Elgin Grade 315: Runs/stops with dial down/up

[nevenbekriev]

7 hours ago, SwissSeiko said:

I work on many high grade watches daily. In fact, those are the ones you definitely don't want to re-poise, as they are most likely adjusted for positional accuracy from the factory.

Only the cheapest watches are 'unadjusted' from the factory.

See, if the watch is 'untouched', then it is worth to mark the position of the roller and to be very careful not to change in any way the hairspring shape and position. After replacing the balance staff just check the performance and probably there will be not big position errors. But having in mind that even unadjusted watches also perform well when the amplitude is proper, check positions with small and big amplitudes.

I work mainly on watches that other people have tried to repair and have failed, this is something like faith... I see alot BAD repairs. There are 'ingenious' ways of 'end shake adjustment' and balance staff repairs I meet I even don't want to speak about (here the term 'previous idiots' is right on it's place). Especially with compensating balances, static poising speeds up alot the adjustment, more - it really helps for the balance truing in round. Of course, everything depends on the aim one wants to achieve.

[Neverenoughwatches]

13 hours ago, docrpm said:

bend it. If I've got the balance shimmed with 5 layers of foil, each roughly 0.016mm thick (standard household aluminum

I would concentrate on a single piece of material at the correct thickness. I have a few rolls of narrow thin steel that I sometimes use for making setting levers. Very cheap to buy and will last all of my life, 0.3mm is the thinnest I have. But there are other common place items you can use, stanley knife blades I think are around 0.4mm, single edge razor blade scapers 0.2mm and beard razors finish at exactly 0.1mm, pretty close to what you require.  Feeler gauges are also a good option. Fiddling around with multiple layers of foil doesn't sound like fun, and aluminium will compress easily,  if not when fitting , then later down the line. Ive found watch paper under a bridge before, probably less likely to compress than tin foil.

Sounds wrong putting paper in a watch, but trapped between the plate and a bridge, I cant see a big problem. Besides it will soak up the extra oil from over-oiling

[JohnR725]

15 hours ago, SwissSeiko said:

I see a lot of posts on here about poising the balance. How often are you guys doing that? It really shouldn't need to be done unless a new balance wheel has been installed. Really if its been adjusted for positions from the factory, it'll be out of poise, but adjusted for accuracy. I have a poising tool, but haven't used it. I do true balance wheels, but re-poising will mess with a watch that's already be set.

poising and the right discussion group with the right people it can become an extremely heated conversation like so many other things in watch repair. but it really shouldn't be an issue as long as everybody understands what they're supposed to be doing and why?

the first and most interesting problem we have is what exactly is the poise balance wheel? Now understand the question we need to clarify and to clarify we need timing specifications that typically do not exist for vintage watches. so the overly simplistic of all of this is you only have a poise problem if the positional timing is outside of the specifications of the particular watch your working on.

now the quote above isn't quite right the reference to a new balance wheel if you purchase a new balance wheel it's supposed to be poised from the factory. but it's only going to be poised to whatever the factory specifications are if you're expecting your whatever to be a chronometer then it's not going to be poised to your specifications. Then on any modern balance wheel without screws you shouldn't have to poise at all?

For instance here's an interesting link on poising. What makes it interesting is that the balance wheel he showing is a modern balance wheel that really does not need any adjusting at all as it would be within the specifications of the factory but he is still going to show you how to do it anyway.

https://great-british-watch.co.uk/how-to-poise-a-balance-wheel/

so is poising necessary sometimes. How do we know poising as necessary? Easy put it on the timing machine look at the watch in four positions is the timing outside of whatever you perceive the specifications are supposed to be for your watch?  I was looking at some of the watches that I do have timing specifications and so far the greatest I've seen is one minute of out of poise would be considered acceptable at a fully wound up watch. so positional errors get all kinds of differing numbers depending upon the specifications. 

[SwissSeiko]

10 hours ago, Nucejoe said:

 Are we talking the same ballgame?  restaffed balance wheel ?   bend, unbend  bridge ?   

If you restaff the balance correctly, there should be no need to repoise the wheel, unless it was messed with by a previous watchmaker. Here is an experiment to try that will explain my findings. Remove the balance from a correctly running watch that is marked "adjusted to 5 or 6 positions" or however many they claim. Place it on your poising tool, and you will find that it is out of poise. But, being out of poise, it has been adjusted for positional adjustments, to average the 5 or 10 seconds a day. It seems like you would want the wheel to be perfectly weighted all around, but it might run better a little heavier CU at the 4:30 position on the wheel. HERE is a good article that can explain it better than me.

8 hours ago, nevenbekriev said:

Only the cheapest watches are 'unadjusted' from the factory.

See, if the watch is 'untouched', then it is worth to mark the position of the roller and to be very careful not to change in any way the hairspring shape and position. After replacing the balance staff just check the performance and probably there will be not big position errors. But having in mind that even unadjusted watches also perform well when the amplitude is proper, check positions with small and big amplitudes.

I work mainly on watches that other people have tried to repair and have failed, this is something like faith... I see alot BAD repairs. There are 'ingenious' ways of 'end shake adjustment' and balance staff repairs I meet I even don't want to speak about (here the term 'previous idiots' is right on it's place). Especially with compensating balances, static poising speeds up alot the adjustment, more - it really helps for the balance truing in round. Of course, everything depends on the aim one wants to achieve.

I also see a large amount of bad repairs, back when it was a cutthroat occupation. So if someone did re-poise a balance before you, you might have to do it again to correct it for positional accuracy. But at that point, its a sunken cost fallacy, as it will take a long time to make adjustments to 6 positional accuracy, so you might be better off finding an unmolested replacement.

[mbwatch]

56 minutes ago, SwissSeiko said:

 HERE is a good article that can explain it better than me.

Ah ok ok his beef is against static poising with bench tools, not with dynamic poising to correct the running watch.

[Nucejoe]

 In conclusion we have discovered,  the reason why a Rolex spends eight months to one year on assembly line is cuz it ain't a WRT member.

 

[mbwatch]

4 minutes ago, Nucejoe said:

it ain't a WRT member

We could give them a Super Diamond Platinum Crown WRT Patron badge if they offered to pay the web hosting fees though...

[JohnR725]

7 hours ago, SwissSeiko said:

if you restaff the balance correctly, there should be no need to repoise the wheel, unless it was messed with by a previous watchmaker

so let's break this down the words restaffed balance wheel correctly would mean no distortion at all on the balance wheel. Which can be done but often isn't. Oftentimes balance wheels end up and times particularly discussing a bimetallic all bent up like the one in this discussion. If the balance wheel is out of round you definitely get a poise issue. You also get an issue if somebody pleases and out around balance wheel and then you fix the out of round now the wheel is poised incorrectly because somebody didn't physically make sure the balance wheel was correct in the first place. Yes people really do poise balance wheels that are neither flat or round it's quite annoying and then it typically with anything vintage there's an extreme likelihood of timing issues. Like the fun I had yesterday the wall thumb was 400 seconds fast and had a 202nd positional problem. So yes it's extremely common on vintage watches you're going to have things to fix from prior inappropriate repairs

7 hours ago, SwissSeiko said:

Place it on your poising tool, and you will find that it is out of poise. But, being out of poise, it has been adjusted for positional adjustments, to average the 5 or 10 seconds a day. It seems like you would want the wheel to be perfectly weighted all around,

typically and it depends upon how it nitpicky you want to be here. In other words if you are statically poising the definition of poised would be if you give the balance wheel of push it will rotate eventually come to a stop and it will not rock back and forth at all is considered perfectly poised. Typically when I poise I don't worry about that I will let it rock back and forth because it still going to be within 15 seconds typically. Now the problem with statically poising absolutely perfectly is that it does not take into account the hairspring. This is why ideally if you are really obsessed you statically poise until your close and then finish with dynamic poising. This will take into account the hairspring collet that's probably not a poised collet. Yes Hamilton had another companies had poised collet's they hairspring itself affects the poise so for absolute perfect you have to dynamic poise. So the suggestion has been that it absolutely poised dynamically done balance wheel will not be statically poise absolutely perfectly. 

 

16 hours ago, nevenbekriev said:

Only the cheapest watches are 'unadjusted' from the factory.

this is an interesting one? In the United States of America over a time span and I don't remember the exact but it ended in the 60s there was a tear off on watches to protect the American watch industry. I've read the books on the tariffs specifications it would bump up the cost of better watches to making them expensive. This is why you will often see a nice Omega watch for instance that was regulated was adjusted and was still keeping really good time will be marked not adjusted. Because the quantity of jewels each jewel bumped up the cost adjustment adjustment and variety of positions bumped up the cost so a huge percentage of watches imported into this country will be not adjusted but they were adjusted at the factory it's just a terra thing

then I suspect that even the cheapest of watches probably were somewhat adjusted because if they had zero adjustments I have no idea how bad the timekeeping would be? It's one of the problems of we don't actually have numbers of what the watches really should be doing we only have numbers of timekeeping basically when timing machines exist or for things like railroad grade watches where they absolutely had to know what the timekeeping is doing. So railroad timekeeping specification in this country is of watch Dependent up has to be within plus or -30/2 per week. Which if you compare it to the specifications of chronometer grade will ask which I think could be up to six seconds a day is better than a Rolex.

