Strange, erratic running (amplitude?) Vostok 2234

[Simeon]

I have just serviced a Vostok 2234 (basically a 2209 with a calendar and hacking). This always ran erratically when cased, periodically stopping. It looked like something was clashing when cased as I noted that touching the crown (pushing it in towards the case) would make it stop. It looked to me like there was some case screws missing and the movement ring might have been touching the stud carrier on the balance cock. When I could keep it running it gave  about +30 / +20 DD and DU. Amplitude was about 180. It definitely needed a service - out with the movement and onto the bench.

After service the watch performed a lot better. Amplitude around 300 and similar dial up / dial down rates. After sitting and watching the timegrapher for a couple of minutes the DD rate suddenly went very erratic - +200 and amplitude dropped to 200 then it would veer to -125 with an amplitude like 316, bouncing around a bit like this before settling back to 295-300 degrees and about +10 SPD. Checking this with a stop watch it has a cycle of running well for about 3 minutes before the amplitude drops off and the rate flies up.

It carries on like this for a few minutes before settling back again. A total cycle (going from 300 degrees down and back to 300 degrees) was timed consistently around 7.5 minutes. 

I can't see anything on, or around the hairspring. Could this be something in the train, on one of the wheels causing it? It's periodic nature strikes me as being train related. Pretty sure everything was clean and undamaged (though that movement  has the strangest train of wheels I have ever seen).

Could it be the spring inconsistently delivering torque? Binding and releasing? 

Two timegrapher shots showing the rate when amplitude dips and one when it is stabilising again (you can see the mess still on the screen). 

 

[mikepilk]

Looks like it could be re-banking from too much amplitude.  Are you sure you have the correct lift angle set ? Can you hear it making a galloping sound when it does it?

I have a cal 2414 which has a 42° lift angle. It also has a high amplitude, but avoids re-banking.

Does it do it if the movement is vertical ?

If not re-banking, could be coils sticking together. Sometimes it can be really hard to get all the sticky gunk off hairsprings.

I assume you have demagnetized ?

Edited July 31, 2024 by mikepilk

[AndyGSi]

The 2209 lift angle should be 46 Deg so expect the 2234 would be the same and the amplitude would then be a lot lower.

[rehajm]

Like mike, maybe sticking or rubbing once it gets a head of steam…or how long has it been running since lubricants? Hours? 

[RichardHarris123]

What is strange is the timegrapher results look better for when it's running fast.

[mikepilk]

40 minutes ago, rehajm said:

Like mike, maybe sticking or rubbing once it gets a head of steam…or how long has it been running since lubricants? Hours? 

Let a few turns off the mainspring and see what happens.

These movements usually have shims under the balance cock, so check that the end shake is OK.

[Simeon]

Thanks all for the responses. 

6 hours ago, mikepilk said:

Looks like it could be re-banking from too much amplitude.  Are you sure you have the correct lift angle set ? Can you hear it making a galloping sound when it does it?

I have a cal 2414 which has a 42° lift angle. It also has a high amplitude, but avoids re-banking.

Does it do it if the movement is vertical ?

If not re-banking, could be coils sticking together. Sometimes it can be really hard to get all the sticky gunk off hairsprings.

I assume you have demagnetized ?

I got the lift angle from a WUS thread which had a .JPG with a massive list of lift angles for Soviet watches. 42 for 24xx and 49 for 22xx it said. Obviously it would be good to get some independent verification of this now, so I will look around. 

It does do it when vertical. The first time I noticed it was when moving from DU, through a vertical position back to DD. It started coincidentally as it was vertical so I was convinced for a while it was a broken pivot (even though it had run happily in vertical a few minutes earlier). I can't hear any change in the sound or detect any knocking. The traces go from nice and clean to erratic, so I guess it may be rattling a bit. Pretty sure it's not the hairspring. It looks pretty concentric and apart from a few slight wiggles in the terminal curve seems untouched. Here is a slo- mo gif of it running, just because. 

I did demagnetise before I stepped it down. It was a little weird though as I found one of the shock settings springs magnetised on its own. I put it in a plastic bag, along with the needle problems I was using and demagnified it again. I will try it again, since I think it is good practice to do it prior to timegrapher testing anyway. 

