Is My Pallet Fork Misaligned?

[Lawren5]

I received a 50’s era watch that was totally dead. After disassembly, cleaning, lubing and reassembly it worked but the amplitude was very low. I was unable to get an amplitude reading with a timegrapher app but I would guess it was about 45 degrees. The BPH measured 30,600 which is high since it should read 18,000 for this movement. Further investigation revealed the balance wheel binding in one direction but I was unable to located the cause. 

Upon further testing, removing the pallet fork allows the balance wheel to rotate freely in both directions.  When the pallet fork is installed without the balance wheel, it moves freely from side to side when lightly prodded.

Since both pallet fork and balance wheel operate normally on their own, could it be there is a misalignment of some type between the two?

[khunter]

Certainly sounds like it.

[rodabod]

Inspect the safety action of the escapement. Check the amount of shake you get when the balance is 90 degrees away from rest in either direction. It should be just perceptible, and the safety dart should not rub against the safety roller as long as the pallets have been sufficently drawn into the escape wheel. Check the action of the pallets and the escape wheel while you are there; go round all escape wheel teeth and check that the pallets draw all the way to the banking pins.

[JohnR725]

I have a little light reading to understand how escapement's work and the things you're supposed to be looking for.

http://raulhorology.com/2012/10/setting-up-the-escapement-eta-6498-end-shakes-divisions/

http://www.rdrop.com/~jsexton/watches/csw/les22/index.html

http://www.timezone.com/2002/09/16/the-anchor-escapement/

[Lawren5]

Interesting reading and it gives me more insight into what the problem might be. I wasn't aware of all the intricacies of the escapement nor how critical setup is for proper operation. It also provided a number of recommendations for what needs to be observed and adjusted if required.

That being said, how does one go about observing the escapement in operation? The pallet fork bridge and balance cock obscure the components. Are there some special tricks-of-the-trade that would allow me to see what's going on?

[rodabod]

Did you try what I suggested by inspecting the pallet fork shake while the balance is held 90 degrees away from rest? I casn explain in more detail if required.

I'd also check the basic things like the the impulse jewel (is it loose, broken?) and generally check the safety dart on the pallets doesn't look bent.

[JohnR725]

Can we see a picture of the watch with the pallet fork in and the balance wheel out close-up would be nice. Then when working on a totally dead watch why was it totally dead?

9 hours ago, rodabod said:

I'd also check the basic things like the the impulse jewel (is it loose, broken?) and generally check the safety dart on the pallets doesn't look bent.

Basic watch repair and terminology differences but the meanings the same. Check the roller jewel always important in older watches they get loose. The guard pin on the pallet fork make sure it's not bent.

 

[Lawren5]

Rodabod - Yes, I would appreciate if you could explain further how to check pallet fork end shake as I wasn't able to find a procedure for it and I suspect that this is where the problem lies. The balance wheel turns freely clockwise but hangs up turning counter-clockwise. As a test, I loosened the screw holding the balance cock and raised it slightly. The amplitude jumped from about 45 degrees to 230 degrees. The timegrapher app was able to read the amplitude and showed the correct BPH of 18,000, neither of which it was able to do before.
 

[rodabod]

Ok, so firstly, your photos look ok, but nothing compares to having a close look with a loupe. Preferably a 10x or 12x for really close inspection.

The issue is not necessarily a fault with the safety action, buit it's always worth checking, and should only take you a few seconds. Look at this image, and try to understand how the safety action works, and why it is absolutely necessary on the lever escapement:

 

The drawing above is showing you a view looking straight down on the balance, but the hairsping, balance wheel and impulse roller have been made invisible as they sit directly above the safety roller and block our view.

The safety roller acts like a gate with a rule, which is that the pallet fork is only allowed to pass when the safety roller is oriented such that the half-moon cutout is in line with the guard pin / dart (in yellow). If the pallet fork tries to move away from either banking pin (say we gave the watch a hard knock) when the cutout is not in alignment then the guard pin will simply crash into the cylindrical body of safety roller, and the pallet fork will return to the banking pin due to the feature called draw. Draw is the effect that the escape wheel has when pressing agaisnt a pallet jewel; it not only holds the pallet jewel static, but also tries to increase its depth of locking slightly and actually causes the entire watch gear train to recoil backwards slightly.

