ETA 2472 dial up snowstorm

[RobS]

I’ve been working on an eta 2472 for a while. But until recently I didn’t own a timegrapher, ignorance is bliss! 
I’ve been regulating watches by eye for beat error, by slow mo iPhone video for amplitude and with a stopwatch over an hour to a day for rate. 
well my eta was resisting all attempts and  eventually I bought a timegrapher, pleasingly, all my seiko attempts were surprisingly good (albeit with trademark low amplitude) but the ETA that had been somewhat problematic turned out to be a mess.  I had thought my date jumper not always engaging was my only real problem, I was wrong!
 

symptoms:

dial down, trace, rate, amplitude, beat error all seem tolerable. 
dial up: snowstorm 

 

what I’ve tested:

Just movement no hands or dial in case of rubbing 

no calendar

no motion works

 all the same  ok DD, bad DU  

completely stripped and cleaned again  

just balance and air puffer balance only span for circa 25 sec DU and DD  

rebuilt watch, same result, DD fine, DU snowstorm  

from my totally clueless analysis, end shake seemed normal enough, results were the same with or without lubricants  In that the DU position is a problem  

I suspect there is something rubbing in the escapement when DU, but I haven’t spotted anything. But that doesn’t mean it isn’t there, I only have a cheap x15 loop for magnification  

 

what is likely, how do I look for it  

 

thanks in advance, Rob

 

 

[Xovius]

I'd be interested to know how it performs in crown down or other sideways positions. 
To me it sounds like there's probably an issue with either a jewel or a pivot on the lower side, when the watch is in a DU position there is a lot more force on the lower side pivots. 

I'd check for broken jewels and bent, broken or beaten up pivots on the side facing away from the dial. 

[JohnR725]

It would be nice to have pictures of the watch and the timing machine results.

[RobS]

Here are some of the photos from the disassembly process. I’m afraid I don’t have any photos of the timegrapher data at present and I’ve left my all my watchmaking stuff at work (my kids fiddle with it if I leave it at home). 

[RobS]

The “crown” is positioned towards the spring loaded black plastic part

[mikepilk]

The DD amplitude is lower than expected.  Is the hairspring perfectly flat and centred? Could it be catching on the balance wheel arms?

Quote

just balance and air puffer balance only span for circa 25 sec DU and DD  

Is that with a strong blast, until stopped, with/without pallet? 

My criteria is : with pallet removed, and with a strong blast from a puffer, I want to see good movement (say 180°+) after 30seconds. On one of these movements it should move until about 1 minute.

[RobS]

As per suggestion, looks like balance spring not flat or central. Side on view suggested possible rubbing but I couldn’t capture on video. 
taking balance off to my untrained eye shows it isn’t central. 

I’m going to try and take the balance off entirely now to see, but my success rate at spring manipulation is at 0/3 so far!
 

25sec was with a fairly strong blast, no pallet fork. 

[RobS]

Once removed I would say it actually looks good. My conclusion is that it’s the very last bit where the final pin goes in that’s not lined up properly. I’m assuming to someone competent the easiest fix….

[RobS]

0/4

 the spring manipulation actually went rather well, I attempted the “stroking” method. But when I tried to put it back in the balance cock, the whole regulator assembly fell apart with one part having snapped in two! 
 

 

[caseback]

Did you take photo's of the trainwreck? 

[RobS]

Now given its location at the end of the train of wheels, train wreck is a lovely description of the mess I got myself into. Alas no, I put all the other parts back together then set it aside in the I should have known better pile until I can find myself a reasonably priced replacement balance cock, and hopefully good balance wheel attached before I repeat this whole debacle! 
 

next mission an old alarm watch that only runs while you shake it, what could possibly go wrong?!

[RobS]

I bought new regulator parts with the hope I can repair it 

[RobS]

So, the regulator arm turned out to be too long so I cut it down. 
Disassembly went rather well just a small pin remove at the back 

Reassembly with somewhat more challenging and getting the pin back in was much harder than getting it out.

and of course getting the retaining pin/clip back in was much harder than getting it out.

my amateur trick as ever with bits that may fly away is a thin sheet of plastic over the top, makes it harder but catches all the parts too!

But I got all the pieces back together and then it came to reinstalling the balance wheel itself which point it all fell apart all over again!  
 

I repeated this twice before admitting I probably didn’t have exactly the right part. 
 

so my bargain  turned out the opposite! 
 

 

 

[Jon]

On 7/12/2024 at 3:34 PM, RobS said:

Once removed I would say it actually looks good. My conclusion is that it’s the very last bit where the final pin goes in that’s not lined up properly. I’m assuming to someone competent the easiest fix….

There's loads of these movements and parts out there to be had at bargain prices. Great movements. I especially like the calendar driving wheel mechanism, if you can call it that, as it grabs and snatches the calendar wheel rather than turning it. Such a snappy and quick date change for an inexpensive movement back in its day!

