Pocket watch balance poising

[Chopin]

Some time ago I got a broken pocket watch that needed a new balance staff. I disassembled everything and put it all back together.

Upon inspecting the bare balance I noticed that some of the screws didn't sit the way they did in the photos that I took before stripping the entire balance ensemble.

I have just installed the balance staff and the roller jewel part. My question is, should I try and poise it now or do I also install the hairspring and then try to balance it out ? (I have to poise it the old school way as I have no special tools or gadgets to do this)

The roller jewel part is uneven so I'm guessing that I should have it attached to the staff and balance, otherwise the uneven weight would mess everything up. Is this correct ?

Edited August 7, 2018 by Chopin

[oldhippy]

You poise it without the hair spring. Correct poising is when it doesn't stop at the same spot.

[Chopin]

Right, so it's safe to poise it WITH the roller jewel part also installed ?

[oldhippy]

You must other wise when you add the roller it will be out of poise.

[Chopin]

Thank you very much! I'll try and poise it tomorrow.

Any tips ? First time doing this. It seems that the balance is hanging with the jewel downwards. (although this might be just a coincidence)

[oldhippy]

Read this from Mark. It will help you.

https://www.watchrepairlessons.com/2011/06/28/poising-a-balance-and-positional-error/

[rodabod]

My first tip would be,  it’s a pocket watch, so positional error is less of a problem than with wristwatches. 

I’d put it on the timing machine first if it were me, and see if it needs adjusting at all. 

[Chopin]

20 hours ago, rodabod said:

My first tip would be,  it’s a pocket watch, so positional error is less of a problem than with wristwatches. 

I’d put it on the timing machine first if it were me, and see if it needs adjusting at all. 

Doesn't it need to be more or less poised so that it can keep up "normal" time ? Can a balance be "unpoised" and still function as it should ?

I'm not asking for COSC standards... Is it possible that this is the normal way it should be ? The heavier area is directly where the roller jewel is.

Edited August 9, 2018 by Chopin

[oldhippy]

Doesn't it need to be more or less poised so that it can keep up "normal" time ?

Correct. If it is un-balanced the timing will be off the scale.

[JohnR725]

besides the definition of pocket watch can you give us a little more details? Depending upon when the pocket watches made or if it's really low-quality poising really isn't an issue. then if it's an American pocket watch with a bimetallic balance wheel you need to true the balance wheel before poising. It's amazing how much screwed up timekeeping can occur when the arms have been squeezed. Then when you put the roller table back on did you put it in the same position it was found? often times people don't mark where the roller table was and then they wonder why they're having a problem with poise.

Then you made the reference to the balance screws being in different positions? did you do anything about that because if the screws are out that screws up the poise. Except you have to be careful you could have mean time screws there supposed to be out although occasionally people don't know what they are and they screw them in nice and tight quite irritating when they do that.

 

[rodabod]

15 hours ago, Chopin said:

Doesn't it need to be more or less poised so that it can keep up "normal" time ? Can a balance be "unpoised" and still function as it should ?

I'm not asking for COSC standards... Is it possible that this is the normal way it should be ? The heavier area is directly where the roller jewel is.

It does not need to be poised accurately to keep good timing in a single position which never moves (think of a platform escapement clock). The purpose of poising is to finely adjust the rate in different positions to keep the timing as consistent as possible.

Have a look here, as it should answer all of your questions:

http://adjustingvintagewatches.com/dynamic-poising-1-3-basic-tasks/

[Chopin]

All the components have been re-installed exactly where they were. The watch is an early 1910s-1920s Longines but the movement itself is probably the cheaper/basic ones.

The interesting thing about the balance is that the added weight is exactly in the center of the arms, right where the roller jewel is situated.

I have actually decided to put everything back together and so far the watch has lost around 30 seconds or so in half a day. (measured dial up)

I have tried all sorts of combinations with the screws of the balance and the weight would still be where it is now so, maybe this is how it was going to be in the first place ? At the moment the screws are, more or less, in the same position that they were when I took it apart.

This is a photo before everything was stripped apart (with the old balance staff basically).

[rodabod]

The silver screws are quarter screws which can effectively be used for poise in four positions. You can refer to that website if you want to make adjustments (screwing outwards is effectively like adding weight). But do not change it if the positional variance is already good. Check it dial-up and pendant-up.

[JohnR725]

2 hours ago, Chopin said:

At the moment the screws are, more or less, in the same position that they were when I took it apart. 

The wording of this is very interesting did you remove the screws on the balance wheel? The screws on the balance wheel are really interesting they all looked the same don't they? The only problem is with the exception of the mean time screws all the rest of the screws conceivably visually look the same but may not weigh the same. Then screws on the balance wheel are in pairs whatever is on one side has to be on the other side they have to match in weight otherwise you will have a poising issue.

So basically you always leave the screws alone. You do want to check though usually after you ran the balance in the cleaning machine that they are still screwed tight in place. If they start to unscrew it causes an uneven a balance wheel.

