My first service, still a lot to learn and chasing low amplitude

[neevo]

I’ve ordered a Weishi 1000 as I’m sure I’ll need it for basic regulation in the future with 3 mechanical watches.

Will see if that reads the NH35 any differently to the basic headphone and iPhone app setup I’ve been using.

[neevo]

My timegrapher turned up today so I set about putting the NH35 on it to see what’s going on. The results are so bad that I think the timegrapher is struggling to even read the amplitude and beat error, but I did get some results after putting a full wind in to the movement:

 

The beat error started off at around 9ms so clearly an indicator of something wrong. I was able to adjust it back to 1.0ms but that leaves the stud very close to the body of the balance which I’m not sure is correct.

I put my other watch on the timegrapher (Miyota 8215) to see how that performed and it was an immediate reading that showed good results.

So now I need a plan of attack to diagnose. Whilst I’d like to get this watch back up and running, I’m fine tearing in to this movement to get my skills up and hours under my belt. However diagnosing the issue will take some guidance.

My current thinking is to strip the movement down again and see if I can assemble the train and check for any damage to wheels. Make sure the train is spinning freely. If that checks out then I assume my trouble is coming from the balance wheel. My reading suggests I can put a very small wind on the train without the pallet fork installed and I should see a smooth action with a bit of reverse spin at the end? If that works then I need to focus my attention on the pallet fork or balance wheel.

Edited December 31, 2024 by neevo

[JohnR725]

 

On 12/7/2024 at 8:15 PM, neevo said:

So I put a small wind in to the movement, got the balance spinning and moved the regulator to the middle position. An immediate improvement in the amplitude that was visible to the eye.

if everything is adjusted correctly moving the regulator should cause a problem like you're seeing. One of the things that really kills amplitude is hairsprings that are not flat. So for instance if you hairspring is touching the balance arms that would be very bad for amplitude. It be really nice to get a picture of the balance wheel in the watch not moving so we can see what it looks like.

On 12/7/2024 at 6:51 PM, neevo said:

I just did the balance chaton on the main plate side and it’s a million times easier than those tiny ones. Dropping the chaton on the end stone also makes it a million times easier.

thinking about the jewels did you notice that one side is flat on the other side has a curve? The flat side goes in the curbside goes out otherwise you have an end shake issue and conceivably loss of power.

On 11/30/2024 at 10:56 AM, neevo said:

have a timegrapher app on my phone and was getting 300 degrees of amplitude beforehand. So I’ve definitely done something to the movement. I was very careful was assembly, so I suspect either incorrect lubrication or magnetisation .

 

then I'm curious about the Phone app the name of it. Then when you look at the oscilloscope you can see what the problem is it's not actually triggering where it's supposed to be.

 

35 minutes ago, neevo said:

The beat error started off at around 9ms so clearly an indicator of something wrong. I was able to adjust it back to 1.0ms but that leaves the stud very close to the body of the balance which I’m not sure is correct.

I'm curious about how you're holding the movement in the timing machine a picture that would be nice. Then some close-up pictures of the balance wheel and hairspring in the watch so we can see what that looks like.

Then just in case you don't have the technical guide I've attached that.

 

 

 

NH35_TG.pdf

[tomh207]

Have you demagnetised it yet? The rate is higher than I would expect unless it is magnetised or the hairspring coils are sticky.

 

Tom

[neevo]

I definitely demagnetised it. Did it twice 90 degrees apart. I also made sure the end stones were flat side in so pretty sure it’s not an end stone problem.  But I might triple check the small ones as I’m not 100% sure on those ones.

I’m in Sydney tonight for the fireworks but will aim to get some balance pics when I get home tomorrow. I suspect I’ve damaged the hair spring or the pallet fork, but will aim to methodically check to try and find the cause.

It’s a good excercise in diagnositics at least. Even if I have to buy some parts to fix it.

[tomh207]

45 minutes ago, neevo said:

I definitely demagnetised it. Did it twice 90 degrees apart. I also made sure the end stones were flat side in so pretty sure it’s not an end stone problem.  But I might triple check the small ones as I’m not 100% sure on those ones.

I’m in Sydney tonight for the fireworks but will aim to get some balance pics when I get home tomorrow. I suspect I’ve damaged the hair spring or the pallet fork, but will aim to methodically check to try and find the cause.

It’s a good excercise in diagnositics at least. Even if I have to buy some parts to fix it.

Then it’s back to @JohnR725 ‘s suggested look at the hairspring and see if it is catching/rubbing someplace.

 

Tom

[Neverenoughwatches]

1 hour ago, neevo said:

I definitely demagnetised it. Did it twice 90 degrees apart. I also made sure the end stones were flat side in so pretty sure it’s not an end stone problem.  But I might triple check the small ones as I’m not 100% sure on those ones.

I’m in Sydney tonight for the fireworks but will aim to get some balance pics when I get home tomorrow. I suspect I’ve damaged the hair spring or the pallet fork, but will aim to methodically check to try and find the cause.

It’s a good excercise in diagnositics at least. Even if I have to buy some parts to fix it.

