Jump to content

Precision Adjustment for Isochronism and Positional Error


Recommended Posts

Precision adjustment for isochronism and positional error, so what's the question? First, the landscape of what I'm working on. 

I have a number of Sellita SW300-1 movements, Premium (Top) grade execution with chronometer grade pallet fork, escape wheel, and 5-position regulation to +/-4 s/d. Attached is the technical cut sheet for reference. 

1715264544_ScreenShot2022-06-09at8_55_32AM.thumb.png.f951556851bdca44d608fc8dddd33f97.png

I am working on fine tuning the adjustment of these movements in preparation to undergo chronometer certification, ISO3159 testing. For reference, the testing protocols: 

392997661_ScreenShot2022-06-09at8_59_43AM.thumb.png.dc76d7135b13d359cc7c22df3b633ef3.png

1273373974_ScreenShot2022-06-09at8_59_50AM.png.2b64bdc4e1c812ce81a6326357bf5c72.png

641965267_ScreenShot2022-06-09at8_59_58AM.png.fd333fecb95f86e960a9ac7c18e115ed.png

710864570_ScreenShot2022-06-09at9_00_06AM.png.7772978517fee33e4aaeaeaccb1450c7.png

Out of the box, the movements are very good. In fact, as an automatic movement that will maintain a near full power reserve over the course of a day while wearing, these will perform exceptionally well. I have been using an MTG-9900A to test each movement at full wind and half wind with a 6-position 20-second stabilization and 40-second timing period protocol. With these numbers, I am capturing the instantaneous rates at 0 and 24, determining the average beat rate for each position, the isochronism for each position, and able to use these numbers to predict the approximate performance when the movement undergoes ISO3159 testing protocol. Here's an example of my data: 

 944450665_ScreenShot2022-06-09at9_22_17AM.thumb.png.56c320f41e094154aecbc7aefea0fb1b.png

So what's bugging me? Like I said, I'm very satisfied with the performance of this movement. On the wrist, it will time very good. Even undergoing ISO3159, it appears it will pass. To give myself some margin of error, I should slow the average rate down 1.5 s/d using the average rate between 0 and 24. Why? Because the test is 24hrs in each position so isochronism matters even though the reality of wearing an automatic doesn't matter as much. 

What's bothering me is the positional error between 6H and 12H. At full wind it is a 6-sec delta, at half wind that is exacerbated to an 11-sec delta. Because of the relative consistency between the other positions this suggests to me the balance needs dynamic poising or inherent to the design of an SW300-1 is a gravitational issue. 

Keep in mind these are factory fresh, top grade movements that, and I confirmed with Sellita, have been dynamically poised in factory. I will test out the theory of dynamic poising to eliminate the delta between 6H and 12H. But should the issue persist (and I think it will since I am seeing this delta on a number of the movements), I believe it is likely a center of gravity issue with the hairspring. Finally arriving at my question, how do you modify the hairspring to shift the center of gravity? 

Looking at Hmilton's book on hairspring manipulations, he recommends a slight tweak to the hairspring at the collet pinning point. However, the SW300-1 collet and pinning design simply does not allow for one to access the start of the coil in the same way it is accessible in a vintage movement. His next advice is to adjust the coil at the other pinning point. But his example is an overcoil, not a flat hairspring. How would I go about making a gravitational adjustment to the flat hairspring at the outer pinning point? And in doing so, how to I determine the direction of the manipulation, eg. opening the arc moves the center down, etc? 

IMG_3389.thumb.jpeg.4e880691dc27c15c06f0c6a68d3bca47.jpeg

IMG_3390.thumb.jpeg.17a4eb02b114918014af3b6e75591a29.jpeg

IMG_3391.thumb.jpeg.6b0a522563343a62f7336dd9740b674c.jpeg

Perhaps the question is moot considering the movements will wear well, the issue is an error between 6H and 12H and since 12H is not factored into ISO3159, maybe it's all a moot point. In any event, I'd still like to learn more about these kinds of precision adjustments to compensate for very small changes to increase the accuracy of a movement. 

