Jump to content
  • 0
Len33

Beat error shows as high, on good working watches.

Question

I have a Unitas 235 pocket watch that needed the balance pivot replaced to get it working again.

After doing so, it keeps excellent time, but the beat error shows as 9.9ms on the timegrapher.

The rate shows  +1 sec (Most of the time) and the amplitude shows 216 degrees.

Whilst it was in pieces, with the balance in place and the pallet removed, I very carefully made sure that the balance jewel was perfectly between the banking pins

when at rest.

I checked it umpteen times, just to be sure. 

So, the question is, why does the beat error show such a high (Unchangeable) reading, and the watch keeps such good time?

I have seen the same thing with a known good Valjoux 72 movement.................9.9ms!

 

Share this post


Link to post
Share on other sites

24 answers to this question

Recommended Posts

  • 0

What or which model of timegrapher Do you have? Then when you're timing the Watch do you time the watch in multiple positions like dial-up and dial down and in any other positions? 

Then with no power applied the pallet fork in the watch where is it in relationship to the banking pins?

 

Share this post


Link to post
Share on other sites
  • 0

I have a model 1500 timegrapher and with 99% of watches that I have used it with, the beat error reading works ok.

I centred the balance jewel between the banking pins with  the pallet removed from the watch and checked it umpteen times.

Being a pocket watch, I don't time face up/down etc, because it's used in a verticle position.

What I do, is to make sure that the watch keeps working when it is twisted and turned, this way and that, so I know that nothing is going to Make it stop.

As I have said, the watch does keep "excellent time."

Share this post


Link to post
Share on other sites
  • 0

I've found sometimes when a watch has a low amplitude like yours, the beat error increases. Also if you looked at the impulse jewel between the banking pins and it appeared central, I'm assuming that was in a horizontal position. You are now measuring the beat error in a vertical position, as you said that is how pocket watches are worn, but your not taking into account the poise of the balance or the isochronal error this can incur, especially as the hairspring may also sag or start touching the curb pin or boot, thus amplifying the beat error and reducing the amplitude

Share this post


Link to post
Share on other sites
  • 0
3 hours ago, Len33 said:

I have a Unitas 235 pocket watch that needed the balance pivot replaced to get it working again.

Sorry I don't know the answer to your question but allow me to remind that the proper section for it is "watch repair help and advice".

Before someone criticizes my observation as coming from the other side of a "language barrer",  is worth to remind that the reason is not just to keep an organized forum, but that you will have more chances of answers, as not everybody browses or checks all sections of this forum, and that is the only one where answer can be rated and even flagged as correct. 

Share this post


Link to post
Share on other sites
  • 0
23 minutes ago, jdm said:

Sorry I don't know the answer to your question but allow me to remind that the proper section for it is "watch repair help and advice".

Before someone criticizes my observation as coming from the other side of a "language barrer",  is worth to remind that the reason is not just to keep an organized forum, but that you will have more chances of answers, as not everybody browses or checks all sections of this forum, and that is the only one where answer can be rated and even flagged as correct. 

Good point... 

Share this post


Link to post
Share on other sites
  • 0

Yes it's really important that discussions occur where there supposed to.  As pointed out above not everyone looks at all the discussions I typically only look at watch repair. Unless it's a really new discussion and it's on the right-hand side and I see it there.

Then the reason for timing watches in multiple positions is it's used for diagnostic purposes. Then if your pocket watch is a vast pocket watch your right you can keep it up right but if it's in your pocket and you should ever sit down then it's going to be resting on the balance pivots or at an angle. Also important when using the timing machine that you should be timing at equivalent of fully wound up but you need to let it run about 30 minutes. Then at 24 hours later as things can change dramatically in 24 hours.

Then the reason I asked which timing machine is on multiple of occasions we've been led down the rabbit hole because somebody's using a phone app or some other device that has a pickup that totally sucks. Can't do proper diagnosis with faulty equipment. So it looks like the 1500 visually looks identical to the 1000.  So it would be really nice to have a picture of the timing machine with this watch so we can see the graphical display just because we like to look at pictures. We've had discussions elsewhere the machines can have rollover errors which might not be the right term where the graphical display will look perfect but the numerical display does not agree with the graphical display. 9.9 is indicating the machine is reached the maximum beat error which isn't good. Amplitude will magnify that but it has to be below 200° Then considerably below that so visually what does the amplitude look like? Then if the amplitude was low enough to make the beat error look this bad it probably wouldn't be keeping time either. 

http://www.ranfft.de/cgi-bin/bidfun-db.cgi?10&ranfft&0&2uswk&Unitas_235

Share this post


Link to post
Share on other sites
  • 0

The beat error is relative to balance velocity and therefore amplitude. 
 

