Running fast. What are the possible causes?

[herg]

I'm working on an AS1194, and after reassembly, amplitude and beat error both look good. It's about +120 s/d though.  I can only get it to about 60 s/d by maxing the adjustment.  I've looked closely at the hairspring, and I don't see anywhere that it's touching itself or any other parts of the movement. 

I do notice that the tail end, where it's attached to the balance cock, it extends past.  You can't see it in this picture, but instead of a nice taper pin, it also looks like a wire is bent in place to hold it.

Is it possible someone shortened the spring by reattaching it?  If so, I have to assume someone before me fixed the beat error. 

 

[JohnR725]

19 minutes ago, herg said:

Is it possible someone shortened the spring by reattaching it?  If so, I have to assume someone before me fixed the beat error. 

One of the problems with anything resembling older watches is people not always doing things correctly. So yes there is a possibility that somebody shortened the hairspring which is why you're going fast and perhaps they even did that to fix the beat error which would not be the correct way to do it.

Sometimes on some vintage watches there can be a little bit of hairspring stuck through the stud but that's usually more vintage. In any case if the stud pinned hairspring doesn't look right it probably isn't right. So underpinning giving a length will slow the watch down and then you'll have to rotate the collet to adjust the beat.

[Knebo]

yes exactly, as John said. 

If you are capable of such a fiddly job, you are lucky that there's so much hairspring left for you to adjust the effective length of the mainspring at the stud. 

The +120 to +60 s/day deviation that you are able to adjust via the regulator pins should give you an idea how many millimeters of hairspring you need. I.e adding the distance travelled by the regulator pins to go from +120 to +60 will slow down the watch by -60.

Alternatively, you can remove some weight from the balance wheel... but go slow, it doesn't take much to remove too much. 

[herg]

Thanks to both of you.  I had not yet grasped that removing weight was another regulation technique.  Either trying to re-pin the hairspring or removing weight from the balance, I'd estimate about a 50% chance of messing it up.  I'm experienced under a microscope, but new to watches, and that hairspring is much more fragile than any surface mount soldering I've done.

I had nearly decided to just live with the deviation when I noticed it stopped running at all after about 4 hours.  There was still wind in the mainspring, and giving the movement a small shake got it running again for a short amount of time.  I guess I'm digging back in to figure out what's causing that.

[nickelsilver]

The way the hairspring is sitting I wouldn't be surprised if the first coil is touching the stud when running. Easy 60+ seconds.

[praezis]

20 hours ago, Knebo said:

Alternatively, you can remove some weight from the balance wheel... but go slow, it doesn't take much to remove too much. 

His watch is already too fast 

[Knebo]

12 minutes ago, praezis said:

His watch is already too fast 

Oh dear, please excuse my brain-fart. Thanks @praezis for the peer-review

Of course, you'd need to ADD weight to the balance wheel to slow it down. 

 

Ironically, I was only doing the exact thing two months ago. Watch too fast, added timing washers. I can post pics on demand. 

[JohnR725]

23 hours ago, herg said:

amplitude and beat error both look good. It's about +120 s/d though.  I can only get it to about 60 s/d by maxing the adjustment.  I've looked closely at the hairspring, and I don't see anywhere that it's touching itself or any other parts of the movement.

One of the things that I always like to see is timing machine results. You tell us the beat looks good etc. but I like to see pictures often times people don't notice the fine details of things and maybe they're not as good as you think maybe they are but it's always nice to show us the timing results.

Then yes you can change the weight of the balance wheel to match the hairspring as long as you're reasonably close. Then if you are changing the balance wheel it does have to have screws. Often times there will be watches that look like there are screws but they don't actually unscrew it all and that's not entirely helpful. Then the definition a reasonably close is if you look at the timing washers for changing timing was adding slowing down one pair one  washer on one side and one on the other to maintain the poise the balance wheel three minutes is the maximum. You don't want to go timing washer crazy and fill the balance wheel with timing washers. Just as he don't want to remove huge quantities of weight from the balance wheel as it will change the characteristics and have other consequences.

EBay's a nice place to swipe the pictures from here's a picture from eBay. Timing washers for slowing the watch down. Typically that will come up to three minutes maximum and thereby the size of the balance wheel. They're not always an exact science in other words you may not get exactly Whatever it says you're supposed to get.

 

3 hours ago, nickelsilver said:

The way the hairspring is sitting I wouldn't be surprised if the first coil is touching the stud when running. Easy 60+ seconds.

On 6/5/2024 at 12:31 PM, herg said:

I've looked closely at the hairspring, and I don't see anywhere that it's touching itself or any other parts of the movement. 

Looking at and actually seeing what the hairspring is doing is very hard to do.  It actually requires lots of practice of staring at hairsprings until you finally grasp exactly what they're doing because they're just really hard to see no matter what. There are out around slightly and they just touched something you'll be fast they touch the armless touch the balance bridge touch anything at all for just a second as all it takes to be fast. So typically with new bees and hairsprings were going to be suspicious that you probably missed something just because it's easy to do.

Another thing I didn't see mentioned anywhere is did you demagnetize the watch? It's amazing what a little magnetism will do to timekeeping even if it doesn't look visibly like there's an issue.

 

[herg]

2 hours ago, JohnR725 said:

One of the things that I always like to see is timing machine results. You tell us the beat looks good etc. but I like to see pictures often times people don't notice the fine details of things and maybe they're not as good as you think maybe they are but it's always nice to show us the timing results.

Here is a screenshot of the timegrapher from today.  Overall, it looks better, but there are details.  One that I notice when being more picky is that the beat isn't all that consistent.  It's between 0.2ms and 1.5ms during the time I watched it.  Maybe the spring is momentarily touching something like you say.  That might also help explain why it's ~ +41 now, instead of +60.  It could be touching less often that it was previously.

2 hours ago, JohnR725 said:

Another thing I didn't see mentioned anywhere is did you demagnetize the watch? It's amazing what a little magnetism will do to timekeeping even if it doesn't look visibly like there's an issue.

That's certainly a thought.  I read the part about wanting to see timegrapher and went down to the desk before reading to the end of your post.  I'll have to go back down to try a demagentize.  I haven't to this point.