To test the safety action of the guard pin and the safety roller, you must have the watch slightly wound; enough for it to run and therefore to ensure that the pallet form is drawn towards the banking pins and held there. Now we need to test the shake (play/freedom) of the pallet fork while the safety roller is not in its rest position. Very carefully arrest the balance wheel with a light oiler, or similar. Now rotate the balance wheel round to approximately 90 degrees away from rest and hold it there. You can use an oiler to do this or if it's a cheap watch then you can sometimes use the edge of your finger agaisnt the balance wheel. Next, use something delicate like an oiler to push the pallet fork away from the banking pin. There should just be the slightest amount of shake detected. You might not be able to see it visibly, but you should be able to feel it. If there is no shake, then the gurad pin/safety pin/ dart may be dragging against the roller. Repeat this test for the other side, ie. rotate the balance around the other direction.

There is an additional aspect to the safety action which is the horns of ther pallet fork (the metal which surround the slot on the pallet fork where the impulse jewel enters) but this can probably wait for just now.

[Lawren5]

Thanks for the procedure. I did not detect any shake, however, it could just be me. I found getting access to the pallet fork difficult and it made me somewhat nervous having to nudge it through the hairspring. Will keep at it until I get the procedure down and will advise whether or not any shake is detected.

[JohnR725]

20 minutes ago, Lawren5 said:

Thanks for the procedure. I did not detect any shake, however, it could just be me. I found getting access to the pallet fork difficult and it made me somewhat nervous having to nudge it through the hairspring. Will keep at it until I get the procedure down and will advise whether or not any shake is detected.

So we can understand better can we get a picture of the overall watch and the make and model.

Rodabod Has really done a super nice job of explaining. Now the problem is you should perhaps practice on something bigger. Then worst-case because this is a learning exercise disassemble the watch with only the balance wheel and pallet fork in place it makes it a lot easier to see things. Then if you work on American pocket watches you get a lot of practice doing this the banking pins are movable and more than likely they been moved. Wristwatches will have a much smaller amount of clearance then a pocket watch. If you can't get at the pallet fork because you have to go through the hairspring you can also move the pallet fork at the pallet stones. Use them as a guide for how much movement you have you can almost always see some part of the pallet fork.

Then this may not be a problem. But I do agree it's a really good thing to check. Problem then as we need to see the watch we really need a 3-D  that we can hold our hand and rotate around it make it a lot simpler. 

 

 

[Lawren5]

The watch belongs to a friend who grew up in Switzerland. It was given to him as a gift in 1957 from a family friend who made the watch and was employed as a watch maker in the town of Neuchatel, a Swiss watch making center. The dial says "Roxton" and the movement is an Adolph Schild, AS 1430 movement. My friend wore the watch for about 15 years including his service in the Swiss military. It was pretty beat up when I received it and was not working. The crystal has several cracks and the dial is quite tarnished. However, he didn't throw the watch away because it was a gift. 
I've included some pictures in addition to one that addresses the issue at hand. It shows the pallet fork lever and right arm, neither of which I could detect any shake on.
 

[Gary]

Looks like a keeper and given the years and history in good shape!
Service and tidy?
Gary


Sent from my iPad using Tapatalk

[JohnR725]

Not requested but attached I found the tech sheets For that movement.

Then not detecting any shake isn't necessarily an issue unless you actually know what you're doing and since I suspect you don't, don't worry about it. It's still be nice to see the other end of the pallet fork 

 

 

 

AS 1430.pdf

[Lawren5]

The watch has a recessed cannon pinion. How do I remove it when a cannon pinion removal tool can't fit underneath it?

[JohnR725]

Most of the time I can just use my tweezers. If that doesn't work as with almost everything in watch repair they make a special tool.  Pictures attached and video part way any shows how to use the tool.

https://www.youtube.com/watch?v=2-DVmSeug-8

 

[rodabod]

Just a quick one: are you sure that you’ve got the upper and lower incabloc jewel holes fitted the correct way around?

[Lawren5]

I don't have the special tool but I was able to remove the cannon pinion fairly easily by gripping it with a pair of diagonal wire cutters.

