Jump to content

Rate difference between flat and hanging


Recommended Posts

So lets say a balance/hairspring is perfectly poised and keeps the same rate in all vertical positions, but the flat positions have different rate than the vertical positions. With a regulated balance wheel I can adjust the index gap to conveniently eliminate this difference. But what if I'm dealing with a freesprung balance? What can be done? I've heard rounding or flattening the balance pivots, but... I don't know about that... In case you want to know I'm dealing with the new style ETA setup with the freesprung flat hairspring and variable inertia balance like in the powermatic 80s. I've gotten it to keep the same rate in all vertical positions but there's about a 5 second difference between horizontal and vertical. Not a big deal I know but if there is a way to improve on this I would like to know.

Edited by CaptCalvin
Link to comment
Share on other sites

If we really had an absolutely poised balance wheel statically and then dynamically poise and there really was zero positional error.. Then free sprung so there's no Regulator  pin issues  the only problem it can be is the escapement.. 

The problem with escapement's is it artificially Is controlling the timekeeping.. This is why Marine chronometers use  detent escapement's  less influence of timekeeping. I think there only in contact with the balance for just a couple of degrees versus 50 for a lever escapement.. It's also why you see some other escapement's out there  less influence.

So amplitude with the escapement is an issue. you want to have equal amplitude in all positions.  I've attached a picture it shows how the ends of the pivots are supposed to be  flat almost. This increases the friction on the ends to hopefully bring it in line with the friction on the sides.

balance pivots End.JPG

  • Thanks 1
Link to comment
Share on other sites

Yes freshly cleaned and oiled, adjusted, cleaned and oiled again.

Free sprung to me at this point seems just a cool gimmick but not all that it's hyped up to be. I've achieved some incredible results with a 7750 and nh35 both of which are plain Jane regulated balance. I have achieved similarly amazing results with a 3135 adjusting only via the microstella screws after a previous watchmaker/tinkerer threw them out of poise, but everything just kinda fell in place for me that time. After I got all the vertical positions to pretty much 0 the horizontal positions also fell in line at around 0, no extra work needed. So this is going to be a road black for me. Afraid I'm just not comfortable enough with rounding or flattening balance pivots yet and seems a very impractical way to try and eliminate a 5 second difference. 

I've also heard the stud mounting point also having an influence in the context of overcoiled HS. Is this true? If so how does it work? Would this also be applicable to flat HS?

Link to comment
Share on other sites

7 minutes ago, CaptCalvin said:

Yes freshly cleaned and oiled, adjusted, cleaned and oiled again

The reason I asked is I assume you have a servicing guide? You notice at the link below that  details seem to be lacking for this watch? Then it is this a chronometer grade movement or the standard grade?

https://calibercorner.com/tissot-caliber-powermatic-80/

Link to comment
Share on other sites

It's not actually the powermatic 80 that I worked on, but rather a Longines L888.02 which is basically a 2892 fitted with a similar balance setup and 7bps train. I would have no idea how to oil or treat the plastic escapement in the powermatic 80s. I just mention powermatic 80 because I assumed that more people would know what I'm talking about in terms of the new balance design then if I specified that I was working on an L888.02. As for servicing guide I don't have one. I don't think there are service manuals for any of the new generation ETA movements out in the public right now so I just loosely follow the one for 2892.

Edited by CaptCalvin
Link to comment
Share on other sites

This is where a microscope really becomes (for me) essential. What looks perfectly acceptable under normal magnification under a microscope a whole new story emerges. Rounding/flattening a balance is not for the faint hearted. Unless this is your watch I would leave alone, trust me it is so easy for this to go wrong. I speak from experience. 

Link to comment
Share on other sites

7 minutes ago, CaptCalvin said:

It's not actually the powermatic 80 that I'm servicing, but rather a Longines L888.02

When asking for help it's really important to tell us  what  you'd need help with. Like for instance  if you had the powermatic 80 You didn't lubricate the escapement correctly.  You probably didn't have the proper tools to adjust the regulation of the balance wheel. Then the positional error is within specifications.

10 minutes ago, CaptCalvin said:

As for servicing guide I don't have one. I don't think there are service manuals for any of the new generation ETA movements out in the public right now

As a wild guess  the modern ETA technical documents will never be out in public.  Both Omega and Longines like everyone in the Swatch group are really paranoid  and all the documents are watermarked on each of the corners of each of the pages  with who downloaded it..

