Watch running too fast

[Ril]

I've been restoring my great grandmother's watch that she had and which was no longer operational.

It's an EB 8800 movement (Seawatch). It had a broken pallet fork (the ones with two wires instead of jewels) which I replaced from a donor movement of the same model.

I did a full clean in my ultrasonic cleaner with elma red, oiled all the jewels, and even used some polywatch to remove the scratches. But after I put the watch together, I noticed that it's running really fast.

One full revolution of the seconds hand took exactly 40 seconds. And it's consistent.

I've read that his could be due to magnetization? But I'm not sure. I don't have a way to demagnetize it.

I'm a bit at a loss how to proceed from here on. Any suggestions? I don't have a timegrapher that I can use.

[rjenkinsgb]

My first though, as a semi-beginner, would be to stop the balance and examine the hairspring under a magnifier - are any turns stuck together due to leftover cleaning fluid or oil?

Or any part of it kinked rather than being in an even spiral??

Magnetism is a possibility, but you would need a demagnetiser in that case..

 

[nevenbekriev]

The first thing that one needs to know before giving some advice is is the amplitude of balance oscillations. As You changed the lever, now the drop lock can be missing and this will make the balance swing really fast and with very lo amplitude. Then, if somethinfg is wrong with the hairspring, the balance will oscillate faster,  with a  bigger amplitude. If the amplitude is small, then no point to check for timekeeping, but the reason must be found and rectified. If the watch works with no stopping, this still doesn't mean that it works well.

What we need to know is if the lever snaps as it should without the balance and if yes, we will need a good picture where the hairspring is seen. Or, please do the free oscillations test, see what the result is and try to caunt the oscillations per minute, they should be exactly 150 if the hairspring is OK.

 

 

[ManSkirtBrew]

22 hours ago, Ril said:

I don't have a timegrapher that I can use.

You can try "Watch Accuracy Meter" on your smartphone. I've compared it to my timegrapher results and it seems very accurate. It doesn't show amplitude, but you can also check that visually using the slow-motion video on your camera and recording the balance as the watch runs.

[Poljot]

Magnetism or oily / dirty hairspring will be easy to detect by simply visually inspecting your watch balance assembly.

[Jon]

On 4/11/2024 at 8:58 PM, Ril said:

I've read that his could be due to magnetization? But I'm not sure. I don't have a way to demagnetize it.

Not to state the obvious, but buy a demagnetiser! If the hairspring is sticking together then this will increase the timing rate. If you demagnetise the movement, at least it takes the problem of being magnetised off the table and being the possible problem and you have a way of demagnetising any other watch you service. This may not be the problem, but watchmaking is all about eliminating everything that couldn't be the problem and ending up with the possible, then the probable.

Edited April 13 by Jon

[nevenbekriev]

I need to justify that actually the hairspring can not be magnetizated. What can be magnetizated and cause problem here is the studd carier and the rate regulator. If they are magnetized, they can attract the hairspring in a way that  it will stick to the cock

Edited April 12 by nevenbekriev

[RichardHarris123]

58 minutes ago, nevenbekriev said:

need to justify that actually the hairspring can not be magnetizated.

Really, modern  hairsprings are made from all sorts of exotic materials to prevent magnetism.  Old, steel ones? 

[nevenbekriev]

Old steel ones are with 'blued' collor.Yes, they can be magnetized

[Ril]

Thank you all for the help and feedback.

4 hours ago, rjenkinsgb said:

My first though, as a semi-beginner, would be to stop the balance and examine the hairspring under a magnifier - are any turns stuck together due to leftover cleaning fluid or oil?

Or any part of it kinked rather than being in an even spiral??

Magnetism is a possibility, but you would need a demagnetiser in that case..

 

I took out the balance and inspected it. All the lines are uniformly spaced out. But when I put it back, I see that two if the inner lines are touching each other. I was unable to get a picture of that unfortunately.

4 hours ago, nevenbekriev said:

The first thing that one needs to know before giving some advice is is the amplitude of balance oscillations. As You changed the lever, now the drop lock can be missing and this will make the balance swing really fast and with very lo amplitude. Then, if somethinfg is wrong with the hairspring, the balance will oscillate faster,  with a  bigger amplitude. If the amplitude is small, then no point to check for timekeeping, but the reason must be found and rectified. If the watch works with no stopping, this still doesn't mean that it works well.