Rodabod - I'm not quite sure what you mean by "correct way around". The bottom part of the jewel has a metal cap that can only go in one way. I made sure that when reassembling the jewel that the round end of the jewel cap was facing outward. I had read somewhere that the upper and lower balance jewel assemblies are not always identical so I was careful not to swap the two.

[JohnR725]

So the message above. Taking the Canon pinion off now that you know how. But the very start of this conversation you indicated you disassembled and service the watch? It seems like your questions are out of order?

So what were trying to do is guess what your problem is from your description. Mixing up the jewels is bad putting the cap stones on upside down  bad but in your photograph at least the one looks right.

So the watch was dead in the beginning why was it dead? So was it wound up tight and not running did the balance wheel move.

Guessing really isn't the best troubleshooting method.

So when the balance wheel is out power is on the watch gently pushing on the fork it snaps from one banking pin than pushing the other way it snaps back the other way? Then when it's doing that each time you push it one tooth of the escape wheel goes by?

Then slightly different angle of photograph is needed? So I've attached a snippet of one your images this is where the regulator pins are. Can you angle the camera so we can see the pins and the hairspring. Then looking in sideways at your watch does the hairspring appear to be flat?

 

 

 

[Lawren5]

Well, I initially disassembled it as far as I could without taking the cannon pinion off. This is actually the first watch I've encountered in which the cannon pinion needed to be removed in order to remove the hour and minute wheels. (Newbie here)

When I first received the watch, it didn't work. It was fully wound and the balance wheel was binding when rotated counter-clockwise.

The drive train is strong with the pallet fork snapping over to the opposite banking pin when lightly nudged. The escapement does advance and in fact the watch does run even though the amplitude is only about 45 degrees.

The movement is now fully stripped and the next step will be to install just the pallet fork and balance assembly in an attempt to get some better photos and isolate the problem.

*** I will be on a two week hiatus due to a family emergency so will provide further details when I return. ***

 

[Deggsie]

The watch has a recessed cannon pinion. How do I remove it when a cannon pinion removal tool can't fit underneath it?

Can I please ask what equipment you use for taking such clear and detailed photos?


Sent from my iPhone using Tapatalk

[Lawren5]

Deggsie - I'm taking the pictures with one of those $20 USB microscopes from Amazon. The specifications for these things are all over the place but in reality these $20 scopes are all about the same. Some claim 2MP sensors and others 5MP but by time the image gets processed they all end up being 640 x 480 which is about 0.3MP. Mine claimed a magnification of 1000x but it's probably more like 50x. The software that came with it did not work but fortunately there's third party software available for free on the web. Anyway, it works for what I'm doing.

[Deggsie]

Deggsie - I'm taking the pictures with one of those $20 USB microscopes from Amazon. The specifications for these things are all over the place but in reality these $20 scopes are all about the same. Some claim 2MP sensors and others 5MP but by time the image gets processed they all end up being 640 x 480 which is about 0.3MP. Mine claimed a magnification of 1000x but it's probably more like 50x. The software that came with it did not work but fortunately there's third party software available for free on the web. Anyway, it works for what I'm doing.

Thanks Lawren. I’ve bought one from eBay and went for the 1000x magnification for 19.99 all delivered. Comes with a small retort stand as well. Looking forward to sending in some equally high standard photos. At that price it seems good value


Sent from my iPhone using Tapatalk

[Lawren5]

OK, I have some new information concerning the original problem of the balance wheel sticking in the counter-clockwise direction. I stripped the watch down to where just the pallet fork and balance wheel assembly were installed but still could not see the interaction of the pallet fork and safety roller. However, I did find it easier to perform the shake test as Rodabod describes above. When the balance wheel was rotated 90 degrees clockwise, there was considerable play between the pallet fork and safety roller. When rotating the balance wheel 90 degrees counter-clockwise, there was no play detected at all.

What is the method for correcting this shake issue?

[rodabod]

5 hours ago, Lawren5 said:

What is the method for correcting this shake issue?

Firstly, recognise what causes the amount of shake; its the interaction between the guard pin and the safety roller.

Most commonly the problem with these is that they are bent. So, look ay the guard pin under high magnification and see if you think it lookms misaligned or curved.