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

[docrpm]

Hi everyone,

I'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.

The balance cap jewel looks fine, as does the jewel at the base of the balance staff. The bottom pivot of the balance staff also seems like it's intact, and the balance moves very freely when I'm working on it face down. If I set it to oscillating, it'll spin for 30-60 seconds even with a powered-down mainspring. However, I removed the wheel and mainspring from the balance cock, and put it under the microscope. The top pivot of the balance staff looks like it's worn, but it's hard to tell without an undamaged one for comparison.

I was hoping there would be some opinions here as to whether (a) I'm on the right track in terms of diagnosis and (b) whether this pivot is in fact worn and I've got an end-shake problem.

Thanks in advance for any thoughts / advice!!

[nevenbekriev]

Hi there,

Yes the pivots seem to be a little worn, and the picture of the hairspring  side pivot doesn't help much to see if the pivot is long enough. It is short, then it's tip may be doesn't reach the cap stone and the coned part rests on the hole stone, thus causing friction. Is the cock cap stone in good order? Did You remove the settings for cleaning? You can remove the hairspring and put the balance in place with no pallet fork to check freedom of it's rotation. It must rotate long tome and stop very very gradually. The same manner in dial up and dial down positions.

[docrpm]

Thanks very much for your thoughts @nevenbekriev...I tried to get a better picture of the side pivot, but it was tricky. I've zoomed in a bit in the attached. I'll need to use my microscope to take a closer look. The cap stone looks to be in great shape (see third picture above).

I was thinking about pulling off the hairspring and seeing how freely it spins...I can give that a shot. I'm tempted to just replace the balance staff.

[JohnR725]

2 hours ago, nevenbekriev said:

Yes the pivots seem to be a little worn

Yes the pivots look a little bit worn. This is where it's really nice to have a timing machine because they work really good at showing Pivot problems. Then yes it's very common to have to refinish the pivots on anything vintage as they will wear with time.

Then there's the other peculiar thing? Notice in your image that I snipped out below the color of the jewels is definitely a concern. The upper jewel is a replacement. That is because it's a modern red color the lower jewel is an original jewel or at least an original color. So if someone has replaced the jewel did they replace with the right diameter? Plus as they replace the jewel was it cracked or broken and that would be bad for the balance staff because crack jewels are not good for pivots as they tend to disintegrate.

Oh and when you're looking at jewels like this it's also good to remove the end stone and check because often times there could be a crack chip flaw or something that you can't see from the side. In addition to having lighting on the other side helps to see cracks as opposed to just looking at the jewel you really need to look through the jewel.

1 hour ago, docrpm said:

I'm tempted to just replace the balance staff.

Just a couple thoughts on replacing the balance staff. Yes you have the grade number of 315 but there is some minor problems typical of American pocket watches. Like which balance staff would you like to use? My favorite place to look up Elgin part numbers is at the link below. Then some of the problems would affect the replacement balance jewel for those people who did not pay attention. So you'll notice that there are two separate balance staffs and a reference to you have to look at something and a serial number. In other words you need the parts book to look at the section on balance staffs. Which typically is still missing a piece of information.

Okay so one balance staff is 1905 but it appears to come in two variations

Then the other staff would be to 2802 and in the fine print is a reference to double roller which is what you have.

Then looking at another reference we can see that the 2802 is for the double roller and the other one is for the single roller that you don't have. So we've narrowed it down except there is the problem that would affect the balance jewel which is one part number and how many different pivots sizes? This is another interesting problem with American pocket watches typically for balance staffs they will come in different pivots sizes not always listed and of course you have to have the right jewel that correspond to the diameter of the pivot.

http://www.elginwatchparts.com/

 

 

 

 

[docrpm]

Great info @JohnR725. Thanks for all the detail.

I actually DO have a timing machine. What should I be looking for? I put this watch on there awhile ago, but don't recall anything jump out at me other than low amplitude, which I attributed to a tired mainspring.

