Jump to content

Jewels are over rated?


Recommended Posts

While I wait for Coca Cola to hopefully free the ratchet screw on my Elgin movement, I tried my hand at servicing a  66 year old dollar pocket watch that certainly wasn’t made with being serviced as a goal.  Tore down the movement, cleaned and lubricated.  Rather fidily getting it put back together given the poor quality of all the parts.

Thought I’d share some pictures and the result.

 

EFA368BB-451C-4BB6-A3C4-B2CE670BEF80.jpeg

852DE335-C6AF-41CB-A361-1F17F73CA502.jpeg

71AC33B0-6435-49DF-9012-049E07D64966.jpeg

07C87127-6016-47F2-ABC9-857F70A04670.jpeg

2E82FABE-6D04-4309-9F3D-D41C9B353A8C.jpeg

  • Like 5
Link to comment
Share on other sites

I have repeatedly said that I have no experience with this sort of equipment but would appreciate someone versed in this to explain photo number five and how the lack of jewels explains the variations illustrated.  Most would say that, quality, jewel count, efficiency, accuracy and longevity are all attributes absent from the above watch.  If the above watch's parts were properly manufactured (to print), assembled and repaired following to factory process, any variations while running should not be expected.  I'm not trying to question the OP's abilities in any way, just what picture five means about jewel counts.

I look forward to a discussion.

Share

Link to comment
Share on other sites

27 minutes ago, Shane said:

I have repeatedly said that I have no experience with this sort of equipment

https://vintagewatchinc.com/using-watch-timegrapher/

I'm not sure that you can infer anything about jewel count from a freshly serviced watch that has no jewels. Jewels are going to minimize metal-to-metal wear as well as providing a surface harder than steel for the steel pivot. I would assume that a freshly serviced watch that had just been lubricated ought to give better readings than shown but I don't know if a lift angle of 36° is correct for the watch shown or if 133° is a reasonable amplitude for the watch. Neither would be close for a modern watch.

All that said, who cares if a so-called "dollar" pocket watch is gaining 38 seconds a day?

The watch was never sold as or intended to compete with a chronometer 🙂 

  • Like 1
Link to comment
Share on other sites

52 minutes ago, grsnovi said:

Jewels are going to minimize metal-to-metal wear as well as providing a surface harder than steel for the steel pivot. I would assume that a freshly serviced watch that had just been lubricated ought to give better readings than shown but I don't know if a lift angle of 36° is correct for the watch shown or if 133° is a reasonable amplitude for the watch. Neither would be close for a modern watch.

All that said, who cares if a so-called "dollar" pocket watch is gaining 38 seconds a day?

Yes, but I am only taking about the variations in how well it is running.  Everything listed should still be consistent throughout the time sample.

I will look at the link above tomorrow.

Shane

Edited by Shane
Link to comment
Share on other sites

What you're looking at is basically an early Timex. The Timex people in the group will get upset that I use it as a reference. The Timex watch is made to be inexpensive easier to manufacture cheap and it was meant to be reliable and work well. These are along the same lines they were designed for easy manufacturing inexpensive and they will work very well.

Although the numbers under timing machine do not necessarily agree with the paragraph I just did. It's not looking too good on the timing machine not too good at all. But who cares what it looks like of a timing machine if the hands move and it keeps time does it keep time?

Really if you want to start to make comparisons you should compare things that are similar. For instance compare a seven jewel American pocket watch to 17 jewel American pocket watch. You'll find that in the seven jewel watch a lot of the holes are no longer around. This results in poor transmission of energy from the barrel to the escapement. The number of jewels does make a difference. So you typically never find it out of round jewel because you just typically don't. You can find crack to jewels broken jewels but they typically don't wear themselves out of round.

Then with jewels there is a limit of at which point in time they are a waste of time and effort other than to impress people. Seven jewels is the minimum typically you would see for a low jewel count watch. Not entirely I've seen less but typically that would cover the balance wheel upper and lower holes. This would be the hole jewel in the end stone on each end that makes up quantity of four. The pallet stones themselves two of them. And finally the roller jewel on the balance that gives you your seven.

They had various combination of jewels in the gear train that would bring up the 17th. Although in the early days of watches you'd see Council 15 sometimes the jewels only be on the backside where you can see them and not on the other side. Yes I always find that amusing on American pocket watch is pretty jewels on one side where the customer sees the watch with a buying it but not on the other side with the dialysis

as soon as you start jeweling the mainspring barrel your brain account up to 21 May be 23.

You look at some modern watches that would give you really high jewel accounts because they put them in really useful places like around the back edge for the automatic weight goes just in case for some reason it decides the balance off the plate which it's ever going to do or they put them in other interesting locations that have absolute no use at all but you bumped account up because the higher the jewel count be more impressive the timepieces.

But as you can see here if you design your timepiece with really bigger place heavier construction do you really need jewels? Well you might need them to please the timing machine it's looking a bit peaked. Plus if you look at clocks clocks typically don't have jewels although some do some even have ball bearings and set a jewels.

On 6/14/2022 at 7:17 AM, EastCoastChuck said:

Rather fidily getting it put back together given the poor quality of all the parts.

Not just the poor qualities but when you get to some of the modern watches or more modern where they were assembled by machinery they were never meant to be assembled by a human person and that can present a challenge of trying to get it together. I remember watching a video of a one the Japanese watches quartz being assembled. Automated machinery put every wheel in place every wheel stayed in place didn't fall over. The main plate came down of course all the pivots went through all the holes which is never going to happen if you're putting a watch together.

3 hours ago, Shane said:

I have repeatedly said that I have no experience with this sort of equipment but would appreciate someone versed in this to explain photo number five and how the lack of jewels explains the variations illustrated.

