Could this incorrectly riveted bridle be the cause of poor amplitude

[VWatchie]

About a year and a half ago I serviced an ETA 2804-2 movement in good condition but in need of service (my service walkthrough here). I haven't worn the watch much since then as it didn't seem to keep time very well. When I measured the amplitude a few days ago with the mainspring fully wound, I got an amplitude of about 260 degrees (dial-up), but already after 24 hours, the amplitude had dropped to about 130 degrees. I also got the impression that the rate can vary greatly even with the mainspring fully wound.

So yesterday I took the movement apart and looked for obvious faults - the thorough troubleshooting I'll do once it's gone through my cleaning machine - but found no glaring faults except for the bridle on the mainspring seen in the above pictures. As can be seen, the bridle is riveted at an angle.

My question is if you think this could be the reason for the poor timekeeping. I should perhaps mention that the bridle is very stiff so it takes quite a bit of force to align it with the rest of the spring, but not so much that it misaligns the barrel lid.

Edited June 24, 2023 by VWatchie

[Klassiker]

I don't see a problem so much with the riveting, but the bridle has a kink in it. That might stop the bridle laying against the inner wall of the barrel with even pressure. It's easy to image the result would be reduced and inconsistent braking.

Please ignore. As pointed out below, I was incorrectly assuming an automatic mainspring.

Edited June 24, 2023 by Klassiker
Apology added.

[nickelsilver]

12 minutes ago, Klassiker said:

I don't see a problem so much with the riveting, but the bridle has a kink in it. That might stop the bridle laying against the inner wall of the barrel with even pressure. It's easy to image the result would be reduced and inconsistent braking.

The 2804 is manual wind.

 

As long as it sits flat in the barrel and the lid can snap on correctly it won't affect the amplitude.

[Nucejoe]

Is this the old/ used mainspring you repaired?     or brand new? 

Anything  that can affect the torque supply  affects  the amplitude.

I would at least file the sharp corners on the bridle. Heat to a glow and file. 

Regds

 

[Neverenoughwatches]

2 hours ago, VWatchie said:

About a year and a half ago I serviced an ETA 2804-2 movement in good condition but in need of service (my service walkthrough here). I haven't worn the watch much since then as it didn't seem to keep time very well. When I measured the amplitude a few days ago with the mainspring fully wound, I got an amplitude of about 260 degrees (dial-up), but already after 24 hours, the amplitude had dropped to about 130 degrees. I also got the impression that the rate can vary greatly even with the mainspring fully wound.

So yesterday I took the movement apart and looked for obvious faults - the thorough troubleshooting I'll do once it's gone through my cleaning machine - but found no glaring faults except for the bridle on the mainspring seen in the above pictures. As can be seen, the bridle is riveted at an angle.

My question is if you think this could be the reason for the poor timekeeping. I should perhaps mention that the bridle is very stiff so it takes quite a bit of force to align it with the rest of the spring, but not so much that it misaligns the barrel lid.

Does the extra height have any effect on the lid not closing fully ? The mainspring should be free floating, we talk a lot about internal damage to a barrel from a deformed spring, any evidence of that  Watchie. If stoning  off the excess to make the spring uniform thickness doesnt effect its catchment on the barrel wall ledge,then i would do that. I would also be worried if it would be damaging the wall of my mainspring winder. Seems like a pretty poor manufacture. 

Edited June 24, 2023 by Neverenoughwatches

[Nucejoe]

How long did your watch run with this spring fully wound. 

I am thinking a spindle this long might slip even in a manual wind barrel. 

I just replace this with a new mainspring for 2804.

rgds

 

[VWatchie]

5 hours ago, nickelsilver said:

As long as it sits flat in the barrel and the lid can snap on correctly it won't affect the amplitude.

OK, thanks, so we'll see what I can find when doing my thorough fault finding. Unfortunately, I didn't examine the barrel lid prior to removing it (another lesson learned).

5 hours ago, Nucejoe said:

Is this the old/ used mainspring you repaired?     or brand new? 

Brand new Generale Ressorts spring, and quite expensive too. I couldn't quite remember if I had replaced the spring when I serviced it but looking at some old Cousins orders I found an "ETA 2801 Mainspring 1.23x0.134x400x10.5" ordered at the time of the service so I'm pretty sure about that.

5 hours ago, Neverenoughwatches said:

Does the extra height have any effect on the lid not closing fully ?

I don't think so, but as mentioned, it didn't cross my mind to have an extra look at the lid before removing it.

5 hours ago, Neverenoughwatches said:

we talk a lot about internal damage to a barrel from a deformed spring, any evidence of that 

No, none that I can see with my bare eyes, but I'll have another look through my scope once I get to my fault-finding.

