Seiko 7625A keeping terrible time

[BrehmerR]

Hello everyone!

I have just serviced this nice Seiko 7625A movement but something is quite fishy about the movement. The rate is flying all over the place. When I had not regulated the watch it was losing 300s. I got it to -60s and when I gave the regulator a 1/10th millimeter nudge it suddenly goes up to +50. When I give it a nudge back it instantly jumps back to -60. Sometimes it even goes back to -60 without me even touching it. It's shifting all over the place. You can see the sudden changes on the timegrapher. Don't mind the low amplitude... It has a pretty tired mainspring that I should change. 

Cap jewels look fine, I have demagnetised the watch several times, the balance pivots are in great condition, the hairspring looks perfect and it has been thoroughly cleaned of course. I can't seem to find the issue... I was thinking that perhaps a pivot on a train wheel is slightly bent but that wouldn't explain the sudden shit in timekeeping when I just nudge the regulator. 

Does anyone have any tips??
Huge thanks in beforehand. 
 

[watchweasol]

Hi  with the power off the movement does the balance spring sit evenly in the center of the regulator pins, I may be as you are moving the regulator,  its bearing against the springs outer or inner edge.  Worth a check before delving further 

[BrehmerR]

31 minutes ago, watchweasol said:

Hi  with the power off the movement does the balance spring sit evenly in the center of the regulator pins, I may be as you are moving the regulator,  its bearing against the springs outer or inner edge.  Worth a check before delving further 

The hairspring seems to be touching the inner pin of the two regulator pins. Should I try to bend the hairspring so that it sits nicely between the pins perhaps?

[watchweasol]

In a perfect world when the regulator is moved end to end the spring should stay centered for the complete arc not touching either pin.  If yours is bearing on the inner pin it will require centering, a tickleish job best done under high magnification as its too easy to cock it up.

[JohnR725]

hard to tell in your picture I expanded it and maybe it's okay. So in addition to worrying about where the hairspring is between the regulator pins you don't have a lot of terminal curve and it looks like where I put the marketing that might deal to expand and hit the regulator maybe it's fine the watch I just can't tell in the picture we have.

also when you're timing the watch look at it in a couple of positions to see if it changes anything like crown down perhaps in addition to whatever position you are doing it in now.

[mikepilk]

20 hours ago, BrehmerR said:

t but that wouldn't explain the sudden shit in timekeeping when I just nudge the regulator. 

You've got a rival @JohnR725 

It looks like the terminal curve is wrong, with different gaps at the start and end - yellow arrows. I would suggest it needs more of a kink at the red arrow.

In your pic it looks like the hairspring is bunched a little tighter on the regulator side.

With an uneven gap, and the spring sat on a pin, moving the regulator left or right moves the hairspring different amounts, putting different side loads on the pivots.

Before trying to reshape the terminal curve, what I'd try to start with, is bend the spring outwards slightly near the bottom yellow arrow, so that it sits in the middle of the pins, then look at the timegrapher plots.

Edited October 12, 2022 by mikepilk

[BrehmerR]

3 hours ago, mikepilk said:

You've got a rival @JohnR725 

It looks like the terminal curve is wrong, with different gaps at the start and end - yellow arrows. I would suggest it needs more of a kink at the red arrow.

In your pic it looks like the hairspring is bunched a little tighter on the regulator side.

With an uneven gap, and the spring sat on a pin, moving the regulator left or right moves the hairspring different amounts, putting different side loads on the pivots.

Before trying to reshape the terminal curve, what I'd try to start with, is bend the spring outwards slightly near the bottom yellow arrow, so that it sits in the middle of the pins, then look at the timegrapher plots.

I'll have a go at that later tonight and come back with the results.
Thanks for all the help from everyone so far!

[mikepilk]

BTW don't worry about the amplitude. Seiko's seemed designed to run with low amplitudes. Even with a new mainspring you might not see much more. I'm happy with anything around 220-240° with a Seiko.

[Neverenoughwatches]

On 10/11/2022 at 4:16 PM, BrehmerR said:

Hello everyone!