If this was in my hands, I’d adjust slightly in either direction to see if both amplitude and BE improved. 

Share this post


Link to post
Share on other sites
  • 0
10 hours ago, Len33 said:

I have a model 1500 timegrapher and with 99% of watches that I have used it with, the beat error reading works ok.

I centred the balance jewel between the banking pins with  the pallet removed from the watch and checked it umpteen times.

Being a pocket watch, I don't time face up/down etc, because it's used in a verticle position.

What I do, is to make sure that the watch keeps working when it is twisted and turned, this way and that, so I know that nothing is going to Make it stop.

As I have said, the watch does keep "excellent time."

ok considering the impulse jewel is dead center between banking pins 3things you should check:

1. Pallet. Check the lock and draw of the pallet stones is a must when changing balance staff, improper lock will cause a beat error. the tooth of the escape wheel should lock no more 1/3 of the way onto the pallet stone and the draw angle should be the same on both stones. Make sure pallet jewels are not chipped, damaged, or loose. In this case pallet jewels may need to be adjusted.

2. Banking Pins.  the safety pin should rest at same angle of degree from the center of the balance pivot jewel on both banks. If the left or right banking pin is allowing the pallet to rest at a higher or lower angle from one banking pin to the other, banking pins will need to be adjusted. I see this to be the case with a lot of pocket watches since in some cases during the watches history a watchmaker will adjust banking pins due to misaligned jewels. Its a quick fix to get a watch running but does not help with accuracy.

3. Hairspring Stud. Did you replaced the hairspring? if so it needs to be vibrated or the spring coils could not be concentric. Did you adjust the roller after or before attaching the balance to the spring? You have to mark the location of the stud on the balance wheel first. You should make all your alignments with balance wheel and cock installed without hairspring. Then while holding the balance wheel with impulse jewel in perfect center of the pins you lay the spring on top of the cock to mark the location of the stud on the wheel. If you did not do this the stud could be in the wrong location, and any adjustments to the impulse jewel after balance being attached will just cause more beat error.

 

Edited by saswatch88

Share this post


Link to post
Share on other sites
  • 0
4 hours ago, saswatch88 said:

ok considering the impulse jewel is dead center between banking pins 3things you should check:

1. Pallet. Check the lock and draw of the pallet stones is a must when changing balance staff, improper lock will cause a beat error. the tooth of the escape wheel should lock no more 1/3 of the way onto the pallet stone and the draw angle should be the same on both stones. Make sure pallet jewels are not chipped, damaged, or loose. In this case pallet jewels may need to be adjusted.

2. Banking Pins.  the safety pin should rest at same angle of degree from the center of the balance pivot jewel on both banks. If the left or right banking pin is allowing the pallet to rest at a higher or lower angle from one banking pin to the other, banking pins will need to be adjusted. I see this to be the case with a lot of pocket watches since in some cases during the watches history a watchmaker will adjust banking pins due to misaligned jewels. Its a quick fix to get a watch running but does not help with accuracy.

3. Hairspring Stud. Did you replaced the hairspring? if so it needs to be vibrated or the spring coils could not be concentric. Did you adjust the roller after or before attaching the balance to the spring? You have to mark the location of the stud on the balance wheel first. You should make all your alignments with balance wheel and cock installed without hairspring. Then while holding the balance wheel with impulse jewel in perfect center of the pins you lay the spring on top of the cock to mark the location of the stud on the wheel. If you did not do this the stud could be in the wrong location, and any adjustments to the impulse jewel after balance being attached will just cause more beat error.

 

I'm a bit confused by what you've said "the safety pin should rest at same angle of degree from the center of the balance pivot jewel on both banks" Do you mean a guard pin when you say safety pin? And when you say balance pivot jewel, how can the guard pin be aligned with what is at the top and bottom of the balance staff in relation to the banking pins?