I get in trouble for saying this but you could do a search we've talked about timing machines before. But that is scattered all across the message board. We have technical documentation. A video link has posted below that I'm going to read in a few minutes. Then you also have a crystal ball reading? So why am I saying that easy I just open my mouth the words come out it's the microphone and magically it appears on the screen sexy quite easy to say. But what it really means is for instance I'm attaching a PDF and basically the last generation of timing machines the paper tape machines you would have a book that would come with it and it lists the standard charts this one is really good because it's little better than the standard charts.

Then depending upon the machine the books are different Greiner wrote a really nice book for their machines as it explains in a lot of detail exactly what the machine does how it works the meeting of the charts everything it's a really big file size which is why I'm not attaching it.

Then I once bought a timing machine manual off of eBay or basically repeat which included a lot of manuals and one of them was something older where I have a suspicion whatever assembled it was probably on drugs or drink a lot of alcohol because are all these peculiar interpretations of what the waveforms mean. This is why make a joke of crystal ball reading it's not something you instantaneously grasp on the timing machine what everything means.

Then there's a problem of this is not a jeweled lever escapement and I haven't played enough with pin lever escapement's on the timing machine to know exactly how they're going to work out as far as the amplitude goes.

The beat is interesting that is a really low beat watch but watches do come in a variety of frequencies so that conceivably looks right. The left angle is really interesting I wonder where that came from? The watch appears to be out of beat but that's easy enough to check if you care and to be honest I wouldn't care if it was mine as long as it appeared to be running.

Then websites running fast that he could regulate four and beyond that my probably wouldn't get too excited about things big thing is if it runs in it keeps time within a few minutes a day overnight then you've done outstanding.

 

3 hours ago, grsnovi said:

I'm not sure I like this pros and cons how can there be a con of a timing machine?

For those reading the article at the link there is one thing that's interesting that anyone repairing watches on this discussion group should think about Is The part about monitoring the health of your watches. This is where if it I like the time before and after so if you have a watch that's running and you run a timing on it and evaluation after you service it it should be running better than it was before or equal. It's running worse you need to reevaluate your servicing procedures.

If you keep track of the watches your service saying and what it was doing and especially as are starting out every few weeks months whatever check your watch again and see how it's doing. If a year from now all of your watches look like they're dying then you may want to reevaluate servicing procedures. In case you think I'm joking about this I'm not if you have interesting choices of lubrication like for instance anchor oil it's very inexpensive oil much cheaper than the Swiss obvious is a good purchase because it's an expensive. When I originally heard about it it had an interesting characteristic of going bad superfast and basically getting really sticky. So sticky that your watch with stop within weeks of using it.

Being curious about that I purchased a bottle and I was very disappointed. The bottle I have is super thin and the test watch I had it did come to a stop but the oil also spread itself all over the place it did not stay where you put it but I was really looking forward to the watch coming to a stop within a couple weeks and it took much longer than that.

Yes the reference to if you're selling the watch show your timing machine resultsImpress the customer and make more money. Or entertain people like me who laugh hysterically at the eBay sales where they show their watch and it looks pretty pathetic at least sometimes. That would indicate to me that the seller doesn't actually know what the timing machine results are supposed to look like but who cares it's entertaining for me.

On the disadvantages is really weird I wonder what watches the author has it looked disappointing on The timing machine? Of course that does bring up the problem of buying used watches or any watch is that it's a sealed up unit and you don't actually know what it's doing. Then you do have to be careful of the timing machine in a sealed watch especially if it's a heavy case Anything resembling a diving watch oftentimes the timing machines have a really hard time picking up the proper signals and the watches can look bad when they're not. I've seen this with Rolex watch is Seiko watch is. Seiko has issues sometimes because the movement ring is plastic and it doesn't transmit the vibration properly. So sometimes it's just the casing of the watch that allows the timing machine the think it's bad

Not that it's going to really matter but they do make a reference to the 1000 models. Is actually more models and what they're indicating there is some minor programming differences also but that would really be a discussion for someplace else because the machines are basically very similar just as somehow some minor enhancements or basically a variation on the programming. And ideally if you can afford at the 1900 is a little better machine. Appears to be made by the same people it make the 1000 machine

Then I've talked enough not got finish reviewing the rest the article there are some YouTube videos on the timing machine out there. There's also additional technical information if you really get obsessed about this.

 

Timing-Machine-Charts.PDF

Link to comment
Share on other sites

On a side note I have just finished a batch of Swiss Ladies watches I had been putting off doing because of their size, or lack of size I should say.

All where listed as 17 jewel and all of them fell well short of 17 when they where actually counted. Yes I did remember the 2 pallet jewels and the impulse jewel.

Only did this with the Ladies watches as I had time to kill full size watches my also overstate their jewel count as well.

Is it a common practice with makers to claim more jewels than they actually fit.

Back to the OP could the erratic timing display be down to the watch using a pin pallet, I have read that timegraphers don't like pin pallets, is that the sort of display you get with one ?

Link to comment
Share on other sites

I did a bit more fiddling after the original post, and the beat error is now showing at .4 ms.

Still plenty of variability across positions, but nothing surprising there.  I did reshoot the balance in slow motion using the 180 degree method and think the lift angle should be closer to 39 degrees. 

I’ve read that these cheaper Chinese timegraphers are not ideal for pocket watches, so that could be part of the mix in how it’s registering amplitude. Even fully wound it shows app. 144 with the gain on its highest setting.  It could also be a tired old mainspring as well, but it does make it around 30 hours on a full wind.