Brochure_technique-SW300-1_2.pdf

  • Like 1
  • Thanks 1
Link to comment
Share on other sites

You can test in 8 vertical positions to see more clearly if you have a poise issue. As overal the rates increase, with the exception of 12*, at 24h, my first thought is the hairspring isn't perfectly centered in the regulating pins, and the pins are possibly too open. This would be a very easy adjustment to make (both), and reversible, so very much worth having a look at before adjusting the poise.

 

* I would bet that at 12 (not sure of your nomenclature, is that crown at 12 or crown right, CR?) the pins are in a position that the hairspring passes through parallel to the ground.

  • Thanks 1
Link to comment
Share on other sites

37 minutes ago, nickelsilver said:

You can test in 8 vertical positions to see more clearly if you have a poise issue. As overal the rates increase, with the exception of 12*, at 24h, my first thought is the hairspring isn't perfectly centered in the regulating pins, and the pins are possibly too open. This would be a very easy adjustment to make (both), and reversible, so very much worth having a look at before adjusting the poise.

 

* I would bet that at 12 (not sure of your nomenclature, is that crown at 12 or crown right, CR?) the pins are in a position that the hairspring passes through parallel to the ground.

12H = 12-o'clock up or crown right

In this position, the regulating arm is pointed at approximately 1:30 which makes the etachron regulating pins almost parallel to the ground when rotated to close the pins. 

  • Like 1
Link to comment
Share on other sites

The thing with the regulating pins, is the basic logic is if you open them, the rate drops, and drops more as amplitude goes down. You could have them open enough that after 24 hours the spring is barely hitting them, so the rate goes way down. But if the spring isn't centered you can get very interesting results, including an increasing rate as amplitude goes down.

Imagining a normal gap on the pins, if the spring favors strongly one side due to poor centering, the spring might pull away more at higher amp, and stay totally or almost totally in contact at lower amplitudes. As you change position, gravity is affecting the spring as well, and so in certain positions it might pull away still (your 12), but stay more in contact in the other ones, including the horizontal positions.

 

This is just one possibility in your case. But the thing any school will hammer and hammer and hammer into students heads is before doing dynamic poising, every thing else must be perfect first. Otherwise, even if you had what looked like a clear poise issue, you can be embarking on a wild goose chase.

 

I'm quite sure you can get your 24h delta lower, and without resorting to some of the techniques in the Hamilton book (which is a truly fantastic book). The pinning points, the way the hairspring leaves the collet, and the terminal curve on an ETA or Sellita has been intensely studied and should need no manipulation, and with the Etachron system you have a very easy way to get the hairspring really behaving top notch between the pins. It could be that in the end your best route is to do some dynamic poise, but do check and recheck and adjust any and everything else first.

  • Like 5
  • Thanks 1
Link to comment
Share on other sites

5 minutes ago, nickelsilver said:

The thing with the regulating pins, is the basic logic is if you open them, the rate drops, and drops more as amplitude goes down. You could have them open enough that after 24 hours the spring is barely hitting them, so the rate goes way down. But if the spring isn't centered you can get very interesting results, including an increasing rate as amplitude goes down.

Imagining a normal gap on the pins, if the spring favors strongly one side due to poor centering, the spring might pull away more at higher amp, and stay totally or almost totally in contact at lower amplitudes. As you change position, gravity is affecting the spring as well, and so in certain positions it might pull away still (your 12), but stay more in contact in the other ones, including the horizontal positions.

 

This is just one possibility in your case. But the thing any school will hammer and hammer and hammer into students heads is before doing dynamic poising, every thing else must be perfect first. Otherwise, even if you had what looked like a clear poise issue, you can be embarking on a wild goose chase.

 

I'm quite sure you can get your 24h delta lower, and without resorting to some of the techniques in the Hamilton book (which is a truly fantastic book). The pinning points, the way the hairspring leaves the collet, and the terminal curve on an ETA or Sellita has been intensely studied and should need no manipulation, and with the Etachron system you have a very easy way to get the hairspring really behaving top notch between the pins. It could be that in the end your best route is to do some dynamic poise, but do check and recheck and adjust any and everything else first.

Thank you!

Are there rules of thumb for reading the data? For instance, if rates in pendant position gain over 24hrs, increase the opening at the pins. 

Is there something similar for deciphering to which side the hairspring may not be centered? 

