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

[docrpm]

On 3/21/2025 at 11:29 AM, mbwatch said:

You can test reducing shake by making a tiny piece of tin foil  and putting it behind the balance cock screw, toward the outside. Screw down the balance cock on top of it and it will push the jewel down on the balance staff to reduce shake. Double the foil for double the effect.

Thanks @mbwatch @nevenbekriev @JohnR725 And apologies for the very long delay in response (Spring break vacation for my son, life, etc.) I haven't given up.

So here's what I did:

1. Fold over a small piece of tin foil multiple times
2. Place under the edge of balance cock
3. Tighten
4. Inspect end shake visually (by moving balance wheel up an down with tweezers, observing the roller movement)
5. Use a puff of air to test rotation

I went through this process multiple times, with three key observation thicknesses:

• NO foil: Lots of end shake visible. Puff of air spins balance, but when I rotate so dial is up (balance down), I can hear metal-on-metal and the balance stops spinning
• Lots of foil: I found the point at which the thickness of the foil was enough to eliminate end shake, but also only allowed slight rotation of the balance wheel, but no ability to spin freely. In other words, just a bit too much, which I then backed off to the goldilocks thickness...
• Goldilocks thickness: A very small amount of end shake visible, but not too much. Balance wheel rotates cleanly with a puff of air. AND (drum roll please), when I rotate the movement to dial side up after a puff of air, there is NO sound of metal-on-metal and the balance wheel continues to spin freely in this orientation.

The conclusion I draw from this experiment is that excessive end shake is at the very least ONE part of the problem.

So what's next?

 

[LittleWatchShop]

Dial washers are useful for this task.

[mbwatch]

12 minutes ago, docrpm said:

So what's next?

Hey now that's a start. Remind us what else is wrong with this watch at this point?

2 minutes ago, LittleWatchShop said:

Dial washers are useful for this task.

Yes, dial washers are what I usually use on pocket watches because I have a lot of them and they're easier to position than foil (but do use foil or watchmaker's paper on wristwatches).

[docrpm]

10 minutes ago, mbwatch said:

Hey now that's a start. Remind us what else is wrong with this watch at this point?

I know! I was excited to make a tiny bit of progress. 

Any advice on how I go about acquiring and using the appropriate dial washers? 
 

I don’t know if there IS anything else wrong, though I suspect the balance wheel is slightly out of true. However, I have yet to reassemble everything with the new balance spring and pallet fork, so it’s hard to know. 
 

Oh and I checked for end shake in the pallet, but it looks solid. 

[mbwatch]

You don't really need to get dial washers, they're just a little easier to push around with tweezers when used as balance cock shims, and more durable than foil. But if you want to look for them, you just have to wait around for an assortment on ebay. Optional.

Truing a pocket watch balance isn't too hard. They made a little wrench tool for it, which is really just a brass rod with a flat end and a notch cut into it - you put the notch over the balance rim and bend it into true. But you can bend it with your fingers too. You obviously just have to be really gentle. To get it perfect, you need a truing caliper ($25-$35) but if you install the balance without the spring or pallet fork and watch it spin, you can probably get a good idea of where it needs to bend. A lot of the time it is a "twist" where one half of the split balance is bent down with the other half bent up.

When you assemble it, make an effort to get it close to in beat before you put the pallet fork in. Do that by sighting a line from the escape wheel jewel, through the pallet arbor jewels toward the balance staff. You want the impulse jewel to be perfectly centered between the banking pins along that sight line. I find getting into beat to be the most stressful thing, because it can mean adjusting the hairspring collet 5 or 6 or 10 times, each time risking damage.You will see some people hang the balance on a tack and reach through the hairspring with a small screwdriver to turn the collet a tiny bit. I no longer do that - it's too hard to see how much you're turning, and just really risky. Instead I undo the stud screw, take the balance off the cock and lay it on a bench block to make collet adjustments. It is easy to see then how far the stud moves in relation to the balance rim; the distance you will typically turn it to center the impulse jewel is really really small. 0.1mm and you may overshoot it.