I have just serviced this nice Seiko 7625A movement but something is quite fishy about the movement. The rate is flying all over the place. When I had not regulated the watch it was losing 300s. I got it to -60s and when I gave the regulator a 1/10th millimeter nudge it suddenly goes up to +50. When I give it a nudge back it instantly jumps back to -60. Sometimes it even goes back to -60 without me even touching it. It's shifting all over the place. You can see the sudden changes on the timegrapher. Don't mind the low amplitude... It has a pretty tired mainspring that I should change. 

Cap jewels look fine, I have demagnetised the watch several times, the balance pivots are in great condition, the hairspring looks perfect and it has been thoroughly cleaned of course. I can't seem to find the issue... I was thinking that perhaps a pivot on a train wheel is slightly bent but that wouldn't explain the sudden shit in timekeeping when I just nudge the regulator. 

Does anyone have any tips??
Huge thanks in beforehand. 
 

Hiya matey. What is the spacing of the hs like between the regulator pins through the full range of the terminal curve ? The jump in the reduction of time keeping could be due to a too large a spacing between the pins at some point on the hs to allow the timing to be taken from the studding point and not from the regulation point. The spacing beween the pins needs to be close enough to allow regulation to be taken at any point on the hs and not from the stud. But not so close that the pins prevent regulation by pulling or pushing on the hs. Hope this make sense.

6 hours ago, mikepilk said:

On 10/11/2022 at 4:16 PM, BrehmerR said:

 

You've got a rival @JohnR725 

 

 

6 hours ago, mikepilk said:

On 10/11/2022 at 4:16 PM, BrehmerR said:

but that wouldn't explain the sudden shit in timekeeping when I just nudge the regulator. 

You've got a rival @JohnR725 

Haha. Don't listen John, taking the wee. Looks like I have as  well

6 hours ago, mikepilk said:

You've got a rival @JohnR725 

It looks like the terminal curve is wrong, with different gaps at the start and end - yellow arrows. I would suggest it needs more of a kink at the red arrow.

In your pic it looks like the hairspring is bunched a little tighter on the regulator side.

With an uneven gap, and the spring sat on a pin, moving the regulator left or right moves the hairspring different amounts, putting different side loads on the pivots.

Before trying to reshape the terminal curve, what I'd try to start with, is bend the spring outwards slightly near the bottom yellow arrow, so that it sits in the middle of the pins, then look at the timegrapher plots.

You beat me to that one Mike.  just making some markings for illustration. 

Edited October 12, 2022 by Neverenoughwatches

[BrehmerR]

On 10/12/2022 at 6:04 PM, Neverenoughwatches said:

Hiya matey. What is the spacing of the hs like between the regulator pins through the full range of the terminal curve ? The jump in the reduction of time keeping could be due to a too large a spacing between the pins at some point on the hs to allow the timing to be taken from the studding point and not from the regulation point. The spacing beween the pins needs to be close enough to allow regulation to be taken at any point on the hs and not from the stud. But not so close that the pins prevent regulation by pulling or pushing on the hs. Hope this make sense.

 

Haha. Don't listen John, taking the wee. Looks like I have as  well

You beat me to that one Mike.  just making some markings for illustration. 

Hello again everyone!
I seem to have gotten the balance working perfectly but the watch is still performing the same. The balance looks perfect, I have checked all the train wheels and they look perfectly fine. I had a similar problem with an Enicar a while back and it turned out to be a faulty pallet fork which was causing the issue. I suppose that that has to be the issue here. I simply cannot imagine anything else. What I am noticing is that the escape wheel teeth are contacting the pallet jewels very far in. On the entry jewel it even looks as though the pallet jewel is touching the escape wheel beyond the tooth because it sits so far forward. The pictures show what I mean. The jewels are however barely able to be moved at all. The jewels can perhaps be moved backwards a maximum of 1/5th of a millimeter. 
Does anyone have any ideas if this could be the issue or?

Thanks again!

 

[mikepilk]

According to Fried, the lock should be about 1/3 the thickness of the jewel. Yours does look much more than that. It's strange that both jewels should look wrong. Could the pallet have been changed to a wrong one?

