Positional beat error delta and its indications

Y.M1619 · Dec 23, 2021

Hi forums, I wish you all a fantastic holiday!

As I purchased a timegrapher recently, trying to deepen my understanding of my watches, I realized how little I know about beat error.

I have got mixed information about BE from the internet but my current belief is that I wouldn’t worry about BE number within 0.8ms for a modern watch unless it changes dramatically which indicates issues that would continuously damage the parts and worsen the movement performance. (Correct me is I am wrong, folks)

I have noticed a quite significant positional BE delta on my Seamaster 300m(with movement 8806), and I can’t tell if it indicates a problem. So I would love to learn from you guys about that.

The thing is, on dial up position, the BE is 0.1ms which is quite nice. But when I turn it to crown right position, the BE will suddenly expand to 0.5(or 0.6 when the power reserve runs low). Some say it indicates wear on balance wheel pivot. But I would be surprised if my 2019 Seamaster started to show wear and tear already.

I will attache a few timegrapher data to the post. Let me know how should I understand the positional delta. Many thanks!

Canuck · Dec 23, 2021

Best way to not enjoy your watch is to invest in a timing machine. How much closer to perfect are you wanting the watch to be?

M'Bob · Dec 24, 2021

I guess your question is, what's an acceptable beat error margin positionally? My sense is that yours is okay.

Leviathan · Dec 26, 2021

How does it run on your wrist? Ultimately, that’s all that matters.

Myron · Dec 26, 2021

Canuck
Best way to not enjoy your watch is to invest in a timing machine. How much closer to perfect are you wanting the watch to be?

^^^ This ^^^. I recently sent my kiddo's Hamilton Scuba into SGUS for a well-deserved service. It had been worn for 4 years on his wrist 24/7 and was starting to lose significant time. The service cost $270 (as I recall) and the watch came back keeping ok time but with crappy beat error (0.8). I ended up deciding that without my silly T I M E G R A P H E R machine I never would have even know this, and would've been happy with the value I received for my service dollar since on the wrist the thing runs ok.

Gruesome · Dec 26, 2021

I think you have to wait longer to get valid measurements. I usually wait until all the numbers on the screen (~10 second updates) are within one digit. Also, I find that I actually have to stop and restart my Weishi 1900 to get it to reset properly. It does not seem to compute a running average properly.

wsfarrell · Dec 27, 2021

If you're not interested in working on your watch, the Timegrapher could be frustrating. If you'd like to open up your watch and tinker with it on the other hand, a Timegrapher comes in very handy. Get yourself a case opener, tweezers, and some nice screwdrivers, and you can have a lot of fun adjusting the timing of your watch, minimizing beat error, removing the movement so you can clean up the case, etc. etc. It's a very rewarding hobby.

SkunkPrince · Dec 27, 2021

I use my timing machine less these days. I have habits that take watches that "should not keep time" and they do, on my wrist.

If I know I need +5 seconds a day, the machine helps a lot with that.

Y.M1619 · Dec 27, 2021

Thank you so much friends for your opinions! I totally not disagree that a timegrapher is not exactly helpful for the most of watch enthusiasts. It could give people unnecessary anxiety about minor imperfections in their watches.

But in my case, I am trying to extending my hobby from enjoying owning and wearing watches to understanding how things works inside a watch and eventually adjusting and servicing them. Buying this timegrapher and a set of screwdrivers, tweezers, etc. is the first step of my new adventure. I will be hunting for cheap old watches on eBay to practice on in the coming weeks.

I asked the question what does the positional beat error variations mean, not because the numbers get me worried(I am not paranoid enough to think this BE will kill the watch LOL) Instead, it is more of my curiosity about how those beautiful tiny machine work.

Again, thank you so much for everyone replied to this thread! I wish I had explained my ask clearer.

felsby · Dec 27, 2021

I tend to focus more on amplitude than on beat error when determining if service is needed. The collection is mostly watches from 1945-60 and most of the watches was used daily; no safe queens, so I sleep good at night if the beat error is 6ms 😁

STANDY · Dec 27, 2021

So you bought the thing you use first to see if you have to take it apart and last to see if you put it together right.