  • Like 1
Link to comment
Share on other sites

2 hours ago, EastCoastChuck said:

I’ve read that these cheaper Chinese timegraphers are not ideal for pocket watches,

I would be curious if you remembered where you read that? I know one of the problems with doing pocket watches is everybody keeps looking for the lift angle list which doesn't exist. Plus of you take the time to make a lift angle list the lift angles vary by quite a bit as opposed to the generalized rule that they should be this.

The reason I find this interesting as I once did a comparison of the 1000 with my witschi machine and the watches I was using were pocket watches. And for the most part the 1000 does really really well. Then later I was able to borrow 1900 and because the programming is basically identical the hardware is very similar if not the same it had the same results. So I don't see pocket watches be in it issue with the timing machine other than the lift angle. The size the microphone can be an issue because they're not designed to hold really big watches. Then as a few enhanced features that which he has that would be nice but the Chinese don't have but otherwise the Chinese machine does quite well with pocket watches and just about anything else.

2 hours ago, EastCoastChuck said:

Even fully wound it shows app. 144 with the gain on its highest setting.  It could also be a tired old mainspring as well, but it does make it around 30 hours on a full wind.

One of the things I'm not sure about is with a different type of escapement that is similar to a jeweled escapement with of the timing machine will understand what it's seeing and do the amplitude.

Then you're tired old mainspring it looks tired in the picture up above. What did you use the lubricate the mainspring the watch and the escapement?

Link to comment
Share on other sites

44 minutes ago, JohnR725 said:

What did you use the lubricate the mainspring the watch and the escapement?

The mainspring was the easy choice, with mobius 8200.

I was at a bit of a loss as to what to do with the escapement.  Given how the whole thing is assemble there really aren’t any low friction areas, so I just opted for mobius 9104.

Link to comment
Share on other sites

1 hour ago, EastCoastChuck said:

I was at a bit of a loss as to what to do with the escapement.  Given how the whole thing is assemble there really aren’t any low friction areas, so I just opted for mobius 9104.

I'm sure the subject of lubrication of This type of escapement's is come up in the past. So my fuzzy memory of lubrication of these type escapement's would be to use the standard 9415 which is what I typically do.

Then lubrication of this watch really doesn't matter at all. But there is a minor problem with the 9104 if you look at the very bottom of the lubrication chart. Notice they are recommending D5 would be the replacement for what you have. Then as typical they don't explain why and of course this conflicts with some of the other Swatch group lubrication recommendations.

HP oil problem on brass.JPG

tableEN.pdf

Link to comment
Share on other sites

12 hours ago, JohnR725 said:

 I know one of the problems with doing pocket watches is everybody keeps looking for the lift angle list which doesn't exist. 

There is a way to determine the lift angle without a list john a little time consuming and not 100%. But very close if you take the time to be accurate in your method. It does require the movement to be able to reach 180 amp though, would old pocket watches struggle with this ? I'm sure you know of it. 

Edited by Neverenoughwatches
Link to comment
Share on other sites

1 hour ago, Neverenoughwatches said:

There is a way to determine the lift angle without a list john a little time consuming and not 100%. But very close if you take the time to be accurate in your method. It does require the movement to be able to reach 180 amp though, would old pocket watches struggle with this ? I'm sure you know of it.

yes even though there is a relatively easy way which we've discussed before on this message board. You still see the continuous asking on message boards for a list or what's the lift angle of my watch. there seems to be an assumption that because the timing machine exists it needs a lift angle that the lift angles have been listed since the beginning of time. The failure to grasp that the modern timing machine inexpensive readily available is relatively new and the lift angle list for the last hundred years does not exist.

there are now more videos on YouTube becoming more common. The one at the link below is my favorite. It's been discussed before on the message board where I had suggestions. Like for instance I use a liquid ink that fluoresces beautifully under UV light. It really makes it easy to see the balance wheel much easier than just a mark on the balance wheel.

https://youtu.be/-Xgcck692js

but is still at minor little question I have this isn't a lever escapement. Well is a lever but it's a pin lever how does that differ by the sounds compared to a jeweled lever. then if it does sound different will let have any affect. Plus a watch of this quality isn't super precise will that interfere with the timing machine? In other words what were seeing may be perfectly normal for watch of this type or it may not be perfectly normal we need to find another example of the watch of this type to put on timing machine the see how it looks.

  • Like 1
Link to comment
Share on other sites

I am probably just being my usual dim self here but watching that video where he talks of the balance moving 180 degree's it looks to me that the red dot is moving about 350 degree's.

How am I seeing that wrong, the red dot starts and finishes at almost the same place surly that's almost 360 degrees how am I seeing that wrong ?

Link to comment
Share on other sites

55 minutes ago, Paul80 said:

I am probably just being my usual dim self here but watching that video where he talks of the balance moving 180 degree's it looks to me that the red dot is moving about 350 degree's.

things get even more confusing if you look at other reference material. I suppose it depends upon which reference book you're looking at? Yes there is some reference books that will say that the  180° we see is really 360°. This means the balance wheel can go about 600 and something basically twice when it can go now. Then there is the definition with the use turns and those numbers can get really confusing.

I have an image showing things not sure if it's going to be confusing or not. This of course is why don't worry about looking at the balance wheel I'm happy that I have a machine that just does this for me sort of. The sort of aspect of course is the watch itself has to agree with whatever the timing machine says in order for the timing machine to be rushed. In other words if the balance wheel looks pathetic in the timing machine is happy with this number is really nice and high the numbers are wrong. It's what happens of the timing machine picks the wrong part of the waveform it will give you the wrong results.

amplitude observation visual balance wheel.JPG

Link to comment
Share on other sites

On 6/19/2022 at 11:47 AM, EastCoastChuck said:

It could also be a tired old mainspring as well, but it does make it around 30 hours on a full wind.