As I currently have a confusion of to watch is in my brain your guy gets stuck with  both being explained. It appears to be from your description and whatever else I'm looking at the balance wheels are more or less the same  it's the escapement's are different. So if you had the Longines Looks like and standard escapement standard lubrication.. The other one though you need special lubrication for the pallet and another lubrication for the pivots.. Both lubrication's do appear to be available.

Timing specifications for the l888 lift angle 53° + or -5s a day. Then for the specification that you're concerned about we need something that only seems to exist for one of the watches. They probably exists is just much harder to find and that is  the manufacturing information sheets provided by eta are where you find all the nifty technical specifications and timing.. So they go by the Watch that you don't have But conveniently I do have access to timing specifications  the maximum positional deviation is18 s/d .

One of the really nice thing about wristwatches  on a wrist versus on a timing machine is? On a timing machine it's in a static position you can observe in their you can get all bent out of shape because your watches off  on the other hand if it's on a wrist it moves around  moving around tends to Average out the timing.  So basically a fine just the way you are leave her alone put her on your wrist see what it does..

 

 

 

 

Link to comment
Share on other sites

3 minutes ago, clockboy said:

This is where a microscope really becomes (for me) essential. What looks perfectly acceptable under normal magnification under a microscope a whole new story emerges. Rounding/flattening a balance is not for the faint hearted. Unless this is your watch I would leave alone, trust me it is so easy for this to go wrong. I speak from experience. 

My brand new watch that I couldn't help dismantling first day :biggrin:. Glad I did too as it's doing fractions of a second a day on the wrist as opposed to around +7 out of the box. The machines they probably used at the factory splashed oil outside the cups and they used that unsightly lubrifar on the escapement so I'd say I feel better about the oil job I done than what they did anyway. 

Link to comment
Share on other sites

15 minutes ago, clockboy said:

Rounding/flattening a balance is not for the faint hearted. Unless this is your watch I would leave alone, trust me it is so easy for this to go wrong. I speak from experience. 

Personally for me  I find it  interesting  that  were trying to reduce friction In a watch and hear you want to have friction..

Then for my own personal experience as a young student I made those pivots nice and   round after all we do want to reduce friction don't wait?  Then on the timing machine  the early days of those paper tape machines  the sound of the galloping horse could be heard all over the classroom.. It's nice when you've quiet timing machines today no one knows what you did right a wrong..

As it looks like were in agreement leave the pivots alone..

 

 

  • Like 1
Link to comment
Share on other sites

In a watch with an overcoil isochronal error can be adjusted by adjusting the overcoil curve, bring the body of the curve closer to the staff to raise the rate at lower amplitude, away to reduce the rate at lower amplitude. This is only ever done with everything else perfect- escapement adjusted, regulator pins (if present) gapped correctly-- and also if the adjustment can't be made with those--. balance pivots perfect and hairspring otherwise perfectly flat and centered at the collet and from the stud.

 

Most of the time a watch that is free sprung will have an overcoil. ETA does a heat treatment of the last part of the outer coil, the "dog leg" which is very clever and allows them to achieve timing that normally is reserved for overcoiled watches.

 

You mention a 7750 and 3135 that timed superbly; these movements are particularly good, even though the 7750 has a flat hairspring, so no surprise. The free sprung versions of the 2892 are fairly new, and a 5 second difference between horizontal and vertical is really no big deal, should be COSC OK.

 

But, if you really want to chase down those 5 seconds, there are some things you can do. Flattening the pivots is one, but you would want to go slowly and check the overall rate both at full wind and after 24 hours to be sure the effect isn't overly amplified at different amplitudes. When dealing with flat pivots you have to be extra extra careful that your balance jewels are absolutely impeccably clean; a microscopic piece of dust will almost inevitably find its way to the pivot end and create sort of a ball bearing effect and your amplitude will go up tremendously.

 

Some makers have used modifed cap jewels to reduce the amplitude in horizontal; Vulcain used to use cap jewels which were conical, this would force the pivot against the side of the hole jewel. Some others use what's commonly called "Daniels" jewels (as they are described in his book), which are the opposite, a divot is created in the cap jewel slightly off center which has the same effect.

 

One thing I have only heard of through knowing some real beasts of regulating work- that is, not in any book- is to ever so slightly through the hairspring out of true at the collet. Like half a hairspring thickness. Whether that is toward the collet or away is simply trial and error. It's not enough to make the hairspring visibly out of true in use, but it really does (sometimes) catch those last seconds between horizontal and vertical.

