Amateur Hour Watch Restoration

Where it’s from

I bought this movement on eBay for $27.

Interestingly, doing a Google Lens search on this movement reveals that it appears with many many different brand names etched on the train bridge. This made identifying the movement quite difficult, but after poking around 17jewels and emmywatch, I think I’ve determined that this is an AS 970. Emmywatch places this movement between 1935-1967; 17jewels says that it used an Incabloc shock protection system, but this movement clearly doesn’t have any shock protection, so I’m guessing this is in the earlier range of the production.

What it looked like

The case is pretty scratched, and the crystal is a bit cracked. The case says “M10K R.G.P.”, which means the case has a 10 karat rolled gold plating. The “M” is likely the maker’s mark. The dial looks in really good shape, but it seems that someone installed the minute hand upside down, since the end curls up, instead of down; the minute hand is also noticeably duller than the hour hand.

How I restored it

Sep 14, 2025

Opening up the case, it seems the movement isn’t actually attached to the top of the case, but it’s attached to the case back. The inside of the case back is stamped with some information:

Maybe the “53” at the end indicates the year 1953? It would match manufacturing dates of the AS 970, at least.

Unfortunately, it looks like the balance wheel is very loose. I’m guessing the balance wheel pin is broken. And surprise surprise, after I took it out, I see that the pin is broken on the watchmaker’s side. I’ll need to find a replacement or donor movement. On the bright side, I might be able to find a replacement with a shock setting.

I disassembled the entire movement, and everything else looks pretty OK to me, BUT! I didn’t realize the ratchet wheel screw was actually reverse threaded, and completely sheared off the head! So, I’ve ordered these replacement parts:

1. Balance complete
2. Ratchet wheel screw
3. Barrel arbor (there’s no way I’m getting the screw out of this one, lol)

Note: “AS 970” is engraved on the main plate, which confirms this is indeed an A Schild 970 movement.

Sep 21, 2025

I’ve gotten a few of the replacement parts I ordered online. Unfortunately, one of the parts I ordered was for the AS976 instead of AS970, oops. I think I got confused, because the part has the same number in both movements (5414). So, I ended up just getting a donor movement, it’ll for sure have the ratchet screw.

Oct 2, 2025

Got the donor movement, so we are good to go! Just spent the last couple hours cleaning the wheels, polishing the pins, and pegging the jewels. It really didn’t have that much dirt and debris, and the jewels look really clean now. Crossing my fingers that I can get this movement running nicely!

Oct 3, 2025

I started reassembly today. Almost forgot that you need to put the center wheel on before the ratchet wheel, since it nests between the barrel and the ratchet. I was able to get the train of wheels on fine, and the train wheel bridge, but it seems that the subseconds wheel doesn’t have enough endshake, and is seizing the movement. Not sure what I can do about it. I guess maybe I can try giving a couple of taps on the corresponding jewel on the bridge?

Oct 5, 2025

The good news: I was able to use a toothpick and give the jewel a couple of taps, so that the seconds wheel spins freely now. The bad news: I may have cracked the jewel, and the wheel prolly has way too much end-shake now. The worse news: the balance also has way too much end shake. It looks like the jewel on the main plate is way too far out, such that the balance rubs against the main plate when sitting dial-side down. It at least kicks up when I put it dial side up though, and I’m even able to get a reading from it. I’m not sure what I can do here though, without a staking set. Do I get a jewel press now? Hm.

Oct 6, 2025

I bought a jewel press. It seems to have caused more problems though. First, I tried it out on some of the jewels in the donor movement. Turns out it is quite easy to accidentally crush the jewels into dust! I managed to successfully extract some of them though. I attempted to replace the watch-side seconds-hand jewel with one from the donor movement, and it literally turned into dust, oops. So, I managed to put the original jewel back, but adjusted the dial-side jewel so that the seconds-hand wheel had a little less end shake.

Then I tried to address the dial-side balance jewel. Oh boy, what a mess. First, I completely nuked the jewel that was in there; I’m not sure what I did wrong, but as soon as I exerted pressure on it, it completely cracked. So then, I tried replacing it with the jewel from the donor movement, but it was slightly smaller, so it wouldn’t fit. Then I tried to use the dial-side escape wheel jewel from the donor movement, but apparently it was slightly too big. I was able to push it in, but now it’s mushroomed up some of the main plate so that I think it’s scraping the bottom of the balance wheel. Ugh.

I ended up ordering another donor movement, since now I’m not sure if I’ll be able to salvage the main plate.

Oct 8, 2025

Alright, so I finally figured out that the dial-side balance holed jewel doesn’t dictate the endshake of the balance, it’s actually the cap jewel! I had pushed the holed jewel as far as I dared, and the balance was still scraping on the main plate. However, I looked closely at the balance staff on the dial side, and I noticed that the pin is quite long. So then I thought to look at the pin while it’s in place, without the cap jewel, and lo and behold it was sticking out a lot. So, I oiled and put the cap jewel back on, and now it’s ticking!

Not the best looking reading I’ll admit, but honestly I got away with worse on my Timex Marlin, which I wore for months without issue. And after giving it a full wind, the amplitude perked up a bit more. The trace is still a little wavy, but I think I’m mostly OK with it.

Oct 9, 2025

I decided to try out adjusting the beat error, which for this movement means taking out the balance and adjusting the hair spring collet. I managed to fiddle with it a little bit, and improve the beat error by a few milliseconds:

To me, that is a really, really good reading. I know the beat error is still kind of high, but the amplitude is much healthier, and the trace is pretty much dead straight. I’m going to resist the urge for perfectionism, and call this done. Next up is cleaning the case, and casing the movement. And then… on to the next watch, lol.

Oct 11, 2025

I cleaned the case in the ultrasonic, and reassembled. I decided to glue the crystal in. The final result: