Potential causes of beat error

[Simeon]

Hi,

 

I have a old Orient automatic with a 46943 movement that I am going to service.  Doing some pre-work inspections on it, it has a low amplitude DU of about 200 degrees / 180 DD a rate of about -10 DU- and +10 DD so not terrible for an unserviced watch.  The problem is that this has a beat error of 8.0ms DU and DD.  I have had a bit of a play with moving the stud carrier through its range of movement but it only seems to get worse, not better - I maybe can't find the sweet spot where it is correct with small adjustments and keep pushing through so that it is out of beat in the other direction.

My question is, other than the position of the stud carrier relative to the neutral line of the pallet fork, what else can contribute to bad beat error that I can look out for when servicing? The overall location of the hairspring collet on the balance staff I guess but what about anything else in the escapement?

Also, I understand that I can find a rough setting by giving the movement a small amount of wind and then move the stud carrier slowly from one extreme to the other, stopping when the balance starts.  Is there any other easy ways to statically set the beat error (or at least get it close with an initial setting)?

[nevenbekriev]

Let me clear something...

Beat Error is not something that can be bad. It doesn't show that there is any problem in the movement and doesn't affect the timekeeping. It only depends on initial (neutral, zero) position of the balance and if this position is wrong, the movement will not start working by itself and will need shaking. Further, if BE is wrong, the movement will come sooner to 'galooping' if the amplitude is very big, than if BE is correct. So the correct BE allows maximal amplitude without 'galoopiung'. Yes, it is possible that other things than initial position may cause BE. For example, artificial limiting of balance movement in one of directions. But, such problem will cause many other things and wrong BE will be the last thing to be concerned about then. Something more - the way BE is calculated in timing machine is such, that the result depends on amplitude. If You have BE different than 0, then it will depend on amplitude. So You don't need to worry if BE is bigger in vertical positions, this is normal.

Generally, You don't need timegrapher at all to adjust the initial position correct. You just need to ensure that the initial position is such that the impulse jewel holds the pallet fork exact in the middle of it's movement (imp. jewel on the line balance bearings-pallet fork bearings) in static when DU/DD. If it is so, then You can ignore what timegrapher shows as what You see in the movement is more relevant than what timegrapher calculates. An easy way to adjust initial position (if the roller is placed in such manner that the impulse jewel is on the bisectrix line of the angle, formed by the spokes/arm of the balance wheel) is to observe the figures that spokes/arm draw when balance rotates. This figures must be symmetrical with line of symmetry  balance bearings-pallet fork bearings.  

[Nucejoe]

On 7/24/2024 at 7:07 AM, Simeon said:

  Is there any other easy ways to statically set the beat error (or at least get it close with an initial setting)?

  With no power in the barrel ,  ink mark ( as a reference point) a suitable point on a bridge where balance usually stops and  another ink mark on rim of balance right in front of the  mark on bridge. You now have marked two point , one  on bridge the other on balance rim , both right in front each other where balance usually stops.  

Manually rotate the balance wheel whilst keeping an eye on the fork, when impulse jewel starts to engage the fork, fork moves, continue rotating the balance until the fork stops moving, indicating no engagement of impulse and fork, ink mark this second point, the two points on bridge where fork started and stopped to move indicate start and end of the arc at midpoint of which your impulse is in beat.

Though not exactly accurate, this two points on bridge indicate  the arc within which you must advance impulse jewel ,in order to have good range of adjustability .  You would observe an interval in which forks interacts with impulse jewel, the midpoint of this interval is where impulse should come to stop, for good range of adjustability.

Take your pick , either rotate the roller table or hairspring coil, to have a balance in beat (or near) in beat. 

An In beat impulse would maximize the amplitude, thus minimize  the ratio of positional variation compared to out of beat balance, movement starts running with minimal wind, and maximized usage of reserved power in the barrel,,,, 

Rgds

Edited July 26, 2024 by Nucejoe

[Simeon]

On 7/25/2024 at 7:21 AM, Nucejoe said:

  With no power in the barrel ,  ink mark ( as a reference point) a suitable point on a bridge where balance usually stops and  another ink mark on rim of balance right in front of the  mark on bridge. You now have marked two point , one  on bridge the other on balance rim , both right in front each other where balance usually stops.  

Manually rotate the balance wheel whilst keeping an eye on the fork, when impulse jewel starts to engage the fork, fork moves, continue rotating the balance until the fork stops moving, indicating no engagement of impulse and fork, ink mark this second point.

Though not exactly accurate, the two points on bridge indicate  the arc within which you must advance impulse jewel ,in order to have good range of adjustability .  You would observe an interval in which forks interacts with impulse jewel, the midpoint of this interval is where impulse should come to stop, for good range of adjustability.

Take your pick , either rotate the roller table or hairspring coil, to have a balance in beat (or near) in beat. 

An In beat impulse would maximize the amplitude, thus minimize  the ratio of positional variation compared to out of beat balance, movement starts running with minimal wind, and maximized usage of reserved power in the barrel,,,, 

Rgds

Thank you for this.  I have a couple of Vostok 2209s with fixed hairsprings that I will need to rotate the hairspring collet to fix beat error - this is a good method.

[Nucejoe]

Just now, Simeon said:

Thank you for this.  I have a couple of Vostok 2209s with fixed hairsprings that I will need to rotate the hairspring collet to fix beat error - this is a good method.

Nickelsilver showed this proprietary tool used for fine adjustment of beat.

Inspiring and I plan to make a simple inexpensive tool for same task. 

 

 

 

 

 

 

 

 

[Neverenoughwatches]

1 minute ago, Nucejoe said:

Nickelsilver showed this proprietary tool used for fine adjustment of beat.

Inspiring and I plan to make a simple inexpensive tool for same task. 

 

 

 

 

 

 

 

 

I'd like to see how that works.

[rehajm]

I’m a collet man myself but the tool is interesting. Turns the roller while mounted on the staff does it?

Edited July 26, 2024 by rehajm

[Simeon]

1 hour ago, Nucejoe said:

Nickelsilver showed this proprietary tool used for fine adjustment of beat.

Inspiring and I plan to make a simple inexpensive tool for same task. 

 

 

 

 

 

 

 

 

Ha! Another tool I can't afford to buy. There's always a tool for every task (or you can cobble something together nearly as good with tweezers, rodico and a plastic bag)

[mikepilk]

Have you checked that the collet/stud/roller and staff are all firmly fixed?

I've had unusual beat errors which turned out to be one of the above being loose.

[Simeon]

Just now, mikepilk said:

Have you checked that the collet/stud/roller and staff are all firmly fixed?

I've had unusual beat errors which turned out to be one of the above being loose.

It's worth a look for sure. It seems to be consistent though regardless of whether it's dial up / dial down etc so I would guess that if it was because something was loose it would possibly vary with position. 

[JohnR725]

I assume you did make sure that they hairspring centered and everything looks fine with the hairspring?

On 7/23/2024 at 8:37 PM, Simeon said:

moving the stud carrier through its range of movement

as nice as the movable stud holder is and timing machine you really should do this the old-fashioned way. The problem with trying to move it around looking at the timing machine there is no plus or minus and typically will sail right past zero and not even grasp you've done that. so basically could spend a lot of time moving a back-and-forth seemingly going nowhere because you really need to be much much closer when you start.

Ideally it be nice if you could see the pallet fork like if it's in between the banking pins for instance. This should get you close enough otherwise you can try the alternative methods?

 

 

 

[Simeon]

1 hour ago, JohnR725 said:

I assume you did make sure that they hairspring centered and everything looks fine with the hairspring?

as nice as the movable stud holder is and timing machine you really should do this the old-fashioned way. The problem with trying to move it around looking at the timing machine there is no plus or minus and typically will sail right past zero and not even grasp you've done that. so basically could spend a lot of time moving a back-and-forth seemingly going nowhere because you really need to be much much closer when you start.

Ideally it be nice if you could see the pallet fork like if it's in between the banking pins for instance. This should get you close enough otherwise you can try the alternative methods?

 

 

 

As far as I can tell, without removing the hairspring from the balance cock, it looks OK. It's not a huge hairspring, so the concentric part, terminal curve and outer coil are all within a small package. Difficult to say it's significantly out of shape.

It definitely feels like I could be pushing through from one side of adjustment to the other trying to correct beat error. Perhaps the geometry of this movement is such that it it really isn't far from one side to the other and makes it difficult?

 

As I say, I haven't stripped the watch down yet, so I can't see the various bits of the escapement through the main plate yet. Ideally I would be able to see both entrance and exit stones on the pallet fork (through holes) to give me an idea where the fork of the pallet fork is. 

[Nucejoe]

@Neverenoughwatches  google this thread on WRT ,

 

Adjusting beat error on a movement without adjustable stud carrier.   

 

 

Reason for giving you the thread title is,

I am doing something wrong , couldn't quote the post on this thread .

Made a mess instead.

 

 

Edited July 26, 2024 by Nucejoe

[RichardHarris123]

@NucejoeI can't delete but I have hidden the post for you. 

[Nucejoe]

Just now, RichardHarris123 said:

@NucejoeI can't delete but I have hidden the post for you. 

Hi Rich , 

I best go hide myself.

