Setting the Seconds Hand

[SloJoe]

The encyclopedia article Setting Watch Hands is clear on how to set the hour and minute hands on a pocket watch, but how about the seconds hand? Since the setting mechanism has no effect on the seconds hand, is it permissible to move it directly with a finger or toothpick? Or is this not advisable? Is there any way to set a pocket watch any more accurately than to within half a minute?

 

[John Cote]

You can stop the balance wheel with a tooth pick or something like it. I would not really advise this if you don't have much experience with watches. This is how a hacking watch (a watch which stops when in setting mode) works.

 

[Kent]

Except for a relatively small number of pocket watches made with a hacking feature for military use (and there might have been a small number made for civilian use), there isn't a 'good' way to set the second hand.

Nevertheless, the following method will do the job. It is NOT recommended for the novice as it risks damage to the staff or balance if not done properly, USE CARE.

With the back open and a good time standard where it can be seen (or heard - if you're tuned in to WWV), use a toothpick to GENTLY stop the balance when the second hand is at 60 seconds. Release the balance when the time standard indicates the exact minute.

If you wish to avoid risk to the balance or staff, there is another way, but this risks breaking the forth wheel pivot (the second hand post) and/or scratching the dial. It was told to me by a young (at the time) watch inspector's assistant, Ted Huguelet. His narrative of 1940s practice contained the following. "I checked his time to the minute and second against our own master timepiece, setting the seconds hand with the point of my tweezers."

I used to use the former method. Nowadays, I'm satisfied to have my watch display the time +/- 30 seconds.

P.S. I see John responded while I was typing.

Good luck,

 

[SloJoe]

Thanks guys, guess I'll just not sweat the seconds.

 

[Luis Casillas]

To put this into historical context, it's worth pointing out that the ability to set the seconds is more a matter of convenience than one of precision timekeeping. Historically, the most precise mechanical timepieces (e.g. chronometers, astronomical regulators) routinely lacked seconds-setting mechanisms, or even a mechanism for precisely regulating the rate.

Precise timekeeping was achieved by periodically comparing the timepiece to a time standard (e.g., the stars), recording the how much it was off, and using math to apply these values as corrections to the times you read off the clock. Once you're going to these lengths to obtain precise time, the ability to set the seconds hand doesn't really do much for you.

In fact in this context it may even make things worse. Many such mechanisms interfere with the stability of the balance, which means that they detract from timekeeping precision. And if you reset the time you then need to record when you did so, the error before and after the reset, and apply these values as a further correction to your data. I.e., you end up doing more work for little or no benefit!

Settable seconds, as I understand it, first became a common feature in aviation and military timepieces, because in those situations you don't generally want to be doing a lot of math. Settable seconds allows you to synchronize your watch to a master clock (or to your colleague's watches, as in the movies) and then be confident that it won't drift too far off during the next day or so.

There's probably some interesting history here also about public reference clocks—accurate clocks that were put in public places for people to set the time by—but I'm not the one to tell it. I do know tidbits like this one: Big Ben is kept accurate by measuring how far off it is from the correct time and adding/removing weights from its pendulum to alter its rate and thus steer it toward the correct time.

And for ordinary laypersons' timekeeping needs, seconds don't really matter, so old watches just don't have anything for this.

 

[Tom McIntyre]

The use of hack mechanisms for military maneuvers and for short term time transport shows up in the English "Center Seconds Chronograph" widely used in the 19th century. Earlier verges also had a hack mechanism but, I believe, that was used for setting since the setting mechanism was relatively crude.

Deck watches also frequently have a hack mechanism to simplify carrying the time from the chronometer to the deck for observations. It was not really needed there since it was just as easy to note the offset when taking the time from the chronometer to do the sighting. It was likely considered more convenient and thus less likely to be overlooked by someone in a hurry. That has always seemed a bit odd to me.

 

[musicguy]

Setting second hand....

I wanted to use Tom's method of using a paint brush to stop
the balance wheel when the second hand is at 60(great idea), but couldn't find
one(my kids must have one somewhere, and I know I must have one too).
I would never push the second hand, so I used the rubber band method. I
usually don't care +- 30 seconds(or even a minute) but I was regulating my
1928 BW Raymond for the last few weeks(no timing machine) I kind of got caught
up in it's accuracyrazz.....and it's been very enjoyable regulating this particular watch.
My older watches I want to be functional, and serviced, but since they
mostly sit in my cabinet +/- 1-2 minutes doesn't bother me.