As I understand, a week mainspring will induce a lower amplitude, due to a reduction in the available force the spring can exert through the train and into the escapement while it relaxes but the over all run time should remain the same due to the length of the spring and number of turns the spring is wound through is unchanged (as long as the watch does not just stop prematurely from it's low amplitude).

 

18 hours ago, Paul80 said:

it looks to me that the red dot is moving about 350 degree's.

My understanding is that what you are seeing is 180 degrees clockwise +- (it then unwinds) and then 180 degrees counterclockwise +- (it also then unwinds) but it is still refered to as just 180 degrees amplitude.

From what I gather from one of the videos I have recently watched about lift angle and timing machines (linked to above... I think).  It should not matter what the amplitude is that the watch is running at just as long as you accurately change the lift angle on the timing machine so it can calculate the amplitude it is actually running at.  The machine just needs to know something in degrees to work out all the math from all the sounds it hears coming from the watch during the time sampled.  They must use 180 degrees because it's much easier to visually determine that angle.

Knowing the lift angle just helps get the other information required about the watch being tested "without the using the clutch" (automatically).

On 6/19/2022 at 12:21 AM, JohnR725 said:

a really nice book for their machines as it explains in a lot of detail exactly what the machine does how it works the meeting of the charts everything it's a really big file size which is why I'm not attaching it.

Thanks for the reply, you really put in the effort and I found the PDF very interesting.  I would really like to see the rest of that machines documentation.

Edited by Shane
Link to comment
Share on other sites

3 hours ago, Shane said:

If I understand correctly, a week mainspring will induce a lower amplitude, due to a reduction in the force the spring can exert through the train and into the escapement while it relaxes but the over all run time should always remain the same due to the length of the spring and number of turns the spring is wound through has remained unchanged (as long as the watch doesn't just stop prematurely from it's low amplitude).

then yes the quoted paragraph above is true. If you took a running watch you put in a weaker spring of the same length it would run the same amount of time everything  would be the same except the consequences of the amplitude.

Amplitude definitely can affect timekeeping unfortunately.  a whole bunch a bad things occur is amplitude decreases and timekeeping gets worse. musingly for amplitude this group loves I amplitude running watches. If you look at service sheets the factory they could care less about the amplitude sort of. They don't want it too high or the roller jewel crashes into the other side of the pallet fork that's very bad. Then they only care about the amplitude at the end of 24 hours and they only really really care what the timekeeping is. The typically everyone wants to see more than 200° of amplitude at 24 hours. Old mega for some other watches will accept less but of course it still has to fit within the timekeeping specifications.

oh and there's a word here you go to Google it isochronism and see how it relates to amplitude.

then as a really didn't want to do something at work today so instead I played with a different watch identical to the one at the link below. You'll notice at the link at says that this watch came in a variety of jewel counts one being the lowest but I think the lowest jewel count for this watch is zero. The difference between this zero jewel timepiece and the one  at the top of the page is this one is much newer doesn't appear to be worn out although the mainspring is set and I think is probably a better condition which allows for an experiment.

You'll notice in the PDF that's attached the one from witschi went against past where everything comes from then they give examples with their machine. Unfortunately there machines are very very expensive. Like in the picture there is one of their machines it's very expensive fortunately I didn't have to buy it it sits behind me at work in a very convenient place to use. It also has a nice feature of you can plug in a USB drive and push the photo button and take images of all the screens versus when I take a camera image which is barely adequate.

In the witschi manual it should explain what all the different screens are but I'm going to give you a few comments. So in the watch came and it was barely running at all and I came in for servicing. The initial evaluation is it can probably be serviced unfortunately by saying that it means that I get the service set. Fortunately it didn't appear to be worn out but it also wasn't running either not running very well.

In modern shops at least some modern shops the employee something called free cleaning. This is where the watch comes out of the case hands and dial come off. Then the entire watch is run through the cleaning machine usually a separate machine in what's called up free clean cycle. This allows the watchmaker to make all the necessary repairs in the case of like a Rolex watch your Omega watch they get obsessed with end shake of the jewels in the wheels. They seem to have considerable excitement over adjusting  these things and free cleaning really helps. Then once I get everything perfect it's disassembled and run to the cleaning machine. Typically I do not preclean because for vintage pocket watches I'm not going to adjust and shake and I is typically don't except.

Exception would be like this watch I want to see how well it was going to run or whether it was even a run at all. So I ran the things through intact and put some fresh oil here and there just to evaluate it. So that means you get that have images today of what a pin lever watch looks like if it's in reasonably good condition.

Regarding the overrating of jewels? Because jewels typically are really hard although in the early days a use garnets the little softer's is not quite as hard but it's much harder than brass. If you have a pivot steel typically with brass you can end up with a hold it's no longer round. In clock repair this occurs all of the time and then they ream out the whole input a bushing in such nice and round again. So for a 17 jewel watch still have a very long life zero jewel watches can wear out and if they're really worn out they can be a real issue to get everything in a running condition. Fortunately this watch looks like it was barely run at all

one of the images is of a time plot PDF page 10 has an example they unfortunately don't really explain how useful this is. Sometimes a notice your watch on the timing machine you'll see the numbers going up and down and if you have a time plot feature they can be really useful for looking for that. Holes that are out of round and cause binding issues uneven power of course. Bent pivots will do the same thing and wheels that are just out of round anyway. So usually on a time plot I'm looking for a pattern patterns are very bad. Fortunately watches a really good it averaging out problems for the most part unless things are extremely bad. But the time plot explains why is your run the timing cycle twice in the numbers are different is because you were in a different place in the time plot so I can be run the bottom of the cycle numbers are to be different than the top of the cycle