7 hours ago, SwissSeiko said:

its a sunken cost fallacy, as it will take a long time to make adjustments to 6 positional accuracy, so you might be better off finding an unmolested replacement.

this is one of my amusements at work. There we have that really nice which he timing machine full automatic microphone we look at every single watch and six positions and my boss once the watches all the keep really good time. So basically the watchmakers have learned to let the boss adjust the watches it seems to make him happy and we can do other things like service watches. So yesterday's fun on the 18 size Waltham to get the 400 timing down to think he got it down to 15 seconds it might've been better than that the Delta was still might've been within a minute much better than what it was before but still took more than an hour of time. When we don't specifically charge for this we charge for servicing of the watch we do not typically charge to make your watch keep time in multiple positions so yes it wastes a lot of time unless you're getting paid for it where it can definitely eat up a lot of time to fix vintage watches that I've had interesting repair histories.

On 4/16/2025 at 3:25 AM, Nucejoe said:

Didn't know shimming conforms to Swiss horolog standards, in watch repair, if its to  compensate for pivot wear, it actually is not a bad practice, for the following reasons. 

sorry I don't remember the brand name we once had a watch that was one of I take 1000 it actually was engraved number or something of something in other words it was a very expensive Swiss watch and I believe there was at least one shim under something. I always think of at the factory level if you have to place a shim to me it's a sign of poor manufacturing but there was a Swiss watch a nice watch that had it done. Of course what it may also be a sign of limited quantity of watches where things or adjusted at the factory

18 hours ago, Nucejoe said:

Are we talking the same ballgame?  restaffed balance wheel ?   bend, unbend  bridge ?   

we end up with interesting problems and I will pick on American pocket watches again with interesting repairs that can happen? For instance pocket watches dropped balance staff is broken possibly the whole jewels are broken and sometimes cap stones get broken. or worst-case I had a Hamilton 992 it actually had little notches taken out of the main plate I always wonder if it got hit by a train and things were dramatically bent in this watch somebody had ground the pallet fork bridge down to accommodate the balance wheel because the balance bridge had been bent to insane proportions it was bent so much it was no longer up and down in alignment so I always wondered if this particular watch got hit by a train. With its ultimate repair it would've been better to just replace the whole movement but instead we cannibalized another movements to replace all of the horribly mutilated parts.

Now back to not watches hit by a train while somebody was crossing the railroad track.

If the balance jewels are broken they are replaced but the same problem like we have on this group where do you get replacements? This means people in the field will get creative if they replace things with size differently than the original things will change. In other words the total length of the balance staff conceivably will change because they've changed the jewels and they're not properly sized.

Then there's a problem of replacement balance staffs I personally have often theorized in the case of  bestfit  that they deliberately made the staffs oversize so the watchmaker can reduce the size. But what if you don't have a lay then what happens? A for instance the Hamilton 992B only has one balance staff as it's of friction staff but often times the friction staff doesn't fit because the replacement staff is slightly oversize which should be impossible but it happens. Often times I have to go through an assortment of staffs to find one that fits because I don't feel like reducing at the size to fit. More common though would be staffs with pivots that are too long and if the watchmaker does not have a lathe again how do they fix that problem well undesirable for us that have to deal with it down the road

so an ideal perfect world the adjusted by the factory to whatever the specification was at the time and on a seven jewel American pocket watch it's not going to be railroad timekeeping. If we replace the staff in a proper fashion without bending or distorting anything the watch should keep time within the specification when it was originally made except of course we don't know what the specification was and they definitely did not have timing machines when this watch was made.

