Demagnetizer question

[spectre6000]

I picked up one of those super cheap blue Chinese demagnetizers, and it was delivered today. I have a pocket watch that had its hair spring magnetized a long time ago, and I never really bothered messing with it until just now. It runs reasonably well, but according to my accuracy app is more than 1200 seconds a day fast. I can see that the hair spring is clearly bunched up on the one side, so all is checking out. 

I put the pocket watch on the demagnetizer, didn't follow the instructions because there are none, but got the gist from a collection of different sources and an idea of what's going on physically. With the watch sitting on the demagnetizer with it on, it hums and even rattles a little. I tried with another ferrous object, and it vibrated and hummed as well, so I guess that's fine and there are just moving parts.

The weird part is that the hair spring is still all bunched up. I guess it could be oil sticking the coils together, but given how it came to be magnetized, that would be quite a coincidence... Not only that, but now it wants to stop. I shake it, and it mostly stays stopped (it may oscillate a second or two), but shaken enough or just the right way and it'll go again. Enough time hasn't elapsed, but I imagine it's still over 1K seconds/day fast.

Thoughts?

[HectorLooi]

The cheap China demagnetizers are ok. But you have to use them correctly. They do not come with a auto power down function like some expensive models.

What you have to do is press the red button, feel the object that you are trying to demagnetize vibrating, then slowly lift it away from the demagnetizer perpendicularly to about 6" - 8" away. Then release the red button. If this is done incorrectly, you could end up magnetizing the object instead.

You can get a small magnetic compass, even those small compasses attached to watch straps, to check for residual magnetism. There are some phone apps that can measure magnetism also.

Practice on your screwdrivers. Magnetize one of them first. Then hold it near the compass, the needle will move. Then demagnetize it. If done correctly, the needle shouldn't move anymore.

Best of luck.

[oldhippy]

Don't just rely on the demagnetizer, remove the balance and clean it and oil the pivots. HectorLooi is correct in how to use that type of tool. 

[spectre6000]

That's precisely what I did, several times. Is it possible the hairspring has a memory for the magnetized state and will require manipulation? Obviously there are other possible issues, but given that it kept great time prior to being placed in front of some computer speakers over night, then suddenly gained obscene amounts of time each day, I figured magnetized hairspring was obvious.

I'll be taking the watch completely down for a service soon, I just wanted to test what effect that had prior since it was obviously an issue so I know I did something positive.

The thing that's confusing me is why the watch would stop entirely and be difficult to restart as a result of using the demagnetizer correctly per the above method. The rattle on top of the hum seems like a clue, but it runs great (just super fast) otherwise.

Edited May 3, 2020 by spectre6000

[oldhippy]

While the watch is being demagnetized it vibrates a little, if dirt is in the movement it can get moved around and might move into the train and cause it not to function.

[spectre6000]

9 hours ago, oldhippy said:

While the watch is being demagnetized it vibrates a little, if dirt is in the movement it can get moved around and might move into the train and cause it not to function.

That strikes me as very plausible. I like it.

[HectorLooi]

JD Richards put up a video on a magnetized pocket watch hairspring. But it turned out to be a dirty hairspring. Watch that video and try cleaning the balance in lighter fluid. It happen to a Seiko watch that I serviced recently. It took a few cleaning cycles in lighter fluid to stop the spring from sticking.

[jdrichard]

I picked up one of those super cheap blue Chinese demagnetizers, and it was delivered today. I have a pocket watch that had its hair spring magnetized a long time ago, and I never really bothered messing with it until just now. It runs reasonably well, but according to my accuracy app is more than 1200 seconds a day fast. I can see that the hair spring is clearly bunched up on the one side, so all is checking out. 
I put the pocket watch on the demagnetizer, didn't follow the instructions because there are none, but got the gist from a collection of different sources and an idea of what's going on physically. With the watch sitting on the demagnetizer with it on, it hums and even rattles a little. I tried with another ferrous object, and it vibrated and hummed as well, so I guess that's fine and there are just moving parts.
The weird part is that the hair spring is still all bunched up. I guess it could be oil sticking the coils together, but given how it came to be magnetized, that would be quite a coincidence... Not only that, but now it wants to stop. I shake it, and it mostly stays stopped (it may oscillate a second or two), but shaken enough or just the right way and it'll go again. Enough time hasn't elapsed, but I imagine it's still over 1K seconds/day fast.
Thoughts?

Sent from my iPhone using Tapatalk Pro

[jdrichard]

JD Richards put up a video on a magnetized pocket watch hairspring. But it turned out to be a dirty hairspring. Watch that video and try cleaning the balance in lighter fluid. It happen to a Seiko watch that I serviced recently. It took a few cleaning cycles in lighter fluid to stop the spring from sticking.

Just posted it.


Sent from my iPhone using Tapatalk Pro

[Vacheron7]

My cheap blue demagnetizer also arrived last week. This week I was looking for a Seagull ST2130 on eBay and came across a guy who was selling a brand new one alongside a generic divers watch containing a genuine ETA 2824 that was running 30 minutes per day fast.

The ETA was apparently "less than 2 years old, but suddenly started running super fast" He had bought the seagull to replace the broken ETA, but in the end decided not to proceed and sold them both together instead.

When they arrived I tested the Seagull which was still in it's transport packaging (+1.3s / day!) and the genuine ETA in the watch (+1600 seconds / day). The fine adjustment had been set all the way to the minus limit.

WIthout even uncasing the watch I ran it over my £11 demagnetizer in 4x90 degree rotations, drawing it slowly away each time until I could sense no vibrations (it is only the 50Hz alternating magnetic field that causes the "feel", there are no moving parts inside) and in less than 1 minute it was running at a consistent -112s/day without any regulation so far.

So it appears that my £11 investment could have saved me over £100 in its first 10 seconds.

Regarding the sticking, I think as others have said, the demagnetizer probably has worked, but the hairspring coils, initially pulled together by the magnetism, are now being retained together by any residue on the hairspring.