Western Union Clocks
Mike Riddle
Western Union Clocks. These were typically installed in train
stations and other public places. They were electrically powered,
self-winding (made by the "Self-Winding Clock Company") and
synchronized with the Naval Observatory on a periodic basis through
a 20 or 60 ma (I'm not sure which right now) circuit to Western
Union.
I've looked in the archives and while I found several of the old
messages, I couldn't locate any detailed information on how to actual
sync them now that WU is no longer in the time business.
So I'm looking for an affordable ("cheap") WWV receiver or other
timing device that could give at least a contact closure daily or
hourly at the top of the hour so I could use one or two of these old
beauties as they were truly designed. A manual circuit is simple and
described in the archives. An automagic one quickly gets past my
knowledge of available products/techniques. ;-(
TIA.
Mike Riddle /"\ ASCII Ribbon Campaign
mriddle%nos...@ivgate.omahug.org \ / Respect for open standards
"To Reply Remove the Obvious" X No HTML/RTF in email
http://www.mikeriddle.com / \ No M$ Word docs in email
[TELECOM Digest Editor's Note: Quite honestly, I am looking *even for
a single clock* right now. I had three (one wooden case, two metal
case ones) several years ago, but some one or more persons stole them
from me around the start of 2000. Mike, *unless you have sweep second
hands on your clock(s) -- not all that common -- those clocks can be
manually calibrated quite easily -- well, it takes a couple days of
very close observation -- to within a few seconds, maybe 10 seconds
per **month**, meaning with the naked eye, any discrepancy in the time
is so negligible it does not matter. They *are* beautiful old clocks,
but hardly the world's most accurate; but to the casual observer or even
some one who stand there and looks at them closely, they can be quite
accurate.
I had all three of my clocks wired in parallel to a 4.5 volt battery
eliminator; since it never happened that all three of them (and quite
rare that even two of them) chose to wind at the same time, that
little battery eliminator was quite sufficient. As the clock(s) would
wind once per hour, I also had the little red light bulb illuminate on
the winding circuit. To set the clocks, I had the setting circuits all
wired in parallel also to a nine-volt battery which was taped under my
desk through a doorbell button. Once a month or so, when I happened to
look at the clocks *and* my radio controlled wristwatch at the same
time and see the digital wristwatch approach an even hour and could
actually see with my eyes where one or more of the clocks was a thin
hair off the correct time, as the proper second approached I would tap
that doorbell button and listen to the ker-chunk as all three clocks
would jerk their minute a tiny fraction of an inch backward or
forward as needed.
How to get the clocks that accurate to start with? Well, they have
to be ***totally level*** both flush against the wall and from the
floor or ceiling. Make sure the hands are screwed on tight. Not so
tight they bend or fall off, but make the little screw fasten on there
snuggly. Starting at midnight or noon, make both hands evenly point
at the '12'. Look again at 12:05; if you can see even the tiniest
discrepancy, then tweak the pendulum set screw just a tiny bit, mark
the *exact* time with a digital time piece and start again. Do that
every five minutes until you can no longer detect any discrepancy with
your naked eye. Then wait 30 minutes or so and check again, and repeat
as needed. No noticeable discrepancies? Then check it again in a
couple hours. Then go off to bed; when you wake up in the morning go
look again. If the discrepancy wasn't really gross to start with,
then after a day or two of looking and making little tweaks of the
pendulum as needed, you will have the clock very, very,very close,
close enough that you can ignore it for at least a few days at a time.
Train your ears also to listen to the beat.
If your clock(s) have a 'sweep second' hand as well, the setting arm
does both the second hand and the minute hand. But even WUTCO techs
had a hard time calibrating the second hand; if it was between 1
second and around 20 seconds it would jump backward; between about
35 seconds and 60 seconds it would jump forward. If it was between
about 20 seconds and 35 seconds more or less when the pulse came down
the line, well then you had a problem. The mechanical finger would
try to pull it one way or the other, but often times miss entirely.
(Remember, WUTCO clocks were more glamorous than they were 'to-the-
split-second' accurate.)
My wooden case clock arrived with a broken glass on the front, and
fifty years of paint covering up an original varnish finish. It spent
the first fifty years of its life hanging in the Board of Education
Building lunchroom (downtown Chicago), then I had it in service for
another *25 years* (1973-1998) in my office. The finest clock I ever
saw was a grandfather clock in a large case which stood on the floor
in the Chicago Temple Building third floor library; it also had
Western Union works in it, and I think its original smooth varnish
finish and the original glass in it, and door latches.
I sure miss those three clocks I had since they were stolen. If
**anyone** is willing to part with one, please let me know and
how much. No total basket cases please, but small repairs are quite
acceptable. Write to ptow...@massis.lcs.mit.edu and thanks. PAT]
Lisa Hancock
> Several years ago our esteemed moderator ran several articles on
> Western Union Clocks. These were typically installed in train
> stations and other public places. They were electrically powered,
> self-winding (made by the "Self-Winding Clock Company") and
> synchronized with the Naval Observatory on a periodic basis through
> a 20 or 60 ma (I'm not sure which right now) circuit to Western
> Union.
The city public schools I attended had IBM clocks. (IBM used to have
a central-clock and timeclock business which was part of its original
corporate merger; they sold it off in the late 1950s).
On a day to day basis they worked ok, but if something went wrong,
i.e. a power failure or especially seasonal change of time, it took
days until they got them right again. If something broke, again it
took days until it was fixed and they'd show all sorts of crazy times.
The reset efforts would have the clocks relatively slowly advance.
I don't know if these troubles was from the system itself or if school
maintenance people had trouble maintaining them. IBM had sold out
when I was in school, although the systems weren't that old (about 12
years old).