 

 

[AndyGSi]

Here's some more details on the 2209 lift angle.

https://17jewels.info/movements/w/wostok/wostok-2209/

[Simeon]

6 hours ago, AndyGSi said:

The 2209 lift angle should be 46 Deg so expect the 2234 would be the same and the amplitude would then be a lot lower.

It could well be. It definitely will be the same as the 2209 so, will check that out.

7 hours ago, rehajm said:

Like mike, maybe sticking or rubbing once it gets a head of steam…or how long has it been running since lubricants? Hours? 

Not hours (at that time), maybe one hour. I have left it running overnight under a dust cover and will check it again this evening with a fresh wind.

 

We all benefit from a rest and recharge!

7 hours ago, RichardHarris123 said:

What is strange is the timegrapher results look better for when it's running fast.

When it's running well, it gives a nice smooth line, when bad it seems to be missing ticks on the graph. You can see it in the second picture after it has stabilised and started running well again towards the bottom of the screen. Nice flat line while the rest remaining on the screen is garbage. 

7 hours ago, mikepilk said:

Let a few turns off the mainspring and see what happens.

These movements usually have shims under the balance cock, so check that the end shake is OK.

Actually, no shims on this one but there is an awful cross scratched underneath the balance cock on the main plate. Like a bad way of raising the surface slightly to give more clearance. The end shake seems fine, the balance runs free when installed with no pallet fork. I puff it up with my blower and it swings cleanly for more than 30 seconds.

You can see the scratched cross under the balance cock here.

[Simeon]

Went back and re-tested it having let out the remaining wind and then rewound again. It is still exhibiting this effect with the erratic amplitude and changes in rate. 

I am going to strip the barrel bridge/ train bridge down again, re-examine everything and maybe swap out the mainspring/ barrel and anything else that I am not sure about. I have donor 2214 and 2209 movements (the same apart from the hack function it seems)

Not sure whether this is relevant but if I wind the watch, put the mainspring under tension with the crown while the amplitude is dropping, the rate improves. When I release the tension on the crown the rate shoots up again as the mainspring relaxes back on the click.

Could this be a sign of resistance within the mainspring/ barrel being temporarily overcome by the winding force?

[Simeon]

Interesting. As an experiment I quickly swapped the mainspring barrel with one from a donor. Untouched (no clean or oil) and it was basically the same.  Still had a circa 300 Deg amplitude before dropping off to about 200 with a crazy rate. Rules out any issues with the mainspring/ barrel. 

I pulled the barrel again, and stripped the train and examined everything. The Vostok 22xx seems to have an unusual train compared to many other watches. One of the stranger wheels is described as a 'double third wheel and pinion' on the parts list. This is literally two wheels, exactly the same stacked on the same staff. They actually can rotate separately but they don't seem to. Basically it provides a double thickness wheel in the train.  Examining all of the other wheels in the train, there didn't seem to be any problems but this double wheel had a slight split between the wheels so that one of them was out of flat and so you could see some light between them. The same wheel taken from the donor had both wheels, perfectly co-planar, no light to be seen between them.

I swapped the wheel and reassembled. I put it straight back on the timegrapher.

It has fixed the problem - the movement now runs consistently but, amplitude is now relatively low at around 240-250 degrees. 

I didn't add any more oil, I am guessing that to do that I should probably tear it down and clean it again. 250 degrees is about what I would have expected to get, reusing an old single curve mainspring in an old Soviet watch. Maybe the 300 degree amplitude that I previously saw was a caused by the fault and it should always have been about 250?

Should I bother cleaning and re-oiling it again in search of those elusive degrees?

Edited August 1, 2024 by Simeon

[AndyGSi]

From what I've seen you loose or gain around 10 amplitude for every 1 Deg of lift angle above or below the correct angle.

As you've now dropped from 49 to 46 this would account for a 30 drop in the amplitude.

[Simeon]

4 minutes ago, AndyGSi said:

From what I've seen you loose or gain around 10 amplitude for every 1 Deg of lift angle above or below the correct angle.

As you've now dropped from 49 to 46 this would account for a 30 drop in the amplitude.

It has definitely dropped with the new lift angle but I was still getting 290-300 using 46 degrees LA before I changed the 3rd wheel.

The amplitude is now creeping up slowly (up to 260-270 now) - perhaps I will let it run for a few hours to let the oil distribute itself again. 

