Full plate watch, Suitable tool for adjusting & testing pallet stones?

[mbwatch]

I find myself in a frustrating situation, and if I have learned anything yet it's that there might be a specific old tool to get me out of it.

I am rebuilding an early grade Elgin 18 size full plate movement, just restaffed and installed a new roller jewel. Next I found that the pallet stones were loose and added shellac to them, reassembled but the watch does not run. Observing the escapement through the holes, it is clear the entry stone needs to move in slightly. So I know what I need to do at this stage, but I fully expect to need a few adjustments of one or both stones. I don't want to move the banking pins because it did want to tick before I noticed the pallets were not secure.

Question is, on a full plate movement this is incredibly inconvenient to make a change, let the shellac harden, and reassemble the entire full plate watch in order to observe the escapement action.  It can only be seen from the dial side through the small holes in the pillar plate, so even the balance must be installed in order to manually operate it from the back. Anyone know how this was done 120 years ago? Is there a tool or fixture that can be used for the job outside the watch? Depthing tool maybe?

You all like pictures, so here are some. Final cleaning has not happened yet, thus the dirt and oil on the 4th wheel jewel.

[Randy55]

Afraid this is where it could get down to old school adjustment and observation.

There are pallet setting tools..but the calibrated ones aren't cheap.

The observation holes in the plate will allow you to see if you have one or both stones hanging up/ not engaging on a tooth, That would ofcourse require you to disassemble, heat/move the jewel in/out as required..by eyesight and experience. It's truly trial and error until you get close enough that you could make small adjustments with the banking pins until they drop to the lock freely.

You can use a Potence tool to help hold the rok in place during reassmbly.

Here's a pic of one...made with a piece of old 18s mainspring

You just fold over a roughly 2 inch piece, then file a slot on the end that engages the pallet fork arbor

Edited April 27 by Randy55

[mbwatch]

1 hour ago, Randy55 said:

Here's a pic of one...made with a piece of old 18s mainspring

Oh that's neat. I typically use a thin strap of rodico laid over the pallet fork to affix it to the upper plate but it can be hard to retrieve sometimes.

And as I did more research, I may have just found your own earlier explanation: https://learntimeonline.com/forums/topic/potence-clip-for-full-plate-p-ws/

[Mercurial]

There is a tool that was made for setting up and adjusting escapements of full plate watches. 

There were two styles, the picture below shows both of them. 

The lower tool held a movement plate and the vertical pointed rods were adjusted to hold the unsupported pivots of the lever and escape wheel. 

There was also a version of this tool that had 3 adjustable safety centres so that the balance pivot could be supported by the tool : 

The other version I’m aware of is the Boynton’s Escapement Matching and Examining Tool came as a set of two or three clamps that gripped the watch plate and held the safety centres for the pivots :

These do turn up on eBay from time to time. 

For some escapement work, you can set up the parts in a regular depthing tool, with the centres set according to the distance between the corresponding pivot holes on the movement. 

I hope this helps,

Mark

[nevenbekriev]

Yes, the specific old tools do exist, but may be having one is not needed as they are not cheap, and also You will be able to do without it well enough.

My advice will be to use regular depthing tool and adjust it for the exact distance between pallet fork and escape wheel bearings from the watch. Then remove the shellac from the pallet that now doesn't pass the ew teeth and move this pallet in. Then put the pallet fork and ew on the depthing tool and check how they lock. They should not lock when the pallet is in, but You will little by little move the pallet out and locking will appear. Then move just an idea out for reliable work and apply shellac, then check if things are still the same. You have to observe where the teeth fall on the pallets - it must be just a little below the edge between impulse and rest planes. Then You must check how everything behaves in the movement

13 hours ago, Randy55 said:

Afraid this is where it could get down to old school adjustment and observation.

There are pallet setting tools..but the calibrated ones aren't cheap.

The observation holes in the plate will allow you to see if you have one or both stones hanging up/ not engaging on a tooth, That would ofcourse require you to disassemble, heat/move the jewel in/out as required..by eyesight and experience. It's truly trial and error until you get close enough that you could make small adjustments with the banking pins until they drop to the lock freely.

You can use a Potence tool to help hold the rok in place during reassmbly.

Here's a pic of one...made with a piece of old 18s mainspring

You just fold over a roughly 2 inch piece, then file a slot on the end that engages the pallet fork arbor

This Potence tool is so ingenious, but actually, the traditional way to do the things is much more simple. Arrange the parts not on the pillar plate, but on the cover plate. Only the central wheel will remain on the pillar plate, secured by the cannon pinion.