[Neverenoughwatches]

1 hour ago, BrehmerR said:

Hello again everyone!
I seem to have gotten the balance working perfectly but the watch is still performing the same. The balance looks perfect, I have checked all the train wheels and they look perfectly fine. I had a similar problem with an Enicar a while back and it turned out to be a faulty pallet fork which was causing the issue. I suppose that that has to be the issue here. I simply cannot imagine anything else. What I am noticing is that the escape wheel teeth are contacting the pallet jewels very far in. On the entry jewel it even looks as though the pallet jewel is touching the escape wheel beyond the tooth because it sits so far forward. The pictures show what I mean. The jewels are however barely able to be moved at all. The jewels can perhaps be moved backwards a maximum of 1/5th of a millimeter. 
Does anyone have any ideas if this could be the issue or?

Thanks again!

 

That does look like the jewel is to deep on the escape wheel, does the side shake seem excessive on either the escape wheel or pallet. Like mike is thinking either could have been incorrectly changed. How do you feel about resetting the pallet jewels ?

[JohnR725]

2 hours ago, mikepilk said:

According to Fried, the lock should be about 1/3 the thickness of the jewel.

no find some variations depending upon who made the watch. It also is determined the locked down by the safety features so you need to do an escapement check.

But it definitely looks like this Seiko is way too deep and that is definitely an issue.

 

 

[BrehmerR]

On 10/23/2022 at 8:28 PM, Neverenoughwatches said:

That does look like the jewel is to deep on the escape wheel, does the side shake seem excessive on either the escape wheel or pallet. Like mike is thinking either could have been incorrectly changed. How do you feel about resetting the pallet jewels ?

I do not have the tools to reset the pallet jewels and I have never done it either. I have however found a pallet fork for this exact movement so maybe I should order that. 

There certainly is no excessive end shake but I do notice now that as soon as I start to move the pallet fork with my tweezers the escape wheel jumps slightly in an unnatural fashion. Just by moving the pallet fork half a millimeter if even that. I think its the pallet jewels...

[mikepilk]

30 minutes ago, BrehmerR said:

I do not have the tools to reset the pallet jewels and I have never done it either. I have however found a pallet fork for this exact movement so maybe I should order that. 

There certainly is no excessive end shake but I do notice now that as soon as I start to move the pallet fork with my tweezers the escape wheel jumps slightly in an unnatural fashion. Just by moving the pallet fork half a millimeter if even that. I think its the pallet jewels...

It's not too difficult to move the jewels in, you don't need any special tools. But, as you say, it doesn't look like they could be moved in enough.

As it's easy enough to find parts/donor movements, I'd change the pallet fork.  Let us know how the new one compares.

[JohnR725]

4 hours ago, BrehmerR said:

I do not have the tools to reset the pallet jewels and I have never done it either. I have however found a pallet fork for this exact movement so maybe I should order that.

with the pallet fork being an extremely key components there's nothing wrong with just replacing it.

it looks like quite a few pallet forks will interchange

http://cgi.julesborel.com/cgi-bin/matcgi2?ref=T_ZF^JS

 

 

 

[BrehmerR]

6 hours ago, mikepilk said:

It's not too difficult to move the jewels in, you don't need any special tools. But, as you say, it doesn't look like they could be moved in enough.

As it's easy enough to find parts/donor movements, I'd change the pallet fork.  Let us know how the new one compares.

I managed to find one locally so it should arrive by tuesday. I will gete back to you then(:

[Kalanag]

I wonder if a bent or wrongly positioned banking pin could be the culprit here? You can see that not only the lock is too deep but also the other pallet is too far away from the escape wheel. This can not be solved by moving the pallet. Maybe the whole pallet fork doesn‘t fit the movement?

[clockboy]

The most common cause of irregular running is the balance. Normally the h/ spring is not centred or the terminal curve is not equal throughout its length or it is not centred between the regulator  pins. Looking at the pics the terminal curve is not equal. 

[mikepilk]

1 hour ago, Kalanag said:

I wonder if a bent or wrongly positioned banking pin could be the culprit here? You can see that not only the lock is too deep but also the other pallet is too far away from the escape wheel. This can not be solved by moving the pallet. Maybe the whole pallet fork doesn‘t fit the movement?

The banking pins don't look obviously bent in the pics. I suspect, like you, that the wrong pallet may have been used.