5 hours ago, Neverenoughwatches said:

Seems like a pretty poor manufacture. 

Indeed, and that's Generale Ressorts for you. Pretty much a hit or miss, and much too often a miss 

3 hours ago, Nucejoe said:

I just replace this with a new mainspring for 2804.

That's my plan, but this time around I'll remove it from the spacer first and have a thorough look at it before getting it into the barrel.

[Neverenoughwatches]

1 hour ago, VWatchie said:

6 hours ago, Neverenoughwatches said:

Does the extra height have any effect on the lid not closing fully ?

I don't think so, but as mentioned, it didn't cross my mind to have an extra look at the lid before removing it.

How about measuring the spring to include the extra part of the bridle sticking out , and also measuring the inside of barrel upto the lid seat. Just to see how much clearance there is.

[VWatchie]

51 minutes ago, Neverenoughwatches said:

Just to see how much clearance there is

Visually, there's zero clearance. So, the bridle is definitely pressed into the barrel by the lid. There's no doubt in my mind that the bridle is exerting pressure but not to the extent that the lid gives, I think. When I measured the height of the spring including the section sticking up I got it to roughly 1.6 mm whereas the rest of the spring is 1.23 mm.

[Neverenoughwatches]

9 minutes ago, VWatchie said:

Visually, there's zero clearance. So, the bridle is definitely pressed into the barrel by the lid. There's no doubt in my mind that the bridle is exerting pressure but not to the extent that the lid gives, I think. When I measured the height of the spring including the section sticking up I got it to roughly 1.6 mm whereas the rest of the spring is 1.23 mm.

Checking the height spec. we should be looking at 1.25mm so thats a big difference of what could be as much as 0.3mm extra height allowing for some error . My understanding of mainspring clearance inside the barrel is usually only a couple of hundredths mm. 

[JohnR725]

20 hours ago, VWatchie said:

but already after 24 hours, the amplitude had dropped to about 130 degrees.

It's a rather dramatic drop in the amplitude over 24 hours.

20 hours ago, VWatchie said:

My question is if you think this could be the reason for the poor timekeeping. I should perhaps mention that the bridle is very stiff so it takes quite a bit of force to align it with the rest of the spring, but not so much that it misaligns the barrel lid.

The bad part of the spring is at the very end? As it's an automatic watch it means you probably don't slip properly sticking up the way it is conceivably means you wouldn't slip at all. So if it's sticking up and it doesn't slip it seems like you should still have more amplitude at 24 hours verses a spring that slips continuously.

Then timekeeping affected by amplitude would suggest a problem with the regulator pins too far apart or extremely Uneven power.

Then a confusion my confusion perhaps.

20 hours ago, VWatchie said:

About a year and a half ago I serviced an ETA 2804-2 movement in good condition but in need of service (my service walkthrough here)

So this is the watch from that walk-through?

 

 

[VWatchie]

On 6/25/2023 at 10:08 AM, JohnR725 said:

As it's an automatic watch

I should have mentioned it, but it is the manual equivalent to the famous ETA 2824-2 but this is the manually wound ETA calibre 2804-2.

On 6/25/2023 at 10:08 AM, JohnR725 said:

Then timekeeping affected by amplitude would suggest a problem with the regulator pins too far apart or extremely Uneven power.

The regulator pins seem fine, but I haven't yet tested if the train of wheels binds (hoping that's the problem). So far I've tested the balance assembly on an otherwise empty main plate and it looks perfect to me. When I pull the balance wheel 180 degrees with an oiler and then release it, it oscillates for about 70 seconds in the horizontal positions and for about half that time (35 seconds) in the vertical positions. During the last 20 seconds or so (horizontal position) the oscillations are minuscule but clearly visible. It's the first time I've tested a balance in this way (got the idea from the Chronoglide Youtube channel) so I really don't know what to expect, but it sort of looks amazingly healthy 

Right now I'm trying to establish if the pallet fork is moving as freely as it should but I'll start a new thread about that to get some more attention.

On 6/25/2023 at 10:08 AM, JohnR725 said:

So this is the watch from that walk-through?

Yes, it is!

Edited July 8, 2023 by VWatchie

[Neverenoughwatches]

2 hours ago, VWatchie said:

I should have mentioned it, but it is the manual equivalent to the famous ETA 2824-2 but this is the manually wound ETA calibre 2804-2.

