Guidelines for double roller jewel depth?

[mbwatch]

I'm working on a low grade 1930's Elgin 12s a family member found in a drawer, my first attempt at replacing a broken balance staff. That part was successful, to my surprise and delight. The new staff is in, the balance rotates freely and I didn't damage the hairspring. However, when I initially removed the double roller, I found it had no shellac and the jewel was floating. I melted in new shellac when reassembling the balance, but I had no reference for how long far the jewel should protrude and some tutorials I had seen single rollers essentially installed them flush. Placing it flush with the top of the table was clearly way too long since it extended well beyond the bottom of the double roller, so I scooted it in about a half mm.

This doesn't seem to work, and I am almost certain the jewel is the problem. With the fork installed I can't seat the balance such that it isn't overbanked and it has to be that the roller jewel is interfering with the guard pin. I don't have a microscope or apparently the experience to get a good enough view of it in situ.

Doing a google image search on double rollers, it does look like in most cases the tip of the jewel is about parallel with the inner surface of the double roller (mine is about parallel with the lower/outer surface as pictured). Should my strategy here be to warm up the shellac and push the jewel in a little farther so it protrudes less? If there is a general guideline for how far a roller jewel should protrude before refining it in situ, I would be interested t know. My de Carle book doesn't directly address this except for a diagram that might help. I wouldn't be surprised if someone tells me the installed jewel is too long overall for this roller to begin with, but I have to assume a professional put it in 60+ years ago if it is not the factory original.

thanks as always.

[nickelsilver]

As it appears in the DeCarle image ia correct. Yours is still out too far.

[mbwatch]

Thanks - I will try nudging it in until it's just about even with the inside edge of the lower part of the roller. Every time I touch the balance assembly it still feels so harrowing.

[Klassiker]

Take the hairspring off and put it to one side until you have the jewel set correctly.

[mbwatch]

4 hours ago, Klassiker said:

Take the hairspring off and put it to one side until you have the jewel set correctly.

Of course I would do this.

[mbwatch]

Okay some success here. Resetting the jewel more like DeCarle's diagram has worked and the movement is now ticking. It won't run dial-up though, and that appears to be due to excessive end shake. I'll have to see if I can correct that - it is also possible I fitted the wrong staff since the ebay seller was a little ambiguous. This was meant to be a practice watch anyway so I'll keep practicing. I had thought the roller table was just not a tight fit on the staff as it wanted to rotate a little too easily, but ultimately discovered I didn't have a good enough rivet on the staff (and that my staking set is missing the exact rounded stake I needed). I did get the rivet set in more solidly and assembled the thing, got the roller aligned between the banking pins, and am pleased that it is at least kind of running since this started out with 2 broken staff pivots.

[Klassiker]

Maybe the jewel is still fouling the guard pin, but only dial up, because of end-shake? Can you push the jewel in just a touch more, or does the jewel then not interact cleanly with the horns dial down?

[mbwatch]

When turning it dial up while running, the sound it makes while stopping sounds like the twang of the hairspring catching on something so my next step is to evaluate its flatness and if the overcoil is twisted up incorrectly. (I do not have a good success rate of fixing issues like that). Still the end shake seems too large and I think the upper balance jewel is seated in as far as it can go but I will check that too.

[RichardHarris123]

Maybe you could shim the balance cock?

Edited April 17, 2023 by RichardHarris123

[Klassiker]

13 hours ago, mbwatch said:

my next step is to evaluate its flatness and if the overcoil is twisted up incorrectly

Sounds sensible. You could also check the shakes on the pallet fork.

[mbwatch]

Watching closely as it stops, it seems very likely that the balance wheel itself is coming into contact with the center wheel when DU. I am guessing this is not just an end shake issue, but is actually a balance truing problem. My next tool pickup will be a truing caliper (I know I was supposed to do that after replacing the balance staff, but was overeager to test out my work)

14 hours ago, Klassiker said:

You could also check the shakes on the pallet fork.

Should the fork pivots have any discernible end shake? This is a 7j movement so they're in steel bushings rather than rubies.

[Klassiker]

If there's no discernible end- (or side) shake, then the fork is probably not moving freely. I would describe the optimum as "the minimum discernible" shake, but things can work fine with a lot more than that, as long as unwanted collisions don't occur. So it was the possibility of too much shake, particularly end-shake (axial movement) but also side-shake (tilting) that I thought might be an issue in your case. You already identified excessive shake in the balance pivots, but the same at the pallet can also negatively affect the sensitive interactions between jewel, guard pin, roller and horns. What can work fine in one position (dial down) can foul up completely in another (dial up), or vice-versa).