  • Like 1
  • Thanks 1
Link to comment
Share on other sites

28 minutes ago, mzinski said:

Is there something similar for deciphering to which side the hairspring may not be centered? 

This is purely visual. I'm a pretty good regleur, but I work with a guy who is amazing, and the absolute number one "trick" he has is to true and center everything up just like in school.

 

Check the centering with no power. The spring should be centered and not touching either pin. Give it a wiggle and see that it bounces back and forth. Tweak at stud to center, twist pin block to open and close.

Edited by nickelsilver
  • Like 3
  • Thanks 1
Link to comment
Share on other sites

On 6/9/2022 at 11:09 AM, nickelsilver said:

This is purely visual.

If you look at the documentation that eta has they show it as being pure visual. The only thing they don't quite explain is You open up so you can center the hairspring between the pins but in the document I was looking at they don't explain when you close it just how close you want to go to the hairspring?

Then down below I have a link to a book explains how to adjust your watch. You will note they're not using a timing machine they hadn't been invented yet So it's going to take a little bit longer to regulate your watch.

Then a PDF extracted from a magazine you might find that of more interest. Then in case the name of the author is not familiar with or to you. He worked his way through the Elgin company becoming the head of research.

 

 The watch adjuster's manual [microform] : being a practical guide for the watch and chronometer adjuster in making, springing, timing and adjusting for isochronism, positions and temperatures

by     Fritts, Charles Edgar

https://archive.org/details/cihm_03260

 

1946-08-HIA-OCR pages 9 to 13 poise Challacombe.pdf

  • Like 2
  • Thanks 2
Link to comment
Share on other sites

Does the Etachron system address the problems outlined above? It does allow the size of the gap of the regulator to be changed and also the centering of the spring between the gap.

Most of the static poising videos I've seen on YouTube show the balance on the poising tool without the roller. Wouldn't the notch in the roller, the jewel and the D shaped hole affect the poising?

Link to comment
Share on other sites

4 hours ago, HectorLooi said:

Does the Etachron system address the problems outlined above? It does allow the size of the gap of the regulator to be changed and also the centering of the spring between the gap.

Most of the static poising videos I've seen on YouTube show the balance on the poising tool without the roller. Wouldn't the notch in the roller, the jewel and the D shaped hole affect the poising?

It's 100% wrong to static poise with the roller off.

  • Like 1
Link to comment
Share on other sites

Thank you all gentleman. the subject of this discusion can cover some of my yet unasked questions.

In what way is a Chrono fork or escape wheel  different from a regulare one?  

Is it just pallet adjustment in case of a Chrono fork? 

Have been doing my best to just read here and not talk. 

TIA 

Link to comment
Share on other sites

On 6/9/2022 at 10:48 AM, nickelsilver said:

The thing with the regulating pins, is the basic logic is if you open them, the rate drops, and drops more as amplitude goes down. You could have them open enough that after 24 hours the spring is barely hitting them, so the rate goes way down. But if the spring isn't centered you can get very interesting results, including an increasing rate as amplitude goes down.

Imagining a normal gap on the pins, if the spring favors strongly one side due to poor centering, the spring might pull away more at higher amp, and stay totally or almost totally in contact at lower amplitudes. As you change position, gravity is affecting the spring as well, and so in certain positions it might pull away still (your 12), but stay more in contact in the other ones, including the horizontal positions.

 

This is just one possibility in your case. But the thing any school will hammer and hammer and hammer into students heads is before doing dynamic poising, every thing else must be perfect first. Otherwise, even if you had what looked like a clear poise issue, you can be embarking on a wild goose chase.

 

I'm quite sure you can get your 24h delta lower, and without resorting to some of the techniques in the Hamilton book (which is a truly fantastic book). The pinning points, the way the hairspring leaves the collet, and the terminal curve on an ETA or Sellita has been intensely studied and should need no manipulation, and with the Etachron system you have a very easy way to get the hairspring really behaving top notch between the pins. It could be that in the end your best route is to do some dynamic poise, but do check and recheck and adjust any and everything else first.

