What have I done to my balance?

[GPrideaux]

I tell the kids I teach that we learn by our mistakes. I must be learning plenty with the Seiko 6105 on my bench. I was reassembling the calendar works and I foolishly failed to remove the balance first. When I turned it over the balance no longer swung freely. I realised that the way I had it in the movement holder left the movement resting on the balance itself. I immediately concluded that I'd broken the balance staff.

I'm discouraged.

Upon inspection, however, the pivots do not appear to be broken, at least to my eye. Slightly encouraging.

Even if the pivots are intact, however, something is stopping it turning freely. Can anyone suggest what I may have done to it, and whether it might be repairable?

I have ordered a replacement NOS balance complete. I jumped the gun and did that before properly inspecting the balance, but I don't currently have the skills or tools required to do any adjustments to the balance myself so even if it is repairable it's likely to be outside my capacity at the moment.

I still need to inspect the Diashock settings and jewels ... I think they look ok in situ, but I haven't taken them out yet.

[oldhippy]

Yes check the jewels can the caps. I agree the staff looks good. It is good practise to remove the balance complete first and put it in a safe place and under cover,  before you start on the rest

[HectorLooi]

Your balance wheel might be out of round and out of flat. If you have a truing caliper, check that the balance is true first.

I find that Seiko balances are very delicate and positioned too close to the edge of the plate. If the movement is placed carelessly in the movement holder and tightened, the jaws of the movement holder can impinge on the balance wheel and distort it.

After truing the balance, check that the hairspring is flat and not rubbing against anything.

[caseback]

The shock protection will/should have protected the pivots..

 

[GPrideaux]

18 hours ago, oldhippy said:

It is good practise to remove the balance complete first and put it in a safe place and under cover,  before you start on the rest

Yes, I know, I made a rookie error through inattention. But then, I am a rookie and I guess this is how we learn. I just wish I could have made this particular rookie error on one of my $10 Russian watches.

18 hours ago, HectorLooi said:

If you have a truing caliper, check that the balance is true first.

I don't have a truing caliper. That might have to be my next tool purchase. If it has been distorted, is it possible to return this kind of balance to true?

 

Thanks all for the encouragement.

Glen

[Neverenoughwatches]

6 minutes ago, GPrideaux said:

Yes, I know, I made a rookie error through inattention. But then, I am a rookie and I guess this is how we learn. I just wish I could have made this particular rookie error on one of my $10 Russian watches.

I don't have a truing caliper. That might have to be my next tool purchase. If it has been distorted, is it possible to return this kind of balance to true?

 

Thanks all for the encouragement.

Glen

Back to basics Glen......function groups.....if you suspect the balance has the problem....then just install the balance, inspect and work that area.

[HectorLooi]

Have a look at this video. Mark shows how to use a truing tool to check for true and how to straighten and true the wheel.

Many Seikos seem to have distorted balance wheels. I always check them. I had to straighten two of them recently. 

[GPrideaux]

Well, I've inspected the Diashock jewels and they look fine. There is a small divot in the lower cap jewel which I will replace, but that was there before my error. I cleaned and oiled them and put them back in then installed the balance with the train and pallet removed, and it appears to be swinging freely.

That's encouraging. My next thought was to check the end shake. I was a bit concerned about that before, and I'm still a bit concerned. The trouble is I don't have the experience to know what is normal vs what is excessive. I've filmed the pivots as I gently move the balance up and down:

Esteemed colleagues and mentors, can you advise me? Is that moving too much?

[caseback]

Looks to me like that's too much.

[GPrideaux]

I've just found this thread:

Super helpful. I have a jewelling tool coming. In the meantime, I'm going to install the pallet and see (a) whether the balance is making contact with the pallet bridge and (b) how the fork lines up with the impulse jewel and safety roller. I'm learning so much here; the biggest lesson is to not be hasty.

[rehajm]

…to backtrack just a bit, when you had the movement dial up did you place pressure on the balance cock and either alter the angle from the screw or bend the cock?

…i ask because I just had a movement stop when I tried to case it. The cock was pressured from the retaining ring and changed the end shake. Instead I looked at the hour wheel, dial washer, the little bend in the dial that might be putting pressure on the train…it was the balance cock moved slightly

[GPrideaux]

5 hours ago, rehajm said:

did you place pressure on the balance cock and either alter the angle from the screw or bend the cock?

Thanks for the suggestion. I don't think I've done anything to the balance cock. If I'd bent it down it would've reduced the end shake, and if anything there's still too much, not too little.

