Line frequency and clocks...
Lasse Langwadt Christensen
>
> glet...@erols.com wrote:
>
> > Hi all,
> >
> > I'm hanging out with a buddy the other night, and he makes a passing
> > reference to the concept that power companies maintain a steady 60Hz line
> > frequency so that people's clocks don't get messed up. Having fooled around
> > with electronics as a kid this flabbergasted me, since it's so simple to run
> > DC power to a clock motor and tune the voltage with a potentiometer to make a
> > simple clock. Why go to the trouble, especially before transistors were
> > available, to try to count up 60 cycles of AC line frequency in order to
> > pulse a stepper motor to move 1/60th of a rotation? Or is there another way
> > this was supposedly done?
>
> Well they don't use stepper motors and your DC potentiometer scheme would
> have crap accuracy by comparison. With an AC clock using a synchronous
> motor you probably will be off by less that a couple of minutes per year.
>
agreed, the potentiometer approch would be verybadcompared to using the
line frequency
> > Then I got really amazed, as nearly everyone around me tried to confirm his
> > theory. Not one of them could give me a possible circuit design that would
> > make this possible, and despite my simple demonstration of why line frequency
> > was unnecessary for clocks.
> >
> > This sounds like another urban legend to me. Anyone else want to take a swipe
> > at this? Please supply authoritative references, if possible.
>
> It is merely opportunistic, equal accuracy could have been designed with any
> AC power frequency. They do jog the frequency from time to time but this has
> nothing to do with the clocks.
>
I once visited the local powerplant,and in the controlroom they have two
clocks
one running at the line freqeuncy of the power they produce, and one
radio
controlled from one of the atomic clocks around the world, and they
adjust the
frequncy to keep them in sync
--L2C
--___--_-_-_-____--_-_--__---_-_--__---_-_-_-__--_----
Lasse Langwadt Christensen, MSEE (to be in 1999)
Aalborg University, Department of communication tech.
Applied Signal Processing and Implementation (ASPI)
http://www.kom.auc.dk/~fuz , mailto:lang...@ieee.org
D.H. Kelly
>It is merely opportunistic, equal accuracy could have been designed >with
any
>AC power frequency. They do jog the frequency from time to time >but this
has
>nothing to do with the clocks.
Actually it does. Since it is impossible to hold exactly 60Hz all the time
as minor shifts in frequency do occur almost continually as loads change, it
is necessary to sometimes run a bit fast or slow to correct system time to
match real time (say WWV). That is, over a period of a day (for example),
the average frequency is 60Hz- very accurately. At a given instant it may be
59.95 to 60.05 typically.
--
Don Kelly
dke...@nanaimo.ark.combull
remove the bull to reply
Doug
> Hi all,
>
> I'm hanging out with a buddy the other night, and he makes a passing
> reference to the concept that power companies maintain a steady 60Hz line
> frequency so that people's clocks don't get messed up. Having fooled around
> with electronics as a kid this flabbergasted me, since it's so simple to run
> DC power to a clock motor and tune the voltage with a potentiometer to make a
> simple clock. Why go to the trouble, especially before transistors were
> available, to try to count up 60 cycles of AC line frequency in order to
> pulse a stepper motor to move 1/60th of a rotation? Or is there another way
> this was supposedly done?
>
> Then I got really amazed, as nearly everyone around me tried to confirm his
> theory. Not one of them could give me a possible circuit design that would
> make this possible, and despite my simple demonstration of why line frequency
> was unnecessary for clocks.
>
> This sounds like another urban legend to me. Anyone else want to take a swipe
> at this? Please supply authoritative references, if possible.
I'm surprised. Were you with a bunch of EEs?
Anyway, if you're really into this sort of thing, go to
http://www.ita.doc.gov/media/1elecur.htm
for the publication called "Current Abroad".
At http://www.rcic.com/cnf/treg/1996/list313.htm
it states the US line frequency is guarenteed to
an accuracy of no more than +-0.06% to +-0.3%.
Or you can go to
http://www.psihq.com/iread/worldpow/ecac/index.htm
which is the same thing only on the web.
It lists the countries with power line frequencies
that are good enough for clocks.