Edited April 28 by nevenbekriev

[JohnR725]

19 hours ago, mbwatch said:

I don't want to move the banking pins

Interesting philosophy I like that but what if, I know this is extremely unlikely after all it's not like they're easy to move what if somebody moved them before you worked on the watch?

[mbwatch]

7 hours ago, Mercurial said:

These do turn up on eBay from time to time. 

For some escapement work, you can set up the parts in a regular depthing tool, with the centres set according to the distance between the corresponding pivot holes on the movement. 

I actually saw this Boynton tool on ebay just a few days ago but had no idea what it was for. Thank you - I'll hunt for it again.

 

1 hour ago, JohnR725 said:

Interesting philosophy I like that but what if, I know this is extremely unlikely after all it's not like they're easy to move what if somebody moved them before you worked on the watch?

Well before I shellacked the stones, the watch did tick long enough for them to become dislodged and long enough that the impulse jewel fell out. So I want to get back to that point first.

[JohnR725]

37 minutes ago, mbwatch said:

So I want to get back to that point first.

in other words you would like the watch to run before doing anything. Yes a full plate watch that's either not running or not running right is not really a fun watch to work with. 

 just a reminder for the future it seems like everyone wants to play with the banking pins and there only for horn clearance. Then once you verified there were there supposed to be then you then you can check the guard pin. then once those things are verified and where there supposed to be then you can play with your pallet stones.

[Mercurial]

42 minutes ago, mbwatch said:

I actually saw this Boynton tool on ebay just a few days ago but had no idea what it was for. Thank you - I'll hunt for it again.

There is one Boynton tool on eBay right now. 

They were sold in pairs or groups of three. One isn’t that helpful alone. 

Many on eBay are incomplete, so check them carefully. The safety centres that hold the pivot must be intact and not blocked by a broken pivot, ask for pictures if needbe to verify this.

They were also made in a few different sizes, so a given set may not be suitable for the movement you want to use it with. 

With patience, I managed to get a matching set of 3 quite cheaply, they sell for much less than the other tool and are often poorly described, and therefore attract few bidders. 

Best Regards,

Mark

[mbwatch]

2 hours ago, JohnR725 said:

in other words you would like the watch to run before doing anything. Yes a full plate watch that's either not running or not running right is not really a fun watch to work with. 

Yes, I know it was able to run before just enough that everything supposed to be affixed with shellac could fall out. I won't get back to work on this watch for several days at least, but one thing you mentioned about the banking pin adjustments being intended for fork horn clearance got me wondering....

The last adjustment I made to the entry stone yesterday did not solve the problem when the balance is installed. The escape teeth can't quite clear the stone when the balance swings. But I thought it was going to work because manually snapping the fork back and forth did allow the escape wheel to advance almost all the time. I wonder if there's another issue with the impulse jewel, that I have to shellac again anyway. I am going to measure the fork slot and the jewel because I am curious if it is the right size or possibly narrower than it should be.

My thinking is if the horn clearance is too great maybe it is not getting enough impulse to clear the escape tooth. But pushing the fork with a tool can move it far enough to clear the escape wheel tooth.  I'll see later in the week. I didn't expect a big challenge from this watch because it ticked for 5 minutes before stopping without shellac. As always I was way wrong.

[JohnR725]

Escapement adjusting always interesting and depending upon the reference always confusing. Okay maybe it's not always confusing but it does lead to confusion.

I have a PDF below it's actually a whole bunch of separate stuff including a hand out that came from a lecture that's on you tube. Then from that we get this image

Consequences of doing things especially if you do things out of order or you do things for the wrong reason. Oh and even if the watches working I made the mistake one so showing my boss how tweaking the banking pins on a full plate on the timing machine made the amplitude get better and now he thinks that's what they're for and I don't think a fully grasped exactly what horn clearance means. Consequence of doing things.

Notice what it says about opening and closing the banking pins and total lock? So yes I've had that on a full plate where it won't unlock at all and that's the banking pins or a combination of things basically.

So banking pins unfortunately get moved. One of the ways to tell if it's been moved is the look straight down at the end of the fork with the balance wheel removed. Power on the fork push at the one side look at it push it to the other side also look at it and compare anything with the center reference the balance jewel and see if both sides of the same. No guarantee after the same there in the right place but at least are the same typically when people play with things one side will be way off from the other because they had no idea what they were doing at all because of course it's a full plate and you really have to paying attention and even then there's still hard to do.