If we replace the balanced staff with a factory original factory staff and everything should fit perfectly if the jewels were broken and we used factory jewels everything once again would be perfect. Although if you look in the parts book typically for American pocket watches for certain parts they recommend sending the entire watch or parts of the watch back because American watch manufacturing at least until the not sure exact date but sometime past the 20s definitely by the 50s watches were made in batches which is why everything has a serial number and the tools over the course of the batch tend to wear a little bit. So conceivably something made at the beginning and the end of the batch may not fit without adjustments. Plus of course there's variations in parts I like the pic on Elgin because they have one balance staff that comes in four different size variations and then a whole bunch of pivot variations all with the same part number

so yes and an absolute perfect world which doesn't exist changing a balance staff with everything factory new would not require poising or truing the balance wheel because everything should still be perfect. But unfortunately in the 100+ year life of your watch things have happened.

If you want to grasp that go back and start on page 1 of this discussion as the balance wheels in this discussion have had issues they're not flat the probably out of round there were issues with the jewels there is issues possibly with the staffs. Oh and yes apparently there's a donor watch and if the donor watch is not within a very close serial number range of the original watch it will have very likely sizing differences especially for anything related to the balance wheel because all the vintage watches were adjusted individually in the factory. If you look at a video on I can find it if you want to watch it on YouTube from I believe the 30s where you will see on escapement related stuff every single watch is adjusted which is why you can't just grab the balance wheel from one watch shove it into another one as they were adjusted at the factory

 

 

 

 

 

 

 

 

 

 

[LittleWatchShop]

3 hours ago, JohnR725 said:

Now back to not watches hit by a train while somebody was crossing the railroad track.

yup, it happens A Pinto pony and a Navaho.

Edited April 18 by LittleWatchShop

[docrpm]

After another long hiatus (while waiting for truing calipers to arrive), I’ve made some more progress.

TRUING IN THE FLAT

I was able to use the calipers I purchased to true the wheel in the flat. There’s still a very slight eccentricity in the round, but I’m going to leave it as is. It doesn’t wobble any more and I’m happy with that. Thanks for the tool recommendation @mbwatch. It worked well, though I must say it takes a delicate touch to get the balance between the calipers. Also, my calipers effectively have four different variations based on the orientation of the arms (2 sides, 2 orientations each). Some are more tapered than others, though to what end, I can’t quite tell. My first thought was that flipping them over yields arms that can be used for poising. They are K&D horizontal calipers, btw.

OSCILLATION TEST

@nevenbekriev @JohnR725 After truing the wheel in the flat, I carefully remounted the hairspring and did your oscillation test. Balance wheel mounted to balance cock, no pallet fork, rotate balance wheel 180 degrees, let it go, and time+count the oscillations. Results as follows for dial down orientation:

• Duration: 1m 15sec (to full stop)
• # of oscillations: About 125, though the amplitude of the oscillations decreased significantly, as one might expect

I should redo the test in at least two other orientations…Forgot to do that.  

In general, it’s safe to say this watch has a pretty low amplitude, likely for a variety of reasons. I haven’t oiled the balance jewels FWIW…I should probably do that, but I really don’t want to keep disassembling and reassembling the balance. Is it possible to oil those jewels (top and bottom) by just removing the capstones and putting in a tiny amount of oil, with balance in place? I can see how having the pivot in the jewel hole could change how capillary action works, and thus might pull more oil in than desired. However, those capillary forces will be in effect anyway once the balance is inserted.

PRACTICE MOVEMENTS

@JohnR725 I agree with your sentiment about using practice movements wholeheartedly, and I’ve done that to some degree. The watch I’m currently repairing is for a friend, and she was willing to have me treat it as a “learn-by-doing experience.” With that said,, I probably should have purchased a few more “beaters” before getting to something where I don’t want to muck things up. I’ve still got a couple of those on the bench that I used as donors, and I’ll try to rehabilitate them.

[docrpm]

@JohnR725 You had also asked about the serial numbers for the watch under repair and the “donor.” Those are 36,764,469 and 19,110,954 respectively, so not that close to each other. However, it was the balance that I was originally going to swap in, and I’ve since purchased both new staff and new hairspring. Pretty sure it’s the original balance cock and wheel currently installed. The only part I wound up using from the donor was the click spring, which had been completely disintegrated by rust in mine.