Share this post


Link to post
Share on other sites
  • 0
4 hours ago, Jon said:

I'm a bit confused by what you've said "the safety pin should rest at same angle of degree from the center of the balance pivot jewel on both banks" Do you mean a guard pin when you say safety pin? And when you say balance pivot jewel, how can the guard pin be aligned with what is at the top and bottom of the balance staff in relation to the banking pins?

Yes guard pin I call it safety pin. A diagram would be better to illustrate what i mean. But when pallet rest on the left bank it should measure the same degree as the other side. Looking at the escapement from the top if you where to draw a half a circle with 90 degree point at center of balance pivot jewel the “guard pin” should run a path of lets say 30 degree then it should be at +15 degrees at one bank and -15 degree on other bank equal on both sides.

Share this post


Link to post
Share on other sites
  • 0
4 hours ago, saswatch88 said:

Yes guard pin I call it safety pin. A diagram would be better to illustrate what i mean. But when pallet rest on the left bank it should measure the same degree as the other side. Looking at the escapement from the top if you where to draw a half a circle with 90 degree point at center of balance pivot jewel the “guard pin” should run a path of 60 degree then it should be at +30 degrees at one bank and -30 degree on other bank equal on both sides. These measurements should be the same from the center of the pallet pivot jewel.

Double post i realized i said 30 degrees but to be actual it should be 60 degrees. This is very tricky stuff and hard to translate to an actual escapement, but with high magnification and a good eye it can be done without any actual measuring.

image.jpg

image.jpg

Edited by saswatch88

Share this post


Link to post
Share on other sites
  • 0

The locking surfaces of the pallet stones span 60 degrees. The fork slot travels approximately 5-6 degrees either side of the line of centers (5ish in a pocket watch 6ish in a wristwatch).  If checking the escapement yes one should check that the guard pin clearance is present and equal on both sides of the safety roller, also the same for fork horn clearance, which should be slightly larger than guard pin clearance, and not allow the pallet jewels to unlock. A quick ballpark method to position the pallet stones is to set them so they are just not unlocking when checking fork horn clearance. 9 times out of 10 this will be sufficient lock and also the minimum lock attainable.

 

But highly unlikely the OP has an issue with the geometry of his escapement, unless the stones are obviously moved (fork horn clearance check can confirm this) or the banking pins visibly bent out of upright.

13 hours ago, saswatch88 said:

3. Hairspring Stud. Did you replaced the hairspring? if so it needs to be vibrated or the spring coils could not be concentric. Did you adjust the roller after or before attaching the balance to the spring? You have to mark the location of the stud on the balance wheel first. You should make all your alignments with balance wheel and cock installed without hairspring. Then while holding the balance wheel with impulse jewel in perfect center of the pins you lay the spring on top of the cock to mark the location of the stud on the wheel. If you did not do this the stud could be in the wrong location, and any adjustments to the impulse jewel after balance being attached will just cause more beat error.

 

Just to clarify here- vibrating a spring gets the length correct. An incorrect spring can be perfectly concentric and level and the watch in beat and a correct spring can be out of round and out of flat and out of beat. Yes you do want to get the stud location correct, but "any adjustments to the impulse jewel after balance being attached will just cause more beat error" is incorrect- you can indeed move the roller table (impulse jewel) to correct for beat error. In some cases it is preferable to moving the hairspring collet and there are tools made just for this.

On 2/11/2020 at 9:48 AM, Len33 said:

 

I centred the balance jewel between the banking pins with  the pallet removed from the watch and checked it umpteen times.

 

 

You want to check the centering of the roller jewel in the fork itself. With no power on the watch, check that the roller holds the fork centered between the bankings. The fork should be pointing directly to the balance center. Even then, if visually "perfect", depending on your perception, you can have several ms beat error. I don't know if your machine has a gain control but adjusting that can help with problem watches. You may have a ghost noise like the hairspring slightly rubbing somewhere or something like that.

 

A pic of the movement with the power off would be helpful.

 

Share this post


Link to post
Share on other sites
  • 0
2 hours ago, nickelsilver said:

I don't know if your machine has a gain control but adjusting that can help with problem watches. You may have a ghost noise like the hairspring slightly rubbing somewhere or something like that.