Thanks for the encouragement. I’ve gotten to a good point in which the spot timing at full wind and half wind looks good in isolation of one another. What isn’t perfect yet is the isochronism. As you can see, it’s gaining as it winds down. 
47EAAF69-F117-4A52-846B-622361C27EF4.thumb.jpeg.73a8608bff0b68305e8ffb251d616226.jpeg
Again, referencing Hamilton, it sounds like one can reduce gain in short arc by widening the regulator pins or increase gain in short arc by closing regulator pins. Am I understanding this correctly? 

Link to comment
Share on other sites

Out of curiosity would you have the book found at the link below? This is because it has several pages on regulator pins spacing and talks a little about the consequences. Otherwise if you don't have that I can scan the relevant pages.

https://youtu.be/op68gwVvsCo

Then I was thinking about this discussion when I was at work on Friday so I took some pictures for you. Vintage American Waltham pocket watch recently serviced. Lift angle probably isn't right is amplitude is too high but it doesn't matter what I was doing. Pivots though probably could've been polished as the dial up and down aren't quite as identical as they could be.

Then images number one shows a problem. Dial up and down look fine but notice what happens when you go to the various crown positions the amplitude drops which is standard and the rate gets really slow.

In image number two notice the Delta site looks great and the watches running fast.

Then image number three timing looks pretty good.

A perfect example of regulator pin spacing. As a reminder with over coil hairspring is the spacing with the regulator pins should be as close as possible as long as they hairspring can slide. In the book I mentioned above and the source by the Elgin watch company they both said the same thing with flat hairspring is the spacing should be about the with or the thickness of the hairspring. I assume that would be split up across both pins.

So in the case of this watch with an over coil hairspring as soon as you go to the crown positions amplitude drops as expected in the watch is slow. Closing the regulator pins now the watch runs fast. Then finally just moving the regulator it looks fine now.

 

 

 

wt-3.JPG

wt-2.JPG

wt-1.JPG

Link to comment
Share on other sites

12 hours ago, mzinski said:

Again, referencing Hamilton, it sounds like one can reduce gain in short arc by widening the regulator pins or increase gain in short arc by closing regulator pins. Am I understanding this correctly? 

Yes, that is exactly what to do. But with rates as close as yours are already, it's a very minute adjustment, and the small change you are seeing between full and half wind could be from something else (still thinking the hairspring could be centered better). With the Etachron, when twisting the stud to close the gap, always do it so the inner pin comes closer to the stud (depends on caliber, and which way the hairspring comes out of the stud- on a 2824 you would turn the stud clockwise). I haven't figured out why it really matters, but they know what they're talking about.

 

The Hamilton book is focused more on older high precision timekeepers, more along pocket watch size, with overcoils; as you've noticed there are fewer adjustment you can make with a flat hairpsring.

 

 

Edited by nickelsilver
  • Like 1
  • Thanks 2
Link to comment
Share on other sites

20 hours ago, JohnR725 said:

Out of curiosity would you have the book found at the link below? This is because it has several pages on regulator pins spacing and talks a little about the consequences. Otherwise if you don't have that I can scan the relevant pages.

https://youtu.be/op68gwVvsCo

Then I was thinking about this discussion when I was at work on Friday so I took some pictures for you. Vintage American Waltham pocket watch recently serviced. Lift angle probably isn't right is amplitude is too high but it doesn't matter what I was doing. Pivots though probably could've been polished as the dial up and down aren't quite as identical as they could be.

Then images number one shows a problem. Dial up and down look fine but notice what happens when you go to the various crown positions the amplitude drops which is standard and the rate gets really slow.

In image number two notice the Delta site looks great and the watches running fast.

Then image number three timing looks pretty good.

A perfect example of regulator pin spacing. As a reminder with over coil hairspring is the spacing with the regulator pins should be as close as possible as long as they hairspring can slide. In the book I mentioned above and the source by the Elgin watch company they both said the same thing with flat hairspring is the spacing should be about the with or the thickness of the hairspring. I assume that would be split up across both pins.

So in the case of this watch with an over coil hairspring as soon as you go to the crown positions amplitude drops as expected in the watch is slow. Closing the regulator pins now the watch runs fast. Then finally just moving the regulator it looks fine now.

 

 

 

wt-3.JPG

wt-2.JPG

wt-1.JPG

I don’t have that Jendritzki book - please share what you can. 