than the oscilloscope images also a really helpful future. PDF page 11 and 12 explains what shows pictures to little bit different than our machine has it. At the left-hand side there is something of interest which is why I have two on oscilloscope images I put a green line if you look at the green line below the green line is a dotted red line done by the machine. as you study your PDF you know what lift angle means. The oscilloscope shows the signal from the time the roller jewel hits until locking occurs. The dashed red line tells us where the timing machine is looking for the locking signal. If your watches running at a super low amplitude witschi did not expect you to be running crappy watches like that on the machine anyway. The timing machine will start looking for the signal someplace else other than where it is. So when you look at my oscilloscope images one of them shows a really nice amplitude and the other one shows a realistic amplitude. The realistic it's using the actual locking and the unrealistic is looking at the middle of the waveform.

The reason I bring this up is it's more common in Chinese machines followed by the machine at work versus the older witschi machines. The older one seemed to be much better with this ad is if the amplitude is super low. Maybe witschi in the old days was more realistic that watches are perfect. In any case in the case of Chinese machines if the amplitude goes too low it will look at the middle of the waveform and give you this really nice amplitude which is entirely unrealistic. Which is why you always have to look at the balance wheel especially if the amplitude is really good on the machine and make sure that whatever the machine says agrees with whatever you're seeing.

In any case here from time to time the timing machine was getting confused and triggering off the wrong part of the waveform

then I did not figure out what the lift angle was because I just didn't care. Ultimately if the watch runs overnight keeps reasonable time then I'm not going to about the lift angle as the amplitude looked good enough as it is

then you can see if the graphical display still wandering around and looks like one of the lines isn't quite as clean as the other one. Plus it's a little out of beat and I'm probably not going to worry about that either.

then if you feel like you need a little more material in timing machines the link below will download a really nice PDF. I would attach it as a rather large file size of the gets around 30 MB and I hate attaching really big stuff to the message board is it clutters up the message board

http://www.historictimekeepers.com/documents/Micromat.pdf

 

http://www.ranfft.de/cgi-bin/bidfun-db.cgi?10&ranfft&0&2uswk&Baumgartner_866

 

 

bfg9.JPG

bfg7.JPG

bfg8.JPG

bfg6.JPG

bfg5.JPG

bfg4.JPG

bfg3.JPG

bfg2.JPG

bfg1.JPG

Witschi Test and measuring technology mechanical watches.pdf

Link to comment
Share on other sites

  • 2 weeks later...
On 6/21/2022 at 3:42 AM, JohnR725 said:

oh and there's a word here you go to Google it isochronism and see how it relates to amplitude.

Isochronism suggests that the rate of an oscillator should not change regardless of the amplitude of it's oscillations.

If I think of a watch just as a power transmission device, Torque and RPM are it's only outputs.  A theoretical frictionless transmission will produce one Torque at one RPM with a given input Torque and RPM.  All imperfections within that system will reduce output Torque.  If the available output falls below the load applied, RPM throughout the system will be compromised.

The presence of an oscillating regulator/escapement, converting the only output "Torque" into amplitude, works with my mental model.  The idea that variations in output due to cyclical and (seemingly) sporadic variations within the system result in mirrored variations in amplitude also seems like a solid concept.

The part that I have not yet wrapped my head around is...

How does an isochronis regulator vary in rate while the watch remains motionless or moved into another position?  The above example shows increases and decrease in rate independent of variations in amplitude.  I'm trying to visualize how the hairspring reacted to positional changes with the 866.  Where all of the positions listed in the photo accurate to how the movement was actually being held?

I have reread the whole thread and the use of the term isochronism seems to argue both sides of this thread.

I have not yet had the time to read and digest all the attached PDFs.

My original question was directed at the variations within the time sample illustrated in the above photo and why should the absence of jewels explain it.  We seem to have moved past that for now.

On 6/14/2022 at 10:17 AM, EastCoastChuck said:

Thought I’d share some pictures and the result.

Was all the data displayed in that photo recorded with the watch in the position illustrated?

Still scratching my head over all of this.

Thanks for all of the discussion.

Shane

Edited by Shane
Link to comment
Share on other sites

13 minutes ago, Shane said:

I'm trying to visualize how the hairspring reacted to positional changes

When vertical the HSt slightly deforms and may start touching the regulator. In addition, the bearing friction shifts from the tip to the sides of the pivots. So it us normal for mov.ts if average quality to deviate 20, 30 secs / day or more. Fully correcting this deviation can be challenging even for top watchmaker, and it's part of the fine art of adjusting, which in the end depends on a lot of variables. 

13 minutes ago, Shane said:

 the use of the term isochronism seems to argue both sides of this thread.

In science and general technology the term merely mean its literal translation from Greek: same (or constant) time. 

However in watchmaking it's almost always synonymous of "small or no variance to power source status". That is because this was one of first problems that had to be solved. 

  • Like 1
Link to comment
Share on other sites

7 hours ago, Shane said:

Was all the data displayed in that photo recorded with the watch in the position illustrated?