Your comment on the cap jewel is very insightful. NO idea if this watch was ever serviced. I'll have to take a close look at the case for service marks. I didn't try to take the cap jewel out, but I can definitely take a closer look there.

For better or worse, I already bought two balance staffs. I bought the 2802 based on referring to the parts listed in the Pocketwatch Database. Hopefully that wasn't in error. If the cap jewel is not the right size, maybe even if I got the right part, it wouldn't work properly. But I can test without the balance spring and see how well it spins, right?

[JohnR725]

28 minutes ago, docrpm said:

What should I be looking for?

When you're looking for pivot issues the most common place where they show up would be dial-up and dial down amplitude differences. If the watch hasn't been cleaned often times it could be just bad lubrication. Otherwise when your watches clean and properly lubricated bad pivots will show up as amplitude differences.

30 minutes ago, docrpm said:

I didn't try to take the cap jewel out

I think where having a terminology confusion here? There are actually two separate jewels on each side. There is the hole jewel and the separate cap jewel.

Let's look at the parts list which gets even more confusing as usual. Notice there is a X in front of some of the parts which I believe usually indicates conceivably a variation of that part. So we see that part 453 is the upper hole jewel and it's also the lower jewel except. Then we see 4806 is the lower jewel with a reference to note 323. This is where we would need to know the serial number because they change the jewels based on a serial number. But I may go find a reference and see what the difference is. You also notice that the upper and lower cap jewels are different typically in the one on the main plate is thinner because it doesn't have all the regulator stuff. Then a lot of times on the balance bridge it might even be a separate color like gold is used because it looks Tyson pretty usually found on higher grade watches.

So as we can see below different hole jewels. Toll diameter is the same which is very common. Next number is the total thickness and the third number is the shoulder thickness. Then of course you'll notice the different diameters for the pivots. Which does not show up as part of the part number or even referenced in the part number. I have seen somewhere in the Elgin catalog which said if you ordered a dozen balance staffs and you did not specify a pivots size they will actually send you an assortment of different pivots sizes.

45 minutes ago, docrpm said:

NO idea if this watch was ever serviced. I'll have to take a close look at the case for service marks.

Several problems with this approach of determining whether the watches ever been serviced. First many times American pocket watches get re-cased and the new case would not have any prior markings. Then some shops don't mark the case at all. So you have no idea if anyone had serviced if they didn't mark. With American pocket watches it makes it easy to keep track of because the movements of serial numbers.

47 minutes ago, docrpm said:

But I can test without the balance spring and see how well it spins, right?

You really do not want to discover you have a problem at this point in time. In other words if your pivot is physically too big for the jewel the best time to deal with that is before it's in the balance wheel. Replacing a balance staff is a very procedural operation. So for instance if you are replacing a balance staff you would measure the old staff just to make sure the new staff you order is the right staff and you also measure the new staff when it comes to make sure it actually corresponds to the dimensions you think it is supposed to be as unfortunately often times there can be variations. It is best to find out if your staff is not sized right before you start attaching things and bad things will happen conceivably like if it's slightly too big lots of things can get damaged or broken trying to force things on the staff is not quite right

 

[docrpm]

3 hours ago, JohnR725 said:

When you're looking for pivot issues the most common place where they show up would be dial-up and dial down amplitude differences.

That's helpful...I'll check that out with a few of my other running watches (e.g., Hamilton 974). Sadly, with the Elgin, the watch completely stops in one orientation, so I can't really do much testing in that regard.

 

3 hours ago, JohnR725 said:

I think where having a terminology confusion here? There are actually two separate jewels on each side. There is the hole jewel and the separate cap jewel.

Yes...A misunderstanding on my part. Apologies. I've not yet disassembled a balance, since I didn't have any of the proper tools. I have a staking set arriving in a few days. Also, I've been trying to avoid doing any jewel work, since I don't have a jeweling tool either. I suspect that purchase might be in my future.

 

3 hours ago, JohnR725 said:

So as we can see below different hole jewels. Toll diameter is the same which is very common. Next number is the total thickness and the third number is the shoulder thickness. Then of course you'll notice the different diameters for the pivots. Which does not show up as part of the part number or even referenced in the part number.