I think the balance wheel itself was resting on the movement holder when I had it DU. It may have moved the lower setting a bit, I guess, but I think that was already too low by perhaps 3 hundredths of a mm. I'll know more when I put the pallet back in. I think the balance had virtually zero clearance from the pallet bridge before so even a few microns movement down would result in it fouling, and I'm guessing that's what I'll find.

[Neverenoughwatches]

On 11/4/2024 at 9:57 AM, HectorLooi said:

Have a look at this video. Mark shows how to use a truing tool to check for true and how to straighten and true the wheel.

Many Seikos seem to have distorted balance wheels. I always check them. I had to straighten two of them recently. 

Good video on wheel truing, i enjoyed it. . Kinda surprised me that Mark did not use any dial protection when lifting the hands. I get that the levers where big to spread the load and polishing the levers will minimise and dial damage ( i didnt notice how polished they were ) but any slight burrs on the underside of the levers from inadvertent created scratches could transfer to the dial, for the sake of a few seconds to lay plastic cover over , yeah kinda surprised me that.

20 hours ago, GPrideaux said:

Well, I've inspected the Diashock jewels and they look fine. There is a small divot in the lower cap jewel which I will replace, but that was there before my error. I cleaned and oiled them and put them back in then installed the balance with the train and pallet removed, and it appears to be swinging freely.

That's encouraging. My next thought was to check the end shake. I was a bit concerned about that before, and I'm still a bit concerned. The trouble is I don't have the experience to know what is normal vs what is excessive. I've filmed the pivots as I gently move the balance up and down:

Esteemed colleagues and mentors, can you advise me? Is that moving too much?

I'd say that looks too much

When adjusting endshake anywhere, i go for the least i can get without causing an ill effect

8 hours ago, GPrideaux said:

I've just found this thread:

Super helpful. I have a jewelling tool coming. In the meantime, I'm going to install the pallet and see (a) whether the balance is making contact with the pallet bridge and (b) how the fork lines up with the impulse jewel and safety roller. I'm learning so much here; the biggest lesson is to not be hasty.

I would try a free oscillation check first, its good to know that the balance will oscillate with adequate free motion in isolation. 

Your video.....as long as the hairspring is in good order and working correctly, does look like poor interaction between the impulse pin and the fork, could be down to the balance endshake or the lever endshake, the pin might be rubbing on the gaurd finger of the fork. That can be the balance is too low...you could hack that by flipping the bottom capstone , or somehow you've lowered the roller table height with the accident, or the fork is too high.....possibly bent up and needs straightening ( not as frightening as it sounds ) . So in summary....free oscillation check and then the lever to balance connection ( forkslot/ impulse pin interaction ). I think you checked for free motion of the train .

Edited November 8, 2024 by Neverenoughwatches

[GPrideaux]

I'm still not sure what happened with the accident, but I'm leaning towards thinking that I displaced the pivots from their settings without actually breaking anything. What I've done since:

1. Checked Diashock jewels and settings: no problems visible. Cleaned, oiled and reinstalled.
2. Installed just the balance and did a free oscillation test. All good: balance oscillates for about a minute.
3. Installed the pallet fork and checked its end shake. Looks good: perhaps 0.02 mm end shake.
4. Inspected the interaction between pallet fork and impulse jewel. The jewel is fully engaged with the fork at both extremes of the end shake and never comes close enough to the guard pin to touch it, and the guard pin lines up well with the safety roller, although the balance may be sitting a little low when it is at the lower extreme of the end-shake.
5. Examined the clearance between the balance and the pallet bridge. At the lower extreme of the end shake there appears to be no visible clearance.
6. Repeated the free oscillation test with the pallet and bridge installed. No good: balance oscillates for less than ten seconds before coming to a complete stop.

I've made a YouTube playlist of the handful of short videos I made of these inspections. https://www.youtube.com/playlist?list=PLP95CUmMHEso3B-6tp3O1Qnqzz1tzdl2W My conclusion is that the balance is rubbing on the pallet bridge. I don't think this was caused by my accident as I was concerned about the lack of clearance before that, and it would also explain the issues the watch had with very different timing dial up vs dial down.

 

I'm aware that turning the lower capstone over might fix this, but I'd rather fix it properly if I can. It's my understanding that the Diashock setting is friction fitted to the mainplate so I should be able to adjust it with a jewelling tool, right?