Fast rates are common to the shortening of the active point of the hairspring.  This is the position on the hairspring from which its oscillation begins, in a correct functioning balance that point is between the regulator pins. The further in towards the hairspring the active point is the faster the hairspring will beat. Once beyond the active curve where the regulator pins reside, If the hairspring's natural coil shapes are interrupted then that is where the active point will begin. So coils touching something or touching each other due to magnetism or contamination will move the active point from its regulated position. I tend to imagine magnetism as an invisible, interfering contamination...that can be bad or good...but bad in this case.

[LittleWatchShop]

Along with the other advice here, if you have a smartphone with super slow motion video, take a slomo video of the balance in action. I always do this as a sanity check.

[neevo]

Today I’m going to spend some time in the shed and answer all (hopefully) of the questions asked.

In preparation for that I had a question about how you manage the movement after it’s been serviced and you have to take it apart again. Does it require a full strip down, clean and re-oil each time or can you take a few parts off and reinstall them without messing up the oiling?

Todays questions to answer are:

1. Get some images of the hairspring and balance wheel without them moving

2. Check end stones are in correct way

3. Demagnetise again (just to be sure)

4. Get picture of how I’m holding the movement in the timegrapher

5. Slow mo of the balance wheel running

Edited January 1 by neevo

[LittleWatchShop]

5 hours ago, neevo said:

Demagnetise again (just to be sure)

How are you demagnetizing? What tool?

[JohnR725]

4 hours ago, neevo said:

In preparation for that I had a question about how you manage the movement after it’s been serviced and you have to take it apart again. Does it require a full strip down, clean and re-oil each time or can you take a few parts off and reinstall them without messing up the oiling?

 

I'm not sure you really want A answer to this question. Ideally in a perfect world you would diagnose the problem in a watch. You would repair the problem and verified that the problem is indeed fixed. Then the watch can be disassembled cleaned lubricated etc. Then you have a completely serviced watch it is now running.

But if you determine that whatever repair you did it didn't work and you have to disassemble which unfortunately happens occasionally. Then yes you have to basically start over as you're going to screw the lubrication up. I suppose it depends upon what you had to take off and what exactly you're doing. So for instance if you change the mainspring and you decided that's not the problem you wouldn't have to take the mainspring out again and replace it with another new one or even just taken out clean it all up and put it back together if you did it properly the first time you shouldn't have to take it apart again. But if you disassemble the watch especially the balance wheel you will screw up the lubrication.

On the other hand if you're new to watch repair you need to practice. Watch repair is all about practicing to get better. In other words you need to practice observing what a running watch looks like. Looking at what the hairspring looks like when the balance wheel oscillating. Plus you just need the practice of disassembling and reassembling the watch and having a functional watch afterwords.

What you do not need to do is practice disassembling a watch reassembling hoping that the outcome will somehow change because it will Not. In other words you if you have a watch that's not running you haven't actually done anything to repair anything other than disassembling reassembling the outcome is not going to change typically.

So yes sometimes you may have to service a watch twice. The first servicing just to fix the problems verify the problems have been fixed. If you had to taken apart multiple of times to get those problems fixed then ill have to service at the second time.

5 minutes ago, LittleWatchShop said:

How are you demagnetizing? What tool?

You don't want to ask that question?  Notice this is a discussion of four pages? But, As you've asked I believe it starts on page 2 I snipped out an image this is what's being used. From electrical engineer's point of view you might find those pages interesting to look at.

 

 

[neevo]

So I’ve done these steps this morning with a hope to trying to understand what’s going on. I might look to buy a new movement to get my watch back up and running and then continue to work on this one to build my skills.

First challenge I have is taking decent photos. The scale is so small and I’m trying to use my loupes on my iPhone, but it’s a bit challenging to get good enough images to help with diagnosis.

1. I demagnetised the movement again in 2 positions. Not sure it’s a magnetism issue as it’s no change

2. Here is how the movement is on the timegrapher

3. Here are my attempts to get some photos of the balance.

The hairspring doesn’t appear to have any damaged sections that my novice eyes can spot. However the coils are definitely all bunched up on 1 side

Here is a slo mo video of the balance in action.

https://youtube.com/shorts/qDfTDn0LrmU?si=g8cnBjKF_vs1wYph

[LittleWatchShop]

Well, the demag tool looks legit. Wow, they are selling these things cheap!

Well, the demag tool looks legit. Wow, they are selling these things cheap!

[neevo]

I think I might strip the movement down. I have an old 50’s watch on the bench too that was my father in laws that was a non runner (FHF Standard 96-4).

Ive been able to strip it, clean it, peg all the jewels and reassemble. It has some drag in the train of wheels but even with that I’ve managed to get it to 270 amplitude and bang on beat error and +2s a day. 18 hours later it’s still got the same beat error and rate with 221 amplitude.

My guess is I have broken or damaged something. So I’ll strip it down and try and understand what I’ve done. I don’t think I’ve damaged the train of wheels but my guess is the pallet fork or the balance wheel.

Edited January 2 by neevo

[nevenbekriev]

On 12/3/2024 at 8:21 AM, nevenbekriev said:

Hi Neevo,

Nothing bad in getting demagnetizer, but don't expect something to change alot when You demagnetize the movement. We need real picture of You hairspring.