So I'm not quite sure what to do here. Should I be using a micrometer to measure the diameter of my pivots? I suspect mine might only be accurate to 0.1mm. I'm not even sure how I'd go about measuring the jewel holes. I'll certainly check my balances when they arrive, to the best of my ability (and the accuracy of my tools).

It's interesting to learn the degree to which, even if you have the grade of a watch, a part is not a part. I had seen with my Hamiltons that depending on serial number they switch between single- and double-rollers. But the old parts catalog I have makes no indications about serial-number variation for other parts.

 

 

[docrpm]

3 hours ago, JohnR725 said:

You really do not want to discover you have a problem at this point in time. In other words if your pivot is physically too big for the jewel the best time to deal with that is before it's in the balance wheel.

I forgot to respond to this bit...It makes total sense. I'll give it my best go to check measurements out before I try the staff in the balance. I can see how it could spiral quickly if the pivot was too large for the jewel at the top of the balance (which is the replacement jewel, as you kindly pointed out).

[JohnR725]

2 hours ago, docrpm said:

But the old parts catalog I have makes no indications about serial-number variation for other parts.

We have unknown variables. For instance I did a project for somebody who used a vast quantity of 12 size Illinois watches. The particular watch was picked because the parts catalog indicated no variations and this watch was made over 30 years roughly. The problem is over 30 years a manufacturing there were variations lots of them. But the parts catalog as far as I remember makes no reference at all to the variations. Typically the older pocket watches will have more variations as they improved and then of course changed things

even modern watches will have parts changes but typically the changes are not in the parts lists. Oftentimes the watch companies will have supplemental information that the watchmakers typically never see which has the updated parts.

2 hours ago, docrpm said:

Sadly, with the Elgin, the watch completely stops in one orientation, so I can't really do much testing in that regard.

One of the things I always like to ask is what is the history of the watch so in other words did the problem occur before you repaired the watch or after? If it's before conceivably somebody was having an issue and just gave up on the watch.

2 hours ago, docrpm said:

I've been trying to avoid doing any jewel work, since I don't have a jeweling tool either. I suspect that purchase might be in my future.

Watch repair is filled with lots and lots and lots of variations. For instance American balance wheel jewels typically held in with two screws. Very tiny screws that oftentimes get stripped. So to disassemble you just have to remove the two screws and push the jewels out. You don't actually need a special jeweling tool although it's nice to have the tool for pushing but you could use peg wood.

For instance here's a picture of the jewels from an 18 size pocket watch we can see the screws that hold them in and the jewels themselves in their brass settings.

Then this is what the tools for pushing the jewels look like. But you could use anything that's round just a push on the edge of the settings preferably not your tweezers as you leave marks on the brass depending upon how hard you have to push. In other words the friction then there supposed to push out by hand you don't need any tools other than something to push on them

3 hours ago, docrpm said:

Should I be using a micrometer to measure the diameter of my pivots? I suspect mine might only be accurate to 0.1mm.

That would be a pretty poor micrometer wouldn't it that can only measure to 1/10 of a millimeter. Micrometers are better than veneer calipers then they do make special watchmakers bench micrometers which would be the best but they're expensive. You do have to be careful in measuring pivots though that you don't squeeze too much and crush the pivots or break it off of your not careful.

For instance here's a picture of a micrometer. Notice it breaks the millimeter up into basically one hundredths. With those big numbers being a 10th of a millimeter.

Then if you like buying tools you can buy something like this a plate with lots of jewels and all you do is stick your balance vivid in the find out what size it is

Then you could always buy something like this if you can find one. Jules on the top to see the diameter of your pivots and the lower part our pivots to try In the jewel although they have a habit to getting broken off

Or there is the other way you don't need any fancy tools you just need the watch and the staff. In other words you see if the staff fits in the jewels. You can also when you have your jewel out see if the staff pivot will fit in the whole of the jewel.