I'm not hundred percent sure what you're asking. But my interpretation is if you look at the automatic microphone timing in six positions two of the positions are reversed. That's because the timing machine expects the watch to be somewhere that I can't quite remember so dial-up and dial down are correct and I don't remember what the left or right crown positions are the other 21 set of those are swapped but it doesn't really matter for the discussion at all. You can change the programming in the machine but for some reason I think the machine thinks it's a case stopwatch even though typically it's an un-case to watch but the up and down at least that's correct

7 hours ago, Shane said:

The presence of an oscillating regulator/escapement, converting the only output "Torque" into amplitude, works with my mental model.  The idea that variations in output due to cyclical and (seemingly) sporadic variations within the system result in mirrored variations in amplitude also seems like a solid concept.

the problem is all of this is really complicated. As you can see from the time plot there's variations in transmission power because the year train all watches will have this to some degree some watches will be way worse than other. The fluctuation in power will disturb things. The timing machine tends to average things that's why normally don't see the powertrain fluctuations. If you're paying attention to your timing machine you will see the numbers go up and down if the power fluctuations longer enough but even on this machine where it just gives you numbers you tend not to see that. It's actually one of the bad characteristics of this timing machine is when you look at the final numbers we often don't go back to look at the graphical display to see if it's moving up and down or see if there's no way as we does look numbers look great the watch is fine and the story

7 hours ago, Shane said:

How does an isochronis regulator vary in rate while the watch remains motionless or moved into another position?  The above example shows increases and decrease in rate independent of variations in amplitude.  I'm trying to visualize how the hairspring reacted to positional changes with the 866.  Where all of the positions listed in the photo accurate to how the movement was actually being held?

so as I stated above you'll have powertrain fluctuations continuously oh and then there's that other little thing watches average out all of their problems so the timing machine will show you have a problem but the watch on your wrist you won't see these fluctuations still tend to average out unless the really really bad

6 hours ago, jdm said:

When vertical the HSt slightly deforms and may start touching the regulator. In addition, the bearing friction shifts from the tip to the sides of the pivots. So it us normal for mov.ts if average quality to deviate 20, 30 secs / day or more.

typically on all watches basically dial-up and dial down we usually be the highest amplitude. Usually the pivots have been flattened because you don't want maximum amplitude here but you always going to have a variation between ends of the pivots and the sides of the pivots. So you always going to CA amplitude difference for the most part. I've seen some watches where their seemingly were no differences and defective watches were sometimes there's an issue was something and it's actually better amplitude in one of the crown positions but that's a defective watch. Ideally the fantasy would be to have even amplitude all the time regardless of position

other things that come in the play is JDM mentions the regulating system like the regulator pins. I have an image notice if they hairspring is centered in providing the pins are close enough together hairspring balances nicely between the pins you get a nice average even rate up until the amplitude gets too low and then you're slow. But what if your pins are farther apart then this effect is much more dramatic. So you actually see going between a higher amplitude a low amplitude you'll see the watch slowed down dramatically timing machine. You can tell immediately that your regulator pins are too far apart.

then there's the other example of if it's touching one of the pins notice the effect on timekeeping.

Oh other things what about the balance wheel poise is out of any concern? for this we get another picture. Notice around 220° is the interesting number were amplitude has no effect on poise. Higher amplitudes are better because there's less of fact low amplitudes basically at low amplitudes everything gets magnified not just poise everything will be much much worse is amplitude goes down.

7 hours ago, Shane said:

My original question was directed at the variations within the time sample illustrated in the above photo and why should the absence of jewels explain it.  We seem to have moved past that for now

by the way L let you have the fun of this but go into Google or whatever your favorite search engine is Google wire jewels used in watches they explaining on various sites are quite amusing. seemingly for the discussion jewels aren't necessary? If the watches knew when it's not worn out the pivots are nice good condition the holes in the brass are nice and smooth and round at least for while it's not going to matter at all. But typically on higher-quality watches in other words that would have jewels they usually work a little harder it making things better. Like typically a 17 jewel watch we better than a seven jewel watch a 21 jewel will be better at least to a certain degree. So in a jeweled watch I'm going to assume that the pivots have probably been polished nicer than in a watch like this where they were made at a price. See of hopefully better pivots nice and shiny but not necessarily you can still polish the pivots on a watch like this and maybe they are polished the same degree as a jeweled watch. The bearings themselves the jewels properly made it very hard very smooth and they stay that way almost forever. Where is the brass the metal on steel wears out the holes become out of round. That's going to change the meshing of the gears uneven power. Typically on no jewel watches they eventually just wear out. But you can wear out 17 jewel watches also the metal on metal and people who don't like to have their watches serviced on a regular interval rust is really good at wearing watches out.

then there's other things like the escapement.

Get the balance wheel oscillating a not usually thinking about the escapement does it have any effect on anything? Like lift angle timing machine needs that but what is that number really mean as far as timekeeping goes? It basically means the time in which the escapement is screwing up timekeeping. Because the roller jewel except in this watch in this example it does never wore jewel has a brass pins sticking out on the balance it's a brass on brass if I remember right. The normal watch at the roller jewel it hits the fork it actually will slow down because it's an impact and then it has to unlock the escapement is a loss of energy. Then some point time he gets a kick that isn't precise time kick it's just a kick of the mainspring power and then it finally locks up in the balance is free to continue as a nice oscillation system but during that time span the escapement is screwing up things. That's why people try to design different escapement's like for instance here's a video. It's really interesting video it shows how the lever escapement works and then they talk about some other escapement will never see. But inattention like is that when the roller jewel hits the fork it's hitting a stationary object it has to unlock it is a lot of friction in the escapement is a lot boss here

https://youtu.be/g5c5RK4WFV8

7 hours ago, Shane said:

absence of jewels

my interpretation of the original purpose of this discussion was look my zero jewel watch is perfect. So that we had a discussion on why jewels are total waste of time in some rare examples where seemingly watch works line without jewels. Although oftentimes watches made without jewels were meant to be easy to make usually very cheap and good put up with a lot of things being out of tolerance and it still run in seemingly still keep time. But unless they get serviced on a regular intervals usually than metal on metal bearings will wear out really fast and the jeweled watches will usually last longer.

so looking at the original title it's actually a question of our jewels overrated and the answer is no they're not.