It is interesting that watches that were made in 1928 (that were
approved for RR use) didn't have a hack to adjust the second hand.
I know it's very easy to do without a hack, but it would seem to be a good idea.

I know when I was a kid, I probably just put my finger on the balance wheel.
That watch didn't last too long.

Rob

 

[pmwas]

"I checked his time to the minute and second against our own master timepiece, setting the seconds hand with the point of my tweezers."

Interesting. In Glashutte watch museum, if I remember right, I saw a movie (to clarify - I did see the movie, I'm just not sure it was in Glashutte ), where - to my amazement - the watchmaker assembling a watch put the second hand on the pinion and just after he moved it to 60s position with the same tweezers.

I think if the second hand is correctly chosen, it should not be tight enough to damage the pinion by moving it. It should be able to rotate around the pinion. However you don't know if it's not too tight or jammed until you try.
Second - if you hold close to the tip, you are more likely to break the hand than the pinion, which is better of the two possible outcomes.

And... I think you can also stop the movement by just stopping the second hand. Place a toothpick in it's way and the movement will just stop. This happens when the hour hand is low or second hand high and the hands meet about 5:20. The second hand touches the hour hand and the watch stops. It should not damage the watch or the hand and I don't think it could dangerous at all... Unless you press and the tool slips damaging the hand or dial.

 

[Paul Raposo]

And... I think you can also stop the movement by just stopping the second hand. Place a toothpick in it's way and the movement will just stop.

On the original Glycine Airman the hack worked when you pulled out the crown and thin wire would stick up from a hole under the 12 o'clock marker and stop the second hand, which would stop the movement, so I think using a toothpick or a piece of peg wood is a good suggestion.

 

[Tom McIntyre]

If someone wanted to build a really precise hacking watch, it would need to bring the minute hand to top dead center and the same with the second hand. It should also move the hour hand to the nearest hour mark. The last requirement is a bit hairy but otherwise the hands are out of register. A flying squad would all gather at the standard clock and set and hack their watches together.

Probably no one really cared about register if it was not grossly off, but the second and minute both being at 12 and 60 was important on close timing.

Cockpit timers set time of flight independently from the civil time and that is all that gets synchronized between instruments.

 

[Paul Raposo]

If someone wanted to build a really precise hacking watch, it would need to bring the minute hand to top dead center and the same with the second hand. It should also move the hour hand to the nearest hour mark.

Doesn't Lange have a modern watch that does this? Very interesting feature--for $20,000.

 

[Dr. Jon]

More on setting pocket watches to the second.

On some pocket watches the cannon pinion is tight enough that pressure to set backwards will stop or even back the train.

A safe approach is let the watch run down and wind is about 1/2 turn and shake it, gently, to start it when the true time is where the second hand is. Then wind it fully.

Hacking has come and gone several times.


Many verge watches lack maintaining power so they stop and even back a bit when wound.

Many early English levers had hack capability even without center second hands. One of mine stops the seconds wheel.

Moving the seconds hand is risky and tricky. If you are the one who serviced it you know how tight the hand is on the pivot. If it is not too tight it is safe to do but unless you have had the hand off and placed it yourself, backing it risks breaking that pivot which is the worst one to replace.

I believe a toothpick blocking the hand is safe provided you do not slip and push the hand backwards, or forwards.

And, yes Lange has a hacking feature and I believe it sets several hands.

 

[musicguy]

whatever method I use within a day my second hand will be wrong when comparing
with official "govt" time, but the seconds will match the minute hand
which is nice.


Rob

 

[Tom McIntyre]

It is a fools errand to try to track civil time. Precision watches have always depended on stability of rate. If you know your gaining or losing rate and it is perfectly stable, you have a perfect watch and can always calculate the correct civil time.

 

[roughbarked]

I've never worried. Apart from being fastidious about chronograph hands, why bother?
No customer has ever complained about where the seconds hand is when they check where the minute hand is.

 

[novicetimekeeper]

I usually set longcase to the second after winding, well if the have a second hand, some of mine don't even have a minute hand. I always set my work wristwatch to the second too as it has a hack.

However it was interesting to me that when I bought myself a more expensive watch it didn't have a hack. The manufacturers say if they wanted it to have a hack it would, but they don't think it necessary or desirable.