 

 

[nevenbekriev]

Pretty good and helpful conversation till here. Just few things to say

OP, before You decide to replace the staff, please show good pivots side pictures. Replacing the staff - are You ready for this? 

In order to take pivots side pictures, You will need to disable 'autofocus' (to lock the focus). In most phone cameras, touching area of the screen with finger activates focusing on this area, and holding the finger for several seconds locks the focus (symbol with locked keypad appears). Once the focus is locked, move the camera to/from the pivot until the image of it is on focus.

As John said, the upper side pivot has to be checked how free it gets in the hole stone. But not only - the other thing is the pivot tip must be able to pop a certain amount above the upper surface of the setting. Thus it will be sure that the tip rests on the cap stone and in the same time there is healthy free play of it's cylindrical part in hole stone.

[docrpm]

Great discussion with lots of insight. 
 

A few points:

1. I should have been clearer that I don’t have an actual micrometer. I have vernier calipers, which are a poor substitute. Maybe there’s a Mitutoya in my future. 
 

2. These pictures have all been taken with a digital microscope. The central challenge was trying to hold the balance vertically to get a side view of the pivot. I’ll try again. 
 

3. It’s unclear to me how I can assess the play of the pivot when it’s inside the cap jewel and supposedly touching the end stone. Isn’t the pivot obscured by the jewel under those circumstances?

4. Should I remove and inspect the balance jewels? Is pegwood adequate to get them back in once I get them out? (Assuming those screws aren’t stripped)

Thanks for all of your patience with my rudimentary questions.

PS: I’m actually kind of excited to replace a balance staff. I know what’s involved. But I’ll exhaust these other avenues first. 

[mbwatch]

12 hours ago, JohnR725 said:

Then if you like buying tools you can buy something like this a plate with lots of jewels and all you do is stick your balance vivid in the find out what size it is

These tools, sadly, now cost more than a new watchmaker's bench micrometer.

 

32 minutes ago, docrpm said:

Should I remove and inspect the balance jewels? Is pegwood adequate to get them back in once I get them out? (Assuming those screws aren’t stripped)

Yes, if you have not removed them to clean and inspect and oil, you definitely should do this. You should be able to push them in with pegwood, making sure you have formed the pegwood so it sits flat on the brass setting rather than push on the ruby.

And if the screws are stripped, they will come out just fine usually just with rodico. The threads are so shallow that there is nothing really holding them in. Putting them back is the bigger challenge (and I sometimes use a tiny bit of blue loctite, but only after I am done with the whole watch and know I won't remove them again). I find a stripped jewel screw on probably 1 out of every 3 watches I take apart. At least.

[nevenbekriev]

Ok, several things more.

There is no difference between cap stone and end stone. They are two names for one thing. The thing is the stone with no hole, with one flat polished side.

You for sure must remove the jewel settings from the cock. Otherwise no way to check what is needed to check.

You don't need micrometer. Probably newer will need one. At this point, You need to check if the pivot can get in the pivot hole of the hole jewel in a manner that the pivot top is at least 0.15 mm above the upper surface of the stone. No matter if it is the old staff or the new one.

 

[docrpm]

My apologies if I'm still getting the terminology wrong. I'm trying to differentiate between the flat jewel with no hole, and the one beneath it with a hole. The pivot goes through the latter, obviously, and then abuts the former.

28 minutes ago, nevenbekriev said:

You don't need micrometer. Probably newer will need one. At this point, You need to check if the pivot can get in the pivot hole of the hole jewel in a manner that the pivot top is at least 0.15 mm above the upper surface of the stone. No matter if it is the old staff or the new one.

Any recommendations on how to measure the desired 0.15mm without a micrometer?