(Sorry about the upside-down photos.) I am going to get a new cap jewel for this setting as the current one has a divot:

I should probably wait until I have the replacement before I do anything about the end shake in case the new stone has a different thickness.

 

[caseback]

If I'm not mistaken, the thickness of the endstone itself won't influence the endshake. This is because the part facing the pivot is flat. Only the fact the new endstone will have no divet will have a small effect.

[nevenbekriev]

@GPrideaux, You have 2 simple problems.

First one is obvious and it is that the main plate antishock device has moved, sunk down. This leads to the great axial free play of the balance and it's rubbing on the pallet fork bridge. This is easy to repair - just press the antishock back. If You have staking se - use it, wide flat punch is needed that will lay on the periphery of the antishock device. Just press, don't use hammer. But, the same thing may be done without the staking set, with proper looking tool There is nothing to adjust there - Just press it to the end.

The second problem is that in the DU position, balance doesn't oscillate free. Well, first rectify the main plate antishock and then try again, but take good look at hairspring side pivot and the jewel setting. Seems that there is something wrong there.

The balance is not completely true, but this will not cause the problem. Only if the upper pivot is bent, then this might be the cause, but I doubt.

Yes, as @caseback says, the endstone thickness doesn't matter at all. Of course, it is true if the spring is able to press on the stone. If the stone is too thick, then the spring will have to be bent alot, if it is too thin, then the spring may not reach it and press the setting.

[Neverenoughwatches]

2 hours ago, GPrideaux said:

I'm still not sure what happened with the accident, but I'm leaning towards thinking that I displaced the pivots from their settings without actually breaking anything. What I've done since:

1. Checked Diashock jewels and settings: no problems visible. Cleaned, oiled and reinstalled.
2. Installed just the balance and did a free oscillation test. All good: balance oscillates for about a minute.
3. Installed the pallet fork and checked its end shake. Looks good: perhaps 0.02 mm end shake.
4. Inspected the interaction between pallet fork and impulse jewel. The jewel is fully engaged with the fork at both extremes of the end shake and never comes close enough to the guard pin to touch it, and the guard pin lines up well with the safety roller, although the balance may be sitting a little low when it is at the lower extreme of the end-shake.
5. Examined the clearance between the balance and the pallet bridge. At the lower extreme of the end shake there appears to be no visible clearance.
6. Repeated the free oscillation test with the pallet and bridge installed. No good: balance oscillates for less than ten seconds before coming to a complete stop.

I've made a YouTube playlist of the handful of short videos I made of these inspections. https://www.youtube.com/playlist?list=PLP95CUmMHEso3B-6tp3O1Qnqzz1tzdl2W My conclusion is that the balance is rubbing on the pallet bridge. I don't think this was caused by my accident as I was concerned about the lack of clearance before that, and it would also explain the issues the watch had with very different timing dial up vs dial down.

 

I'm aware that turning the lower capstone over might fix this, but I'd rather fix it properly if I can. It's my understanding that the Diashock setting is friction fitted to the mainplate so I should be able to adjust it with a jewelling tool, right?

(Sorry about the upside-down photos.) I am going to get a new cap jewel for this setting as the current one has a divot:

I should probably wait until I have the replacement before I do anything about the end shake in case the new stone has a different thickness.

 

As caseback says . The capstone thickness does not influence the height of the balance. The capstone sits with its flat side at the same position, i thicker stone would merely put more pressure on the shock spring. I would still try flipping the stone out of curiosity, the center of dome side will protrude past the capstone's normal seating position and will still givel good oil retention capillary action. Just an experiment to try for whenever you're in a hole. If the shock spring wont lock into position with the same ease then obviously dont force it.

[GPrideaux]

Ok, I now understand that the Diashock setting does not have parallel sides that can be height adjusted: it has a fixed height and needs to be pushed all the way in. I have a Seiko 7S26C donor movement and I used it to experiment with my recently acquired staking set. I pushed the Diashock setting out of the mainplate, then pushed it back in. Then I swapped to the 6105 and gave the Diashock a good push. I couldn't feel any movement, but with the pallet and balance reinstalled I can now see light between the balance and the pallet bridge when it is dial down and I'm hoping it has moved it enough.

That's enough for today, but when I get to it next I'll take the pallet out again and re-do the free oscillation test with the pallet bridge in place. With the pallet in it is swinging for about 22s in DD and something similar (as best I can tell) DU and about 18 or 19s in crown up. I was hoping for more but I guess the pallet takes a little of its inertia each swing so the real test will require that removed.