The circuitry I would guess is fairly simple and common.
Off hand, tap off the secondary side in parallel with
power input, overdrive to something like 5 volt pulses,
put through a divide by 120 (or whatever accuracy you want)
for the full wave rectifier, and presto, one second pulses.
I've never designed one nor seen the schematic of one,
but I wouldn't think there's any mystery to it.
----------------------------------------
Return Address is dmckean'at'gte'dot'net
----------------------------------------
Pete Fraser
>On Sat, 27 Feb 1999 23:02:09 -0800, Doug <d...@not.spam> wrote:
>>The circuitry I would guess is fairly simple and common.
>
>It's simpler and more common than you think. It isn't done with stepper
motors,
>it's done with synchronous motors. You youngins can only think digital, and
>think it's always the best way.
Also, on the designs I've played with, the synchronous motor can start
in either direction. Typically there's a clutch which will reverse
the travel if the motor starts off backward. The clutch can easily be
tied back with thread, and the clock started in reverse.
--
Pete Fraser
rob...@bestweb.net
Books), pp. 102-3. Especially interesting in terms of how strains on power
supply during World War 2 caused errors in radio network synchronization,
causing stations to switch to tuning-fork clocks.
BTW, line lock is no longer used in video. On old TVs, 60 Hz QRM was
visible as a pair of stationary horizontal bars; now the bars scroll slowly
upward. I think computer monitors are in synch with TV video, but I could
be wrong.
Speaking of TV & frequency, I think I've already remarked on
alt.fan.cecil-adams that today's crystal frequency control at TV stations is
very good, as revealed by the slow inversion of video polarity of a distant
signal which is co-channel and not offset to a local station. The
heterodyne is usually considerably less than 1 hertz, and remember the
transmitted signal's in the tens or hundreds of megahertz. We're talking
parts per billion here! It's been said that had these oscillators been
standard in the early days, the +10/0/-10 kHz offset system of TV frequency
assignments in the USA would not have been instituted. Offsets were used
because a 10 or 20 kHz beat was considered less offensive than a heterodyne
of, say, 20 Hz, which would've been considered a "zero-beat" in those days.
Robert
Net-Tamer V 1.11 - Registered
Adam Wozniak
>glet...@erols.com wrote:
>
>> Hi all,
>>
>> I'm hanging out with a buddy the other night, and he makes a passing
>> reference to the concept that power companies maintain a steady 60Hz line
>> frequency so that people's clocks don't get messed up. Having fooled around
>> with electronics as a kid this flabbergasted me, since it's so simple to run
>> DC power to a clock motor and tune the voltage with a potentiometer to make a
>> simple clock. Why go to the trouble, especially before transistors were
>> available, to try to count up 60 cycles of AC line frequency in order to
>> pulse a stepper motor to move 1/60th of a rotation? Or is there another way
>> this was supposedly done?
>>
Potentiometers drift with temperature.
(not that it can't be done. Pententiometers are a bad choice, however)
--Adam
--
ad...@mudlist.eorbit.net
ICQ 27422092
Kevin D. Quitt
wind-up clock, and when the spring unwound, the battery powered a motor that
wound it back up.
--
#include <standard.disclaimer>
_
Kevin D Quitt USA 91351-4454 96.37% of all statistics are made up
Per the FCA, this email address may not be added to any commercial mail list
Nick Spalding
> The first battery-powered clock I had still went tick-tick-tick. It was a
> wind-up clock, and when the spring unwound, the battery powered a motor that
> wound it back up.
The clock in my 1956 Pontiac worked like that. Generically that sort
of arrangement is called a remontoire.
--
Nick Spalding
Don Klipstein
: Also, on the designs I've played with, the synchronous motor can start
: in either direction. Typically there's a clutch which will reverse
: the travel if the motor starts off backward. The clutch can easily be
: tied back with thread, and the clock started in reverse.
I have seen a couple where the motor is an induction motor with a
special rotor design favorable to "locking in" as a synchronous motor once
it comes up to speed. Induction motors normally run a little slower than
synchronous speed.