Many thanks .

 

2 hours ago, Simeon said:

Ha! Another tool I can't afford to buy. There's always a tool for every task (or you can cobble something together nearly as good with tweezers, rodico and a plastic bag)

Not sure but think this one is propriatory. 

A few was made for use on manufacturer's assembly line.

Rgds

 

Edited July 26, 2024 by Nucejoe
correction

[JohnR725]

6 hours ago, Simeon said:

Perhaps the geometry of this movement is such that it it really isn't far from one side to the other and makes it difficult?

yes in the past I've noticed that it is extremely easy to get lost with something that's supposed to be very helpful.

In the absence of seeing the escapement when you take the balance wheel out note where the roller jewel is. Sometimes you have a three arm balance wheel on the roller jewel corresponds to one of the arms so you use out to figure out where you need to be. or you can put a Mark on the balance rim where the roller jewel is in use at the site off of.  But yes on some watches it's hard to do which is why they have movable stuff because you can't see anything.

another option is try again. But this time go really really slowly perhaps moving a tiny bit and stopping. Also allow time for the timing machine to catch up with what you're doing? Then out of curiosity which timing machine are you using? Another helpful thing is to find a picture of your watch and see where they put the stud that will give you an approximation.

https://17jewels.info/movements/o/orient/orient-46943/

then technical information for some watch companies like Orient can be difficult to find. But sometimes were lucky. It also looks like from the image of the three arm balance wheel so if you're lucky maybe the roller jewel will correspond to one of the arms you can site off of that.

Orient-46943.pdf

[dnhb]

On 7/24/2024 at 4:37 AM, Simeon said:

"......I have had a bit of a play with moving the stud carrier through its range of movement but it only seems to get worse, not better - I maybe can't find the sweet spot where it is correct with small adjustments and keep pushing through so that it is out of beat in the other direction"

 

I wonder if, instead of "pushing" the stud carrier - & thereby always overshooting on alternate sides - you've tried tapping it (cf how one positions items accurately prior to milling operations) as this will produce much smaller movements.

[dnhb]

On 7/26/2024 at 10:30 AM, rehajm said:

I’m a collet man myself....

I find I'm dealing with a watch in which the stud carrier is fixed so I'll have to adjust the collet to put it in beat. As I've not attempted this before I'd be grateful for any tips on how to go about this, particularly with regard to risk reduction as the watch is not mine !

Edited August 14, 2024 by dnhb
Typo

[Neverenoughwatches]

1 hour ago, dnhb said:

I find I'm dealing with a watch in which the stud carrier is fixed so I'll have to adjust the collet to put it in beat. As I've not attempted this before I'd be grateful for any tips on how to go about this, particularly with regard to risk reduction as the watch is not mine !

There are a few different tools that hold and seperate the balance and the cock that help to make the adjustment with more safety than a free hand adjustment. Try a googly of hairspring collet adjusting tools. Some attempt free handing with the use of a balance tack some by sitting the balance in a staking block and lifting the cock to one side. I like this kind of staking block for the lifting aside method

[caseback]

I've used something simular, but cruder, cobbled together from what I already have. The advantage of this setup is that all parts are stable/supported, reducing the risks.

 

Edited August 14, 2024 by caseback

[Neverenoughwatches]

1 hour ago, caseback said:

I've used something simular, but cruder, cobbled together from what I already have. The advantage of this setup is that all parts are stable/supported, reducing the risks.

 

Any good sized staking block by itself can be used with a simple tapered pin set in one of it's hole to hold the cock.

[rehajm]

4 hours ago, dnhb said:

I find I'm dealing with a watch in which the stud carrier is fixed so I'll have to adjust the collet to put it in beat. As I've not attempted this before I'd be grateful for any tips on how to go about this, particularly with regard to risk reduction as the watch is not mine !

I paid up for the bergeon collet tool .45mm but others make something with a little tooth…others are happy with the little screwdriver or a flat needle or something…and I like to remove the balance from the stud and place it on a cheap nylon balance support that is round with various diameter holes- looks like the metal ones above. If you work on swiss from the era of fixed stud you can get away with one sized hole and one tool. Place the tooth in the collet gap from above and rotate…I leave the balance on the cock only when I have to. Usually I’m talking it apart for cleaning anyways…

To determine the amount of adjustment…others have their methods but for me, with the balance in the watch I look down between the pins or bridge and carefully rotate the balance wheel to center the roller jewel. like to use the stud as a reference relative to the wheel. Sometimes there’s a factory mark on the wheel on or near the right place but you can use the balance arm or a scratch or mark already on the wheel.

You should practice before the friend watch

Edited August 14, 2024 by rehajm