Makes sense. 👍

ChrisN · Dec 27, 2021

To understand better what a timegrapher does, have a look at the first 15 pages of this document:

https://www.witschi.com/assets/files/sheets/Witschi Training Course.pdf

Witschi are the experts in this. I would say that unless your timegrapher has a selectable option in the menu that is specific to the co-axial movement then then any results you get are not going to be 100% correct for all variables. I doubt your machine can do a co-axial correctly regardless of what you've been told.

The beat error variations that you are seeing are usually just down to gravity. Omega typically allow up to 0.8ms as you'll be unlikely to get the same value in every orientation.

I should note that you're not putting the watch in the correct way for the measurement. The crown should always be against the microphone end (opposite the sprung slider) and this may give more consistent results as the microphone will hear the movement more clearly. Wait 30 seconds after turning the microphone so that the movement settles down.

For an older watch where the balance doesn't have a moveable stud, I correct beat error by measuring it with two opposite configurations where the balance is directly left and directly right of the pallet. With these two values, when I remove the balance to turn the collet (that holds the hairspring to the balance), I know which way to turn it - this method is using that gravitational effect.

As others have said, you can make yourself nervous about your watch with these machines so bear that in mind.

Good luck, Chris

Y.M1619 · Dec 27, 2021

ChrisN
To understand better what a timegrapher does, have a look at the first 15 pages of this document:

https://www.witschi.com/assets/files/sheets/Witschi Training Course.pdf

Witschi are the experts in this. I would say that unless your timegrapher has a selectable option in the menu that is specific to the co-axial movement then then any results you get are not going to be 100% correct for all variables. I doubt your machine can do a co-axial correctly regardless of what you've been told.

The beat error variations that you are seeing are usually just down to gravity. Omega typically allow up to 0.8ms as you'll be unlikely to get the same value in every orientation.

I should note that you're not putting the watch in the correct way for the measurement. The crown should always be against the microphone end (opposite the sprung slider) and this may give more consistent results as the microphone will hear the movement more clearly. Wait 30 seconds after turning the microphone so that the movement settles down.

For an older watch where the balance doesn't have a moveable stud, I correct beat error by measuring it with two opposite configurations where the balance is directly left and directly right of the pallet. With these two values, when I remove the balance to turn the collet (that holds the hairspring to the balance), I know which way to turn it - this method is using that gravitational effect.

As others have said, you can make yourself nervous about your watch with these machines so bear that in mind.

Good luck, Chris

Thank you so much Chris! Your post is very informative and the Witschi training course material is really helpful for me. Appreciate it!

I do agree that my entry-level timegrapher(a Witschi 1900) may not be able to measure an co-axial 100% correctly. It doesn’t have an option in the settings for co-axials. I can only set the lift angle.

Anyways, I think it’s time for me to move on. Thanks everyone posted in this thread!

SkunkPrince · Dec 27, 2021

ChrisN
I should note that you're not putting the watch in the correct way for the measurement. The crown should always be against the microphone end (opposite the sprung slider) and this may give more consistent results as the microphone will hear the movement more clearly.

This has always confused me. On the two machines I've had, the slider has a cutout for the crown and the metal piece seems shaped for the case. With the gain midway, I almost always have plenty of amplification. Turning it the other way is a lot less stable and I actually had a watch fall off the microphone dial-down.

Joe_A · Dec 27, 2021

One of the industry standard precision machines is made by Witschi, a Swiss company.

Weishi sounds a bit similar, but is an inexpensive machine made in China and also sold under other name brands. One should not confuse the two.

I have a Weishi timegrapher and I use it fairy often as a tool to do the following things:

* Provide an opportunity to judge the overall health of a watch.
* Help me to decide when sending a watch out for service will be a worthwhile investment.
* Benchmark the operational parameters when I first receive a watch.
* Confirm whether a good job was done when the watch was sent out for service and has been returned..

There are a number of good videos to watch.

Here's one I watched for this post. it refers to other videos which discuss beat error and indications that a watch may have been magnetized:

I won't comment concerning your watch in particular as I am no watchmaker 😉

However . . .

Generally, an amplitude of 300 or above in the dial up or dial down position is excellent for most watches. Anything over 250 degrees is acceptable. One should expect a drop when the watch is vertical. Order of magnitude perhaps 30 to 50 degrees?

