Diagnosing problems from Timegraph results

[Bonzer 1]

As a noob, I've serviced a Longines watch recently but there are a couple of issues I have noticed when putting it on the timegrapher (screenshot attached):

1) There is a variation that occurs every 5-6 seconds - what could be the possible cause?

2) The amplitude, although much better than before the service, is still low - again, what could be the cause/s? 

The servicing included a complete strip-down, cleaning all parts, reassembling and oiling.

 

[nickelsilver 1,234]

You can't get into dissecting anything on the graph until the amplitude is up, you'll want at least 250 degrees, better 270-280. What could be causing the low amplitude is a laundry list of things.

-was the train free, did it move freely with just 1 or 2 clicks of wind and did the escape wheel reverse when it came to a stop?

-endshakes all good?

-mainspring and barrel in good condition?

-pallet fork snaps cleanly to its bankings  with just a few clicks of wind?

-balance pivots in good shape, all jewels in good shape, hairspring in good shape?

-did you peg jewels, especially the pallet fork jewels (and not oil the pf jewels)?

-how did you clean? Is it oiled correctly?

 

About any one of the above things could rob 100 degrees, with almost certainly the exception of the one thing everyone always goes at first- the mainspring.

[WellAdjusted 22]

Agreed, the first step is understanding the low amplitude, which is a kind of global health marker.

In the spirit of starting at the mainspring, was it replaced? Does it have a white alloy mainspring?

A common cause of poor amplitude is when the mainspring barrel binds against the arbor. It's a good idea to insert the arbor into the empty barrel, close it up, and check out the clearances. The empty barrel should spin smoothly around the arbor.

[VWatchie 462]

1 hour ago, WellAdjusted said:

It's a good idea to insert the arbor into the empty barrel, close it up, and check out the clearances. The empty barrel should spin smoothly around the arbor.

I had been doing this for years before realizing the importance this. Good point, and a very important point to start with!

[Bonzer 1]

2 hours ago, nickelsilver said:

 

-was the train free, did it move freely with just 1 or 2 clicks of wind and did the escape wheel reverse when it came to a stop?

They all moved freely  - although I didn't check with just one or two clicks of wind

2 hours ago, nickelsilver said:

-endshakes all good?

They seemed ok, but I will re-check

2 hours ago, nickelsilver said:

-mainspring and barrel in good condition?

The mainspring and barrel are the two things I did not open - although you indicate that the amplitude probably hasn't got much to do with this?

2 hours ago, nickelsilver said:

-pallet fork snaps cleanly to its bankings  with just a few clicks of wind?

Again, I didn't check with one or two clicks but it seemed ok - another thing to go back and check

2 hours ago, nickelsilver said:

-balance pivots in good shape, all jewels in good shape, hairspring in good shape?

Yes, they are fine - removed shock springs and cleaned the endstones and oiled (9010)

2 hours ago, nickelsilver said:

-did you peg jewels, especially the pallet fork jewels (and not oil the pf jewels)?

I dd peg them but probably not very well. Need to do them again. They weren't oiled afterward

2 hours ago, nickelsilver said:

-how did you clean? Is it oiled correctly?

Probably not as well as I could as I haven't got the right gear yet. Soaked everything except balance and PF in Naptha over night. Agitated it all several times. Two rinses in IPA.  PF and balance received same treatment but just for minutes rather than overnight.

As far as I could ascertain - it was all oiled reasonably ok. I used 9010 on most pivots. 9415 on the PF exist stone and D5 on centre wheel pinion and high friction parts.

All in all, an amateur job but I was trying to be diligent! I will re-check the other things you mentioned and re-peg the PF jewels.

 

[Bonzer 1]

2 hours ago, nickelsilver said:

-was the train free, did it move freely with just 1 or 2 clicks of wind and did the escape wheel reverse when it came to a stop?

I've just check this and the train runs very well with one or two clicks but the escape wheel doesn't reverse when it stops. It sort of continues creeping in the same direction for a bit. Does this mean the mainspring could be a problem?

[JohnR725 1,359]

Can we also get a oscilloscope view and more than one position. Also the model number of the watch.

[WellAdjusted 22]

If the mainspring barrel wasn't opened, that is the place to start! As statisticians say, "Rare things happen rarely," so it's best to look at the obvious things.

