On My Bench - A Normana

JimInOz · Oct 8, 2020

😡
I HATE FIXED STUDS!

In the photo below you can see the clear impulse jewel (just blow the hairspring) is not quite centred between the two upright banking pins (it's a bit to the left).
So the simple solution is to move it to a central position, thus evening out the "beats" as the balance swings back and forth. This is done by moving the location of the impulse jewel in relation to the outer end of the hairspring. I most modern watches, this is done with a movable stud that allows movement of the impulse jewel by moving the stud end of the hairspring.

However, older watches with fixed studs, need to have the balance assembly removed and the balance wheel and hairspring removed from the cock. Then the collet that holds the hairspring can be rotated on the balance stuff causing the outer end of the hairpring to move either left or right.
So starting with a beat error of 3.2ms, I went through the process and got a beat error of 0.3ms.
I can do better than that I thought, so everything pulled apart again, this time 2.8ms 😡. Turned it too far!
Once again everything off and adjusted and now I have it at 0.9ms, which isn't perfect, but I thought I'd better take a break before I lost my cool with this.

Maybe tomorrow I'll be in a better frame of mind and can get it closer to 0.0ms.

Maybe......................

TDBK · Oct 8, 2020

JimInOz
I went through the process and got a beat error of 0.3ms.
I can do better than that I thought

I'm curious as to why: simple pursuit of perfection? Or is there a timekeeping-related reason? I am of the uninformed impression that beat errors under 1ms have no practical effect.

Thanks as always for your informative and entertaining tales.

Archer · Oct 8, 2020

JimInOz
The high beat error (≈3.0ms) that I thought might go away by itself is still there, so I must take it apart so I can view the position of the balance impulse pin in relation to the banking pins.

With a big pocket watch it's an easy task as there's lots of gaps to peer through, but with a small (10½ ligne) caliber there's now way I can see where I need to.

Well, if you are expecting beat error to fix itself, you will be waiting a very long time. 😁

And "small 10 1/2 ligne"? Okay...if you think that's small, you might not want to repair any ladies watches, where the whole movement is smaller than your balance wheel in this one. 😉

JimInOz
However, older watches with fixed studs, need to have the balance assembly removed and the balance wheel and hairspring removed from the cock.

I rarely remove the balance from the balance cock/bridge for this. Pretty sure you have a balance tack, so if you place the balance on that, and let the wheel hang down, there is often plenty of room to get the tool in between the coils to rotate the collet without taking the balance off the stud. At least on most watches I have to do this with (a lot of them). Of course you have to be quite careful, so if you are not confident doing this that's fine, but if you can it saves a bit of time for sure.

TDBK
I'm curious as to why: simple pursuit of perfection? Or is there a timekeeping-related reason? I am of the uninformed impression that beat errors under 1ms have no practical effect.

Beat error won't affect timekeeping until it is fairly large, and most consider under 1 ms acceptable. Most Omega's have a tolerance of max. 0.8 ms for beat error. So 0.3 ms would be within any brand's tolerance for beat error, but yes I have done the same in making things worse in the pursuit of getting it "just a little bit better" so I can't fault Jim for that.

Cheers, Al

wsfarrell · Oct 8, 2020

Naive question: is it not possible to adjust the beat error by moving those tiny screws around the perimeter of the wheel?

Archer · Oct 8, 2020

wsfarrell
Naive question: is it not possible to adjust the beat error by moving those tiny screws around the perimeter of the wheel?

No, this would change the poise, but not the beat error. Beat error is about geometry and the relationship between the roller jewel and the location of the balance spring stud.

Canuck · Oct 8, 2020

Since the screws on the perimeter of the balance wheel on the subject watch are tight, moving them would mean loosening them, causing them to fall out after a time. Plus, moving the screws (in or out) would affect the rate (regulation) but certainly not the beat error. Mean time screws on high grade watches can be moved in or out without worrying about them falling out. But the only reason you would move the meantime screws would be to make small adjustments in the poise of the wheel, or to affect the rate. There are no meantime screws on the balance wheel of the subject watch.

JimInOz · Oct 13, 2020

Well, after four more tries I got it below 1.0. Due to the risk of damaging the hairspring I figured that 0.8ms is within specs.

Now running at +2 to +4 secs/day with amplitude at 305º.

Just have to wait for a low dome crystal to arrive, the one from my parts box looks too high.

François Pépin · Dec 25, 2020

Nice job Jim!

To me, the worse is not regulating the beat error, but doing dynamic poising with little screws that’s can break in the balance! It happened to me several times, and though I like poising and try to have the best possible rate, I am always scared to break balance screws while putting them back...