For a first restoration, I would be totally happy with an impulse jewel that visually "looks" pretty centered. That is as good as an old-timer would have gotten it 100 years ago anyway, even if a timegrapher then tells me it has 2-3ms beat error it will run fine. It is possible to dial them in to 0.0ms beat error on the timegrapher, but it takes trial and error moving the collet back and forth until hitting it on the nose mostly by luck. Only worth it for a really nice watch.

Edited Sunday at 03:12 AM by mbwatch

[Nucejoe]

Good free oscilation in both DU & DD positions indicates good pivot on both sides of balance. 

 And you hear no sound of  balance rim rubbing indicates you have superhuman ears.

Excessive side shake can't make balance wheel wobble. Broken pivot and broken  jewel hole might wobble.

Possible faults are ,

Roller table rubbing on fork horn,

Balance rim grossly out of flat.

Balance rim rub on fork cock,

Guard pin out of adjust.

Regs

 

[docrpm]

22 hours ago, mbwatch said:

A lot of the time it is a "twist" where one half of the split balance is bent down with the other half bent up.

It’s pretty safe to say this is how the balance wheel is out of true. I’ve tried to correct this issue by hand, but I think my touch just might not be delicate enough. I might pick up one of those calipers, though I’m not quite sure how to pick from the many vintage pieces available on eBay.

I think I’m just going to get the balance spring back on there and then try the other pure rotation test suggested by @JohnR725  @nevenbekriev. I’m interested to see the results, though I’m dubious as to whether this balance will be up to the task.

If I can get it aligned well enough that it runs dial up and down, I’m willing to accept that as a “good enough” outcome. I’ll do my best to align the jewels how you’ve described @mbwatch, but honestly, this watch only has some sentimental value to my friend; it’s not valuable beyond that. Your advice, however, won’t be forgotten in future repairs where precision is more important.

20 hours ago, Nucejoe said:

And you hear no sound of  balance rim rubbing indicates you have superhuman ears.

Trust me – the balance wheel sounded like an old rickety bike rubbing against the center wheel. So the difference between noise and no noise was pretty obvious.

The wobble on this balance is because it’s out of flat…The pivots are fine (it’s a new balance staff) and the jewels are in good condition, as far as I can tell.

 

[mbwatch]

10 minutes ago, docrpm said:

I might pick up one of those calipers, though I’m not quite sure how to pick from the many vintage pieces available on eBay.

I think the type with a thumb screw in the middle where the two jaws stay parallel are the best, easiest and safest to use. If you can find one that comes with the little balance rim wrench, all the better though you can get by without it. A lot of them are incomplete so you have to make sure it has the adjustable pointed blade attached, that is placed just very close to the balance rim so you can judge by eye the comparatively high and low spots. But that piece is frequently missing. Levin made one that looks pretty much the same as the Vigor model pictured, that is really nice.

Other types that you open and close jaws without the thumb screw in the middle that open and close like scissors are a little harder to use. 

[nevenbekriev]

OK, I read the new posts diagonally...

Now You have to regulate the axial free play and true the balance.

The free play first: I would recommend to remove everything from the main plate and to put the balance cock only (no balance). But before that, if there are traces of 'repairs' like this

then smooth and flatten all the surface under the cock or the cock bottom surface beforehand.

Then fix the cock with it's screw and press it down on the stones area in order to bend it down a little. Then insert the balance and check the free play. You aim is to achieve what You call 'goldilocks' situation - there mus me almost no free play and the balance must turn absolutely free and all this without shimming. Of course, if the bending is too much, then You will use some lever to lift the cock or if the bending is not enough, repeat the pressing with more force.

Now, the truing. The truing is truing in plain and truing in round.

Truing in plain first.