[JohnR725]

3 hours ago, docrpm said:

Those are 36,764,469 and 19,110,954 respectively, so not that close to each other.

yes the pocketwatch database shows quite a spread in those serial numbers. With the watches typically made in batches and even in a batch from the beginning to the end there be some variations. Then you have approximately a 10 year difference between the two yes some of the stuff will probably not fit especially escapement components as there usually adjusted at the time of manufacturing and of course manufacturing tolerances were improving with time.

https://pocketwatchdatabase.com/search/result/elgin/36764469

Grade:    315   
Model:    3    
Class:    114  
Estimated Production Year:    1937 
Run Quantity:    4,000    
Grade/Model Run:    310 of 329

https://pocketwatchdatabase.com/search/result/elgin/19110954

Grade:    315    
Model:    3    
Class:    114   
Estimated Production Year:    1917  
Run Quantity:    2,000    
Grade/Model Run:    127 of 329

 

[docrpm]

More data on the oscillation tests:

• Vertical: 15-20 seconds…Abysmal
• Dial up: 1 min 15-30 seconds

Any thoughts @nevenbekriev @JohnR725?

My next thought is to oil the balance jewels to see if that helps. It’s certainly one major thing left that I haven’t done.

[nevenbekriev]

Yes, You have to oil the stones. This will of course make some improvement. With such size of balance, a little bit more free oscillations is expected to happen, like 150 -200. The amplitude decreases exponentially, this are the laws of nature. See what will happen with oiling. You can add oil thru the hole of the hole stone, without disassembling the jewel settings. The main concern when a newbie doesn't have good enough free oscillations test is the hairspring. But since You have almost equal results DU/DD, I am not sure.

Edited Wednesday at 05:50 AM by nevenbekriev

[docrpm]

Well, I oiled the jewels and now it’s worse. I give up.  

I think it’s time to just re-assemble and call this project a learning effort that didn’t work out. Thanks for all the help and advice @mbwatch @nevenbekriev @JohnR725. I’ll be back with another project soon, I’m sure.

[mbwatch]

Before you give up, have one last look at the hairspring to make sure it didn't land outside the regulator pins or something like that after oiling the jewels and reassembling. That happens easily and is easy to miss.

[JohnR725]

6 hours ago, docrpm said:

Well, I oiled the jewels and now it’s worse. I give up.  

I think it’s time to just re-assemble and call this project a learning effort that didn’t work out. Thanks for all the help and advice

One of the problems for people helping he is we rely on your eyes on the watch. So it's worse but what does that mean? Is it on a timing machine how did you do the before-and-after so it be nice if you give us a little better answer

then since apparently this discussion is coming to an end in your giving up I thought I go back and look at the very first page let me quote something

On 2/3/2025 at 2:06 PM, docrpm said:

'm working on an Elgin Grade 315 vintage pocket watch. Generally speaking it's in great condition. I broken it down, cleaned it, re-assembled it, and got it running, sort of. The problem is that it runs when it's dial side down, but doesn't when it's dial side up. After doing a bit of sleuthing, I decided to take a very close look at the balance wheel and staff.

So generally speaking it's in great condition?

For future reference what you think about several things we tell newbies to work on pocket watches because they're big. Typically I tell people to get a Chinese clone of a 6497 as it's a new pocket watch and it's running. Newbies need to work on a running watch to see what a running watch looks like to evaluate to disassemble and reassemble not just one time but a lot of times to learn how to disassemble and reassemble and then you practice oiling you do this a lot and after you completed doing it a lot of times you do it again because you really do need the practice

The problem with starting on a pocket watch like the one that you chose its old and it's probably not going to run if you clean it. As you can see they need repairs and adjustments conceivably lots of them and hear the person on the field you're making the observations that you still don't have the skills to make the observations it's going to be a problem for all of us. Like for instance it's in great condition when exactly did that mean? It's always nice to evaluate the watch before disassembly so did your great condition pocket watch run initially did it run in all positions what exactly was the assessment before you disassemble It

5 hours ago, mbwatch said:

Before you give up, have one last look at the hairspring to make sure it didn't land outside the regulator pins or something like that after oiling the jewels and reassembling. That happens easily and is easy to miss.

Excellent device because at happens to all of us. Put a watch back together take the balance wheel in and out of multiple of times put it on the timing machine and how that looks really bad? Then notice the hairspring has magically relocated to someplace else.

6 hours ago, docrpm said:

think it’s time to just re-assemble and call this project a learning effort that didn’t work out.

I do find it really sad that it didn't work out? Six pages of I guess we wasted our time you learned nothing?

The problem with watch repair is when we don't always see this in popular YouTube videos would be the word repair is involved in an ordered you repairs you have to do diagnostics and as you can see learning how to repair in doing diagnostics on a watch takes a long time the learner I would like to think that you may be learned something and six pages of discussion if nothing else repairing a vintage pocket watch isn't as easy as the videos look like on YouTube

[docrpm]

3 hours ago, JohnR725 said:

I do find it really sad that it didn't work out? Six pages of I guess we wasted our time you learned nothing?