Yes the Chinese machines all have gain adjustment, in some case it helps with ghost ticks, but in my experience these show on the screen and beat error is not affected. I think the OP should post a screen picture of what he's seeing.

Share this post


Link to post
Share on other sites
  • 0
3 hours ago, saswatch88 said:

Double post i realized i said 30 degrees but to be actual it should be 60 degrees. This is very tricky stuff and hard to translate to an actual escapement, but with high magnification and a good eye it can be done without any actual measuring.

image.jpg

image.jpg

Got it... thanks

 

Share this post


Link to post
Share on other sites
  • 0
3 hours ago, nickelsilver said:

But highly unlikely the OP has an issue with the geometry of his escapement, unless the stones are obviously moved (fork horn clearance check can confirm this) or the banking pins visibly bent out of upright.

. Yes you do want to get the stud location correct, but "any adjustments to the impulse jewel after balance being attached will just cause more beat error" is incorrect- you can indeed move the roller table (impulse jewel) to correct for beat error. In some cases it is preferable to moving the hairspring collet and there are tools made just for this..

 

This is an old pocket watch so issues with banking pins can 100% be the issue i have seen it more often than not on old movements. Bent or previously moved. I would say damage to the pallet jewels if anything but i have seen loose jewels.

you can make adjustments to the roller/jewel after attachment, but its not ideal. Going by the collet is best and I have the tool but this procedure is done with hairspring attached to the wheel but not the cock, and hence “making the adjustment before attaching the balance and spring” which is how i should have worded it. These pocket watch balance wheels are split and have weights and the jewel should rest at 90 degrees between the arms, having the roller turned too far to one side to compensate a misaligned stud can actually cause the jewel to slip out of the fork and lock the escapement. If the amp is at 270 degrees ideally then it should be equal on both ends in relationship to the arms esp. on a split wheel.

Edited by saswatch88

Share this post


Link to post
Share on other sites
  • 0

Of course the roller should be perpendicular to the arms, the adjustments sometimes made to the roller for beat error purposes are minute and unremarkable as far as visual alignment. I'm talking about maybe 1 degree.

 

28 minutes ago, saswatch88 said:

 These pocket watch balance wheels are split and have weights and the jewel should rest at 90 degrees between the arms, having the roller turned too far to one side to compensate a misaligned stud can actually cause the jewel to slip out of the fork and lock the escapement.

It seems you're saying that moving the roller to compensate for a misaligned hairspring (i.e. putting the watch in beat) can cause the escapement to lock up? How?

 

32 minutes ago, saswatch88 said:

Going by the collet is best and I have the tool but this procedure is done with hairspring attached to the wheel but not the cock, and hence “making the adjustment before attaching the balance and spring” which is how i should have worded it.

Ok, I think you're saying to get the hairspring in the right position when installing; this is true of course, but there will always be some fine adjustment and this is made with the balance and cock out but the balance still attached to the cock. At least that is how it is taught in school, and works for 18s pocketwatches down to a microscopic LeCoultre 101.

Share this post


Link to post
Share on other sites
  • 0
16 minutes ago, nickelsilver said:

It seems you're saying that moving the roller to compensate for a misaligned hairspring (i.e. putting the watch in beat) can cause the escapement to lock up? How?

Ok, I think you're saying to get the hairspring in the right position when installing; this is true of course, but there will always be some fine adjustment and this is made with the balance and cock out but the balance still attached to the cock. At least that is how it is taught in school, and works for 18s pocketwatches down to a microscopic LeCoultre 101.

Maybe iam using the wrong nomenclature but.....

Yes if the roller/ jewel is turned too far it will slip out of the fork causing the balance to stop And fork will stay banked to one side and escapement will lock up.

And as you stated a fine adjustment “1 degree” that should be the Margain of error, but if a hairspring is not aligned well prior to assembly it will take much more than just a 1 degree adjustment to get everything to work right. The point iam trying to make is to get everything lined up as close as possible before attaching everything, checking pallet entry and draw with power on prior to installing a balance that has already been adjusted is also good practice. i went way too far into it with angles and geometry but it does play a big part in understanding how to diagnose issues with the escapement and what would need adjusting to solve it.