I liked how you tried to get the upper index to fit in the recess of the cock. Great thinking!

One thing, I made an observation of the hairspring when you removed the balance and hairspring from the cock. There is no end curve on the spring, so it won't work how you expect it to. This is a side issue to the problem yopu are having, but knowing the end curve needs forming and trying in situ to know it is correct is vital before you go forward from here.

Also, taking the hairspring off the balance and attaching it to the balnce cock as in the last pic will show you if the hairspring is true because it shpuld be parralell to the cock when the stud is screwed up tight

Here's some pics of hairsprings with endcurves formed correctly. The last pic is with the hairspring removed from the balance and placed and screwed in position (very important to do, as the screw locates with the indentation in the stud and positions everything properly). You can see that the collet needs to be exactly over the jewel, otherwise when the index on top of the cock is moved from the extremes of '+' to '-' the curb pin and boot won't touch the hairspring. If it does this will affect timing especially in vertical positions because the hairspring may 'hang' on the curb pin and make an increase in the timing in that vertcal position, all because the hairspring hasn't been centred shown in this last pic. That's where I would start, otherwise you are trying to diagnose and alleviate a problem when there is a potentially galringly big one in front of you because the hairspring is not shaped correctly. This is incidental to the DU and DD problems you are having BTW

A good way to check that the collet is or isn't centred over the balance cock jewel is to move the index from right to left, from '+' to '-' to change the timing from fast to slow and if you see and big difference in the beat error. If there is, then that is strongly inferring that the endcurve isn't formed properly and the curbpin and boot are pushing and pulling on the hairspring when the index is in different positions from '-' to '+'

If the hairspring is ever badly pushed or pulled by curb pin and boot there will be be loss in amplitude.

 

Edited July 17, 2024 by Jon

[nevenbekriev]

4 hours ago, RobS said:

...But I got all the pieces back together and then it came to reinstalling the balance wheel itself which point it all fell apart all over again!

I repeated this twice before admitting I probably didn’t have exactly the right part.

so my bargain  turned out the opposite!

As i see, actually the stud carrier is the original one, isn't it? If so, then no reason to be not able to fit the stud in the carrier.  But You should not aply force on it. Put hte cock on the table upside down, put the balance on it hairspring down, align the stud above the hole in the carrier and press gently the stud to get in, that's all. the carrier has to be in contact wit the table top, this will prevent it from detouching from the cock.

It is usually not needed to remove the incablok just to fit the regulator and carrier in place. They should just snap in place with a little pressing.

[VWatchie]

7 hours ago, Jon said:

Such a snappy and quick date change for an inexpensive movement back in its day!

 

[RobS]

Thank you all for the super detailed replies. 
the regulator came off in the first place as I had held the balance cock still to install the balance wheel with rodoco, when I lifted it up it had stuck to the regulator arm and pulled it right off! I tried to press it back into place and it snapped in two, from bad to worse!

i bought a new regular pointer thing, meaning the other two components are original to the best of my knowledge. Once bitten twice shy I assumed you had to remove the incabloc to get it all together. 
 

the date does look good indeed, I actually had this in a totally clear case for a while with a clear dial, I liked the juxtaposition of aged mechanism with “modern” case. I have had issues with the jumper spring arm not engaging though, but considering my current predicament with it that’s the least of my worries.

 The terminal curve issue is interesting, as far as I can tell the pin was already pointing too far from the centre. 

[RobS]

You couldn’t make it up, the complete pigs ear I made of this movement. So after the regulator (balance) cock up I finally got it all back together again only to realise I lost the screw for the balance stud in the process. Trusty bag of spare screws to the rescue (getting that screw in was tough though). 
I remounted the improved balance spring. Popped it on the Timegrapher and things were better hard not to be better than a snow storm really. Well when I moved the regulator around it didn’t behave predictably I think the afore mentioned lack of terminal curve to blame. Just as I decided to set about trying to form one. The movement pinged out of the Timegrapher then refused to run…

stripped movement. Only to discover the fall bent centre pin

[JohnR725]

On 7/19/2024 at 12:26 AM, RobS said:

I had held the balance cock still to install the balance wheel with rodoco, when I lifted it up

Interesting technique don't think I've seen that one before.

 

 

[RobS]

Dear John, I wouldn’t recommend it! Today I attempted to shape a terminal curve on the balance. Needless to say it ended in my ordering a new balance wheel…

i think I’ve sorted the bent post though!

in a similar amount of time that I spent on messing up the hairspring further I managed to strip, troubleshoot, fix, clean and reassemble the sekonda alarm watch shown below. Although it does have a beat error to sort and that means dancing around the hairspring again, I’m avoiding it for now!