The main problem in compensating balances is that the rim is cut. Thus it has more freedom to take improper shape than the rim of 'regular' balances. The first distortion usually happens right with punching the old staff out. This make the balance arms bend where the hole is and this immediately brings the two halves of the rim to two different plains. So, the first thing to do if such thing happens is to immediately straighten the arms. Never use punches and hammering, do it just with hands. Then, similar bending may happen when riveting the staff. Again, straighten with hands only and try to bring the rim halves to one plain. Now, the real truing.

I prefer to use this kind of tools.

The left one is what we call 'eight, 8' in my language. It has no parallel jaws and only steel conical cup bearings. The idea of this tool is only to hold the balance and let it turn free. It is not used for bending the rim and arms, but only to rotate the balance and help visually find lo and hi places. There must be a pointer there, but I do without it pretty well. When You know where the lo (or high) point is, then You grasp the balance arm possibly closer to the staff with the pliers shown and bend in order to make the point higher (or lower). 

You bend only the arm which You hold. Bending may be lifting or lowering up/down or twisting, depending where the point is. Twisting is only the arm, not the arc with screws. If the arc is twisted in it's length, then it has to be untwisted in order entire arc to be in one plain. Thus You work on the one half of the rim only and when it is ready, work on the other half. Finally, You must have balance rim that rotates true in one plain. The main idea of not using truing calipers but such pliers is never to force the joint between the staff and the rim where the staff is riveted. This may be in a manner makes the work more complicated, but prevents from creating other problems.The pliers shown are prepared from long nose small pliers like this,

which are usually sold in child/youth sets of 5. The jaws have to be annealed in order to be bent and shaped.

Then truing in round is to be checked. If every person that have worked in the past on this balance did His work well, there should be no issues here. But, if truing is needed, then first try to understand where the problem is. It is needed to know if the arcs are bent or not. There is no logical reason for the arks to be bent unless someone did this without knowing what they do. If they are not distorted, bringing the rim to round should be done just by bending right where the arcs are connected to the arms. Otherwise bending may be needed in different points of the arcs. For this purpose the tool that @mbwatch describes is usually used

Edited Monday at 08:51 AM by nevenbekriev

[mbwatch]

2 hours ago, nevenbekriev said:

The free play first: I would recommend to remove everything from the main plate and to put the balance cock only (no balance). But before that, if there are traces of 'repairs' like this

Confusion! @nevenbekriev this was not the OP's photo, it was my photo to demonstrate an example of why the balance shake could be faulty.

[nevenbekriev]

4 hours ago, mbwatch said:

Confusion! @nevenbekriev this was not the OP's photo, it was my photo to demonstrate an example of why the balance shake could be faulty.

Yes, that's why I said 'If there are traces...'

[docrpm]

Lots to consider here…With regards to the tool, I’ve now got two pieces of conflicting advice @mbwatch and @nevenbekriev: horizontal calipers or figure-8 calipers. Also, true with calipers, don’t true with calipers. It seems to me that I can get the horizontal calipers to spin the balance freely, even if I don’t use them to affect repair, so I think that’s the route I will go. I found a pair for $20 on eBay.

In terms of hand- v. caliper-truing, not sure what to do here…My hands were not precise enough before, although I think I may have been applying force in the wrong place (rim as opposed to arms).

Regarding the balance cock, there’s no sign of damage or tinkering on the base plate. Bending the cock seems like a really easy thing to mess up in an irreversible way. The benefit of a shim is that I can get it precise without any application of force. What’s the downside to shimming @nevenbekriev?

Oh and @nevenbekriev, your suspicion was pretty close in terms of what caused the wheel to come out of flat. I think the greatest harm was done removing the hairspring. The collet was on very firmly and I think I applied too much downward force to remove. Also, in the category of being a total noob, I think I tried to remove the hairspring with the balance resting on a piece of pith wood, as opposed to an anvil. Not smart, in hindsight. 