Link to comment
Share on other sites

24 minutes ago, nickelsilver said:

There's a reprint available from Editions Simonin, 74 bucks, and worth multiples of that.

 

https://www.booksimonin.ch/fr/3706-watch-adjustment.php#corps

I will buy a copy. 
 

More questions on the Etachron system: 

1. Let’s assume the Hairspring is perfectly center between the regulator pins. What happens to the timing of the pins are rotated closed? What happens if they’re rotated open? 
- My understanding is this will increase or decrease the rate in the pendant positions relative to the horizontal positions. Will it equally increase/decrease the rates to the pendant positions or does this action favor one of the four? 
2. Let’s assume the regulator pins are rotated closed just right. What happens to the beat rate when the stud is rotated inwards/outwards? 
- Does one of the pendant positions increase/decrease more pronounced than another? 
 

I’d like to understand if there is a general action and reaction when adjusting each etachron. But where this is leading, let’s say CH, FH, 6H, and 3H are beating at a consistent rate (within 1 s/d) but 9H and 12H are significantly out of beat (10 s/d) is there a adjustment to the etachron that will align the vertical rates? 

Link to comment
Share on other sites

28 minutes ago, mzinski said:

I will buy a copy. 
 

More questions on the Etachron system: 

1. Let’s assume the Hairspring is perfectly center between the regulator pins. What happens to the timing of the pins are rotated closed? What happens if they’re rotated open? 
- My understanding is this will increase or decrease the rate in the pendant positions relative to the horizontal positions. Will it equally increase/decrease the rates to the pendant positions or does this action favor one of the four? 
2. Let’s assume the regulator pins are rotated closed just right. What happens to the beat rate when the stud is rotated inwards/outwards? 
- Does one of the pendant positions increase/decrease more pronounced than another? 
 

I’d like to understand if there is a general action and reaction when adjusting each etachron. But where this is leading, let’s say CH, FH, 6H, and 3H are beating at a consistent rate (within 1 s/d) but 9H and 12H are significantly out of beat (10 s/d) is there a adjustment to the etachron that will align the vertical rates? 

1- Rotate closed, all rates increased*. Rotated open, all decreased*.

2- If you rotate the stud, you shift the centering of the hairpspring. This can have multiple and surprising effects.

 

** Changing the pin gap can have vastly different effects from horizontal to vertical depending on the size and overall quality of the watch. The flat hairspring can deliver astonishingly terrific rates, in positions and isochronally. Again, I mention my friend who methodically trues and tidies up the hairspring before even looking at the machine. The gap at the pins becomes the big thing when all else is good. Play with the gap when you're sure everything else is perfect; when you get it too tight everything goes awry, same with too loose. As John mentioned above, the gap is as a rule of thumb the thickness of the hairspring, spread over the gap. But tiny variations on that can have big influences on rate when hunting down chronometry.

 

And to a question from Joe above-ETA has some sort of lower grade ruby they use for pallet stones on lower grade watches. The Chronometer spec ones have real "natural ruby", which is actually produced and not out of this earth. Both work great, as many have observed "pedestrian grade" 2824s running like  marine chronometers with the crappy pallet jewels and all.

  • Like 2
Link to comment
Share on other sites

6 hours ago, nickelsilver said:

There's a reprint available from Editions Simonin, 74 bucks, and worth multiples of that.

Yes the reprint is really nice and it slightly updated from one of the earlier editions.. I used to have a copy but I made a mistake of loading it to my boss who has decided never to return it. Amusingly he wasn't paying attention 100% and has embraced something it's probably in the book which has to do with the centering of the hairspring. So he likes they hairspring not to be centered because of something in the book. Except that also goes against what the book says. Fortunately I have an earlier edition and tomorrow all skin is basically just two pages of important stuff for the regulator pins. Not that the rest of the book isn't really important but there just to sections so they should be easy to scan in as a sample of the book So you can see what it looks like

I was thinking because you have all those nice watches. Take the one that is running the worst if there is such a thing and experiment. I was curious if you open the regulator pins all the way with the hairspring centered whether it would still touch the pins? I know if you had a witschi timing machine you can go to frequency mode if you didn't have the regulator pins at all and you did multiposition timekeeping whether that would be a way of telling that you have an actual poising error versus a regulator pin issue? It be interesting experiment to try but probably not on your chronometer grade movements of for you probably may be able to remove the regulator pin assembly as I'm assuming it snaps in the same as the stud?

