pins. But if you look at a lot of the modern hairsprings the way they open and close without a terminal curve they hairspring would be bumping into the backside of the regulator pins.

If that is the question I think you're trying to ask is the terminal curve necessary Yes it is clear the regulator pins.

Yes eBay if you're patient sooner or later just about everything shows up and if even more patient sooner or later it might show up at a price you can afford.

Haha no i haven't read through it, i like to ask questions as they arrive in my brain

7 minutes ago, JohnR725 said:

The problem with acquiring new knowledge is new questions come from that. I thought you would've read through the hairsprings by now how complicated could that be? Don't answer that question answer pretty complicated paragraph is why a lot of the stuff really only applies if you're making a watch as we just have to deal with the watch that we been given were not here to reinvent the wheel and modify and change it to be possibly better or destroy it trying to make it better.

If you look at vintage American pocket watches 18 size typical you'll find that there is no terminal curve for the hairspring. They hairspring is also much longer than usual and it's one of the rare times we can purchase a flat hairspring with  collet and no stud. Typically can only purchase over Quill hairsprings premade to exacting specifications. But in those early days you could purchase a flat hairspring and there was no terminal curve and it didn't seem to be a problem with the regulator pins. But if you look at a lot of the modern hairsprings the way they open and close without a terminal curve they hairspring would be bumping into the backside of the regulator pins.

If that is the question I think you're trying to ask is the terminal curve necessary Yes it is clear the regulator pins.

Yes eBay if you're patient sooner or later just about everything shows up and if even more patient sooner or later it might show up at a price you can afford.

Yes John thats the question i was asking, still a bit confused though. If the dog's leg bend has a purpose as you describe then that means it has an effect on the spring. It feels like it has an effect on the active point of the spring so an effect on the rate. The shape of the spring is altered from its natural manufactured spiral shape past its active point into the main body of the coils, so the spring is also oscillating from that bend as well as the pins, 2 points of activation ? But in practice it doesn't seem to effect the rate. 

[Neverenoughwatches]

On 11/28/2023 at 10:43 AM, Neverenoughwatches said:

Haha no i haven't read through it, i like to ask questions as they arrive in my brain

Yes John thats the question i was asking, still a bit confused though. If the dog's leg bend has a purpose as you describe then that means it has an effect on the spring. It feels like it has an effect on the active point of the spring so an effect on the rate. The shape of the spring is altered from its natural manufactured spiral shape past its active point into the main body of the coils, so the spring is also oscillating from that bend as well as the pins, 2 points of activation ? But in practice it doesn't seem to effect the rate. 

Could oscillation activation to some degree come from all 3 of these points ? 

[JohnR725]

5 hours ago, Neverenoughwatches said:

Could oscillation activation to some degree come from all 3 of these points ? 

Don't remember where I saw it but with the etachron system where you basically have one and two they did a modification in later generations. There was some reference to they changed I have quite remember why they made some change there for reasons that I can't remember and I don't remember where I saw that. Possibly in some of the Swatch group documentation I obtained but finding that would be really difficult. So yes they made modifications right at that instant location because there must've been some reason why. It might have even had to do with heat treating from my memory but just not quite sure

that I course regulator pin spacing is important. Especially when you can adjust it things become more interesting because then you can hopefully get things better.

Don't know if that's answering your question at all other than basically everything that moves in the watch probably affects timekeeping after all it's a mechanical watch. But as I said before over time they will tend to average out all their problems hopefully.

[Neverenoughwatches]

1 hour ago, JohnR725 said:

Don't remember where I saw it but with the etachron system where you basically have one and two they did a modification in later generations. There was some reference to they changed I have quite remember why they made some change there for reasons that I can't remember and I don't remember where I saw that. Possibly in some of the Swatch group documentation I obtained but finding that would be really difficult. So yes they made modifications right at that instant location because there must've been some reason why. It might have even had to do with heat treating from my memory but just not quite sure

that I course regulator pin spacing is important. Especially when you can adjust it things become more interesting because then you can hopefully get things better.

Don't know if that's answering your question at all other than basically everything that moves in the watch probably affects timekeeping after all it's a mechanical watch. But as I said before over time they will tend to average out all their problems hopefully.

Thanks John,  i just wondered if the change of shape/direction of the spring where the start of the terminal curve begins might have a different effect on the rate compared to a spring where there is no curve. By coincidence i looked at the watch i am working on at the moment,  it has no terminal curve.

[nevenbekriev]

6 hours ago, Neverenoughwatches said:

 i just wondered if the change of shape/direction of the spring where the start of the terminal curve begins might have a different effect on the rate compared to a spring where there is no curve. By coincidence i looked at the watch i am working on at the moment,  it has no terminal curve.

Yes, all this things: the end curve shape, it's position against the begining of the spring, and the beginning curve shape and distance from beginning point to center (staff axe) have special inpact on the frequency as function of amplitude. And yes, the frequency depends on amplitude and this function is complicated one. But all this needs to many writing to explain, and it is easier to point where is written in a book, as many books have already been written on the matter.

Edited November 30, 2023 by nevenbekriev