First attempt at tuning with a timegrapher...what am I looking at here?

Richb77 · August 14, 2022

Hi all.

So. I have built a couple of seiko NH35/36 based customs and wanted to get more into the servicing of older watches (the NH is great b6t we all know its almost a throw away).

I bought a Weishi 1000 clone (MTG brand) and thought I'd look at my two first. The first watch I made (PVD Samurai) had a movement bought from a reputable ebay seller in the USA. The second (Turtle) was build 6 months later with another movement from the same seller. I have ZERO reason to believe these are anything but genuine Seiko.

The Samurai has been bugging me as I can see its about 3-4 minutes fast buy the end of an 8 hour "day". The second seemed reasonable.

So I threw the Samurai and its all over the shop. +438 S/day amplitude of 160 and beat error of 8.7ms. (I should add. Beat Detection was auto and lift angle set to 53°).

I was puzzled at why it was so bad so put on the Turtle. A tiny nudge to the adjustor and it was reading -1S/day, amplitude of 240 and beat error of 3.7ms.thats OK for me (my Omega seamaster Chrono was spot on! 0 error)

So I went back to the Samurai and this was honestly the best I could get. Amplitude suggestions its become magnetized? Look at the adjustor. I cannot get anymore delay!

Does it need demagnetising? Is it a bad movement? Part.

Any advice would be great.

Edited August 14, 2022 by Richb77
Added info

jdm · August 14, 2022

1 hour ago, Richb77 said:

+438 S/day amplitude of 160 and beat error of 8.7ms

Move the end stud arm first to minimize beat error first, then regulate rate at the regulator arm, repeat these steps as many time as necessary until the best reading is obtained. Even if you have tried already you should try again, maybe you didn't do it right. The two movable arms must never be so close to each other as in your pictures. If you can't get it back to normal performances that is at least 240 deg amplitude, and a good pattern in all position, being it a brand new mov.t it should be returned and refunded.

1 hour ago, Richb77 said:

Amplitude suggestions its become magnetized?

Magnetism does not really affect amplitude. You can, of course, try to demagnetize anyway, and check yourself.

steve1811uk · August 14, 2022

5 hours ago, jdm said:

Move the end stud arm first to minimize beat error first, then regulate rate at the regulator arm, repeat these steps as many time as necessary until the best reading is obtained. Even if you have tried already you should try again, maybe you didn't do it right. The two movable arms must never be so close to each other as in your pictures. If you can't get it back to normal performances that is at least 240 deg amplitude, and a good pattern in all position, being it a brand new mov.t it should be returned and refunded.

Magnetism does not really affect amplitude. You can, of course, try to demagnetize anyway, and check yourself.

It does if the hairspring coils stick together. That's when the rate goes extremely fast too.

Edited August 14, 2022 by steve1811uk
Typo

Richb77 · August 16, 2022

I'll give the seller the benefit of the doubt as this was my first build but I have tried and tried but it's nothing like close to accurate.

I can see (under 60x magnification) the spring is damaged (flattened section where it should not be)so I guess this is my issue. Shame a new spring is the same money as a new movement.

Edited August 16, 2022 by Richb77

JohnR725 · August 16, 2022

On 8/14/2022 at 8:43 AM, jdm said:

Move the end stud arm first to minimize beat error first, then regulate rate at the regulator arm, repeat these steps as many time as necessary until the best reading is obtained. Even if you have tried already you should try again, maybe you didn't do it right. The two movable arms must never be so close to each other as in your pictures. If you can't get it back to normal performances that is at least 240 deg amplitude, and a good pattern in all position, being it a brand new mov.t it should be returned and refunded.

4 hours ago, Richb77 said:

first build but I have tried and tried but it's nothing like close to accurate.

One of the problems with the regulation system is it allows lots of adjustment. Specifically can allow a lot of bad adjustments. For instance putting the watch in beat there is not a negative or positive number. This means you can easily go right past zero all the way to the extreme like you've done now and never find the watch being in beat at all. The same with the regulator arm although there is a plus and minus for rate but you can we get in a situation like you are now where you are very hopelessly out of adjustment.

Then there is the problem of the Etachron system In addition to having those nice movable arms are going to screw up like you have the components on each arm can be rotated. So if you're not careful in addition to putting the arms were the not supposed to be even start rotating stuff and you can destroy the entire hairspring assembly.

Just in case you don't have the technical guide for your movement I've attached that and an image out of the guide. Notice his comments about how to adjust the watch. The first thing I would do is put the stud in the regulating arm back where there supposed to be a reasonably close then you put on the timing machine but also look at they hairspring as more than likely you have been it. As you have nothing to lose you can try to straighten their hairspring. Ideally you'd want to practice on scrap balance wheels that are disposable but seeing as how you can't get a replacement balance wheel who the new watch you now have a balance wheel the practice if it's not flat.

Depending upon how bad things look you can open up the regulator pins to the maximum openness that will help for adjusting things. Often times there adjusted to tighten if you start moving things you have issues.