There's a good chance that you have an old, "set" mainspring that simply isn't generating any force. Removing it, cleaning the barrel and arbor, and replacing it with a white alloy mainspring might cure what ails this watch.

Many things make a small difference, but replacing a tired mainspring is the low-hanging fruit of amplitude.

[JohnR725 1,359]

6 hours ago, Bonzer said:

The mainspring and barrel are the two things I did not open - although you indicate that the amplitude probably hasn't got much to do with this?

 

3 hours ago, WellAdjusted said:

There's a good chance that you have an old, "set" mainspring that simply isn't generating any force. Removing it, cleaning the barrel and arbor, and replacing it with a white alloy mainspring might cure what ails this watch.

With a set blued steel mainspring that's in desperate need of being replaced you can still verify that a watch is functional and you would have more than 160° of amplitude. At least for a little while it's not going to have the running time of a mainspring that's in proper condition.

10 hours ago, Bonzer said:

There is a variation that occurs every 5-6 seconds - what could be the possible cause?

You repeating pattern could be magnetism or something in the gear train. I'm attaching an image it will differ depending upon the gear ratios of the particular watch but ill give you an idea. Then the picture came out of a manual for timing machine. Even though it's an older machine 90% of the manual is still relevant for modern timing machines well worth the read I'm attaching that also.

Then your amplitude and the way the rest of the watch looks just doesn't look quite right. Usually a really low amplitude everything gets magnified beat looks really bad everything will just look bad euros almost looks to  good for that low of amplitude. Which is one of the reasons I'd like to see timing and more than one position which you should always do anyway. Plus the oscilloscope view.

 

1556614207_GreinerMicromatTimingmachine.pdf

[nickelsilver 1,234]

I agree with John, even with a set steel spring you might be missing 10-20 degrees of amplitude, but not much more. At WOSTEP they spend a lot of time on barrel inspection*, adjusting, and maintenance. It is very important. But I find that most watchmakers, both professional and amateur, go straight to changing the spring or barrel complete when 49/50 times it's something or combination of things else (one of the combination may well be the spring too of course). I've never seen a watch that was in just decent but worn condition miss 100 degrees from a tired mainspring. Wrong mainspring, yes, if it's 0.02mm thinner than it's supposed to be, or 0.1mm too high and rubbing the lid and barrel bottom.

 

*they also spend a lot of time on adjusting the escapement, which is another thing some folks who have a little more experience go straight to.  The issue, or a large part of it, is often in between these two extremes.

8 hours ago, Bonzer said:

I've just check this and the train runs very well with one or two clicks but the escape wheel doesn't reverse when it stops. It sort of continues creeping in the same direction for a bit. Does this mean the mainspring could be a problem?

This is totally an indication of the train freedom, and very much not an indication of mainspring condition. I don't know which caliber of Longines you are working on; some from the mid 20th century have friction spring for the sweep seconds which may cause what you describe. But if you don't have a friction spring, after a few clicks of power the train should spin up, then watching the escape wheel it will reverse. This is from the momentum of the train and the backlash in the gearing. It's normally checked with the train unlubricated; the oil can have just enough damping effect to keep the reversal from occurring, but the train should come to a gradual halt, not in spurts.

 

If this is an actual Longines caliber, they are some of the nicest movements of the last century.

 

And the old Greiner manual that John posted is excellent. Well worth printing out and reading all the way through. Even if it seems like a lot of it is hard to grasp, it will work its way into your thinking and you'll find yourself going back and understanding more each time.

Edited February 23 by nickelsilver

[mikepilk 149]

4 hours ago, nickelsilver said:

I agree with John, even with a set steel spring you might be missing 10-20 degrees of amplitude, but not much more. At WOSTEP they spend a lot of time on barrel inspection*, adjusting, and maintenance. It is very important. But I find that most watchmakers, both professional and amateur, go straight to changing the spring or barrel complete when 49/50 times it's something or combination of things else (one of the combination may well be the spring too of course). I've never seen a watch that was in just decent but worn condition miss 100 degrees from a tired mainspring. Wrong mainspring, yes, if it's 0.02mm thinner than it's supposed to be, or 0.1mm too high and rubbing the lid and barrel bottom.