Edited by saswatch88

Share this post


Link to post
Share on other sites
  • 0

Ignoring poise, the position of the roller only matters when comparing its position to the hairspring stud. 
 

On the later Smiths slimline movements (6046xx) the roller jewel isn’t centred between the balance arms anyway. It’s poised to take this into account. 

Share this post


Link to post
Share on other sites
  • 0

My "thanks" to everyone that contributed to the question.

Here is a picture of the test results that I obtained on my 1500 timegrapher.

I hope it's clear enough,  I took the picture with my ipad.IMG_4260.thumb.JPG.60e439300e9441b318817c08c32eaab1.JPG

Share this post


Link to post
Share on other sites
  • 0
2 hours ago, rodabod said:

Ignoring poise, the position of the roller only matters when comparing its position to the hairspring stud. 
 

On the later Smiths slimline movements (6046xx) the roller jewel isn’t centred between the balance arms anyway. It’s poised to take this into account. 

its a big difference when dealing with split wheels with weights,

Share this post


Link to post
Share on other sites
  • 0

The weishi timegraphers are great but not infallible.

Don't assume that the readings are definitely correct until you've established all other factors with the running of the watch are.

Disregarding the 9.9ms reading currently. (This essentially is as relevant as it reading out of limits). There's a notable rise and drop in b.e. (where you see the lines separate).

Very first thing I'd suggest is remove the balance and check that the roller jewel is centred and secure, I've seen similar results with a loose roller jewel. Still tight enough to fit but with a bit of side to side play.

If that is shellac ed securely. Check the lock and drop of the pallet fork jewels and their security of fit, Check the swing between guard pins is even. (I've seen so many old pocket watches with bent guard pins it's almost become an expected sight).

Check that the roller jewel is centering after a gentle rotation in both directions, check the hairspring is securely pinned and it's flat and reasonably concentric.

Double check this with the balance installed viewed dial up. Dial down and pendant up etc.

If your phone has a slow motion video facility. (Most do these days). Take a short clip of the balance and visually verify that it is oscillating roughly at the amplitude shown. Again don't assume it is 216 degrees, I've seen watches barely pushing 160 where the timegrapher reports a much higher amplitude.

I know a chunk of this is repeated from the above good pointers but assuming you've checked all that and jewels are secured, endshake and side shake are OK, motion is acceptable and hs are OK then this should see a consistent line gap and a b.e. of 3.0 ish ms at the most.

Also be sure to check the timegrapher reading with the watch on other positions especially the love you checked the roller jewel visually in, DD? I'd assume. See if there's a noticeable difference.

 

Edited by m1ks

Share this post


Link to post
Share on other sites
  • 0
36 minutes ago, m1ks said:

Very first thing I'd suggest is remove the balance and check that the roller jewel is centred and secure, I've seen similar results with a loose roller jewel. Still tight enough to fit but with a bit of side to side play.

YESSS very important! Great input MK.

this is why i never rely on machinery I have a timegrapher, but always do the slow motion video, i actually have a video editing software where i have all the geometry mapped out. I import the video under the template and when i slow it down i can see the amplitude very precisely.

Edited by saswatch88

Share this post


Link to post
Share on other sites
  • 0

Before ripping your watch apart and doing all those checks you were supposed to do in the first place can you do the visual check of the pallet fork between the banking pins without power? Then yes for everybody else it's not a perfect test but it still should be reasonably close versus the goofy numbers the timing machine is showing. It's important for people to learn what to look for. In other words the classic check for seeing if your watch is beat is the look at the pallet fork at rest because it gets you really really close.

Timing machines are interesting devices for diagnosing adjusting watches etc. but they rely on certain things. They have to have a good clean signal and the watch has to be running correctly. Then it's always interesting to see how the timing machine attempts to communicate that it's having a problem. Typically the Chinese machines will attempt to give you something no matter what with zero indication that's having a problem. The number 9.9 is a clue that it's having an problem. The expensive switch's witschi machines will usually get mad and give you an error message and tell you you're either out of range or some other silly message. You can usually get around that by changing the parameters but they can still be fooled. One of my favorites is if the amplitude is too low the timing machine will read the wrong part of the waveform and show you a nice amplitude that's not there at all. 

 

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Answer this question...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...

×
×
  • Create New...