 

Rolex timing regulator pins.JPG

poise of balance wheel versus amplitude.JPG

  • Thanks 1
Link to comment
Share on other sites

I will start out by saying that I do appreciate the time everyone took to try and answer my questions but I do not think that my questions was fully understood.

On 7/2/2022 at 10:04 PM, JohnR725 said:

Google wire jewels used in watches they explaining on various sites are quite amusing.

I think I already have a reasonable understanding of why and where jewels are used and unless it is needed here by others, this part of the discussion can end.

On 7/2/2022 at 10:04 PM, JohnR725 said:

but it doesn't really matter for the discussion at all.

If we are discussing positional inaccuracies, than knowing whether or not the information you are presenting is accurate and consistent does matter.

How the different parts within your BFG866 are reacting to gravity is exactly what we are talking about and the reason why you are testing it in different positions.

On 7/2/2022 at 10:04 PM, JohnR725 said:

so looking at the original title it's actually a question of our jewels overrated and the answer is no they're not.

I never intended to to suggest that the use of jewels are overrated.

On 7/2/2022 at 2:58 PM, jdm said:

However in watchmaking it's (isochronism) almost always synonymous of "small or no variance to power source status". That is because this was one of first problems that had to be solved. 

 

On 7/2/2022 at 10:04 PM, JohnR725 said:

that other little thing watches average out all of their problems

All of the above discussion about out of round holes, rough and flattened pivots, inadequate lubrication and poorly meshing train wheels all comes down to variations in amplitude.

Everything (except bent pivots) should be consistent from one TIC to the next and most variations should be somewhat averaged away throughout a larger time sample.  So amplitude should be somewhat consistent (within any given watch, at it's particular level of wear/damage) if it remains positionally unchanged.

This... This is the part that I still don't get.

How do changes in amplitude affect isochronism if the watch is motionless (in one position) and all forces exerted against it are consistent?  How does any one cycle take a different amount of time than any other cycle?

 

JohnR275,

The problem with using images to illustrate your point is that not everyone understands what they mean (you do, that's why you are using them) but without a reasonable explanation their meaning might not extend to your audience.

Your last two images seam to have interesting information within them.  Can you explain what they are trying to say?

Thank you all for your attention.

Shane

 

Link to comment
Share on other sites

2 hours ago, Shane said:

The problem with using images to illustrate your point is that not everyone understands what they mean (you do, that's why you are using them) but without a reasonable explanation their meaning might not extend to your audience.

okay let's look at the first picture what does the title say? Monetize pictures don't come with titles or prints at all these ones have text in them which is helpful otherwise some nice pictures with no text or unhelpful. Effect of the regulating pinson timekeeping. So this tells us were discussing regulating pin so the picture is showing the hairspring between the regulating pins found in typical watches.

the upper picture shows the classic where they hairspring is supposed to be dead center between the regulator pins. This is why regulator pin spacing is very critical to timekeeping and also why amplitude changes affect timekeeping so what it's showing S is that as long as they hairspring bounces between the regulator pins because the amplitude is high enough it maintains a even timekeeping. As soon as the amplitude drops low enough that they hairspring is no longer actually touching the regulator pins then the regulation point becomes the end of the hairspring or farther down to you how often it's actually touching and the watch becomes really slow.

Now if the regulator pins are farther apart then this effect is seen much sooner. I can actually see this often times with the watches I time when they go from the dial positions with a higher amplitude to a crown position with the lower amplitude the watch dramatically slows that so it's an example of the regulator pins have to really close together but they hairspring isn't supposed to touch at least on a flash hairspring on over coil hairspring it's supposed be as close together as long as you can slide they hairspring through otherwise this effect is really dramatic on and over coil which is typically when I work on.

Now the second half of the picture shows if they hairspring is not centered and touching one of the pins as I said above variations can occur here it is on how tightly it's pushed against the P and also so it's touching the pen that's the regulation point. As the amplitude increases it will lift off the pin as a C timekeeping will change you'll and then it amplitude picks up enough so it's bouncing between the pins it will even out again.

These are just examples of why hairsprings have the dead center between the regulator pins the spacing has to be right otherwise you get timekeeping affected by amplitude or if the amplitude in the watch changes timekeeping changes is why typically around a lot of Rolex watches don't even have regulator pins eliminate the problems of that

second picture doesn't have any title but it does have numbers and has what looks like balance wheel pictures three of them there giving you examples of what happens if you have a poise error with the balance wheel. Such a chart of basically three separate examples of what happens with the positional air and rate versus amplitude

then basically this is a at rest of the was the watches are running the heavy spot is in the location there showing and then as it's moving from the various amplitudes then what effect would it have on timekeeping or what happens is you feel the positional air and watch a wristwatch and you move around in different positions you get all kinds of different timing wrists just from the poise error.

This is also how you can look this up this is also how people dynamically poise their watches they make use of this information by rotating the watch around and seeing where the heavy and light spot is depending upon the amplitude and whether the watches running faster slow they can figure out where the heavy spot is and then reduce them down and even up to timekeeping hopefully providing it really is a poise zero not something else.