If this is an actual Longines caliber, they are some of the nicest movements of the last century.

Useful info (as always) @nickelsilver  I've also been guilty of blaming low amplitude on the mainspring too often, and changing it as a matter of routine. Thanks to your info, I now rarely change them. BUT I now always assemble the barrel without the spring to check the arbor is free.
(Though I have just serviced an ETA 2789 : amplitude 280° but only about 30 h power reserve, which could be a tired mainspring).

I agree about the quality of Longines movements - I have a few from the 60s-70s and they are just so nice to work on. They just fit together so well, and I've always had great amplitude. I rate them at least as good as Omega. My Cal 284 is probably my most accurate watch.

Edited February 23 by mikepilk

[canthus 145]

Re the Greiner pdf download, I see that pages 2-27, 2-28, 2-29 are missing.  Is there a reason why?

[Nucejoe 1,425]

 Have you checked the pinon on escape wheel under high magnification?

[Bonzer 1]

I'll reassemble later today and get some timegtaph shots in different positions and get a scope shot. Thanks for the attached docs too 

The movement is a longines cal 6942

The barrel says that the spring is 'self lubricated'. Obviously I need to do some maintenance on the barrel even if it's not the cause of all the loss in amplitide so what should I do to it if it says self lubricated, apart from check the arbour clearance ?

I haven't looked closely at the escape wheel pinion, I'll do that later..thanks. 

So the lack of backlash on train wheels is possibly because I'd already oiled then? It does come to a gradual halt..it's very smooth.

 

Thanks all!

[JohnR725 1,359]

12 minutes ago, Bonzer said:

The movement is a longines cal 6942

What's interesting about cousins is that sometimes are technical bulletins are not complete you'll notice the word smaller in the title the complete document is 20 pages and the smaller version is five pages. Unfortunately the complete document is watermarked with where it came from which always makes me nervous so I will just extract interesting things out of it. Although I might have a non-watermarked version if I could figure out what I did with my longines Physical manuals but the looks of it I don't think we need it.

Mainspring and barrel is labeled as having a stainless steel spring that basically requires no lubrication is almost unbreakable and it doesn't say this but basically perfect forever. I doubt that's where your problem is although you could probably put a little oil on the Arbor pivots. Like some HP 1300 which I believe is what Omega recommends further brand-new dry barrels before you put them in service.

Bananas nothing really remarkable in the technical documentation nothing it jumps out as potential problems or anything else. Unfortunately came before nifty timing machines with amplitude so that isn't mentioned at all but I am going to attach the rate specification I don't even think they said how many hours this thing supposed to run.

If you wind the watch up how long does it run before it comes to a stop? That's also a test for the mainspring at the amplitude you have it really shouldn't take very long if the amplitude is really correct.

 

3804_Longines 6902-6972_Smaller-1.pdf

[JohnR725 1,359]

1 hour ago, canthus said:

Re the Greiner pdf download, I see that pages 2-27, 2-28, 2-29 are missing.  Is there a reason why?

To be honest I never paid attention to the document at least as far as the pages go. I had assumed that it was correct. Fortunately it looks like for this version of the manual it's basically identical to an earlier version of their manual. This  manual actually came from the company it scanned it looks really visually nice. I have another version that I scanned from one of the earlier manuals is black and white it's not quite as nice but the file sizes smaller. The missing pages are only of interest if you have their special pickup for magnetic fields generated by electric watches. So yes you have the entire manual you just missing the electric watch section.

[JohnR725 1,359]

3 hours ago, mikepilk said:

(Though I have just serviced an ETA 2789 : amplitude 280° but only about 30 h power reserve, which could be a tired mainspring)

According to the link below you're supposed to hit 47 hours you're running a little short.

What I find interesting about companies like ETA  is if you Can  find their manufacturing information documentation It gives you all sorts of interesting specifications unfortunately I don't have anything for the 27 series so currently  I'm looking at a 2660. The run time is Not as long but what's interesting is all they care about is what is it doing at the end of 24 hours. So even though they specify these long run times in the specifications they only care about 24 hours. In this particular case this watch could be 190° With vertical 6 o'clock up.