[nevenbekriev]

OK, now I see You believe the balance got out of flat while removing the hairspring. It is not probable, but not impossible. But no matter, You need to true You balance. After riveting the balance staff, this is normal to expect the balance will need truing. Actually, we don't have idea how bad the balance is out of true, we only know the balance touches the center wheel and thus we can suppose as one of the reasons for it is balance out of true.

You can use truing calipers of whatever type You have. The 8-type I prefer is easier to use as it takes no time to open and close it. It can not  be used for the truing itself, as the pivots will immediately break. The other type can be used for the truing as the pivots do not contact with the tool and are thus preserved, but the riveting of the staff this way is subjected to significant pressure and if it is not riveted so firmly, the joint can move and it is very bad when trying to true the balance as every little move there will strongly affect truing. If the staff is not riveted properly, then for sure removing or placing the hairspring can easily bring the rim out of true

We need to see some video that shows how the balance turns in the movement and how it is positioned against the central wheel.

Shimming the cock is OK, but You have to check end shake every time You put the cock in the movement. The way You put the foil can not be the same every time and the foil itself will get thinner and flattened with every tightening of the screw. Regulating by bending the cock gives reliable result that doesn't depend on chance. Why do You think that bending the cock will  be irreversible? Of course, You can decide which way to use, both are possible.

 

[LittleWatchShop]

7 minutes ago, nevenbekriev said:

You can use truing calipers of whatever type You have.

Do you have an opinion about Trupoise tool? Rare and no longer made. I have one.

[nevenbekriev]

7 minutes ago, LittleWatchShop said:

Do you have an opinion about Trupoise tool? Rare and no longer made. I have one.

I don't know what it is, can I see a picture?

[LittleWatchShop]

18 minutes ago, LittleWatchShop said:

Do you have an opinion about Trupoise tool? Rare and no longer made. I have one.

 

[nevenbekriev]

7 hours ago, LittleWatchShop said:

 

I tried to answer, but the answer is gone in that thread now...

 

Edited 23 hours ago by nevenbekriev

[docrpm]

@LittleWatchShop I've seen one of those Trupoise tools on the ChronoGlide YouTube channel...Pretty amazing tool, but I went looking for one, and there were none on offer. It strikes me as similar to one of those JKA Feintaster bench micrometers: really neat, really expensive.

@nevenbekriev Regarding shimming v. bending the balance cock, the permanence of the solution is something that I hadn't considered. It totally makes sense if one is going to be removing and replacing the balance cock more than once or twice. I'm just not sure (a) how I would go about bending and (b) how I wouldn't bend it too much. It seems like an approach best left to the professionals. 

[nevenbekriev]

23 minutes ago, docrpm said:

@nevenbekriev Regarding shimming v. bending the balance cock, the permanence of the solution is something that I hadn't considered. It totally makes sense if one is going to be removing and replacing the balance cock more than once or twice. I'm just not sure (a) how I would go about bending and (b) how I wouldn't bend it too much. It seems like an approach best left to the professionals. 

Yes, I understand. But one has to try, only this way the fear of the unknown disappears. I promise it will be not so hard to do and of course, if bending is too much, then lift it with some lever (bend it up) and with several iterations everything will be perfect. Never try to lift or bend down when the balance is in place, this is the important rule. If you have some useless movement, try and exercise on it.

[LittleWatchShop]

12 hours ago, nevenbekriev said:

I tried to answer, but the answer is gone in that thread now...

 

The clamping is brilliant. The jewels holding the pivot are spring loaded and retract as you tighten on to the arms.

[Nucejoe]

On 4/13/2025 at 5:20 AM, docrpm said:

 

1. Fold over a small piece of tin foil multiple times

• Goldilocks thickness:

 

 Your approach limits your adjusting  to integer number of folds. 

One fold 

Two folds,     ......so on.

You don't have 1.3     or     1.3758 folds.

 Just insert one fold under the cock,  then moving the foil with respect to cock screw, gives you infinite slopes of the cock, and  shoot for the  smallest end shake. 

