Should I bother with this beat error?

[thor447]

Hello and good day everyone.  I'm still getting my feet wet with watch repair, and would appreciate your advice regarding some timegrapher readings of a service I completed yesterday.

I bought an older Jerome Piquot watch off of eBay.  I know nothing about the brand, I just thought the watch looked unique and thought it'd be fun to rebuild.  Considering I paid under $25 for it, I figured if I botched it completely it wouldn't turn out to be an expensive way to learn a lesson in watch repair.   It has an AS 1686 movement.  The service went really well yesterday and it has been running in for around 9 hours at this point.  I'm getting a beat error of 0.8ms.  This AS 1686 does not have an adjustable beat error stud, I know that I would have to remove the balance wheel and adjust it manually if I wanted to.  I'd like to avoid that if possible and just wanted to call upon the experience of here to see if 0.8ms is something they'd try to improve upon on this watch.

Here's dial up running at +5 to +6 s/d, and has a very strong amplitude considering I didn't replace the mainspring, but I was meticulous with the cleaning and did polish the pivots with my recently acquired Jacot tool and remove a bit of extra side shake in the mainspring barrel.

 

Here's dial down - running at -2 to -3 s/d with a slight amplitude drop from the previous position.

 

I feel pretty good about the service, and was encouraged to see single digits after a short period of regulation.  Would you bother with the 0.8ms beat error in this 60 year old movement?  Also, does the amplitude drop of 21 degrees between dial up and dial down seem within reason to you?  There is a bit more end shake in the 4th wheel than I'd ultimately like, but it is not so much that it would cause any part of the wheels to not be in contact with one another.  It just meets the escape wheel not in the center of it's pinion, but it is still in full contact throughout all end play testing between both the wheels.  As of right now though, I do not have a jeweling/Seitz tool needed to adjust for that.  I think it is as good as I can get it with my tools & skill level.  I'd appreciate any feedback you may have.  Thanks.

 

[nickelsilver]

Companies I've done work for on both new builds and vintage repair stipulated a beat error under 1ms on watches with fixed stud holders (in 6 positions, haha). I recall in some Vibrograf literature, written before the advent of digital timing machines, they said anything under 2ms will have essentially zero effect on the running. So you look fine there!

[thor447]

1 minute ago, nickelsilver said:

Companies I've done work for on both new builds and vintage repair stipulated a beat error under 1ms on watches with fixed stud holders (in 6 positions, haha). I recall in some Vibrograf literature, written before the advent of digital timing machines, they said anything under 2ms will have essentially zero effect on the running. So you look fine there!

Good information to know.  Thank you!

[Shane]

I would say that .8ms, your worrying about nothing. 

On the other hand, 315 degrees, especially with a used mainspring,  sounds a bit high.  Are you sure that you have the correct lift angle entered?

I do not think that it would have any effect on time keeping but it would be nice to know if the information it was providing was accurate.

Good luck. 

Shane 

Edited September 20, 2022 by Shane

[praezis]

3 hours ago, thor447 said:

Also, does the amplitude drop of 21 degrees between dial up and dial down seem within reason to you? 

I suppose, your pivot polishing caused that difference. If a pivot end of the balance staff is falsely too high domed, this will happen in the corresponding flat position. How is your amplitude in vertical positions?

3 hours ago, thor447 said:

and just wanted to call upon the experience of here to see if 0.8ms is something they'd try to improve upon on this watch.

If you want to know, if xx ms is ok or not, that number does not yet tell too much (0.8 however will be ok in any case).
Essential is, how many degrees out of the middle is the balance in rest? That is the reason of the beat error.
The very same error of 5 degrees will show:
1,3 ms @ 28800, 300° Ampl. (B.E.= 5°)
2,1 ms @ 18000, 300° Ampl. (B.E.= 5°)
3,6 ms @ 18000, 180° Ampl. (B.E.= 5°)

That is why I regard the beat error in degrees only, which my PCTM shows. 2 degrees and lower will be near to perfect. Your 0.8 ms corresponds to 2.0 degrees!

Of course you cannot expect such sophisticated measurements from Weishis and Witschis 

Frank

[mikepilk]

5 hours ago, nickelsilver said:

Companies I've done work for on both new builds and vintage repair stipulated a beat error under 1ms on watches with fixed stud holders (in 6 positions, haha). I recall in some Vibrograf literature, written before the advent of digital timing machines, they said anything under 2ms will have essentially zero effect on the running. So you look fine there!

For watches with fixed stud holders, 2ms is my "reference point" as to whether to adjust or not. 

I think people get too fixated with beat error.

BTW the correct lift angle for the AS 1686 is 52° - which is the default on that timegrapher.

[thor447]

I just googled the lift angle for the 1686 and found it to be 52°, so I left the machine on it's factory setting.  Great info from everyone, I learned a few new things.  Thanks!

[JohnR725]

Rolex is another interesting company for timing specifications. All sorts of timing specifications fully wound up half wound even have paperwork you're supposed to fill out but when it comes to the beat error that's pretty simple. A maximum of 0.8 ms test in any one of the five positions they normally test.

 

 

[Neverenoughwatches]

11 hours ago, nickelsilver said:

5 hours ago, mikepilk said:

they said anything under 2ms will have essentially zero effect on the running. So you look fine there!

 

I must be being very hard on myself in this case,anything over 0.3 and I'm starting to feel agitation .

8 hours ago, praezis said:

I suppose, your pivot polishing caused that difference. If a pivot end of the balance staff is falsely too high domed, this will happen in the corresponding flat position. How is your amplitude in vertical positions?

If you want to know, if xx ms is ok or not, that number does not yet tell too much (0.8 however will be ok in any case).
Essential is, how many degrees out of the middle is the balance in rest? That is the reason of the beat error.
The very same error of 5 degrees will show:
1,3 ms @ 28800, 300° Ampl. (B.E.= 5°)
2,1 ms @ 18000, 300° Ampl. (B.E.= 5°)
3,6 ms @ 18000, 180° Ampl. (B.E.= 5°)

That is why I regard the beat error in degrees only, which my PCTM shows. 2 degrees and lower will be near to perfect. Your 0.8 ms corresponds to 2.0 degrees!

Of course you cannot expect such sophisticated measurements from Weishis and Witschis 

Frank

I'm  thinking power reserve would be a little less with a higher beat error than acceptable. The balance wheel would want to come to rest at a sooner time period ?

49 minutes ago, JohnR725 said:

Rolex is another interesting company for timing specifications. All sorts of timing specifications fully wound up half wound even have paperwork you're supposed to fill out but when it comes to the beat error that's pretty simple. A maximum of 0.8 ms test in any one of the five positions they normally test.

 

 

Five positions ?  Do they leave out DD ?