2824-2 movement running very fast?

[Mnk1]

Hi guys, 

Glad to be a new member and thanks for the welcome. 

I’m sort of new to working on movements and Ive just cleaned my 2824-2 movement which the balance was not running. After finding dirt and hair inside near the pallet fork the balance is now running. Once the movement was assembled I used the Timegrapher app to get an estimate of the rate and beat error, it measured 0.2ms and around +2 per day... my issue is that the movement gains about 35-40mins over 24hours. 
would appreciate some advice as where I could have gone wrong or causes of this kind of increase? 

i assumed the hairspring maybe magnetised? It is new so I wondered how but anyway , would love to hear some thoughts and thanks in advance for your help  

[Nucejoe]

Is the hairspring new?  

Edited July 14, 2020 by Nucejoe
OR THE WATCH?

[Marc]

4 hours ago, Mnk1 said:

It is new so I wondered how but anyway

If your mobile phone is in a flip case with a magnetic catch then all you have to do is leave your watch sitting on top of you phone for 5 minutes and bingo....  it can be as simple as that unless the watch has magnetic shielding.

Either way magnetism or sticky residue on the H/S are you most likely causes.

[watchweasol]

Hi  I would agree with Marc as to the likely cause of the problem. Have a close look at the balance spring in action .

[JohnR725]

17 hours ago, Mnk1 said:

Timegrapher app to get an estimate of the rate and beat error,

what exactly are you timing the watch on? We've had issues in the past with applications and microphones in non-timing machines that don't always pick up the correct signal and give incorrect results.. Then especially if you're trying to diagnose a problem  you need to develop a proper timing procedure. Like timing the watch in multiple positions and did you give us the amplitude?  also as a reminder timing machines are giving you instantaneous results of what the watches doing now not necessarily what is going to do on your wrist..

17 hours ago, Mnk1 said:

I’m sort of new to working on movements

what exactly does sort of new mean?  If you totally new servicing a watch is a lot of things that can make it run fast besides magnetism.  Need to visually look at your hairspring make sure it's flat  make sure it's not  bumping into anything.. Make sure it's not touching anything it's not supposed to..  Depending upon how you cleaned the hairspring might be sticky.

 

[Mnk1]

Thanks for your help Guys, 

The movement was bought new it looks to be Chinese made although not a really bad one. St2130. It ran well when purchased but suddenly stopped. After I stripped it down I found it to be pretty dirty so gave it a clean, oil and now it runs. But I didn’t clean the hairspring it looked fine there was just debris under the pallet fork ( a big black hair lol) 

 I checked the hairspring and all looks fine so last night I replaced the balance Cock and balance with a new spare I had and it’s still running fast. 
just a note I removed the date change mech and spring and any associated parts on top of the movement as date function was not needed.  I also changed the running seconds wheel part 227 as the original was rubbish ( badly cut teeth on one side) Could this part be different from ST part? 
ps. I used Timegrapher app which is apparently one of the better ones I adjust to 0.1ms beat error and it was around 0.9-2 seconds per day. 
I will try to demag the other balance as well see if that helps. 
 

 

[Mnk1]

22 hours ago, Marc said:

If your mobile phone is in a flip case with a magnetic catch then all you have to do is leave your watch sitting on top of you phone for 5 minutes and bingo....  it can be as simple as that unless the watch has magnetic shielding.

Either way magnetism or sticky residue on the H/S are you most likely causes.

Unfortunately no flip case haha, I changed the balance out for a new one to see if that was the issue but still runs very fast. 

[JohnR725]

1 hour ago, Mnk1 said:

Unfortunately no flip case haha, I changed the balance out for a new one to see if that was the issue but still runs very fast

would you know if the St2130 is an exact clone of the 2824-2?  unfortunately in the review below the gears themselves are not physically check for being identical.

timing machines or timing apps tell you what frequency or seconds per day error the balance wheel is oscillating at. your timing apps that you claim is wonderful says your balance wheel is oscillating basically perfectly although we don't see the graphical display which I always like to see . Sometimes I find people rely too much on numbers without looking at the graphical display which might indicate a problem.

Did you physically look at the replacement second wheel and compare it to the wheel you are taking out? In other words the number of teeth on the gear and pinion are they identical? 

watches are interesting mechanical devices they have a balance wheel that oscillates at a frequency. A gear train that has gear ratios in other words the number of teeth on the gear versus pinion and that gear ratio has to match the oscillation of the balance wheel otherwise you get strange timing issues. So is a wild guess I would go and look at your second wheel and make sure the number of teeth on both the gear and the pinion are exactly the same otherwise you're going to have the problem you have now a gear train running at a different ratio expecting a different rate of balance wheel oscillation.

https://watchguy.co.uk/comparison-sea-gull-st2130-eta-2824-2-peacock-sl3000/

[Mnk1]

@JohnR725 thanks for the advice mate I will definitely check if there’s a difference between the wheels sounds like that could be the problem, I have so many parts I loose track sometimes or misplace in the wrong part box. I probably should get a proper Timegrapher. The app does show the graph with and the two graph lines run straight and consistently but I’m still learning all this so thanks for all your help I really appreciate it.  I’ll post results here anyway. 

[Marc]

10 hours ago, Mnk1 said:

I used Timegrapher app which is apparently one of the better ones I adjust to 0.1ms beat error and it was around 0.9-2 seconds per day.

So your Timegrapher is indicating correct balance frequency, but the movement is gaining 40 minutes in 24 hours.

If the TG is correct then magnetism is unlikely to be the issue as magnetism screws up time keeping by messing with the oscillator frequency, in which case as @JohnR725 suggests the gear train ratios must be incorrect. As you have exchanged some of the wheels this is entirely possible.

Alternatively if the gear train ratios are correct then the Timegrapher must be giving false data, since if the oscillator frequency and the ratios are correct, the time keeping has to be correct.

[Mnk1]

22 hours ago, Marc said:

So your Timegrapher is indicating correct balance frequency, but the movement is gaining 40 minutes in 24 hours.

If the TG is correct then magnetism is unlikely to be the issue as magnetism screws up time keeping by messing with the oscillator frequency, in which case as @JohnR725 suggests the gear train ratios must be incorrect. As you have exchanged some of the wheels this is entirely possible.

Alternatively if the gear train ratios are correct then the Timegrapher must be giving false data, since if the oscillator frequency and the ratios are correct, the time keeping has to be correct.

Thanks mark, my thoughts exactly. I’ll check over the parts I added to see if there’s differences, it seems to be the best explanation as to what’s going on.