What we need to know is if the lever snaps as it should without the balance and if yes, we will need a good picture where the hairspring is seen. Or, please do the free oscillations test, see what the result is and try to caunt the oscillations per minute, they should be exactly 150 if the hairspring is OK.

 

 

I checked the leaver, it is snapping to the other side when pushed. But I noticed that it only snaps after I push a bit. There's a slight degree where if I push it a tiny bit, it stays there (doesn't go back to its previous position), and if I nudge it a bit further it snaps to the other side.

I'm not familiar with the term drop lock, what does it refer to?

The inner part of the hairspring seems to be touching each other. I also noticed that the balance stops sometimes. I'll try to get a picture of the balance installed.

4 hours ago, ManSkirtBrew said:

You can try "Watch Accuracy Meter" on your smartphone. I've compared it to my timegrapher results and it seems very accurate. It doesn't show amplitude, but you can also check that visually using the slow-motion video on your camera and recording the balance as the watch runs.

The scan points look like they're scattered around at random. There is no single branch like points that I could see, even with scrolling through the samples.

2 hours ago, Jon said:

Not to state the obvious, but buy a demagnetiser! If the hairspring is sticking together then this will increase the timing rate. If you demagnetise the movement, at least it takes the problem of being magnetised off the table and being the possible problem and you have a way of demagnetising any other watch you service. This may not be the problem, but watchmaking is all about eliminating everything that could be the problem and ending up with the possible, then the probable.

I suppose this is the next tool to get.

[nevenbekriev]

What You decribe as lever behavior is more or less common for pin lever movements. Actually it is normal, but may benefit from proper lubricating the pins/escape wheel teeth.

So You problem is the hairspring. Probably it is only 'sticky' and washing it in accetone or pure petrol will bring it to normal state. Drying with hairdryer helps much after the washing. Of course You can try to demagnetise the balance and the cock. The problem will be solved when the sping doesn't stick on ocasion, e.g. when one grasps several coils together with the tweezers, they should get separated by themself imediately after released. Some chance exists that the spring is bent where the coils touch, but try firs what i have writen above.

[Mercurial]

The fact that you have witnessed coils of the hairspring clinging or sticking together explains why the watch is running so fast. 

If you reduce the length of a hairspring, the watch will run faster. When hairspring coils stick or cling together, this effectively reduces the active length of the hairspring, so the watch runs fast. 

As nevenbekriev has said, the most likely culprit here is oil or other contamination on the hairspring causing coils to stick together. Cleaning the hairspring as he describes will solve the problem.

Occasionally the contamination causing the hairspring coils to stick can be very stubborn to remove, in which case I suggest ultrasonic cleaning in acetone, naphtha or a commercial watch cleaning solution.

Magnetism can also cause coils of the hairspring to stick together, if the hairspring is made of an alloy that can be magnetised.

Finally, your watch has what is called a pin pallet escapement. They are considered to be less refined, but quite rugged compared to a traditional Swiss lever escapement. They are quite forgiving.

They do require lubrication, if you haven’t done so already, find instructions for lubrication of these escapements to get optimal performance from this watch.

Hope this helps,

Mark

[Ril]

On 4/13/2024 at 8:18 PM, nevenbekriev said:

What You decribe as lever behavior is more or less common for pin lever movements. Actually it is normal, but may benefit from proper lubricating the pins/escape wheel teeth.

So You problem is the hairspring. Probably it is only 'sticky' and washing it in accetone or pure petrol will bring it to normal state. Drying with hairdryer helps much after the washing. Of course You can try to demagnetise the balance and the cock. The problem will be solved when the sping doesn't stick on ocasion, e.g. when one grasps several coils together with the tweezers, they should get separated by themself imediately after released. Some chance exists that the spring is bent where the coils touch, but try firs what i have writen above.

I did some several wash cycles and it seemed to have done the trick. No more sticking and the watch is running fine for now, full 60 seconds!

It hasn't stopped working. I have also re-oiled the capstone jewel on the balance, and demagnetized it.

I got the following results using the watch accuracy meter app. I tried adjusting it a bit since the rate was 210s/d and got it down quite a bit. I'll monitor the watch over a few days and see how it's holding up.

I didn't find much information on the accuracy of the EB8800 movement. So I can't tell if this is a good result or it can be better. But I think it's good for the time being.

Thank you all for the support. I'm very grateful.