 

Back to the 3135- I have a friend who is perhaps the most fastidious watchmaker I know. He had a 3135 that had a 2 second difference between the two flats. I know, I know... Anyway, he decided that must be rectified, if only for the experience. He tried tweeking the hairspring, recleaning, no good, finally he put a microflat on the pivot of the "fast" side and boom that did it. Didn't even really show up as an amplitude difference on the Witschi either.

  • Like 1
Link to comment
Share on other sites

5 hours ago, JohnR725 said:

As a wild guess  the modern ETA technical documents will never be out in public.  Both Omega and Longines like everyone in the Swatch group are really paranoid  and all the documents are watermarked on each of the corners of each of the pages  with who downloaded it..

Strangely enough current ETA docs can be downloaded by anyone, so can acces  e.g. the below and others 2824-2.pdf

Link to comment
Share on other sites

47 minutes ago, jdm said:

Strangely enough current ETA docs can be downloaded by anyone, so can acces  e.g. the below and others

your only sort of correct? If you go to Eta website you can download current documents but the original question related to some specific numbers which I'm going to give you. So as soon as you find them I'd be curious to know where I could download these from. Although thinking about it I didn't check cousins they might have them  but in a case  I still be curious if you can find  them.

so this is what you're looking for L888.2 = ETA A31.L01 Acier and Tissot Powermatic 80.111 even though it says that it is a tissot the tech sheet is actually made by eta.. They also have the really useful  manufacturing information sheet for the same watch. 

  • Like 1
Link to comment
Share on other sites

5 hours ago, nickelsilver said:

One thing I have only heard of through knowing some real beasts of regulating work- that is, not in any book- is to ever so slightly through the hairspring out of true at the collet. Like half a hairspring thickness. Whether that is toward the collet or away is simply trial and error. It's not enough to make the hairspring visibly out of true in use, but it really does (sometimes) catch those last seconds between horizontal and vertical.

Spot on!

It only needs to be out a fraction and that's where those few seconds are gained or lost.

Personally, 5 seconds between vertical and horizontal is good for me. Because it is really 2.5 seconds difference. So let's say you gain 2.5 seconds on vertical and lose 2.5 seconds on the horizontal reading, that's a total of 5 seconds difference. Or am I missing something @CaptCalvin?

2.5 seconds difference statically on a timegrapher, may be something less on the wrist. I concur about timing 7750's, which I have found amazing to regulate and adjust. The one I'm wearing gains 30 seconds a month... Perfect!

  • Like 1
Link to comment
Share on other sites

if you want to understand how to adjust the watch there's a book. Go to the site below  do a search for this number 3706 which is easy way to do it and the book comes up. the reason why you want to go to the website below for the book is it doesn't seem to be in stock on Amazon anymore what you're looking for would be Watch Adjustment  by H. Jendritzki (Author) the reason you're going to the site below  is he's the publisher.. 

then what sort of wonderful stuff can you find in this book?  The image for changing the pivot shape that came out of the book.. Lots and lots of interesting stuff on everything affecting timekeeping.. But a lot of it is totally irrelevant unless you're designing the watch. Basically you have what you have  and  its within specifications. other than adjusting the timing screws on the balance wheel not much else you can do other then if you adjust the screws incorrectly can probably damage something which is why you're supposed to use a special tool which I assume you don't have..

 

http://www.booksimonin.ch/

  • Like 1
Link to comment
Share on other sites

7 hours ago, nickelsilver said:

In a watch with an overcoil isochronal error can be adjusted by adjusting the overcoil curve, bring the body of the curve closer to the staff to raise the rate at lower amplitude, away to reduce the rate at lower amplitude. This is only ever done with everything else perfect- escapement adjusted, regulator pins (if present) gapped correctly-- and also if the adjustment can't be made with those--. balance pivots perfect and hairspring otherwise perfectly flat and centered at the collet and from the stud.

What would be the theory behind this? I've read books and browsed forums that tell me what needs to be done to achieve what results, but having a hard time finding out the why's of the relationship between procedure and result especially when it comes to hairspring adjustment.