In practice, smallest end shake is, one that,

if cock screw is  99% tightened !!   balance is not free to turn.

But 

When fully  100% tightened, balance frees up to turn. 

It is a very practical approach.

Mark teaches it in one of his videos. 

Regs

   

 

 

[SwissSeiko]

Get a piece of brass plate stock and mark on it where the indicating pins on the balance cock are, and drill them out. Then mark out where the screw would go, then drill and tap it. Then do as nev said, and bend the cock gently. In my opinion, shimming is an unacceptable repair, as its only a timesaving method. Too many bad previous repairs I've seen, I'd rather do it right the first time. Maybe that's why I have the clients watch now, and not the repairman who shimmed the balance.

[JohnR725]

3 hours ago, nevenbekriev said:

then lift it with some lever (

Personally I prefer to bend the balance bridge when it's out of the watch. Although typically I'm doing American pocket watches and it actually takes quite a lot of force to bend the bridge. But if you would like a lever and it's a wristwatch or something smaller than a pocket watch the Swiss have you covered as they have a lever or two.

3 hours ago, docrpm said:

Trupoise tools on the ChronoGlide YouTube channel

I remember somebody pointed this out to me and I looked at the tool and can't remember why but I was not impressed.

 

3 hours ago, docrpm said:

egarding shimming v. bending the balance cock, the permanence of the solution is something that I hadn't considered. It totally makes sense if one is going to be removing and replacing the balance cock more than once or twice. I'm just not sure (a) how I would go about bending and (b) how I wouldn't bend it too much. It seems like an approach best left to the professionals. 

I suppose the real question is why do we either method? In other words the watch left the factory without a shim unless of course it's a Russian watch and then it definitely had one plus a few other companies also had them but typically the watch doesn't need them at all. But with anything of vintage things have happened things have well things aren't always where there supposed to be. Then people change balance staffs and don't realize that pivots can be shortened. I saw a picture of a Russian balance staff and noticed they have extremely long pivots but apparently nobody has a lathe Or fail to grasp why the pivots are so long and shimming. But it would really be best if you didn't have to.

Then for doing any of the things like we talk about you really should be practicing on a practice watch. Or even pieces of watches you don't even need a complete watch you just need the balance wheel on the bridge in the main plate see them practice bending things and see how much worse it takes to break etc. As everything requires practice and practicing on a watch you want to keep when you have no experience isn't necessarily what you want to do as it will lead to probably what were seeing where things are going to get much much worse before they hopefully get better hopefully

Then I'm attaching a PDF on balance wheel truing. If you work on American pocket watches with bimetallic balance wheels your destiny will be to always have to true them. Typically there almost always out of flat or they been squeezed which has undesirable  timing consequences and really screws up the poise. Plus typically American pocket watches have just had a lot of undesirable repairers over the years.

Joseph Bulova School of Watch Making - Unit 2 balance wheel truing.pdf

[Nucejoe]

 Disagree with Neven and JohnR, bending the bridge or cock is simply bad watch repair.  

  Would you bend  a cock in patek Phillip or audemars piguet watch?   

 Good watch repair is one that conforms to exact factory specification. 

 

 

 

 

 

 

 

 

[JohnR725]

3 minutes ago, Nucejoe said:

Good watch repair is one that conforms to exact factory specification. 

Exact factory specifications does not require a shim.

Bending of the balance bridge is only required if prior idiots bent the bridge for you and you need to restore it back to where it's supposed to be. Which unfortunately is quite common on American pocket watch repair.

Oh and what about the Swiss tool that I mentioned above why would the Swiss have a tool? Or even more important why would a school  like wostep use such a evil barbaric tool as that is actually how I know about the tools existence there must've been a reason why?

So yes in absolute perfect world you would never bend the bridge. You would also never use a shim as typically that's a sign of poor manufacturing skills ideally you would adjust the balance pivots and do a proper repair. Unless of course the repair requires you to restore the bridge back to where it was supposed to be in the first place.