I believe in the late 1960s, the Pennsylvania Railroad or early Penn
Central replaced the old style clocks with modern looking digitial
clocks. They had problems with those. In the waning days of the Penn
Central RR, they replaced their station clocks with individual
stand-alone units that each had to be set manually. They were always
a few minutes off from each other which used to be a big no-no in the
railroad business.
I don't know why it's so hard to keep time coordinated. In the
Philadelphia area, the transportation authority tells riders to use
Bell (Verizon) time, 215-846-1212, as a standardized source. My $20
Casio "50M" watch is pretty reliable. Bell has offered that service
since at least the 1960s (846 was TIme 6).
mike_riddle
There are usualy a few for sale on Ebay. You might have to gut and
paste this URL.
http://search.ebay.com/search/search.dll?cgiurl=http%3A%2F%
2Fcgi.ebay.com%2Fws%2F&krd=1&from=R8&MfcISAPICommand=GetResult&ht=1
&SortProperty=MetaEndSort&query=Western+Union+Clock
Mike
DevilsPGD
hanc...@bbs.cpcn.com (Lisa Hancock) did ramble:
> I don't know why it's so hard to keep time coordinated. In the
> Philadelphia area, the transportation authority tells riders to use
> Bell (Verizon) time, 215-846-1212, as a standardized source. My $20
> Casio "50M" watch is pretty reliable. Bell has offered that service
> since at least the 1960s (846 was TIme 6).
This always confuses me. One of my servers pulls from NTP sources on
the next, the rest synch from there. My phones all set themselves
(Analog, via CID/CND data). My bedroom clock not only has a backup
battery, but it also receives over the air signals, and is able to set
itself as well.
Given that the data is out there, via a network, or being broadcast, why
is it such a challenge in the business world?
A well-dressed man walks into a bar and asks a woman to sleep
with him for $1M. The woman is excited and she gives immediate
consent: "Of course I'll sleep with you!".
Then the man asks, "will you sleep with me for $5?". The woman
indignantly replies, "Of course not! What do you think I am?".
The man replies, "We've already established what you are; now
we're merely haggling over the price."
[TELECOM Digest Editor's Note: I use ntp2.kansas.net for my computers
here. The Windows 2000 pulls from there, or navobs, and then serves
the Windows 98 and Windows 95 laptops over the LAN. PAT]
Tony P.
lalala...@crazyhat.net says:
> In message <<telecom...@telecom-digest.org>>
> hanc...@bbs.cpcn.com (Lisa Hancock) did ramble:
>> I don't know why it's so hard to keep time coordinated. In the
>> Philadelphia area, the transportation authority tells riders to use
>> Bell (Verizon) time, 215-846-1212, as a standardized source. My $20
>> Casio "50M" watch is pretty reliable. Bell has offered that service
>> since at least the 1960s (846 was TIme 6).
> This always confuses me. One of my servers pulls from NTP sources on
> the next, the rest synch from there. My phones all set themselves
> (Analog, via CID/CND data). My bedroom clock not only has a backup
> battery, but it also receives over the air signals, and is able to set
> itself as well.
> Given that the data is out there, via a network, or being broadcast,
> why is it such a challenge in the business world?
Most PBS stations (All if I'm not mistaken as both WGBH Channel 2
Boston, and WSBE Channel 36 Providence both do so.) send a time signal
in the blanking interval.
My Philips VCR can sync the time by tuning to a PBS station. Pretty
cool feature as I never have to manually set the damned clock.
Chuk Gleason
http://www.piexx.com/imp1/imp1dc.html
Chuk G.
Cary, NC
Reed
Back in the 60's I worked as an engineer in a TV/FM broadcast
facility. We had about a dozen of these clocks all over the building,
all synchronized by the hourly pulse from Western Union. Sometime
around 1970 we designed a circuit based on the NE555 (?) PLL chip that
would listen for the hourly tone from our dual-redundancy WWV receiver
and generate the sync pulse to all of the clocks. It was a very
simple design, but I don't recall the specifics.
More recently someone else had the great idea of using a computer,
synchronized to one of the online NBS time servers, that would send an
hourly X10 signal to an X10 relay module to synchronize his WU clocks.
Look here for his article:
http://www.ubr.com/clocks/sync/syncwu.html
-Reed
Howard S. Wharton
systems in service on our south (Main Street in Buffalo) campus.
Sometime in the 1950's IBM sold their clock service plus the fire
alarms to Simplex Time Recorder Company. Today they are called
SimplexGrinnell merging with Grinnell Sprinkler Co. after they were
purchased by Tyco few years ago.
Howard S. Wharton
Fire Safety Technician
Occupational and Environmental Safety Services
State University of New York at Buffalo
Mike Riddle
Mike
Chuk Gleason wrote:
> http://www.piexx.com/imp1/imp1dc.html
> Chuk G.
> Cary, NC
>> Mike Riddle <mriddle%nos...@ivgate.omahug.org> wrote:
Mike Riddle /"\ ASCII Ribbon Campaign
mriddle%nos...@ivgate.omahug.org \ / Respect for open standards
"To Reply Remove the Obvious" X No HTML/RTF in email
http://www.mikeriddle.com / \ No M$ Word docs in email
[TELECOM Digest Editor's Note: Your esteemed moderator running those
articles on those clocks caused so many people to look for them that
today they are considered collector items. Bah, humbug! Me and my
big mouth, I guess. Now I cannot find one anywhere for less than
about three hundred dollars. I made a big leap (for me!) and bid on
one slightly banged up one on EBay last night going at auction for
a 'mere' $125, and after I worked the bid up to $150 (which I can
ill-afford but somehow would have paid for) some bozo came along and
out-bid me on that. I was too scared to bid any higher. PAT]