Since the motor is a single phase induction motor, it needs a little
extra something to make it start - in this case, the motor had "shaded
poles". Each pole had a shorted turn of wire or a shorted few turns of
wire around one of the two side protrusions that each pole had. This makes
the motor start in one particular direction as an induction motor.
Without the pole shading wire, the motor usually did not start, but I
imagine it could start randomly.
- Don Klipstein (d...@misty.com)
Kevin D. Quitt
>Kevin D. Quitt wrote:
>> ...wind-up clock, and when the spring unwound, the battery powered a motor that
>> wound it back up.
>
>The clock in my 1956 Pontiac worked like that. Generically that sort
>of arrangement is called a remontoire.
That's the device that adds power to a pendulum; I can see how that apply. (And
to pick a nit, I think it's 'remontoir'.)
William L. Bahn
> Hi all,
>
> I'm hanging out with a buddy the other night, and he makes a passing
> reference to the concept that power companies maintain a steady 60Hz line
> frequency so that people's clocks don't get messed up. Having fooled
around
> with electronics as a kid this flabbergasted me, since it's so simple to
run
> DC power to a clock motor and tune the voltage with a potentiometer to
make a
> simple clock. Why go to the trouble, especially before transistors were
> available, to try to count up 60 cycles of AC line frequency in order to
> pulse a stepper motor to move 1/60th of a rotation? Or is there another
way
> this was supposedly done?
No supposedly about it - have you ever heard of synchronous motors?
Your notion of tuning the voltage to a DC motor using a potentiometer as a
counter argument is laughable. Sure, you can make a very crude clock this
way - and the performance is just as crude. Let's say that you are able to
hold the DC voltage to within 0.1% over the course of a day and that the
input voltage is the only thing that affects the speed of the motor. Keep in
mind that for a 12V supply that means holding the voltage to within 12mV.
What is the accuracy of your "clock"? A minute and a half a day. Would you
call a clock that gains or loses 43 minutes a month a good clock? In
practice, you would be hard pressed to get such a clock to stay within
several hours a month.
With a typical watch crystal you have a frequency stability of 20ppm. That
puts you on the order of 2 seconds a day or about 11 minutes a year. Not
bad, but take a cheap AC clock that they give away for filling out a credit
card application at the mall and plug it in and if you have no power outages
that clock will be within a couple of seconds a year later. That's a long
term accuracy of 50 parts per billion. Do you really think that the
necessary components to achieve that level of accuracy are contained in a
freebie clock?
>
> Then I got really amazed, as nearly everyone around me tried to confirm
his
> theory. Not one of them could give me a possible circuit design that
would
> make this possible, and despite my simple demonstration of why line
frequency
> was unnecessary for clocks.
>
What demonstration? You "described" a clock that you "claimed" made line
frequency uncessary for clocks. At what point did you "demonstrate" this? Do
even have any notion of what it would take to suitably demonstrate such a
claim?
What do you consider adequate performance for a clock (and I'm talking about
not only the clock on your kitchen wall but the clocks used by railroads on
the east coast as they send trains off for the west coast wanting to
schedule the traffic so that they don't have to stop and wait as sidings for
other trains that started at the west coast to pass - or the clocks used by
airlines to schedule and track aircraft flying between different cities).
Wouldn't it be nice for such clocks to be stable enough that they don't
drift relative to each other by more than a fraction of a minute a month? In
reality, that is no where near good enough - but let's just stipulate it.
Let's say that we specify that each clock is to drift no more than 15
seconds a month. That's about 6ppm. Your "simple demonstration" would have
to hold the voltage (using the 12V nominal supply mentioned earlier) to
within 70uV. Do you really think that your potentiometer can do that?
> This sounds like another urban legend to me. Anyone else want to take a
swipe
> at this? Please supply authoritative references, if possible.
Let me get this straight - you think something is an urban legend, so you
post something to Usenet asking for total strangers to give you
authoratative references. Why should you believe them any more than you
believed your buddy or any of the others.
Did the thought ever occur to you to go straight to the horse's mouth? Take
your local phonebook and contact your local power utility and ask to speak
to an engineer involved in power generation and ask him or her.