Beat error below 1.0 msecs is generally acceptable though below 0.5 msecs is ideal.

If the watch has a mobile (adjustable) stud carrier one can get the beat error below 0.5 msecs for a healthy watch. With a fixed stud carrier, I pretty much accept anything at or below 1.0 msec as I am not qualified to fiddle with the collet that allows adjustment in the case of a fixed or immovable stud carrier. One of these days I will give it a go.

I can't say for certain whether a trained watchmaker will agree with everything I've said; I am always open to correction.

In evaluating the health of a watch I look for the following things:

A relatively small difference between the beat rate of a watch over the six most commonly considered positions:

Dial Up
Crown down (especially for people who wear the watch on the left wrist)
Crown up (especially for people who wear the watch on the right wrist.)
Crown to the right
Crown to the left
Dial down

Since you adjust the watch with the dial down, it's nice to be able to predict what an adjustment there will do to the results of the other positions.

I have often been accused of over-thinking things . . . and for good reason. 😀

Edit: Corrected the right-handed, left-handed positions.

sleepyastronaut · Jan 26, 2022

ChrisN
The crown should always be against the microphone end (opposite the sprung slider) and this may give more consistent results as the microphone will hear the movement more clearly. Wait 30 seconds after turning the microphone so that the movement settles down.

Thanks very much for this info, Chris, it's extremely helpful. I've been using an Android app to check accuracy, and the instructions did not mention putting the crown to the mic.

justacpa · Jan 26, 2022

Joe_A
One of the industry standard precision machines is made by Witschi, a Swiss company.

Weishi sounds a bit similar, but is an inexpensive machine made in China and also sold under other name brands. One should not confuse the two.

I have a Weishi timegrapher and I use it fairy often as a tool to do the following things:

* Provide an opportunity to judge the overall health of a watch.
* Help me to decide when sending a watch out for service will be a worthwhile investment.
* Benchmark the operational parameters when I first receive a watch.
* Confirm whether a good job was done when the watch was sent out for service and has been returned..

There are a number of good videos to watch.

Here's one I watched for this post. it refers to other videos which discuss beat error and indications that a watch may have been magnetized:

I won't comment concerning your watch in particular as I am no watchmaker 😉

However . . .

Generally, an amplitude of 300 or above in the dial up or dial down position is excellent for most watches. Anything over 250 degrees is acceptable. One should expect a drop when the watch is vertical. Order of magnitude perhaps 30 to 50 degrees?

Beat error below 1.0 msecs is generally acceptable though below 0.5 msecs is ideal.

If the watch has a mobile (adjustable) stud carrier one can get the beat error below 0.5 msecs for a healthy watch. With a fixed stud carrier, I pretty much accept anything at or below 1.0 msec as I am not qualified to fiddle with the collet that allows adjustment in the case of a fixed or immovable stud carrier. One of these days I will give it a go.

I can't say for certain whether a trained watchmaker will agree with everything I've said; I am always open to correction.

In evaluating the health of a watch I look for the following things:

A relatively small difference between the beat rate of a watch over the six most commonly considered positions:

Dial Up
Crown down (especially for people who wear the watch on the left wrist)
Crown up (especially for people who wear the watch on the right wrist.)
Crown to the right
Crown to the left
Dial down

Since you adjust the watch with the dial down, it's nice to be able to predict what an adjustment there will do to the results of the other positions.

I have often been accused of over-thinking things . . . and for good reason. 😀

Edit: Corrected the right-handed, left-handed positions.

Thanks for that video. One day soon I'll buy a timegrapher and put it to use.

Positional beat error delta and its indications

Y.M1619

Canuck

M'Bob

Leviathan

Myron

Gruesome

wsfarrell

SkunkPrince

Y.M1619

felsby

STANDY

ChrisN

Y.M1619

SkunkPrince

Joe_A

sleepyastronaut

justacpa

Similar threads

Omega Cal 564 Beat error

The X-33 Flightmaster Prototype and its mission to MIR

DSoTM Apollo 8 has large Beat Error

Horology 101: Positional Error

A Unique Vintage Longines, and its Connection to One of the “Princes of Precision”

Beat error at 12 o'clock up on 1861...