Alleluia! We have a picture...

You need to open the angle marked with red and may be a little the green one too in order to bring the spring centered. Now it touches the stud or touches it when oscillating.

 

[neevo]

1 hour ago, nevenbekriev said:

Alleluia! We have a picture...

You need to open the angle marked with red and may be a little the green one too in order to bring the spring centered. Now it touches the stud or touches it when oscillating.

 

Ok, sounds like I was a bit heavy handed when I cleaned it first time. Let me have a go at gently teasing that coil out. It looks like that bend in the green section shouldn’t be there.

Will report back when I have updates. Should I attempt that with the balance on the cock still or take it off?

Edited January 2 by neevo

[nevenbekriev]

7 hours ago, neevo said:

Ok, sounds like I was a bit heavy handed when I cleaned it first time. Let me have a go at gently teasing that coil out. It looks like that bend in the green section shouldn’t be there.

Will report back when I have updates. Should I attempt that with the balance on the cock still or take it off?

Do not remove the balance from the cock. When it is on the cock You will be able to see the result of what You are doing. Both the red and the green sections must be there

[neevo]

13 minutes ago, nevenbekriev said:

Do not remove the balance from the cock. When it is on the cock You will be able to see the result of what You are doing. Both the red and the green sections must be there

And I need to make sure that last section is always centred in the pins of the regulator arm?

[Neverenoughwatches]

9 hours ago, neevo said:

It looks like that bend in the green section shouldn’t be there.

That bend on the green section should definitely be there, it's by no accident.  Those two offset bends are the start of the terminal curve finishing at the stud, the space on hairspring where the regulation of the rate takes place. But as Nev says they do need tweeking a little to center the coils, just dont remove them.

[Neverenoughwatches]

15 hours ago, nevenbekriev said:

Alleluia! We have a picture...

You need to open the angle marked with red and may be a little the green one too in order to bring the spring centered. Now it touches the stud or touches it when oscillating.

 

I nearly gave you a laughing face instead of a thumbs up Nev....for just the Alleluia !  alone

[nevenbekriev]

11 hours ago, neevo said:

And I need to make sure that last section is always centred in the pins of the regulator arm?

Yes, but leave this for later. Now You need to bring the spring centered and not touching the stud.

5 hours ago, Neverenoughwatches said:

I nearly gave you a laughing face instead of a thumbs up Nev....for just the Alleluia !  alone

Well, see how many posts and 4 pages, when it was clear from the beginning that the problem has nothing to do with magnetization

[Neverenoughwatches]

27 minutes ago, nevenbekriev said:

Yes, but leave this for later. Now You need to bring the spring centered and not touching the stud.

Well, see how many posts and 4 pages, when it was clear from the beginning that the problem has nothing to do with magnetization

I forget and it's a difficult habit to get out of suspecting an alloy hairspring has magnetism.  I think that comes from being told when I started that the main culprit of sticking coils is magnetism when it obviously can't be.

[RichardHarris123]

14 minutes ago, Neverenoughwatches said:

I forget and it's a difficult habit to get out of suspecting an alloy hairspring has magnetism.  I think that comes from being told when I started that the main culprit of sticking coils is magnetism when it obviously can't be.

As the say " old habits die hard " or more relevant, instilled knowledge is difficult to forget, even if it's incorrect.  

[neevo]

3 hours ago, nevenbekriev said:

Yes, but leave this for later. Now You need to bring the spring centered and not touching the stud.

Very clear. Thanks.

Heading away with the family today for a couple of days. So I’ll leave this for when I return so I don’t rush it and make more of a mess of it.

[neevo]

This morning I made myself a coffee and set about looking at the hairspring. Wow are these things tiny!!!

Originally my plan was to see if I could address the issue on the balance cock itself, but I’m finding magnification is a real problem. I bought some cheap loupes and they really are a struggle to use effectively. I was using the 25x and found it really fuzzy making seeing what I was doing almost impossible. Plus the working distance was so small I couldn’t get my tweezers in there to do any work.

So I’ve bitten the bullet and bought a replacement balance wheel. I then proceeded to take the balance wheel off the cock so I could see it a lot better and have a bit more of a play with it.

I put the wheel in some rodico and used my loupe to snap a photo:

 

My completely novice guessing thought that maybe the last bend was wrong, causing the terminal curve to wrap too far inside?

So I grabbed my tweezers, a yellow oiler and gently set about seeing if I could adjust the last bend:

Whilst it’s not as easy to see, I found using the 15x loupe much easier to use. Partly because it’s a much cleaner image. Holding the bit just before the bend with the tweezers I was able to gently tease the bend with the oiler to something that looked better to my eye.

However I have no idea if this is correct. At this point I’m using it more as a practice exercise. I’m definitely on the look out for a scope though. Not having cheap loupes around my head I’m sure will make a big difference.

As a consequence of adjusting the bend I have put a twist in to the spring. But before I fix that I’ll wait for the experts to advise if I need to do anything else beforehand.

Edited January 10 by neevo
Spelling