I think there are a few big takeaways for me at this point:

• Pivots really need to match their corresponding jewel holes to within some small tolerance. Too small is bad and so is too big.
• Jewel variation is a real thing...you can't go based off specs, especially if your jewels have been replaced (possibly with the wrong ones)
• Measurement tools for pivots are very expensive (e.g., Seitz jewel gauges). Based on some additional outside reading, it seems like the use of micrometers to measure pivots is discouraged because it's too easy to cause damage. Plus, bench micrometers are too expensive anyway.
• BOTTOM LINE: Test the pivots in the actual jewels in which they're supposed to be seated 

How well would the following work:

1. Mount the pivot vertically in some pith wood in the same orientation it would have looking top down on the balance cock
2. Put the hole jewel on top of the pivot and check to see if (1) it fits on the pivot easily and (2) the pivot is sticking out the top an adequate amount so that it will abut the cap stone (0.15mm?)

If the above are true, and the pivot also fits in the bottom jewel, then I could safely mount the staff in the balance wheel, without mainspring, attach the balance, then test to see how freely the wheel spins.

Is that a workable plan? Also, as an aside, assuming I would use Moebius 9010 as a lubricant for these particular jewels?

[mbwatch]

10 minutes ago, docrpm said:

How well would the following work:

1. Mount the pivot vertically in some pith wood in the same orientation it would have looking top down on the balance cock
2. Put the hole jewel on top of the pivot and check to see if (1) it fits on the pivot easily and (2) the pivot is sticking out the top an adequate amount so that it will abut the cap stone (0.15mm?)

Yes this should work. The 0.15mm you need to judge by eye if you can. It takes experience and I am sure I would not guess accurately beyond a tolerance of +/-0.1mm. A wizard like Nev sure would get it on the nose though.

12 minutes ago, docrpm said:

If the above are true, and the pivot also fits in the bottom jewel, then I could safely mount the staff in the balance wheel, without mainspring, attach the balance, then test to see how freely the wheel spins.

Yes.

12 minutes ago, docrpm said:

assuming I would use Moebius 9010 as a lubricant for these particular jewels?

Yes you can use Moebius 9010 for these. I believe JohnR has a preference for 9020 on pocket watches and I would definitely defer if he says No to 9010, given his experience with American watches, but I have good success with 9010.

[docrpm]

10 minutes ago, mbwatch said:

Yes this should work. The 0.15mm you need to judge by eye if you can.

Great! I like to have a working plan that I can execute with the tools at my disposal (though I really do love cool tools, and may acquire more in the future).

Regarding the 0.15mm, I'm certain I'll be able to judge the difference between 0 and something (i.e., if the pivot at least sticks out of the hole jewel). Beyond that, no guarantees.  

The great thing is, I have one donor movement with a pivot that's heavily worn if not just sheared off on the top end, so I'll be able to try things out with that. I might even be able to cannibalize the hole jewel if it's original.

 

13 minutes ago, mbwatch said:

Yes you can use Moebius 9010 for these. I believe JohnR has a preference for 9020

Sincerely praying that I don't have to buy more Moebius lubricants.

[docrpm]

I pulled off the jewel(s) in the balance cock and grabbed a few pictures...Interesting results (to me, at least).

As you can see in the attached, the jewel(s) appear to be in good shape. I looked in direct light and on top of a light table, and couldn't see any cracks or signs of wear. The interesting thing was that the hole jewel and end jewel seemed to be fused together...At least, I was unable to get them apart. Do these jewels ever come as a single assembly, or are they just stuck?

[docrpm]

I also pulled the jewels from my donor movement, and those appear to be (a) original (they're almost clear) and (b) slightly worn. The hole jewel looks ok, but the cap jewel seems to show some signs of wear (though it's not cracked). 

I also got a nice close-up of the pivots on that balance for comparison. The upper pivot on this balance staff does appear to be a bit longer. The lower pivot is pretty much gone. This is the staff I'll replace as a test of my 2802 part and new staking set.

[JohnR725]

6 hours ago, mbwatch said:

Yes you can use Moebius 9010 for these. I believe JohnR has a preference for 9020 on pocket watches and I would definitely defer if he says No to 9010, given his experience with American watches, but I have good success with 9010.