[AndyGSi]

You could try letting the power down and winding till you get the 180 Deg
visually, then see what the timegrapher shows at 46Deg lift angle.

[Nucejoe]

Doesn't  this double third wheel and pinion work as a clutch ? 

You were getting high amplitude with the one which wasn't clutching well , so I think you'd see lower amp by clean& oil ing  the new one.

 

 

[nevenbekriev]

See, the 3th wheel has the one table riveted to the pinion, but the other must turn free. If not, this is the problem with the original wheel. Check it and clean the wheel well and ensure that the table rotates free. A little oil where the the free table is on the pinion will not be useless

@nickelsilver has explained well the purpose of this train system in another thread, don't remember where, but try and search. The russians got it from Zenith 135 calibre, which they produced as Восток/Волна 2809 and used it in Poljot 2209 and 2615 and then in this Wostok2209. I will only say that this system actually looses power and in some degree tends to stabilze (equalize) the torque of the main spring in the train. The main idea is to solve the problem with the additional seconds "free" pinion, which would make the second hand jump 2-3 seconds in the limits of the free play of the engagement of the pinion with the 3th wheel. Now the pinion is not free, as the riveted table is engaged with the pinion, then the pinion transmits the movement to the free rotating table and it then transmits to the 4th wheel pinion.

It is not normal that reducing the amplitude (due to drop of torque) leads to such significant increasing of the rate. I would check the depth of the escapement. You may try to replace the pallet fork from donor to see the difference..

The amplitude on the video is about 275.

[mikepilk]

1 hour ago, nevenbekriev said:

@nickelsilver has explained well the purpose of this train system in another thread, don't remember where, but try and search.

I remembered that explanation. And found the thread.

I've come across a few watches with the double wheel and couldn't figure out the reason. Thanks nickelsilver

 

 

[nevenbekriev]

7 hours ago, nevenbekriev said:

The amplitude on the video is about 275.

Wrong, actually about 245

[Simeon]

10 hours ago, nevenbekriev said:

See, the 3th wheel has the one table riveted to the pinion, but the other must turn free. If not, this is the problem with the original wheel. Check it and clean the wheel well and ensure that the table rotates free. A little oil where the the free table is on the pinion will not be useless

@nickelsilver has explained well the purpose of this train system in another thread, don't remember where, but try and search. The russians got it from Zenith 135 calibre, which they produced as Восток/Волна 2809 and used it in Poljot 2209 and 2615 and then in this Wostok2209. I will only say that this system actually looses power and in some degree tends to stabilze (equalize) the torque of the main spring in the train. The main idea is to solve the problem with the additional seconds "free" pinion, which would make the second hand jump 2-3 seconds in the limits of the free play of the engagement of the pinion with the 3th wheel. Now the pinion is not free, as the riveted table is engaged with the pinion, then the pinion transmits the movement to the free rotating table and it then transmits to the 4th wheel pinion.

It is not normal that reducing the amplitude (due to drop of torque) leads to such significant increasing of the rate. I would check the depth of the escapement. You may try to replace the pallet fork from donor to see the difference..

The amplitude on the video is about 275.

Thank you and thanks @nickelsilver for the explanation - Thanks to @mikepilk for finding the thread as well.

Basically instead of a spring pushing up against the sweep seconds and holding it in mesh and provide tension, with this system the sweep second pinion leaves are 'gripped' by the leaves of two wheels acting together.  One directly driven by the 4th wheel and one just dragging to give a slight force like a pair of scissors acting on the seconds pinion leaves.

In my case, with what looks like the free wheel being out of true compared to the fixed wheel, this might be varying the force applied to the seconds pinion leaves as it goes through a full revolution - hence the periodic variance in amplitude?

 

12 hours ago, AndyGSi said:

You could try letting the power down and winding till you get the 180 Deg
visually, then see what the timegrapher shows at 46Deg lift angle.

Yes, I will have to do this.  Searching around on the internet gives conflicting results. I could always split the difference and test at 44 

[AndyGSi]

Everything I've seen for the 2209 shows 46Deg with the 24** being 42 Deg.

[Simeon]

28 minutes ago, AndyGSi said:

Everything I've seen for the 2209 shows 46Deg with the 24** being 42 Deg.

Yeah, 46 degrees seems to be the consensus. I will stick with that.