5 hours ago, jdm said:

Strangely enough current ETA docs can be downloaded by anyone, so can acces  e.g. the below and others 2824-2.pdf

Current in the sense that they're still made, but nothing can be found for any of their newer stuff. No doubt they're going to try to phase out their older stuff somewhere down the line as patents expired. A shame too because I've had much better results with etachron than their new freesprung stuff. Sure in the bigger picture both are held to pretty much the same specifications out of the factory and average consumers put up with them and are probably dazzled by the novel gimmick, but for a tinkerer like me who likes to squeeze every bit they can out of their toys etachron has definitely been giving me more for my money's worth up to now.

1 hour ago, Jon said:

Spot on!

It only needs to be out a fraction and that's where those few seconds are gained or lost.

Personally, 5 seconds between vertical and horizontal is good for me. Because it is really 2.5 seconds difference. So let's say you gain 2.5 seconds on vertical and lose 2.5 seconds on the horizontal reading, that's a total of 5 seconds difference. Or am I missing something @CaptCalvin?

2.5 seconds difference statically on a timegrapher, may be something less on the wrist. I concur about timing 7750's, which I have found amazing to regulate and adjust. The one I'm wearing gains 30 seconds a month... Perfect!

To me regulation is trivial. The real challenge is to eliminate the difference in positions. 0/+5 is just as good as +2.5/-2.5 in my eyes in terms of movement quality.

40 minutes ago, JohnR725 said:

if you want to understand how to adjust the watch there's a book. Go to the site below  do a search for this number 3706 which is easy way to do it and the book comes up. the reason why you want to go to the website below for the book is it doesn't seem to be in stock on Amazon anymore what you're looking for would be Watch Adjustment  by H. Jendritzki (Author) the reason you're going to the site below  is he's the publisher.. 

then what sort of wonderful stuff can you find in this book?  The image for changing the pivot shape that came out of the book.. Lots and lots of interesting stuff on everything affecting timekeeping.. But a lot of it is totally irrelevant unless you're designing the watch. Basically you have what you have  and  its within specifications. other than adjusting the timing screws on the balance wheel not much else you can do other then if you adjust the screws incorrectly can probably damage something which is why you're supposed to use a special tool which I assume you don't have..

 

http://www.booksimonin.ch/

Does this book explain the theories? I've read another book called "Practical Watch Adjusting" by Donald de Carle and it's basically "if the watch does this you do this." No real explanation as to WHY certain procedures would have certain effects. If the book you mentioned goes in more depth I would definitely like to have a look.

Link to comment
Share on other sites

 

1 hour ago, CaptCalvin said:

Does this book explain the theories?

unlike the book that you mentioned which I do have  this thing is way more detailed. Goes in the details about  every thay I kinda consider at the must-have book if you want to understand timekeeping..

  • Thanks 1
Link to comment
Share on other sites

CaptCalvin, there are numerous books on regulating, but they are geared towards geometrically and proportionally "correct" timepieces. I.E. for small wristwatches things break down. I had an A.S. 1012 ladies movement in once that I was unable to break 80 seconds delta with. It was used in a high end watch from a "major player", but back in the 70s dark period. I did it directly for the manufacturer as a "restoration" , and man, I changed the pallet hole jewels to olived jewels, nothing. Switched the balance hole jewels too as they appeared to be straight under 40x mag. Nothing. Escapement pushed to its max nothing. New balance staff. Nothing. Reduced balance staff pivots by 5 microns- nothing. 0.01mm, nothing. New staff with maybe 0.002mm clearance in the jewels- - - - nothing. Spoke to perhaps one of the best living watchmakers about it and this is what he said:

 

"Take the movement out of your movement holder, and with careful aim, throw it in your trash can. The proportional dimensions of everything in the escapement are such that the balance pivots are far too big, same for fork, the hairspring collet is waaaay to big, and you'll never do better."

 

Despite the fact that I've seen some (Bulova) derived A.S. 1012s that had a delta closer to 30 in 6 positions I tend to agree with his sentiments.

 

I had a Zenith deck watch in once, this thing was a beast. Initial test it had a delta of like 250 seconds. This is a watch so precise the regulating pins had a screw to adjust the gap.

 

Start from zero. Poise the balance. It sunk like a brick to one position, ha, there were like 5 timing washers there. Took them off, a little light tweeking and boom perfect. Service, adjusted escapement, hairspring gap (some genius in LeLocle made that easy) and I swear it was in something like 5 sec/6 posotions 0h and 24h.