Yes 9010 is the universal light oil found on balance staffs. Yes it really is the universal balance jewel oil with the exception of the Swiss company Piaget Uses 9020 on its balance pivots except their smallest caliber which uses 9010. Then there's me I use 9020 on the balance pivots and most of the other pivots. As far as I can tell I've never seen the loss of performance and yes if you really wanted to know I would explain a technical reason why like 9020 but that would require a whole bunch of paragraphs not related to the current subject. So the simplistic is you're going to use 9010 to fit in with the rest of the world.

6 hours ago, docrpm said:

• Mount the pivot vertically in some pith wood in the same orientation it would have looking top down on the balance cock
• Put the hole jewel on top of the pivot and check to see if (1) it fits on the pivot easily and (2) the pivot is sticking out the top an adequate amount so that it will abut the cap stone (0.15mm?)

So for instance here's a slightly misleading image because the jewel is basically floating in place in real life it would drop down until it touches the curved part of the staff with quite a bit of the pivot sticking out.  

Did you notice the image I had up above? Let's look at part of it again. Notice that the end shake is controlled by the end stones not the hole jewels. Notice how in the example below the pivot does not stick up as that's the spacing of your end shake.

Here is a another example of balance pivots in the jewels. So basically you need the side clearance so the pivot can spin and you need a little bit of end shake at the end.

 

Oh and this does bring up the peculiar problem of? I remember when I was in school and I replaced a probably balance staff and I brought it up to the instructor minus the end stone. I thought this way he could see how things looked and I was basically sent back to my bench to install the end stone. Because you cannot check end shake by how much is sticking up only with the end stone in place can you check. In other words you just gently lift up and see if the balance wheel has a little bit a play up and down.

7 hours ago, docrpm said:

If the above are true, and the pivot also fits in the bottom jewel, then I could safely mount the staff in the balance wheel, without mainspring, attach the balance, then test to see how freely the wheel spins.

Another way to do this would be to remove the pallet fork as that's not needed at this time. Balance wheel with its new balance staff without hairspring. Then with just a puff of air the balance wheel should spin very nicely. Then with the balance wheel spinning you can pick up the watch and just verify that it continues to run as you rotate the movement around.

[caseback]

8 hours ago, docrpm said:

Any recommendations on how to measure the desired 0.15mm without a micrometer?

A set of feeler gauges might be useful to be used as a reference. I've used them as a visual reference when doing lathe work as well. Shouldn't cost more than a few $'s.

[Klassiker]

Apologies for the late contribution, but could you check the lower pivot again? That's the one that the balance is sitting on dial down, i.e. when the watch runs. It could just be the lighting, but to me it looks bent as well as worn.

[GPrideaux]

8 hours ago, docrpm said:

the hole jewel and end jewel seemed to be fused together...At least, I was unable to get them apart. Do these jewels ever come as a single assembly, or are they just stuck?

Probably just stuck with old, dried up lubricant. They'll probably come apart after a few seconds (or minutes) in solvent.

If that IS the case, it's likely a major contributor to your balance not moving freely.

Edited February 7 by GPrideaux

[mbwatch]

3 hours ago, GPrideaux said:

Probably just stuck with old, dried up lubricant. They'll probably come apart after a few seconds (or minutes) in solvent.

Yep. And if the solvent doesn't separate them, you can try a tiny bud of rodico on each side then pull apart. I don't think I've ever had to do this with pocket watch jewel settings, only with shock setting chatons. These mate at flat surfaces so there is nothing to hold them together other than lubricant. No friction.

[docrpm]

Lots of great stuff to respond to. Thanks everyone. 

@JohnR725Those diagrams are extremely helpful. Now I see how the end stone will control end shake. 

I think I’m going to let those jewels sit in some solvent for awhile, pull them apart, add some 9010, then reattach my balance sans hairspring. Then I’ll take out the pallet fork, install stripped balance, and see how well it spins. 

[docrpm]

Well, the fused assembly has been sitting in 99% IPA for two hours now, and no luck. Next up: paint thinner. Given how tightly bound they are, not sure if Rodico would have enough adhesion, even though it is the magical watch-repair substance.