I would like to do the experiment of flipping the capstone at some time, but I'm not sure about doing it on this movement just now. Perhaps I'll try it when I'm doing the final reassembly and lubrication, but at the moment its looking like it won't be necessary.

[Neverenoughwatches]

Do your free oscillation test with the pallet bridge fitted but not the pallet fork.  

As you said

8 minutes ago, Neverenoughwatches said:

Do your free oscillation test with the pallet bridge fitted but not the pallet fork.  

As you said

Just thinking, bottom pivot might be short, you have excessive endshake for a reason.

Edited November 9, 2024 by Neverenoughwatches

[GPrideaux]

On 11/9/2024 at 11:50 PM, Neverenoughwatches said:

Just thinking, bottom pivot might be short, you have excessive endshake for a reason.

Perhaps. The divot in the capstone came from somewhere so presumably the pivot in contact with it might have experienced similar damage. Although I don't know by how much the end shake is excessive now. I think I moved the setting in by perhaps one or two hundredths of a mm, and the divot in the capstone might be another hundredth. According to Jon's lesson I referenced above, the balance should have about 0.09 end shake; I wonder if it might be approaching that now. I need to think of a way to calibrate my microscope to quantify how much the balance staff is moving from the images.

As I mentioned, I do have a NOS balance complete on order. When it arrives I may be able to compare its pivots with the one in the watch without opening the packaging. (If I don't need to use it I'll probably re-sell it, so it would be nice to be able to decide that without opening the packaging.) We'll see.

[nevenbekriev]

Well, it is not needed to measure the end shake, seeing it is enough. If You make a video, like the previous, we will tell if it is OK or not.

What looks like that divot in the endstone? I doubt that it will lead to enlarged axial free play, but may be it will eat some amplitude. The stone can be replaced, but can be lapped too. Didn't You try if the movement is working now???

The axial free play is regulated by bending the cock. Of course, bending is just a little and the balance must be out of the movement. The bending is at the cock base, not where the antishock is. There are some other ways like kerning the place where the cock lays, but I will not recommend that. The antishock settings (as far as I know) always seat flush to the end and are not shifted for regulation in any watch movement.

[rehajm]

On 11/7/2024 at 11:30 PM, GPrideaux said:

If I'd bent it down it would've reduced the end shake, and if anything there's still too much, not too little.

A bend on the back edge of the cock could cause an increase in end shake, and downward pressure on the cock could flatten burrs someone dug up to reduce shake. It could have also bent or altered the position of the alignment posts sufficient to create a problem…

[Neverenoughwatches]

6 hours ago, GPrideaux said:

According to Jon's lesson I referenced above, the balance should have about 0.09 end shake 

Working down through the train as arbors becomes pivots getting smaller and smaller, endshake required reduces also. 0.09 ......almost a tenth of a mm is way too much for a balance staff.

Nothing is gained by having this much endshake here.

4 hours ago, nevenbekriev said:

Well, it is not needed to measure the end shake, seeing it is enough. If You make a video, like the previous, we will tell if it is OK or not.

What looks like that divot in the endstone? I doubt that it will lead to enlarged axial free play, but may be it will eat some amplitude. The stone can be replaced, but can be lapped too. Didn't You try if the movement is working now???

The axial free play is regulated by bending the cock. Of course, bending is just a little and the balance must be out of the movement. The bending is at the cock base, not where the antishock is. There are some other ways like kerning the place where the cock lays, but I will not recommend that. The antishock settings (as far as I know) always seat flush to the end and are not shifted for regulation in any watch movement.

The staff is too low Nev.

As i understood it....... the balance wheel is too low rubbing on the lever bridge.

5 hours ago, nevenbekriev said:

Well, it is not needed to measure the end shake, seeing it is enough. If You make a video, like the previous, we will tell if it is OK or not.

I can't say i have even attempted to measure endshake, like you say Nev seeing it is enough.

I'd say you just need to know that endshake exists and that it isn't too much. If its barely visible at x10 and the pivots are still free then you've probably achieved correct endshake.

[nevenbekriev]

As it is hard to explain to a someone with no experience how much endshake is enough, it is easier to tell that the ensheke should be about the same as the one of the escape wheel. Then He will be able to move the escape wheel up/down and see how much it moves and compare to the same kind of movement of the balance staff. Here we need to say that the balance staff movement is to be observed, not the balance rim movement.

[GPrideaux]

Thanks all. That gives me a clearer understanding of what I'm looking for. I still haven't got back to the movement yet. Life intervenes.