Then the other thing that comes up like it shows below is people often adjust the banking pins to do all those other things as opposed to horn clearance which is all that it's therefore and maybe bonus Guard pin clearance although you're supposed to deal with the guard pin is a separate thing like single roller gets bent in Or out or sometimes physically gets moved in and out. Some full plates older escapement's typically pallet forks held together with screws and you can actually unscrew and move the entire assembly in Or out more complications to deal with.

 

 

Escapement handout wostep nscc.pdf

[mbwatch]

3 hours ago, JohnR725 said:

One of the ways to tell if it's been moved is the look straight down at the end of the fork with the balance wheel removed.

You shared this test with me a few months back on a Hamilton project and so I had checked this watch to find the pins were still equally spaced. Their screw slots were undamaged and very tight hard to move so maybe no one has messed with them before.

I can't work on it much today but I did measure the fork slot again and it seems like 0.44 is the right size. I measure the jewel that had been installed at just 0.42. would an undersized too narrow impulse jewel impact the impulse drastically? I ask because I have had it at a state where flipping the fork manually with a tool would unlock some of the time, but the balance was never capable of unlocking.

[JohnR725]

5 hours ago, mbwatch said:

You shared this test with me a few months back

just a reminder about this test is it's not a perfect test it's a quick test. In other words you can adjust the banking pins that are both the same and visually this test will pass everything looks the same but both banking pins can be in the wrong place. although the majority of time when people are playing with banking pins I don't put them in the same place.

5 hours ago, mbwatch said:

I can't work on it much today but I did measure the fork slot again and it seems like 0.44 is the right size. I measure the jewel that had been installed at just 0.42. would an undersized too narrow impulse jewel impact the impulse drastically? I ask because I have had it at a state where flipping the fork manually with a tool would unlock some of the time, but the balance was never capable of unlocking.

one of things have to be careful of is I believe some of the pallet fork measuring tools that actually give you the roller jewel size are actually size so that I get confused? What I mean by this is if you inserted a whatever size in and are always told to go a slightly smaller I thought that the gauge itself its number corresponded to give you the exact number. So a lot of it depends upon the gauge itself I think you do want the roller jewel slightly smaller because it does have to fit in the slot and it does have a little bit a play. But if it's too small you will lose energy so does have to be sized right.

Oh other things to check is? I'm attaching an Elgin sheet on checking the escapement I've seen references in the past to making sure that the slot in the four corn is nice and smooth and apparently you can end up with a rough slot and then the recommendation is to polisher clean that up. Not sure how well that's really going to work even if it has been the recommendation of other reference materials.

yes art full plate watches fun when they don't work. This is where it's nice to have another set of eyes sometimes as maybe they'll find something you didn't  although that can add other issues. The owner of the shop provides himself on his skills of so now two of my watches have relocated to his bench to solve problems both real and imaginary I'm sure that will get fixed eventually hopefully. But still sometimes another set of eyes might see something that you're not.

5 hours ago, mbwatch said:

flipping the fork manually with a tool would unlock some of the time, but the balance was never capable of unlocking.

yes this sort of thing can be quite frustrating. Also makes for an interesting problem unless of course you're the one trying to solve the problem that it's a

Escapement Elgin setting up the escapement.PDF

[mbwatch]

4 hours ago, JohnR725 said:

one of things have to be careful of is I believe some of the pallet fork measuring tools that actually give you the roller jewel size are actually size so that I get confused?

Oh I didn't think of that.  I measured my roller jewel gauge with a screw micrometer and it is all over the place. The 46 measures at 49, 44 -> 46, 42 -> 43, 40 -> 40(!) on the nose, 38 -> 38, 36 -> 37....

So most of them are 0.01 to 0.02 larger than labeled but some of them are not so I have learned this gauge is unreliable and I need to measure it when I use it. Still my jewel is 42, my gauge at 44 (actual 46) just fits in the slot without binding so maybe I will try going up a size if I have one.

[JohnR725]

I suppose this will add to the confusion I have a roller jewel assortment. It lists out American pocket watches for Elgin 18 size and even 16 size it's a 50. But not all the various companies used 50-50 does seem to be common one company had a 51 and the smallest is 43.

American parts are always interesting? Francis Elgin for mainsprings will tell you the thickness of the spring other companies will not even though the spring for the same number could come in a variety of thicknesses. But if we actually had the model number of your watch we would find it probably makes a reference that the roller jewel came in different dimensions.

So overlook the parts book we find that? So it appears to be 18 and 16 size would be the same sort of the arson different catalog numbers and as I said we don't have your Mongol know which Log number were supposed to be using. Variety of materials garnered her sapphire single or double but zero mention about diameters.