I was surprised, but I've got the hang of that now. It's more accurate than my watch with a hack anyway.and I've got used to not worrying about it.

 

[musicguy]

why bother?

Like the mountain climber said, because it's there.

Do I really care, no.

Rob

 

[musicguy]

It is a fools errand to try to track civil time.

I'm a fool sometimes. When I was young(and a runner) I had a Casio watch
that I would try to set to the exact time (using the telephone time phone number).
I was always frustrated that this new Casio digital watch couldn't keep
what I thought of(then) as exact time. My friends and I would
synchronize our watches, and the watches would never match up, even a few days later.
As you say above, "it is a fools errand" but I still chase it when I regulate a
mechanical watch.



Rob

 

[Accutronica]

I'm still trying to figure out why anyone would care about a few seconds off with a PW...

Robert

 

[musicguy]

I'm still trying to figure out why anyone would care about a few seconds off with a PW...

Robert

Tell that to the RR worker who had his watch rejected at the RR inspector office

 

[Accutronica]

Tell that to the RR worker who had his watch rejected at the RR inspector office

I meant now-a-days. If a couple of seconds mattered that much I wouldn't be depending on a PW in the first place.

Robert

 

[musicguy]

I was kidding.

Rob

 

[novicetimekeeper]

I meant now-a-days. If a couple of seconds mattered that much I wouldn't be depending on a PW in the first place.

Robert

It's quite fun when all the clocks in the house start striking at the same moment, if a little deafening.

 

[Accutronica]

It's quite fun when all the clocks in the house start striking at the same moment, if a little deafening.

I don't have any clocks. It seems like many youtube videos are done when all of their clocks will sound also. But that's better than listening to an annoyingly loud squawking bird.

Robert

Robert

 

[topspin]

It was quite fun on Sunday, walking into the local watch & clock shop to tease the guy about all the watches & clocks he'll need to reset because of BST.

I have several PWs on my desk at work - some much better timekeepers than others. Instead of worrying about how much they might diverge by over a given period, I like to play a different game... I know which ones run fast and which ones run slow, therefore the game is to set them all on Monday morning such that by Friday afternoon they'll all be telling the same time.

 

[musicguy]

therefore the game is to set them
all on Monday morning such that by Friday afternoon
they'll all be telling the same time.

I like that.



Rob

 

[Tom McIntyre]

Looking at the 200 year old duplex sitting on the table next to my chair, I decided to determine if the minute and second hand were synchronized. It is easy to determine and almost as easy to correct without touching the second hand.

The trick is to watch the minute hand move across a minute on the dial when the minute hand is on the mark, the second hand should be at 60. On this particular watch, it is consistently 15 seconds late getting there. To correct it, the minute hand needs to be moved forward by 1/4 of the mark. It is easy enough to do that you can do it a few times to get it right.

Of course, the watch has gained 17 minutes since last night, but the hands are still in synch (since that is a function of the gearing, not the timekeeping).

 

[GeneJockey]

It is a fools errand to try to track civil time. Precision watches have always depended on stability of rate. If you know your gaining or losing rate and it is perfectly stable, you have a perfect watch and can always calculate the correct civil time.

And unless I'm mistaken, even most high quality, adjusted movements will vary slightly in rate over the course of 24 hours, so you do your final regulation based on performance over 24 hours. COSC standards only require the movement to stay within a very narrow range, rather than not to vary at all.

I wonder how much of the desire to set to the exact second, rather than +/- 30 seconds, is driven by expectations from quartz watches.
You could drive yourself mad, trying to keep a mechanical timepiece absolutely dead on with NIST, or any other atomic clock.

 

[GeneJockey]

The other thing that will drive you crazy? Trying to synchronize the second hand and minute hand on a rectangular wrist watch with no minute chapter, only hour markers. I try to do that only at :00, :15, :30, and :45!

 

[musicguy]

I wonder how much of the desire to set to
the exact second, rather than +/- 30 seconds, is driven by expectations
from quartz watches.

Accuracy using the second hand has nothing to do with Quartz with me.



Rob

 

[Tom McIntyre]

I think it is a matter of cognitive dissonance. If the hands are not synchronized, the watch looks wrong to some people.

The hour hand is a worse offender in that synchronization is always very visible. I have a curtain blocking the view of the washer and dryer in my downstairs bathroom which has pictures of clock dials on it. The artist who drew the pattern must never have seen a clock since all of the clock dials (6 unique ones) are out of synchronization.