In any case without going so dramatic you could open up the pins and see what happens on both a high and the low amplitude. The good find out what happens if you adjust the centering off so you're on one in versus the other. All the tests have to be done at both a high and the low amplitude. This way can see what the effect of doing all this is

5 hours ago, nickelsilver said:

ETA has some sort of lower grade ruby they use for pallet stones on lower grade watches.

I've seen the lower grade and probably one of the manufacturing information sheets specifies what it actually is. The lower grade sapphire is pink in color it's not translucent. To me it looks like they took sapphire dust and somehow made it solid. So it has a nice pink color but not a nice transparent pink.

I did look up out of curiosity when Joe asked the question and on the julesborel.com Website They don't have many of the watches from this company but they did have the 200 I was curious as to whether there were any parts specifications differences in other words is there a different pallet fork. B is often times when you look at the parts list you don't actually indicate that the chronometer grade is different. What becomes interesting though on the website is you get a balance wheel complete for $40 and a balance for the abbreviated And are using this words  BALANCE, CHRONO Which I really assume means balance chronometer grade that will set you back $10 more. So otherwise if you believe the parts list that would be the only parts difference. At least for the  SW200.2209.01.A.

 

  • Like 1
Link to comment
Share on other sites

I think I got all the stuff on regulator pins plus a little bonus material

then kind of as a reminder perhaps? The book is filled with everything you need to know to do precision timekeeping. It's filled with theory is ideas concepts everything which is nice except? You have a finished product the product is supposed to be chronometer grade? Okay went back to the tech sheet it's a top grade. Basically identical to the chronometer grade except it hasn't passed chronometer testing. Cleverly it doesn't say what the difference is? In other words what exactly do you really have to do? I was in the impression it was just possibly little tweaking here and there. Is where you want to be careful not to start to do things that maybe aren't necessary even though you have a perception that you need to do them to achieve the goals you desire?

Won't be interesting is is the track down the companies were using this movement who get the movement chronometer certified then see if he can find the person who does the actual adjusting. Person is probably locked away in some safe fault somewhere is the most valuable possession in the company at finding that person might be a challenge. Then yes someone one of the vintage books it talked about the person who could do the adjustment to get the watch to keep precision time basically was the most valuable on the company.

Normally in the images I trim things a little tighter but I gave you bonus material. Then each paragraph is a number which usually corresponds the picture which isn't necessarily on the same page.

Then I forgot to ask I noticed the print outs up above? What you think of your nifty new timing machine?

Image pt4.jpg

Image pt3.jpg

Image pt2.jpg

Image pt1.jpg

  • Like 2
Link to comment
Share on other sites

On 6/14/2022 at 4:10 AM, JohnR725 said:

Then I forgot to ask I noticed the print outs up above? What you think of your nifty new timing machine?

I've been very pleased with it. It makes testing full and half wind in six positions a breeze having that automatic mic. I feel fairly confident in the timing too. Upon receiving it I did a timing test on my Tudor BB58. I captured the full wind and half wind in six positions. I then tested the watch in three of the positions to see if it performed as expected based on the timing machine readout - it did. So at this point I believe the numbers. 

Link to comment
Share on other sites

4 minutes ago, mzinski said:

It makes testing full and half wind in six positions a breeze having that automatic mic.

Yes you really get spoiled with an automatic microphone. I actually have a timing machine sitting right in front of me at work and I'd rather use the one behind me with the automatic microphone.

5 minutes ago, mzinski said:

So at this point I believe the numbers. 

Just in case you need a comparison I was commenting to Sam about your machine and its paper tape Print out. I really think a predator would be a nice addition to our machine. In case you're curious about the numbers Sam did give you an invitation bring a watch by and put it on the machine see what it's up to. Although I have a suspicion he has another motive of offering you a precision timing adjustment. But then I suggested if he does a precision timing adjustment would he guarantee that the watch would meet and pass the chronometer specifications?

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.



×
×
  • Create New...