I said it was a learning effort...I learned a TON!!! It was a great learning effort. Your time was much appreciated and led to a much deeper understanding on my part of many many things. I'm sorry if what I said gave a different impression. When I said it didn't work out, I meant in the sense that I wasn't able to repair the problem on the pocket watch I'm trying to service!

More later...

[docrpm]

@mbwatch @JohnR725 I checked the hairspring to see if it’s between the regulator pins, and indeed it is, so that’s not the culprit, unfortunately. Thanks for the suggestion; I had forgotten to double-check that when I re-installed the balance.

In terms of what’s not working @JohnR725, I should have been clearer. For the oscillation test, in the dial-down position, the time dropped from over a minute to something more like 20-30 seconds. It’s markedly worse than it was before I lubricated the balance jewels. Which opens up a number of different questions:

• Did I mess something up in oiling the jewels? Too little oil? Reinstalled incorrectly? The jewels are in good condition, btw. I looked closely at the cap jewels and the hole jewels. They’re all original and fine. No scratches, dings or cracks.
• Has something changed with my shim given the repeated removals of the balance? I checked the end shake as best I can, and it seems ok.

In terms of starting out with a practice movement @JohnR725, I did that. In fact, I worked on a Seagull ST-36, which unless I’m mistaken, is a Unitas 6497 clone. I stripped and re-assembled that movement three times, familiarizing myself with all of the parts, taking notes, photos, etc. I then oiled on my last assembly, but didn’t attempt to regulate. Performance was slightly worse on the timegrapher after oiling. In hindsight, what I should have done was spend even more time disassembling, re-assembling and attempting to regulate the movement. Lesson learned!

After the ST-36, I tackled my grandfather’s Hamilton 974. It took me a little while, but I restored it to working order after replacing a few parts and troubleshooting one performance issue. I didn’t oil and regulate, because I wanted to practice more before I tackled that task.

At that point, I was confident enough to try another pocket watch. Enter the Elgin in this thread. I had a misplaced sense of confidence after my previous efforts, and dove in with disassembly, cleaning, fixing obvious issues (missing crystal, rusted click spring). Then the DU/DD problem cropped up, and I entered the rabbit hole.

Again, in hindsight, what I should have done at this point was NOT go down the rabbit hole. Instead I should have gone back to the ST-36 and spent some more time with deeper work and practice there. With that said, I would not have had any practice with balance staff repair, because there would have been no need. Working with balance staffs seems like a Catch-22 to me. You can only get the experience by doing it, but you need the experience to do it.

If I had not found this entire process interesting, helpful, and a great learning experience, I wouldn’t have continued. I’ve tried repeatedly to express my thanks for the time all have invested in this lively thread, and I’ll do it again: Much appreciated @mbwatch @nevenbekriev and others!

And to be clear, many thanks to you (that was implied  ).

[mbwatch]

Nahhh you didn't do the wrong thing. You started out on the ST36 and that's great, and you would have fallen down this rabbit hole eventually anyway. Maybe another 12 size Elgin to restore from more dire condition before attempting the harder repairs on this one, but you have a result of some kind here.

Oiling should not have a negative effect on your free oscillations, so I would look into a change to the end shake, however tiny, caused by moving your shim. If the shake is just a little bit too much, amplitude can drop in a major way which would show up early in degraded free oscillations. Sometimes it is only a matter of tightening the balance cock screw a little more or shifting the shim a tiny bit, or replacing it if it got compressed too much.

All the stuff you have since learned, you can take back to your grandfather's Hamilton and do a true restoration. But practice on 3 or 4 other unimportant pocket watches first.

[docrpm]

I learned one other thing yesterday watching a video from Alex Hamilton, which was to test the end shake of the balance staff BEFORE mounting it in the balance wheel. He also suggested setting it in the baseplate jewel and looking at the tilt angle (5 degrees being ideal). I wish I had done this with the balance staff currently installed!

So I tried adjusting the shim a few times with no success (I.e., no change in oscillation test). Then I looked at the end shake again and wasn’t happy with the amount. It seemed maybe a bit too much, because the balance wheel can tilt back and forth quite a bit when I’m trying to get the pivots seated in the jewels.

Then I remembered something you said @mbwatch, which was that occasionally one needs to adjust the depth of the hairspring “nut” (I don’t remember the exact name). So I removed the balance, flipped it over, loosened the hairspring screw, and tapped that nut slightly to seat it more deeply in the balance cock, thus raising the hairspring slightly (I think).