 

http://www.ranfft.de/cgi-bin/bidfun-db.cgi?10&ranfft&0&2uswk&ETA_2789

[mikepilk 149]

14 minutes ago, JohnR725 said:

According to the link below you're supposed to hit 47 hours you're running a little short.

What I find interesting about companies like ETA  is if you Can  find their manufacturing information documentation It gives you all sorts of interesting specifications unfortunately I don't have anything for the 27 series so currently  I'm looking at a 2660. The run time is Not as long but what's interesting is all they care about is what is it doing at the end of 24 hours. So even though they specify these long run times in the specifications they only care about 24 hours. In this particular case this watch could be 190° With vertical 6 o'clock up.

 

http://www.ranfft.de/cgi-bin/bidfun-db.cgi?10&ranfft&0&2uswk&ETA_2789

Hi John, I use  Ranfft to get the reserve times. I just tried manually winding the barrel with a screwdriver to see when it slips - I got 6.5 turns (I don't know if ETA give a figure?) but that seems OK to me.

After 24 hours I'm getting about 230° DU, so it's OK.  Looks like the mainspring is a little tired.

[Bonzer 1]

4 hours ago, JohnR725 said:

What's interesting about cousins is that sometimes are technical bulletins are not complete you'll notice the word smaller in the title the complete document is 20 pages and the smaller version is five pages. Unfortunately the complete document is watermarked with where it came from which always makes me nervous so I will just extract interesting things out of it. Although I might have a non-watermarked version if I could figure out what I did with my longines Physical manuals but the looks of it I don't think we need it.

I searched the Cousins website and only found the 6 page version you speak of. Is the longer watermarked version on there?

Regarding the length of times the movement runs, I haven't really tested it closely but I  think I pretty much get 36 hours out of it but I will see, once it's back together.

Edited February 23 by Bonzer

[mikepilk 149]

36 hours is fine. Ranfft gives a power reserve of 40 h

http://www.ranfft.de/cgi-bin/bidfun-db.cgi?10&ranfft&&2uswk&Longines_6942

[Bonzer 1]

Ok, I opened the barrel and cleaned the lid and oiled the shoulder on the arbor but as I don't have a winder, I decided not to risk removing and handwinding the spring. I reassembled the movement at took some readings at full wind minus half hour (as the spec posted above stated).

I recorded three positions - DU, PD & PL. I also took scope snapshots at each position but using the program TG, as the signal amplitude on WOS is quite low. TG and WOS seem to give fairly close results most of the time.

[JohnR725 1,359]

18 hours ago, mikepilk said:

I got 6.5 turns (I don't know if ETA give a figure?)

I decided that your watch doesn't exist. That's because I can get manufacturing information for the 26xx and 28xx the 27 just doesn't seem to exist. There is a tech sheet out there but that's it. So I'm attaching a different manufacturing information sheet that somewhat resembles your watch. This way you can see what is in there and what's not in their.

The only thing that I remember telling you how many turns the ratchet wheel is Seiko  their OEM division time module and that specifies eight turns the ratchet wheel is fully wound up.

So as you can see all I really care about is at the end of 24 hours does the watch keep time and is the isochronism number correct. The longer running time helps with that. So real life if your watch Time it was running at 199° throughout the 24 hours but I The timing's backs it would be fine. It would be fine on this discussion group everybody be unhappy unless it's really close to 300° but if the watch keeps time over 24 hours that's important thing. Unfortunately to keep time of the amplitude needs to be higher.

 

Which for the original question here your amplitude is still too low which I bring up down below the real question is what is the amplitude at the end of 24 hours and what is the timekeeping look like. That's really the clue because at 194 dial up and you also should do A dial down just to compare 24 hours probably aren't going to be running very well.

 

14 hours ago, Bonzer said:

I searched the Cousins website and only found the 6 page version you speak of. Is the longer watermarked version on there?

Two ways to get the longer document know the right people who have access to the Swatch group website and ask really nicely or have access to the physical document. I find scanning documents really not my thing. Then if the papers really thin like in this document it's easy to miss something. This means you end up with two  PDFs to get the entire document.

Do you have access to some way to demagnetized the watch?