 

So someone messed with a good thing, wrongly. But the gist is a bigger watch is easier to get really precise. A Rolex 3130 is right at the edge of good proportions, ETA 2824 and 92 are pushing it more but both of those movements have lineage dating back 50 or more years with a lot of innovation (all the ETA stuff got it) and even if boring the ETA engineers are darn smart, really smart.

 

 

 

 

Link to comment
Share on other sites

To give an update: The situation improved itself after having it on my wrist these few days.

Out of the box:20200529_213046.thumb.jpg.a09f386b2234ffbc7f2c3ce3cd82317f.jpg20200529_212937.thumb.jpg.02152a19a35870c3775632c81b29c49a.jpg20200529_212747.thumb.jpg.a5452fd7cb4f7e5db13ebcd0e5f5f8df.jpg20200529_212532.thumb.jpg.52345777d77fc0aea3df3a440e1fc9df.jpg20200529_212346.thumb.jpg.b4717d747362e7a2ec09651a1688f3e0.jpg20200529_212233.thumb.jpg.cf8b88d8a8fdd6a76f6f8c578fc9a6a9.jpg

Not sure how the gifter would've felt about this but I just couldn't help myself so... bits and pieces one day one:D:

20200604_190240.thumb.jpg.cd363692ddb24aa961d25a597927dcd0.jpg

Shhhh... not planning on telling her any time soon:ph34r:.

Now:

20200616_141043.thumb.jpg.3b083c88d3782ae9b59d305ee8da55bb.jpg20200616_135957.thumb.jpg.e66df9a1b151bfa794b855958d5022c6.jpg20200616_135716.thumb.jpg.52442307106480504c88fd28ea5cdd85.jpg20200616_135621.thumb.jpg.6df2f3c0992917faee670263cbce7ed9.jpg20200616_135452.thumb.jpg.add82051a71bd83a98fd74f1a9da25c7.jpg20200616_135214.thumb.jpg.954ed420d07a9c4a61ae4bab5c999228.jpg

Pretty happy with that. Dead accurate on wrist:thumbsu:. Healthy boost in amplitude as you can see. What do you guys think?

Link to comment
Share on other sites

On 6/13/2020 at 4:32 AM, CaptCalvin said:

I'm dealing with the new style ETA setup with the freesprung flat hairspring and variable inertia balance

Impressive results! :Bravo:

Edited by VWatchie
Link to comment
Share on other sites

On 6/13/2020 at 1:17 AM, CaptCalvin said:

I've also heard the stud mounting point also having an influence in the context of overcoiled HS. Is this true? If so how does it work? Would this also be applicable to flat HS?

I've posted this video a handful of times. Great lecture. Addresses this question directly and in great detail with lots of charts and probably some formulae:

https://www.youtube.com/watch?v=DpDH6fLumE8

Link to comment
Share on other sites



  • Recently Browsing

    • No registered users viewing this page.
  • Topics

  • Posts

    • That would be something! Which brings me back to;  
    • you think you're going to sleep tonight you're not, you're going to ponder the question of what makes you think those of the right parts?  
    • Does anyone knows what size case a need for a dial diameter 20.60mm?
    • Sounds like the story with my Rolex. Poor (expensive) job done by an official Rolex dealer with an "in-house" watchmaker, hence I learned watchrepair and did the servicing myself. Same story as I learned with the Omega 861, again poor job by an "in-house" watchmaker by an official Omega dealer. Once your watch goes through that back-door, you have no idea what is going to happening to it 🫣   Quite nice that they sent back the parts which had been replaced !
    • yes that's definitely not right at all. I have a picture one of my friends has a Omega coaxial there was having issues to lose asking me where he should send it. As that's a specialty watch I suggested the service center. When he got it back he sent me a picture so the replace the dial as you can see the hands the mainspring barrel and I think the price was really quite decent considering all the stuff they can replace. So I do know they do change the barrels but the other person I worked at the service center when I would ask questions and unfortunately I can't remember all the answers. I think a lot of the changing a parts is at the discretion of the watchmaker. Plus I don't know enough about the chronographs and whether that would be considered a vintage watch? I take some of the vintage watches may have been sent directly to Switzerland or another service center. Obviously with a watch like the one down below they probably have a infinite supply of parts is its relatively modern vintage stuff becomes more interesting even the watch companies don't have necessarily infinite supply of parts. But no matter what the watch shouldn't disintegrated six months that's definitely an issue.        
×
×
  • Create New...