Then on these early balance wheels when they were manufactured the wheel itself may not actually be poised they use the screws to do that in which case the screws have to go back to where they are found. If this was a bimetallic balance wheel then the screws are in a very exacting position for temperature compensation. Moving these around would cause interesting timing problems with temperature you probably never be able to get them back to where they were..

Dynamic poising as great except remember you can’t dynamically poise until you statically poise. Then if you do a really good job with static poising especially on pocket watches usually don’t have to dynamically poise unless you’re trying for railroad timekeeping.

The mean time screws ideally visually they should be in the same position. In other words one on one side should be in the same position as the other one once again they need to be matched in weight to keep the balance wheel poised. They can be used for minor poising that’s not the real purpose it’s really for timekeeping. Then try to remember not to play with these too much they have a habit of getting loose if they been over adjusted.

 

 

[Chopin]

I never removed any of the screws. They were just more screwed in or less screwed in but now they are in the same position.

Put it on a timegrapher app and the watch losses about 2 and a half minutes per day. What could cause this ?

I have regulated using the only regulator that the movement has and it's at -14 seconds per day at the moment. Still would be curious to find out as to what causes it to lose time with the regulator in the center.

As I said, the added weight is exactly in the middle, where the roller jewel is. I would assume that this is far better than having added weight somewhere else on either side. Right ?

I'm also thinking that, since I tried balancing it by tightening or loosening the screws and the weight was still where it is now, maybe this is just how the balance was originally ?

Could the lost time be caused by the fact that, maybe, I didn't center the hairspring the way I should have ? I placed it exactly where it was before, but then again, I don't know if it was the correct position before.

As can be seen in the photo the stud isn't exactly in the center of the balance arm. Nor is the small slot of the hairspring collet in the center, either.

 

Edited August 10, 2018 by Chopin

[JohnR725]

First off we assume that the watch always had the regulator in the zero everything was always perfect it may not have been. If it's been worked on especially the older watches that is the pocket watches things get changed the chance of the regulator going back to zero is almost 0. This is also where I really like before and after timing results which unfortunately can't be done with the balance staff issue.

Added weight? If you added weight to poise the roller jewel issue the watch will now run slow. Typically when you're poising your removing weight because that's usually the easiest thing to do. Then the balance wheel is to light you'll have to add timing washers but? before adding timing washers make sure the regulator is back at zero.

then the mean time screws in poising? The mean time screw purpose is for regulation I'm attaching a chart. True it's a chart for Hamilton but you can see the effect of moving the screws in it out changes the rate of the watch. so they can be used for tiny poising issues but in a dramatic moving you're going to screw up timekeeping.

All of the normal screws need to be in tight not so tight that they break off which they will do occasionally. They need to stay stationary. then depending upon the cleaning machine sometimes they loosen up and they may just a bit loose before. It's always good just to check their down.

In a ideal perfect world when restaffing you mark where the hairspring stud is and the roller jewel is. Typically though I do not mark where the hairspring is. If I knew they hairspring stud was where it's supposed to be because the watch was in beat yes but I find it's easier to figure out where they hairspring study is supposed to go by visually verify that the balance is in beat and then putting the hairspring there.

then the collet slot it doesn't have to go in any exacting position other than its position determines where the stud goes. On some watches more common on modern the shape of a collet is done for poising purposes.

[Chopin]

John thank you for all of this information, learning some new things.

I will wait for the movement to stop completely and check that the pallet fork is in the center. It's something simple that I can do and I'm thinking that it might also affect timekeeping.

If it's not in the center it means that the hairspring collet needs to be repositioned a bit, right ?

[JohnR725]

I found Mark has two separate videos explains things a little bit differently. He uses a slightly different method to determine where the hairspring goes.   then he shows in the second video using a balance tack I don't like that as I worry about stretching the hairspring so I remove the balance wheel when rotating the collet

I use a different method it into it with a hairspring in place it's easier if it's not in place.  I have the balance wheel in the watch with its bridge you rotate the balance wheel until the watch is visually in beat as Mark shows in his video. Then using a felt pen  place a mark on the balance rim corresponding to where the stud will go. then the balance wheel out of the watch you can now rotate collet until the stud lines up with the mark you made. then if the balance wheel is out of the watch when you're rotating the collet makes it really easy to see that the stud is now lined up with the mark you made.

then for rotating the collet they actually make a special tool which I show in the pictures. I preferred the tool over the screwdriver method. But more than likely you don't have the tool. the main difference between the tool and the screwdriver blade is the tool has a long taper and is almost flat at the end. This allows the tool to be inserted into the collet slot without spreading the collet. there's a danger with screwdrivers of pushing it into the slot and spreading the collet making it loose.

https://www.youtube.com/watch?v=C9tOHiPQpbw

https://www.youtube.com/watch?v=uvMCNyrDZQo