The regulator pins seem fine, but I haven't yet tested if the train of wheels binds (hoping that's the problem). So far I've tested the balance assembly on an otherwise empty main plate and it looks perfect to me. When I pull the balance wheel 180 degrees with an oiler and then release it, it oscillates for about 70 seconds in the horizontal positions and for about half that time (35 seconds) in the vertical positions. During the last 20 seconds or so (horizontal position) the oscillations are minuscule but clearly visible. It's the first time I've tested a balance in this way (got the idea from the Chronoglide Youtube channel) so I really don't know what to expect, but it sort of looks amazingly healthy 

Right now I'm trying to establish if the pallet fork is moving as freely as it should but I'll start a new thread about that to get some more attention.

Yes, it is!

Did you get to the bottom of the skewed bridle ? It looked like it was adding height to the mainspring. 

[VWatchie]

6 hours ago, Neverenoughwatches said:

Did you get to the bottom of the skewed bridle ? It looked like it was adding height to the mainspring. 

Nope, I've bought a new spring, but this time around I think I'm going to remove it from the spacer so that I can properly inspect it before I install it. It's a bit of a hassle of course as I'll have to use a mainspring winder (or do it by hand) but I'm thinking since we're inspecting everything else why not the mainspring as well?

[HectorLooi]

Don't forget to inspect the barrel and cover too. Check for scratches, concaving of the cover, etc.

Also, check the barrel arbor for damage, sideshake and endshake. While you are at it, check the barrel pivot hole for wear.

[VWatchie]

I'm sorry to have kept you in agony for this long which must have given you many sleepless nights, but finally, to my great joy, I have identified and fixed the source of the low amplitude issue, resulting in excellent improvements.  

Upon close inspection (40X magnification) I found some very slight rust and oxidation on several of the train wheel pivots. Don't know how I missed that when I serviced the movement the first time around (really a bit embarrassing). Perhaps I simply wasn't looking close and/or long enough, and no, the rust didn't get there after the service as I wasn't wearing it due to its poor timekeeping.

So, my beloved Steiner Jacot tool to the rescue and what do you know, amplitude in the horizontal positions is now slightly above 300 degrees and around 275 degrees in the vertical positions. I'll never get rid of my Jacot tool for as long as I'm able to breathe and am capable to use it! 

I haven't had the time to fully assemble the movement yet, but for anyone interested, I get back with the measured amplitude after 24 hours of fully wound.

Edited August 7, 2023 by VWatchie

[VWatchie]

So, the numbers are in.

Fully wound:
Dial up: 309°
Dial down: 303°
Crown Left: 281°
Crown Down: 280°
Crown Right: 282°
Crown up: 283°

Fully wound minus 25.5 hours:
Dial up: 268°
Dial down: 264°
Crown Left: 242°
Crown Down: 242°
Crown Right: 251°
Crown up: 247°

A truly dramatic difference compared to the 260° fully wound and 130° after 24 hours in the dial-up and dial-down positions. I really don't think it was the mainspring that caused the problems although I did change it. Burnishing the pivots is in my opinion what did it, and I think it goes to show how crucial it is that all pivots are perfectly free of any impurities however small. I didn't even notice them the first time a serviced this movement.

[mikepilk]

48 minutes ago, VWatchie said:

. I really don't think it was the mainspring that caused the problems although I did change it. 

Well done, those are impressive figures. 

I have come to realise that the mainspring is rarely the problem*. Even 'tired' looking springs can give reasonable performance.  I use to change them as a matter of course, but now only change them if they are really shot. 

50 minutes ago, VWatchie said:

Burnishing the pivots is in my opinion what did it, and I think it goes to show how crucial it is that all pivots are perfectly free of any impurities however small. I didn't even notice them the first time a serviced this movement.

I'm still not confident about burnishing the pivots. I guess you need full magnification to see what's going on, but then you have very little working depth.

( * One watch was driving me crazy, just couldn't get it working well. Turned out that the new mainspring I fitted hadn't been hardened at all. It was just a piece of bendy wire)

[VWatchie]

On 8/19/2023 at 11:18 AM, mikepilk said:

I use to change them as a matter of course, but now only change them if they are really shot. 

I've started to adopt this approach more and more as well, and listening to the pros like @nickelsilver they do not overestimate the impact the mainspring has on the amplitude.

On 8/19/2023 at 11:18 AM, mikepilk said:

I'm still not confident about burnishing the pivots. I guess you need full magnification to see what's going on, but then you have very little working depth.

As I mentioned in a previous post in this thread "upon close inspection (40X magnification) I found some very slight rust and oxidation on several of the train wheel pivots", and as I didn't find anything else looking suspicious I feel confident that it was the burnishing of the pivots that took care of the problem. Of course, you can never be 100 per cent sure about anything, but that is my best assessment. So, from now on I think I'll be using my Jacot tool a lot more often. Since we are meticulous about the condition of the jewel holes it makes sense that we should be equally meticulous about the condition of the pivots and pivot shoulders.

Edited August 22, 2023 by VWatchie