Then in a section of rollers in this case rollers with jewels we do get this down in the notes section

Roller specifications but of course zero reference to the jewel size.

I was really hoping the roller jewel assortment would give us sizes it doesn't really. But it does show a picture of how one particular roller jewel gauge is used

 

[mbwatch]

1 hour ago, JohnR725 said:

I suppose this will add to the confusion I have a roller jewel assortment. It lists out American pocket watches for Elgin 18 size and even 16 size it's a 50. But not all the various companies used 50-50 does seem to be common one company had a 51 and the smallest is 43

Yeah I have seen 44-56 documented for Elgin 18 size elsewhere. I have the style of gauge that's like a set of feeler gauges and I never noticed before today it is faintly stamped "Elgin". Pictured is the way I have been measuring.

I am going to go to a larger jewel than the 42 that had been in there. I recently won an auction on a big set of Fitrite jewels all in the little bottles, to discover that the what's actually in the bottles bears no relation to the chart of sizes printed on the box. In most cases it seems to be all mixed so in fact I have about 24 little bottles and thousands of assorted jewels of all diameters and lengths, which is better than nothing. But since I don't have a micrometer with a table, measuring for the exact diameter will be a big job. Instead I will test a lot of them in the fork slot the same way I measure with the gauge.

But this is still the next thing. I have to get the escapement to unlock first.

[JohnR725]

6 hours ago, mbwatch said:

But this is still the next thing. I have to get the escapement to unlock first.

I was thinking it might be nice to see a picture up above or the picture I'm going to paste down below from up above with the escape wheel just setting their. Then it almost looks like the lower jewel is out quite a ways? But maybe it's an optical illusion with the shellac

6 hours ago, mbwatch said:

Fitrite jewels all in the little bottles, to discover that the what's actually in the bottles bears no relation to the chart of sizes printed on the box. In most cases it seems to be all mixed so in fact I have about 24 little bottles and thousands of assorted jewels of all diameters and lengths,

yes the assortment set over time have become something other than what they once were. Like it work with a assortment of Waltham jewels for burnishing in and the last time I tried to find one I discovered there all randomly assorted. Then of course often times the assortments had a tiny quantity of things and the entire container got re-tasked for other things and down the road somebody sells it totally clueless of what they're selling. But in the case of roller jewel's especially after the longer ones of pocket watch size for a single roller then randomly and bottles is definitely good. If you really careful you may be able to place the roller jewel on a flat surface and use calipers although a bench micrometer really is preferred.

 

Edited April 29 by JohnR725

[nevenbekriev]

On 4/28/2025 at 5:54 PM, mbwatch said:

Yes, I know it was able to run before just enough that everything supposed to be affixed with shellac could fall out. I won't get back to work on this watch for several days at least, but one thing you mentioned about the banking pin adjustments being intended for fork horn clearance got me wondering....

The last adjustment I made to the entry stone yesterday did not solve the problem when the balance is installed. The escape teeth can't quite clear the stone when the balance swings. But I thought it was going to work because manually snapping the fork back and forth did allow the escape wheel to advance almost all the time. I wonder if there's another issue with the impulse jewel, that I have to shellac again anyway. I am going to measure the fork slot and the jewel because I am curious if it is the right size or possibly narrower than it should be.

My thinking is if the horn clearance is too great maybe it is not getting enough impulse to clear the escape tooth. But pushing the fork with a tool can move it far enough to clear the escape wheel tooth.  I'll see later in the week. I didn't expect a big challenge from this watch because it ticked for 5 minutes before stopping without shellac. As always I was way wrong.

This here is just thinking in wrong direction.  It is like putting the cart before the horse.

For sure, the problem You have is not connected with the impulse jewel size. I can not claim that the jewel size is correct. But, since the watch was ticking, this means the impulse jewel is not wider than needed. It can be either  correct (which is most probable), or narrow. in both cases, the watch will work if the pallet stones/banking pins are correctly adjusted. What You describe is not enough 'run to the banking' at least on one pallet stone. This side 'run to the banking' is practically '0' and the highest EW teeth can't pass. This is not miss of unlocking. You have to adjust the escapement as there to be (greater than 0) run to the banking in both sides. As I understand, it is more or less OK in one of the sides, but it is not enough on the other. When You snap the fork manually, the teeth pass, but You used 'almost all the time', and this confirms what i say. So You still have to move this pallet stone back, and it is safe to move it back until reliable drop lock exists. The other way to solve the problem is to open just a little the banking pin, just enough a little 'run to the banking' to appear. Of course, this will make the watch work, but the horn and roller clearance on this side will get a little bit bigger and the escapement efficiency will decrease. But, this decrease will be very small, and if You fill not confident enough to correct the pallet position, then leave as it is and open the banking pin as little as it will make the watch work flawlessly.