How did you oil the balance jewels? Did you disassemble them? Reason I ask is if you disassemble them to clean them did you make sure you put the and stones the right way. Towards one side is flat that goes in the other side is curved that goes out if you mix them up you'd have no end shake her little and shake and you would not run the best.

Casually since both timing programs look the same everything sounds wonderful that you've done and your amplitude still sucks there's something still not right here.

 

I've also attached a another timing machine  information interpretation sheet.

Longines No2a.PDF Longines No2b.PDF Timing-Machine-Charts.PDF Eta 2801-2a to 2836-2 MANUFACTURING info.pdf

[Bonzer 1]

8 hours ago, JohnR725 said:

Which for the original question here your amplitude is still too low which I bring up down below the real question is what is the amplitude at the end of 24 hours and what is the timekeeping look like. That's really the clue because at 194 dial up and you also should do A dial down just to compare 24 hours probably aren't going to be running very well.

I've attached screenshots of the watch after 24 hours of running and included the DD position

8 hours ago, JohnR725 said:

 

Two ways to get the longer document know the right people who have access to the Swatch group website and ask really nicely or have access to the physical document. I find scanning documents really not my thing. Then if the papers really thin like in this document it's easy to miss something. This means you end up with two  PDFs to get the entire document.

Many thanks for the Longines documents!!

8 hours ago, JohnR725 said:

Do you have access to some way to demagnetized the watch?

No, not yet. I notice from one of the documents that you attached that the variations in output I'm experiencing could be due to magnetism on the escape wheel (or possible because it's out-of-round). Would this also affect amplitude?

8 hours ago, JohnR725 said:

How did you oil the balance jewels? Did you disassemble them? Reason I ask is if you disassemble them to clean them did you make sure you put the and stones the right way. Towards one side is flat that goes in the other side is curved that goes out if you mix them up you'd have no end shake her little and shake and you would not run the best.

I opened the diashock springs and took out the endstones. I've double-checked and they are definitely the right way round. I'm not 100% convinced that the diashock springs are in good condition though. I might try to get a photo. If they weren't seated correctly I suppose there could be incorrect pressure on the balance pivot.

[JohnR725 1,359]

5 minutes ago, Bonzer said:

(or possible because it's out-of-round). Would this also affect amplitude?

If you have uneven power fluctuations you'll see a change in amplitude and usually that results in a line spacing difference. Your lines are perfectly spaced at least consistently spaced indicating something else? I'm attaching some images I was once doing a comparison of timing machines and the person who loaned me his Chinese machine loaned his watches. Normally watchmakers would demagnetized the watch before placing it on a timing machine or even before working on them here I just timed of exactly as they were found.

So you'll see the images the effect of a magnetic field and then after it was demagnetized.

6 minutes ago, Bonzer said:

If they weren't seated correctly I suppose there could be incorrect pressure on the balance pivot.

If the stones are upside down you'd have lack of end shake. If the Springs were weak or possibly not quite right I don't think it be an issue.

If you look at fully wound up versus 24 hours it looks like? The typically dial up and down should be the same when you go to the pendant or crown positions you always lose amplitude. If we ignore the the numbers kind of everything looks right in that 24 hours later you have less amplitude. It just looks like you're losing a heck of a lot of power consistently someplace or you just don't have power.

[Bonzer 1]

Ok, so would I be right in saying that the main areas of suspicion for the loss in power would be:

Balance wheel pivot/jewels

Escape wheel pivot/jewels

Mainspring

?

Would it also be correct to say that the rest of the train not as influential when it comes to power?

[Nucejoe 1,425]

On 2/22/2021 at 10:20 PM, Bonzer said:

 There is a variation that occurs every 5-6 seconds - what could be the possible cause?

This corresponds to the period of the escape wheel. The fault includes power transmition to & from the escape wheel, a list of such fault;

-Bent escape wheel  pivot, rusty escape pinion, wobbling( third gear, escape wheel or damaged teeth ).

-Faulty jewel or end stones, no end shake so end stones be pushing on pivots.

As said any such fault eats amplitude.

Use a wooden stick to assist turning the barrel, if little assistance gives you good amplitude, the fault is likely to be inside the barrel. 

TG shows you something is wrong, interpretation of T-graph hardly gets better than what John is doing, visual inspection of parts is an instrument of diagonstics.

Good luck

Regs Joe