 

[hamiltonrrwatch]

There is another way. Less scientific but hey, if it works?

 

[musicguy]

I think it is a matter of cognitive dissonance.

For me(today) not cognitive dissonance, but getting a RR watch accuracy
to be within 30 seconds a week(US official time) . And if the second hand is
synchronized with the minute hand, it's much easier to regulate.

Currently, I'm at plus 4-5 seconds fast a day on my daily carry. I don't think I
can dial it in any closer. I would rather it be a tad fast, than slow.

I realized it was 12:00pm so I added this photo maybe I have
cognitive dissonance


Rob

 

[GeneJockey]

Good job getting the pic in the precise 1/5 second when the second hand was at 60!

If you have a timing machine, and your watch runs reliably 5 sec fast, regulating it closer should be easy. Without one? Trial and error!

I see it's an Elgin, my favorite! - got a shot of the movement?

 

[musicguy]

GeneJockey:

I do not have a timing machine, but I recently downloaded a watch timing
application on my phone that does a rudimentary job. (The app is called Hairspring)

Mostly I just look at US official time everyday and adjust the regulator as needed.
After a while, I usually get fairly close.

Elgin PW's are also my favorite.
I checked out your Elgin Blog, very nice!
The watch pictured above is a 1928 Grade 478 B W Raymond.


Rob

 

[musicguy]

If you have a timing machine, and your watch runs reliably
5 sec fast, regulating it closer should be easy.

I wasn't going to regulate it anymore, but when you said, "regulating it closer should be easy"
I decided to give the regulator screw another tinny nudge toward the slow side.
With my luck it will now be running too slowly. It reminds me of trying to adjust our old
turntable.



Rob

 

[musicguy]

I've regulated it down to 2 seconds a day(or less).




Rob

 

[musicguy]

My watch seems to be almost exact now. It's running 5 seconds faster
than official Govt. time, but not gaining/loosing anything per day(so it's consistent).
After two weeks(or more) it has not lost/gained any time.
The only negative that I see is that the regulator is now set fairly close to F(fast) and the watch
was serviced. As a daily wear watch, I am very happy with it.


Rob

 

[musicguy]

Just for fun....
I re-set this watch about 18 minutes ago to 11:57 AM and I did not
set the second hand but when it was getting close to
12:00 I realized all three hands would indicate exactly(very close) 12:00pm





Rob

 

[Tom McIntyre]

That's cool! So they line up roughly every 5 years plus or minus a few months.

 

[Rodney Leon]

Another one that has hacking is the Seikosha Railway Watch, This one shown was made September 1958, Is the type 19RW model. When the crown is pulled up the second hand on the sub-dial will continue around and then stop at the 60 second mark. This hacking functionality was first introduced in 1955, and makes it easy for railway workers to easily sync the watch to a standard time. All Seiko railroad pocket watches have a hacking feature by pulling the crown up at the 12 set the hand and push the crown back to start. Quartz and manual wind. I have many in the collection.

 

[Kent]

... This hacking functionality was first introduced in 1955, and makes it easy for railway workers to easily sync the watch to a standard time. All Seiko railroad pocket watches have a hacking feature by pulling the crown up at the 12 set the hand and push the crown back to start. ...

Rodney:

That's a good story, but railroaders weren't supposed to set their watches. That would defeat the purpose of many watch features (to ensure accuracy) and for having the watch inspector track the error.

 

[Rodney Leon]

That's a good story, but railroaders weren't supposed to set their watches. That would defeat the purpose of many watch features (to ensure accuracy) and for having the watch inspector track the error.

Thanks Kent, I got some information from a website called Plus9Time and from various Seiko sites. I think they must have had a different standard for watches then we had but I do not know for sure. I have 15 in the collection most for Anniversaries and Commemorating of train stations. I know they have standard clocks at the stations that are all synced. And the Drivers and conductors set watches from them. The older Seiko watches I have, have only a couple of watchmaker marks in the back for cleaning. I do not know what kind of inspections they had to follow. Something to look into. I do know the train Engineer use to have a place built into the dash for the watches they carry. (See photo.) That was years ago so I am sure they do it different now but don't know. I will have my brother check it out next time he is in Japan. Seiko is still making Railroad Pocket watches.