What you will find is that somewhere on the grid is a control station (for
the Western Area Power Administration it's Southern California Edison - I
don't know if I've remembered the name correctly, but they're in southern
CA). With regards to this topic, the control station basically does two
things. First, they monitor the requirements for every generating plant on
the grid and issue speed orders telling plants to slow down or speed up
their generators in order to keep the grid in synch. This is necessary
because at any given time a certain utility may not be able to hold their
frequency due to short term load demands (either going on or coming off) and
the other stations must compensate to prevent contention. There are other
devices that are in place to deal with this issue as well (such as phase
adjusting transformers at the grid nodes).
The other major role of the control station is to maintain the long term
frequency stability of the grid. They do this by running one clock that uses
their frequency as a time base and another clock that acts as a standard. At
one time it was a pendulum clock but now it is a clock that receives it's
timebase from the atomic clocks (via WWV) that are the world wide standards
for time. They have a readout that gives the error of the two and they issue
time-factor correction orders several times a day to all of the power
utilities telling them to speed up or slow down their generators for
specific amounts of time in order to drive this error toward zero. The
accululated error at any time is seldom more than a small fraction of a
second since that last time the grid-referenced clock was reset. If the
error ever exceeds two seconds (which generally only happens when a major
disturbance such as a widespread blackout occurs) then they basically admit
defeat and reset the clock. At that point, the error in your cheapo clock
(assuming you weren't in the blackout area yourself) has a permanent error
in it that you would have to remove yourself.
At one point in time, if long term time stability was really important to
you, you could actually subscribe to a list where you would be notified
anytime there was a reset of the clock and what the introduced offset was.
This was basically a service for users that couldn't afford the equipment
necessary to pull there own timebase off of WWV directly. I think that the
costs of this have come down enough that this notification service is no
longer offered.
Bret Wood
"William L. Bahn" wrote:
<Excellent post on clocks snipped except for the following line>
> or the clocks used by
> airlines to schedule and track aircraft flying between different cities
Does anyone have definitive proof that airlines actually use clocks?
I'm kind of skeptical myself. :)
-Bret Wood
-bret...@cs.uoregon.edu
D.H. Kelly
in Europe of course) use a tiny stepper motor (a toothed wheel between the
poles of an electromagnet) directly.
------------snip------------
I've never seen this - but have seen many synchronous motors used for this.
I'm not sure that this is a true stepper but may be a variation of the
synchronous machine - probably had shading poles on the stator- The result
would be a smooth rotation rather than hops. Use of a strobe would determine
this.
D.H. Kelly
>wire around one of the two side protrusions that each pole had. This >makes
>the motor start in one particular direction as an induction motor.
Commonly used in hysteresis synchronous motors as well as induction motors
(these develop nearly constant torque at all speeds from start to
synchronous). The use of the shaded pole produces a rotating component in
the magnetic field. Without it, there would be a pulsating field (same as
two equal but opposite rotating fields) and the motor would not start. Given
a spin in either direction- it would then accelerate and turn in that
direction. The shaded pole unbalances these fields to produce a non-zero
starting torque. In larger single phase motors a starting winding serves the
same purpose, with better torque and efficiency, at a $ price.
Kevin D. Quitt
>The circuitry I would guess is fairly simple and common.
It's simpler and more common than you think. It isn't done with stepper motors,
it's done with synchronous motors. You youngins can only think digital, and
think it's always the best way.
--
Peter Ceresole
Doug <d...@not.spam> wrote:
>The circuitry I would guess is fairly simple and common.
>Off hand, tap off the secondary side in parallel with
>power input, overdrive to something like 5 volt pulses,
>put through a divide by 120 (or whatever accuracy you want)
>for the full wave rectifier, and presto, one second pulses.
But why go for one second pulses? All the mains locked clocks I knew (50Hz
in Europe of course) use a tiny stepper motor (a toothed wheel between the
poles of an electromagnet) directly. They drove the hands through a simple
gear train, apparently completely smoothly- 50Hz hops are hard to observe
directly... That was what distinguished them from later, battery powered
electronic clocks that were based on an oscillator and divider; those
jumped and clicked every second. The mains clocks were silent and
mysterious.
