H/S centering problem

[bsoderling]

Hi all,

 

Struggling with a h/s problem that I can’t get my head around.

 

The movement is a Tavannes 38 that belongs to an old (30’s?) art deco tank watch. I have serviced the movement and at first (well after some tinkering in the h/s area) it ran beautifully with good amplitude and nice straight lines on the timegrapher.

 

But then it started to behave weird, speeding heavily and stopping.

 

I had noticed that the first h/s turn outside the regulator runs very close to the regulator pin and this is where I had pushed and pulled a bit to get it to run free. Which apparently worked for a while....

 

Removing the h/s from the balance wheel and mounting it back on the bridge, I seem to find that the h/s (in my eyes) looks quite decent with overall evenly spaced turns etc.

 

But the collet is clearly offset and centering it will inevitably push the turns tighter and closer to the regulator pin.

 

With my limited h/s adjustment experience, I can’t see a way to move the collet to the correct position without, in some way, influencing the overall h/s shape and possibly create unwanted touching between h/s turns or the regulator pin.

 

A photo is attached. I hope it makes sense to explain my problem.

 

 

 

Sent from my iPad using Tapatalk

[rogart63]

If you straighten it a little where the stud is . And make the curve a little tighter after the stud up to the adjusting arm. I think it will be much closer to the center . 

[watchweasol]

Hi as mentioned the B/s is not right , The spring between the regulator and the stud is a little out of concentricity,  The arc should allow the regulator to move round without touching the spring and be centered between the pins. If you move the regulator towards the stud the collet will move from the center. Ideally you want the collet centered and the reglator moving towards the stud without moving the spring..   easier said than done.  Take your time.

 

[bsoderling]

Hi and thanks for feedback.

I have tried to alter the shape from the stud towards regulator to move the collet more towards center. And as such it works...but since anything I do close to the h/s circumference also move the entire h/s and ending with that first turn outside the regulator to move against the reg pin. It’s as if I should really need to only move the collet without touching any of the other h/s turn positions. Not possible, I guess.

To me it feels like the h/s is overall a tiny bit too large diameter to fit the three points; center, regulator and stud.

Sigh....







Sent from my iPad using Tapatalk

[praezis]

4 hours ago, bsoderling said:

To me it feels like the h/s is overall a tiny bit too large diameter to fit the three points; center, regulator and stud.

Exactly.
If the collet is centered, the spring will touch the regulator pin. Probably this hairspring was changed in the past.

Frank

[bsoderling]

Probably this hairspring was changed in the past.

The thought has struck me as well that this may be the case.

I will try some more, maybe there’s a compromize that works?

And I was also thinking that maybe I can find a regulator from a scrap movement with a slightly longer distance from center to reg pin that offer a tiny bit more space for the first h/s turn?

Any comments to that idea?


Sent from my iPad using Tapatalk

[bsoderling]

Hi again,

Running out of ideas, I decided to try a somewhat counterintuitive approach.

With the main trouble being the h/s touching the reg pin from the outside, I decided to remedy that to start.

I slightly straightened the h/s right across from the stud i.e. Moving the entire h/s in the ”wrong” direction and effectively increasing the error between collet and jewel hole.

What then happens when ”forcing” the collet to be over the jewel hole is that the hairspring gets a bit distorted with tighter turns on the regulator side.

But....the error introduced by the straightening gets spread out over many h/s turns and there’s still space between the spring and the the reg pin.

Looks a bit ugly but runs very well again and hopefully interesting for some h/s nerds out there.






Sent from my iPad using Tapatalk

[Klassiker]

Would you mind posting a photo for comparison?

[bsoderling]

Ouch...I assembled the whole thing and right now I can’t bring myself to tear it apart again, I’m afraid.


Sent from my iPad using Tapatalk