[Neverenoughwatches]

On 4/29/2025 at 2:19 AM, mbwatch said:

 would an undersized too narrow impulse jewel impact the impulse drastically? I ask because I have had it at a state where flipping the fork manually with a tool would unlock some of the time, but the balance was never capable of unlocking.

So a larger impulse pin will have more contact time with the fork than a smaller pin ....more time to build up energy to be thrown, but  not so much size that the safety clearances become too small. I think you are best to sort out other things first, in case you change something that didn't need changing.

Edited April 30 by Neverenoughwatches

[nevenbekriev]

You can read in all textbooks that reducing the size of impulse stone will reduce the amplitude. I have tried to confirm this, but with limited success... Actually, if any, the reducing of amplitude will be small. But, another problems appear, which may lead to not reliable work instead.

[mbwatch]

12 minutes ago, nevenbekriev said:

You can read in all textbooks that reducing the size of impulse stone will reduce the amplitude. I have tried to confirm this, but with limited success...

I have another watch where I recently replaced the impulse ruby, it has been running with unsatisfactory amplitude after all problems addressed to the best of my abilities. In that watch, I believe I measured the fork with may gauges pictured in the thread at "36" then ordered a jewel 0.02mm smaller than the measurement 0.34mm. But at John's suggestion I checked that the sizes printed on the gauges match their actual thickness and found that mostly the printed size is 0.02mm greater than actual size so they were intended to show the jewel size to install not the size of the fork.

On that watch, maybe I will install a different jewel that matches the printed gauge size - that should be a jewel 0.36 instead of the 0.34 installed now to see if it affects the amplitude measurably.

But as you suggested earlier on the full plate watch that is subject of this thread, I need to fix the lock before I make another change to the impulse stone.

[JohnR725]

3 hours ago, nevenbekriev said:

You can read in all textbooks that reducing the size of impulse stone will reduce the amplitude. I have tried to confirm this, but with limited success... Actually, if any, the reducing of amplitude will be small. But, another problems appear, which may lead to not reliable work instead.

There is something in all of this that was bothering me because I have seen a huge loss of amplitude and I couldn't remember why? But then I remembered why wasn't the roller jewel. So the theory of the roller jewel would be if it's too small when it engages with the fork and causes the fork to unlock then the fork as a greater distance to travel until it pushes on the fork to give an impulse to the balance. But I suspect the above statement is actually correct. Because? Just how small would you have to make the roller jewel before that loss is a huge amount probably insanely small like 50% of the original size or one quarter the original size that would make impact on amplitude probably although getting the roller jewel stay in place might be fun

now something that would cause a huge impact to amplitude is the pallet stones are out too far. This is because when the roller jewel hits the fork normally would hit and there'd be a small amount a motion to unlock but if that locking depth is way too deep then it's going to be unlocking for way too long and impulse will occur too late and yes that does cause a huge decrease in amplitude

2 hours ago, mbwatch said:

I have another watch where I recently replaced the impulse ruby, it has been running with unsatisfactory amplitude after all problems addressed to the best of my abilities.

Problem with the above quote is another watch each watch as far as I'm concerned is unique sort of. The sort of aspect is if you work on enough watches they all become the same but in a discussion each watch is unique and you can't lump two or three watches into the same problem because they may not actually have the same problem and we don't have the entire history of this other watch. Then what is the definition of an unsatisfactory amplitude?

2 hours ago, mbwatch said:

But as you suggested earlier on the full plate watch that is subject of this thread, I need to fix the lock before I make another change to the impulse stone.

Then yes too much a lock and you will lose a heck of a lot of amplitude.

[mbwatch]

5 minutes ago, JohnR725 said:

Then what is the definition of an unsatisfactory amplitude?

175° in the horizontals with a new mainspring 

[JohnR725]

12 minutes ago, mbwatch said:

175° in the horizontals with a new mainspring

I assume that the lift angle is set correctly for that watch? Then I suppose the next question is is it running the next day?

 

[mbwatch]

26 minutes ago, JohnR725 said:

I assume that the lift angle is set correctly for that watch? Then I suppose the next question is is it running the next day?

Yeah - it's a 42 degree lift angle, measured visually. And remarkably it runs with the same amplitude, a weak 175 for >30 hours. Just chugs along and keeps time in 2 positions. I'll figure that one out another time.