When I attached and ran the oscillation test again, my result improved by maybe 20 seconds (which isn’t bad, when you’re starting with a baseline total duration of 20-30 seconds. It was then that I noticed a ding on the TOP of the balance cock near this nut. See attached photos.

This looks suspicious to me, and when I look very closely at the balance cock arm, it doesn’t look exactly flat. Attaching photos of that as well, but it’s hard to get the perfect angle in the microscope.

 

I just had a #facepalm realization: Why not try swapping in the balance cock from my donor movement? If anything it will provide more data points. Worst case, it’s worse (in terms of oscillations), best case it’s better. I’ll give this a shot.

[JohnR725]

3 hours ago, docrpm said:

Too little oil?

Out of curiosity which oil did you use?

Oh and yes oil does have a dampening effect. It's actually quite interesting if you take like your 6497 clone clean it lubricate all the key lists mainspring etc. do not lubricate the watch. But on the timing machine see how it looks start oiling the watch see what happens you probably see a loss of amplitude at least until you get to the escapement and then you'll see a spectacular improvement so oil doesn't always have a spectacular increase in amplitude but you definitely have to have it

2 hours ago, docrpm said:

I learned one other thing yesterday watching a video from Alex Hamilton, which was to test the end shake of the balance staff BEFORE mounting it in the balance wheel. He also suggested setting it in the baseplate jewel and looking at the tilt angle (5 degrees being ideal). I wish I had done this with the balance staff currently installed!

One of the ways I explain learning watch repair would be similar to learning to be a doctor. Have you ever noticed how doctors say they're practicing or they're in a practice why are they still practicing shouldn't they get in the game? That's because they are still practicing their getting better the same as were getting better each of our watches is like a patient guess they're similar but there are also different but all different and we are practicing every single day

but typically in watch repair people watch a video they practice on one watch taken apart a couple of times just like you did then they want to get into the game they want to work on a live watch to learn experience. Problem is experiences learned over time a very long time in the case of watch repair and as you can see working on a vintage watch requires a heckuva lot of knowledge and egg a lot of remembering things like for instance that 5° angle thing that you mentioned up above wider to go look at the first page of this discussion somebody already showed you that. The problem was too much knowledge being gained too fast and you didn't absorb it because you can't. Knowledge of watch repair is learned over time. Knowledge of vintage watch repair is learned over a considerable quantity of time because there's so much stuff going on so many things to deal with

oh and we are learning watch repair like at least in the old days when their learning to be a surgeon they got the work on dead bodies I believe now they get to work with computer simulation. They do not jump in the field and start cutting apart people like we do in watch repair where your practicing on.

2 hours ago, docrpm said:

I just had a #facepalm realization: Why not try swapping in the balance cock from my donor movement? If anything it will provide more data points. Worst case, it’s worse (in terms of oscillations), best case it’s better. I’ll give this a shot.

Outstanding idea as you're here to learn.

3 hours ago, mbwatch said:

All the stuff you have since learned, you can take back to your grandfather's Hamilton and do a true restoration. But practice on 3 or 4 other unimportant pocket watches first.

To be quite honest I think practicing on 3 to 4 unimportant pocket watches is outstanding but realistically you're going to need more experience before you really start restoration work and you're going to need more equipment like a watchmaker's lathe would be nice.

Then ideally try to find some pocket watches that aren't totally destroyed. Because ideally with learning you actually want to learn and grasp what you did or did not do to end up where you are as opposed to where we are now

Now that you have greater knowledge of pocket watch repair man make a suggestion is going to require considerable patience though.  Go back to page 1 of this discussion and read through the discussion do not act upon anything you see just read the whole discussion again.

[nevenbekriev]

Well, as I have said not once, for the beginners in watchmaking, the hairspring, especially when there is overcoil, is the source of the most of the problems. This is the case here too.

The end shake... I told You to regulate it by bending the cock. If it was done, then would be no wondering if the shim is seated correctly every time. You still have to understand how much end shake is normal and correct. One exercise for this purpose:  remove the hairspring from balance. Put the balance in the movement with the cock. Use thin sharp pointed tweezers to grasp the balance staff  where the hairspring collet seats and try to move the staff up/down. The amount of movement should be more or less as much as in escape wheel or pallet fork arbor. In the same time You will see how easy the balance rotates when it is free and without hairspring. Moving the balance up/down by holding it by the rim is not the same, as it forces it to tilt and some shaking may be result of the radial free play of the pivots in the  bearings even if there is no endshake at all. Of course, You can grasp the roller too. You can also grasp the rim, but observe the staff movement, not the rim shake. One should be able to check and say if there is end shake and if it is normal without hesitation, no matter if the movements works well or not at all, and not to wonder if the reason for the not normal working is endshake.