--
Peter
Alan Hamilton
>The circuitry I would guess is fairly simple and common.
>Off hand, tap off the secondary side in parallel with
>power input, overdrive to something like 5 volt pulses,
>put through a divide by 120 (or whatever accuracy you want)
>for the full wave rectifier, and presto, one second pulses.
>I've never designed one nor seen the schematic of one,
>but I wouldn't think there's any mystery to it.
Oh, you kids. The line-frequency-based clocks did it with something
called an "electric motor", which through a series of gears moved
analog indicators called "hands".
The motor was designed so that its speed depended on the line
frequency. As long as that was stable, it would keep good time.
--
/
/ * / Alan Hamilton
* * al...@primenet.com
Doug
>
> On Sat, 27 Feb 1999 23:02:09 -0800, Doug <d...@not.spam> wrote:
>
> >The circuitry I would guess is fairly simple and common.
> >Off hand, tap off the secondary side in parallel with
> >power input, overdrive to something like 5 volt pulses,
> >put through a divide by 120 (or whatever accuracy you want)
> >for the full wave rectifier, and presto, one second pulses.
> >I've never designed one nor seen the schematic of one,
> >but I wouldn't think there's any mystery to it.
>
> Oh, you kids. The line-frequency-based clocks did it with something
> called an "electric motor", which through a series of gears moved
> analog indicators called "hands".
You're talking past tense. TODAY, most line frequency based clocks
don't have those little analog indicators called "hands". Some have
these little things called "LEDs". And the guts of them have these
things called "ICs". Most analog clocks TODAY run off batteries.
The motors today being SOOOOO much more efficient than they were
back in the stone age. :)
> The motor was designed so that its speed depended on the line
> frequency. As long as that was stable, it would keep good time.
Yea. They've had a name for some 90 or so years.
A reeeeeal old name for them - "synchronous motors". :)
Nick Spalding
> That's the device that adds power to a pendulum; I can see how that apply. (And
> to pick a nit, I think it's 'remontoir'.)
You are absolutely right. I dredged the wrong word out of my memory.
It is nearly 45 years since pendulum clocks were part of my every day
life. There is a word, or phrase, for the type of mechanism where the
going train is driven by a low power spring or weight which is
automatically rewound at intervals but I cannot for the life of me
remember it.
--
Nick Spalding
Greg Goss
>In article <36e6ff4b....@news.pacificnet.net>, Ke...@Quitt.net wrote:
>
>>On Sat, 27 Feb 1999 23:02:09 -0800, Doug <d...@not.spam> wrote:
>>>The circuitry I would guess is fairly simple and common.
>>
>>It's simpler and more common than you think. It isn't done with stepper
>motors,
>>it's done with synchronous motors. You youngins can only think digital, and
>>think it's always the best way.
>
>Also, on the designs I've played with, the synchronous motor can start
>in either direction. Typically there's a clutch which will reverse
>the travel if the motor starts off backward. The clutch can easily be
>tied back with thread, and the clock started in reverse.
Somewhere around here I have an early electric clock. There is a knob
on the back labeled "start -->" that you twirl to get the clock
started, You can start it backwards if you wish.
Greg Goss
>At one point in time, if long term time stability was really important to
>you, you could actually subscribe to a list where you would be notified
>anytime there was a reset of the clock and what the introduced offset was.
>This was basically a service for users that couldn't afford the equipment
>necessary to pull there own timebase off of WWV directly. I think that the
>costs of this have come down enough that this notification service is no
>longer offered.
I've seen ads for a wristwatch that resynchronized to WWV at regular
intervals. I think it's on the order of $70.
Peter Ceresole
"D.H. Kelly" <dke...@nanaimo.ark.com> wrote:
>? All the mains locked clocks I knew (50Hz
>in Europe of course) use a tiny stepper motor (a toothed wheel between the
>poles of an electromagnet) directly.
>------------snip------------
>I've never seen this - but have seen many synchronous motors used for this.
>I'm not sure that this is a true stepper but may be a variation of the
>synchronous machine - probably had shading poles on the stator-
Yes, that's what I meant. I just didn't know what it was called- 'stepper'
came to mind because of the jump of the teeth from pole to pole.
