Need help understanding major amplitude difference

[VWatchie]

The following pictures show a just serviced and for the first time fully wound ETA 2804-2.

Dial up

Dial down

Pendant right

Pendant down

Pendant up

Pendant left

As can be seen in the last two pictures the rate and amplitude drop dramatically in the pendant up and the pendant left positions.

The only thing I wasn’t too pleased with during the service was the pierced jewel sitting inside the jewel housing in the lower shock block, as the edge of the hole in the pierced jewel looked slightly frayed/chipped. Also, the cap jewel was ever so slightly grooved (hardly visible under strong magnification). The question is if this, less than perfect, pierced jewel could cause such a dramatic difference in amplitude and rate?

[clockboy]

Maybe but it one of faults that really the only any is to change the jewels. But first it could be a balance poise issue. I know our Mark advocates dynamic poising. I have not tried dynamic poising as I am not willing to experiment on a customer’s watch. The only poising I have done is the traditional method (static poising) after fitting a balance.
Personally if it was not my watch I would not fiddle but on the other hand if it was my watch then yes I would check the poise. 
See below an article on static vs dynamic poising.

https://adjustingvintagewatches.com/smooth-alloy-balance-wheel-poise/

[nickelsilver]

My first guess if this were my piece would be the safety roller touching the guard pin in those positions, and/or the roller jewel touching a fork horn slightly. This can be because someone closed the banking/s, or that the balance jewels aren't seated correctly (off center)* or sometimes a bent balance pivot. If it was a bent pivot you would almost certainly see a much bigger change in rate along with the amplitude drop- same if the poise was out enough to knock down the amplitude the rate would be off the charts.

 

If you do have a damaged hole jewel, that would definitely need to be changed, and the staff pivots examined very closely. The horizontal positions look good enough I wouldn't worry about the cap jewel.

 

*with most Incablocs and many Kifs if for some reason the jewels aren't centered the diameter of the staff just after the pivot and conical section can rub on the "chimney" or hole in the setting. There's only 0.02-0.03mm clearance so it doesn't take much.

Edited August 30, 2021 by nickelsilver

[Nucejoe]

 I would replace the jewel before spending time to diagnos anything, then you shoud check for possible wear/ scratches on the corresponding pivot as well, you wouldn't know if you don't check it.

 Could you get a TG reading with pendent at or close to  4,22 and 10.22, at these positions you have balance down and up so, so you get an idea if guard pin, fork and  the roller is rubbing. Expectedly you should see low amplitude at position where pivot is on top of the jewel fault. 

Lucky the fork is not rubbing on the roller table,which usually means excessive pivot wear or moved jewels.

And Clockboy has a very valid point , oscilator runs aweful in vertical positions if the balance wheel is not poised.

Regs

[Nucejoe]

 Instead of pendent down position try hr 4.22 down, this would put the oscilator farthest distance from the guard pin and vice versa when 10.22 is up like in  pendent up.

This might vary for different calibers, depending on where the oscilator is located. 

I put my two cents on the pivot riding on faulty part of the jewel where you see the lowest amplitude.

[JohnR725]

a experiment to try would be to lift up the Springs rotate the jewels without taking them out in other words put them in a different position and see if your problem moves.

[VWatchie]

Well, I’m super happy to let you know that the problem has been resolved! 

Pendant up

Pendant left

And I should add that all other positions perform equally well (see the first post in this thread).

Before starting my investigation, I decided to replace the cap jewels. For the sake of it I took long hard look at the old cap jewels under my stereo microscope (40X) and realised that the grooves were much deeper and wider than I had perceived during my initial inspection. I don’t know why I missed that, perhaps a bit of negligence. It really is important to get the light to reflect on the cap jewel surfaces from the correct angle to truly see the grooves.

Inspecting the holes in the pierced jewels in the jewel housings revealed, to my surprise, no signs of being frayed or chipped, as mentioned in my first post. I probably inspected those jewels before having cleaned them and I suspect that what I actually saw was dried up oil scattered around the hole rather than the hole being frayed or chipped.

Consequently, I kept the jewel housings and replaced the cap jewels. These jewels were also used but in perfect condition (thanks again @rogart63!)

So, having replaced the cap jewels and before continuing my investigation I placed the movement on my TM, and to my surprise the problem was solved. So, what exactly caused the problem? I don’t know, but my guess is that the balance jewels weren’t seated correctly, as pointed out by @nickelsilver, and perhaps aggravated by the grooved cap jewels. Perhaps the grooves would make the staff tilt that "0.02-0.03mm" and make "the diameter of the staff just after the pivot and conical section rub on the 'chimney' or hole in the setting". What do you think?

Thanks a lot guys, your help is (as always) invaluable! 
 

[clockboy]

Great news indeed. I can remember a few years ago when after servicing  I had a terrible reading on my TM . THEN I realised I had forgot to oil the cap jewels.Silly mistake but as they say l learn  from your mistakes. 

[VWatchie]

3 hours ago, clockboy said:

Silly mistake but as they say l learn  from your mistakes.

Good to know we're not alone! Thanks clockboy!