Balance Amplitude And Speed - Comparison

[WillFly]

I was winding my pocket watches yesterday - I give them a once-a-week wind to keep them supple and check condition - and was struck by the apparent balance slowness of my Elgin 571 "B.W. Raymond" as compared with one of my Illinois "Bunn Special" watches. Both RR Grade. So, I made a little video of them side by side with a close up of both balances. I don't know whether you can tell from the video shots but, to me, the amplitude and speed of the Illinois balance (on the left) seem so much greater than that of the Elgin (on the right).

 

So take a look and tell me: which runs fast, which runs slow, and which is keeping dead accurate time.

 

https://www.youtube.com/watch?v=4wBoctO5qOE

 

Answer: They're both keeping dead accurate time - to the second in 24 hours running.

 

My technical question is therefore - how can two similar watches with different amplitudes and speeds run accurately? Is it a question of movement design? Balance wheel size? Length of hairspring, etc?

 

As you can see, my technical knowledge is limited in many respects!

Edited September 26, 2014 by WillFly

[matabog]

I suppose they have different BPH. I think the amplitude may also be different. I believe the amplitude affects timing only if very low. But if it decreases, it's good sign the watch needs a service.

 

Try Wildspectra or tickoprint on Android. They will tell you the bph. Those are strong ticks, the phone mic won't have a problem "hearing" them.

[bobm12]

Does pretty has anything to do with it?

[DJW]

Hi Will,

I've noticed this before across different manufacturers and grades of watches. My first inclination would be to pin the difference on the spring strength, either by design or just plain wear. Most of the Hamilton's 10s that I service run at an amplitude around 275-295. The Howards and South Bends can peak at 330+! This is even when a new mainspring is installed. So this is why I say design first. Then, I would point to the mainspring wear as the next reason for difference between similar watches of the same size.

 

I believe these are both 18,000bph? And, as was mentioned before, in general, lower amplitude will not have a significant effect in timing unit it is very low, near the end of a wind.

 

That's just what I've observed in my limited experience! Someone else may be able to give you a more detailed(or correct) explanation. :P

Edited September 28, 2014 by DJW

[WillFly]

Thanks for that - a very useful comment. And, yes, both watches run at 18,000.

[guidovelasquez]

Hi Will, I have seen his posts and have learned much from their contributions to the forum. I wanted to ask your help on something I find hard to understand. I am studying the amplitude of balance of a clock, and I would like to show a video that I need to help me determine the approximate amplitude of the balance of this watch. It is an Omega chronometer Lemania 9160A with 41 degrees of lift angle. horizontal. I'll put the video to download it.

 

Amplitude O.9160A 41 deg.doc

 

Thank you so much Guido Velasquez

 

PD

 

I could not post the video. But I'm uploading file is video. Please change the file extension mp4.

 

[clockboy]

Hi Will, I have seen his posts and have learned much from their contributions to the forum. I wanted to ask your help on something I find hard to understand. I am studying the amplitude of balance of a clock, and I would like to show a video that I need to help me determine the approximate amplitude of the balance of this watch. It is an Omega chronometer Lemania 9160A with 41 degrees of lift angle. horizontal. I'll put the video to download it.

 

Amplitude O.9160A 41 deg.doc

 

Thank you so much Guido Velasquez

 

PD

 

I could not post the video. But I'm uploading file is video. Please change the file extension mp4.

Your Doc does not open on my Mac. There is a vid on youtube my Mark who explains amplitude & beat etc.

[cdjswiss]

A pendulum clock keeps good time with very low amplitude.

The beat time (timekeeping) of a hairspring balance is independent of amplitude - Google 'simple harmonic motion'.

Large amplitude is to a certain extent a fad of the purist who likes to see that the watch has plenty of power.

And that is an indication of its quality, the state of its mainspring, and its cleanliness.