Now about the hairspring. You spring is touching somewhere. The fact that this slight lifting of the stud changed the situation proves it. The spring probably touches the arms of the balance, but You are the one that can look at the spring, so it is up to You only to find where it is touching. The spring should be parallel to the plane of balance and thus on equal distances from the cock  or from the balance arms in every position of the balance when it rotates. The coils of the spring must not touch each other. The overcoil also should be parallel and not touching the cock. The spring should not touch (be on some distance from) the regulator pins or the stud bottom. All this must be true also in DU position, not only in DD.  This is achieved by slight bending and twisting close to the stud  or of the overcoil.

[docrpm]

1 hour ago, JohnR725 said:

Out of curiosity which oil did you use?

Moebius 9010. A small drop on each end cap jewel (top and bottom)…enough to cover at least half the jewel, but not all the way to the edge. I was conservative.

 

1 hour ago, JohnR725 said:

That's because they are still practicing their getting better the same as were getting better each of our watches is like a patient

I like this analogy. And yes, it’s imperfect, because surgeons don’t start with live people. It’s definitely one of the things I learned with this watch, where I’d fix one issue, but then introduce 1-2 others, often not necessarily getting at the root cause of the “illness.”

1 hour ago, JohnR725 said:

Then ideally try to find some pocket watches that aren't totally destroyed. Because ideally with learning you actually want to learn and grasp what you did or did not do to end up where you are as opposed to where we are now

Well, this watch wasn’t totally destroyed…Unfortunately, I wasn’t able to do enough up-front diagnosis before I started pulling everything apart. This was one major lesson learned to me, but tbh, I don’t think I would have learned it had I not pulled everything apart and seen the interactions of all the different systems (just inside the balance assembly itself!). I have a few other Hamilton 974s that I’ll be able to work on more thoughtfully.

1 hour ago, nevenbekriev said:

The end shake... I told You to regulate it by bending the cock. If it was done, then would be no wondering if the shim is seated correctly every time.

I definitely heard your advice, and I think a few others echoed it, but I’m fairly certain I would have introduced more problems by attempting that fix. I don’t have the necessary skill or tools to (1) bend the cock and (2) assess exactly how much it has been bent. I can’t machine a brass plate to attach the cock while bending, I can’t measure the amount of deflection I’ve introduced. And it’s the kind of thing where botching it seemed like it would be irreversible.

1 hour ago, nevenbekriev said:

Now about the hairspring. You spring is touching somewhere. The fact that this slight lifting of the stud changed the situation proves it. The spring probably touches the arms of the balance, but You are the one that can look at the spring

I’ll attempt to look at the hairspring to see if it’s touching anywhere. The central challenge is that I don’t have a reference point. I don’t have that for end shake or for anything else. I could compare to my ST36, but that seems like an apples-oranges comparison to me (new movement v. 100+ years old). That’s one of the main challenges for me when learning a lot of this stuff – you create reference points yourself!

2 hours ago, JohnR725 said:

you're going to need more equipment like a watchmaker's lathe would be nice.

The next thing I was thinking of getting was a JKA Feintaster bench micrometer, tbh. After watching Alex Hamilton measure staffs and pivots, it seemed like it would be a good purchase. They’re just expensive. I’ll need to sell a watch or two first.  

A lathe seems like something where I need to collect more experience before I pick one up. Perhaps you convince me otherwise.

In the category of ideas good in theory, but not in practice, I swapped the balance wheel onto the balance cock from my donor, only to discover that the balance cock doesn’t quite fit into the baseplate. The pin on the cock that aligns with the baseplate is either slightly too large or displaced relative to the hole in the baseplate (I suspect the latter).

I actually took off the balance wheel and double-checked by trying to screw down the balance cock by itself, just in case there was an alignment issue with the pivot, but no…The picture shows the sizable gap that exists even with the screw for the cock tightened all the way. 

This is probably one of those circumstances where I could use a lathe to reduce the diameter of the pin, right @JohnR725?  

[docrpm]

Now that I think about it, I can probably just file down the pin carefully to the point where it fits. Is that too much of a hack?