The chart also shows something else at 220° approximately poising errors are not a problem they cancel out as the amplitude increases you still have poise errors but they are as much of a problem. At under 220° then the poise errors are much more dramatic and everything else is bad in the watch will be dramatic which is why watch companies typically like the sealer watch running at higher amplitudes but most watch companies will take a watch at 200° at the end of 24 hours some may take at a lower amplitude as long as the watch action maintains timekeeping. Omega will do that but they still have timekeeping specifications

this is where all these charts are kind of approximations to explain concepts. In other words of your balance wheel is really closely poised you wouldn't see the extremes of the end of the amplitude timekeeper was still be really good and timekeeping are probably be just about flat past 220°

so simplistic is both charts show you that amplitude affects timekeeping. If your balance wheels out of poise you have timing changes from that they hairspring isn't where it's supposed to be between the regulator pins you'll get timekeeping variations from that. And as I said that something regularly I see with American pocket watches are put it on the machine arrive cycling 00 and is close regulator pins and timekeeping improves considerably. A lot of people even when they possessed timing machines typically only look at their watch in one position which is dial down so they never see the effect of moving it to another position. We even had that occur in this group somewhere where someone did there I think Omega watch and is it was perfect timing machine are horrible on the wrist once they put it in some other positions I noticed they had some issues. Can't remember what the problem was low.

now while I'm looking for something else to the computer is a nice timing of the Hamilton 936. Unfortunately I don't take detailed notes as to what the watches condition was whether this was before servicing after servicing for whatever reason I ran the numbers and captured them on a USB drive just don't know why but does give us an example something

you'll notice the amplitude dial-up and dial down the amplitudes are identical which is what you should typically expect on a watch because both pivots should be identical. Timekeeping you see a little bit of variations that tends to be common. I've seen on a wristwatch is where the hairspring is much softer than a pocket watch hairspring that dial-up and dial down to the regulator pins art tapered that definitely causes a timing issue because in one position the pins are slightly farther apart than another it can see less of a problem a pocket watch because the hairsprings made a steel or so stiff they never move up and down that much.

Then soon as it goes the crown positions amplitude drops which is typically what you always see. Then as the amplitude drops you can see the effect of probably the regulator pins were too far apart watches running slow timing variations because of positional errors also. This might have been timing of the serviced watch at the end of 24 hours but like you said have no idea why I sometimes capture these images other than there were important at some time.

was kinda hoping to find a more dramatic time plus but the first one that comes up there is a pocket watch and we can see very clear issues you really shouldn't see sine waves in the amplitude you should see more like random noise ideally. The more pronounced a more visible problem is the greater effect it has usually on timekeeping. So here we can see two separate wheel issue is probably the center wheel meshing with the barrel that's what's occurring over a much longer interval and some other wheel having a issue. So not a huge problem then a course you can see the effect on timekeeping. Of course normal people wouldn't notice this because I don't have a time plot feature all you would notice normally a timing machine would be variations as they trace is running across the screen

I suppose I should really label images better but I'm not going to is another time plot because I found time plots fascinating and one of my coworkers did we would do a variety watches just because were curious of the effect on timekeeping. So he had a brand-new system 51 watch not sure what that is right now but it was supposed be a reasonable grade watch. You can see very clear amplitude variation issues almost no timekeeping variations. Usually what happens is the higher the quality of the watch the lesser of fact amplitude has on timekeeping at least in the time plots

I was trying to find an extreme this is not an extreme it is labeled though some of the images got labeled there is a Rolex Explorer you can see very clearly powertrain problem but timekeeping is really good of course I'm pretty sure this watch does not have regulator pins.

then reading time plots is more of a crystal ball thing pretends be a lot of variations did find my example of a bad watch somewhere in the discussion group someone was having an issue with this particular watch the discussion went on long enough that I was able to purchase one off of eBay to see if I can figure out why there were so many timing issues the final conclusion is the watch was basically a really crappy watch and that they can end the person decided not to worry about it and gave it to his aunts and she was happy as can be. Being a small ladies watch no one was really concerned about how much time keeping it had or how well it kept its time keeping

not sure if I'm interpreting your questions correctly the problem we have with the mechanical watch is variations in the powertrain which depends upon each watch how well the watches actually set up how well it actually runs how the regulator pins are positions the poise of the balance wheel lubrication on the pivots can have dramatic effects on timekeeping depending upon things to ends up being a whole bunch of variations that can cause timekeeping changes even with the watch just sitting there.

 

 

Hamilton 936 regulator pins.JPG

pocket watch time plot.JPG

time plot system 51 automatic watch.JPG

time plot Rolex Explorer.JPG

time plot bad watch.JPG

Link to comment
Share on other sites

4 hours ago, Shane said:

How do changes in amplitude affect isochronism if the watch is motionless (in one position) and all forces exerted against it are consistent?  How does any one cycle take a different amount of time than any other cycle?

In short, the natural rate of oscillation of a (flat spiral) spring is largely independent of the amount of motion (amplitude). Think also of a musical instrument string (even the name is different only by changin p to t), no matter how strong is hit, it plays the same note.

You may find this interesting to read: https://www.forgottenbooks.com/en/download/IsochronismofBalanceSprings_10290824.pdf

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

2 hours ago, jdm said:

In short, the natural rate of oscillation of a (flat spiral) spring is largely independent of the amount of motion (amplitude). Think also of a musical instrument string (even the name is different only by changin p to t), no matter how strong is hit, it plays the same note.

Frequency is just frequency.  This is what I have been assuming all along.  I knew that gravity could slump or stretch more or less hairspring through the regulator, I just didn't see that the length of the spring changes while the balance is oscillating and how amplitude changes the percentage of time the spring is unregulated.  I was assuming that most of, if not all amplitude errors were in the slow direction (as suggested in John725's latest post.  That was why I wanted to know if the positions indicated in anearlier post were accurately represented.

Thanks, I will need to chew on everything that JohnR725 said.

 

Thanks for the link to the forgotten books.  I will try digest that soon.

Shane

Edited by Shane
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...