[JohnR725]

To understand the relationship of amplitude versus timekeeping look at page 6/14 of the PDF below. Using a sine wave to represent balance wheel timekeeping/amplitude Makes it a lot easier to understand why one balance wheel it looks like it's moving really fast versus another one moving slow can still keep the same time.

http://www.witschi.com/assets/files/sheets/Test%20and%20measuring%20technology%20mechanical%20watches.pdf

 

Then amplitude does affect timekeeping if the watches in any position other than dial up or dial down. To understand this I've attached an image which shows timekeeping and positional errors versus amplitude. You'll notice on the chart that around 220° positional errors are not a problem.

 

Then comparing pendulum clocks to balance wheels in the watch has a minor technical issue in that clocks are usually stationary with pendulums, watches usually move around. A watch with a high amplitude tends to be influenced less by external influences such as motion.

 

John

 

[clockboy]

Amplitude is an interesting subject.  I have noticed with most of the Seiko,s I have serviced although all is OK with good readings I get a low amplitude & the watch runs fine & keeps good time.

[guidovelasquez]

Thank you very much for your help. I thank the documents they share. Well the experience. John, the image in the attached page 45 indicates that it is, is it part of any publication that can get? On this graph, in the illustrated case does it indicate that above 240 degrees the clock is less accurate than 220 degrees? On the comment that makes cdjswiss, I agreement on the "purist" attitude on the extent, and if both the pendulum as the balance wheel is governed by the simple harmonic motion is a waste of energy greater extent where less so direct clock accuracy. While I'll upload the video to yuoutube because I think it was not possible to see it. Clokcboy comment on, I can also say that many clocks go well with a relatively low amplitude. no "as they say in books."

 

Here the link Amplitude.  

 

https://youtu.be/HaZV1kownXM

[cdjswiss]

 

Then amplitude does affect timekeeping if the watches in any position other than dial up or dial down. To understand this I've attached an image which shows timekeeping and positional errors versus amplitude. You'll notice on the chart that around 220° positional errors are not a problem.

 

John

 

I stand corrected provided that you can give an explanation of the effect of position error! It must be to do with the dynamics of the hairspring. Maybe higher amplitude (faster motion) contributes to maintaining the hairspring shape when it is not horizontal and even higher angular velocity does something else. But surely the curves do not always, for all hairsprings, cross the zero at 220deg?

 

The clock example was merely to illustrate shm.

 

Colin.

[guidovelasquez]

On the relationship between the amplitude of the balance and accuracy, here's an interesting analysis performed 10 hours a stopwatch Omega Lemania 9160A.

They may be some small fluctuations in the curve Daily Rate and Beat Error, but not reflected nothing in amplitude. Notes that although as wind loses its tension, and the amplitude decreases from 270 degrees. up to 220 degrees, the rate is maintained.

It is also important to say that the nominal elevation angle was changed from 41 degrees to 45 degrees in this experiment. The Watch-O-Scope program is an excellent tool in this analysis.

 

[Michael]

Hi Guidovela, I like your graph. Looks like a winner.

 

When working on the beat error, like on Marks last video, with the movement where one has to turn the collet manually with a screwdriver to adjust beat error, how to do get it so accurate? It seems to me like a process of guess work, and a long dangerous process to adjust each time (could take 10 attempts to get it to approx 0.1ms). To get it really accurate by turning the collet could end in misery, or could make the previous adjustment worse (when working below for example 1ms). Am I missing a trick here?

[guidovelasquez]

Michael is right. There is no Trick. First precise hands. Second patience because if you overdo it can take several attempts to bring its optimum setting. I think for watchmakers should be a little easier. But even for them should be a moment of great concentration and applied all his experience. In my case, some watches little value, not worth trying to bring them to such high levels. I leave you in 1.1- 1.9 of beat error. It has its irony, because if you do something wrong with a watch Omega will have a high price.

What I do is I take to practice running watches. So when I have a better quality, I feel a little safer.

You can see the links where I have other graphics streams watches.

 

http://www.watchrepairtalk.com/uploads/monthly_09_2015/post-670-0-71156500-1442513637.png

Edited September 25, 2015 by guidovelasquez

[Michael]

Thanks for the reply. I suppose it is like a lot of things in this game ..... lots of practise