--
Peter
William L. Bahn
the dubious claim that airlines use schedules. ;-P
Bret Wood wrote in message <36D9F61B...@cs.uoregon.edu>...
>
>
>"William L. Bahn" wrote:
>
><Excellent post on clocks snipped except for the following line>
>
>> or the clocks used by
Ken Redman
The stability spec is only long-term. Short-term the frequency can
change considerably: it tends to reduce (and the clocks run slow) during
the commercial breaks - when everyone in the UK rushes off to make a cup
of tea ;-)
The utilities compensate later (during the night) so the clocks catch up
again.
--
Ken Redman
To reply remove dot. from email address
glet...@erols.com wrote:
> ... power companies maintain a steady 60Hz line
> frequency so that people's clocks don't get messed up. ...
Ken Redman
Plugged in in and it ran forwards for me...
A little investigation showed that with the (2-pin) plug in one way, it
usually runs forwards, plugged in other way it usually runs backwards...
It contains a little sealed motor: very odd??!! ;-)
--
Ken Redman
To reply remove dot. from email address
Kevin D. Quitt
>Kevin D. Quitt wrote:
>
>> That's the device that adds power to a pendulum; I can see how that apply. (And
>> to pick a nit, I think it's 'remontoir'.)
>
>You are absolutely right. I dredged the wrong word out of my memory.
Oh, drat! You mean I have to unremember that, and the train of rationalization
that led to me accept the new usage of the word? Phooey.
Doug
> >you, you could actually subscribe to a list where you would be notified
> >anytime there was a reset of the clock and what the introduced offset was.
> >This was basically a service for users that couldn't afford the equipment
> >necessary to pull there own timebase off of WWV directly. I think that the
> >costs of this have come down enough that this notification service is no
> >longer offered.
>
> intervals. I think it's on the order of $70.
I think there's a horizontal line of the tv that's
synched to the primary standard of the main station.
John Francis
Kevin D. Quitt <Ke...@Quitt.net> wrote:
>On Sat, 27 Feb 1999 23:02:09 -0800, Doug <d...@not.spam> wrote:
>>The circuitry I would guess is fairly simple and common.
>
>It's simpler and more common than you think. It isn't done with stepper motors,
>it's done with synchronous motors. You youngins can only think digital, and
>think it's always the best way.
I believe my mother *still* uses a mains-powered synchronous motor clock
as her primary timekeeper. It's about as simple as an electric clock
can be - just a small motor, and some step-down gearing. I know for a fact
that it is forty-five years old, and I believe it is at least seven years
older than that.
To add first-hand evidence to some speculations: Yes, you could start
the clock running backwards. If you had to restart it after a power cut
(it didn't always start up again) there was a little knurled shaft on
the back of the clock. You gave it a spin between your fingers to start
things going again. It was marked with an arrow to show the direction
to spin it - if you spun it the other way, the clock ran backwards.
[A source of much amusement to a young, inquisitive child. My parents
didn't seem to find it quite so amusing]
Provided there were no power cuts the clock never had to be adjusted.
Accuracy to that level (let's say one minute a year, which works out
to around two parts in a million) doesn't happen by chance. If the
long-term frequency was not adjusted [to precisely 50Hz, in this case]
the clock would have needed periodic corrections. But the only time
we ever had to adjust the clock was at the start and end of daylight
saving time (British Summer Time, to be pedantic). We set it exactly
to the second as best as we could tell, and it stayed bang on time.
Dan Drake
> The first battery-powered clock I had still went tick-tick-tick. It was a
> wind-up clock, and when the spring unwound, the battery powered a motor that
> wound it back up.
The operating principle, so I was told at the time, of clocks in the
better sort of car.
Was there ever a device more notorious for failure (before personal
computers, of course) than the pre-digital car clocks?
--
Dan Drake
d...@dandrake.com
http://www.dandrake.com/index.html
Nothing so moves men to the condonation of crime as the spectacle
of a vindictive judge, whose relish in pronouncing sentence seems
sharpened by personal spite.
--Dorothy L Sayers
Nick Spalding
> On Sun, 28 Feb 1999 18:10:36, Ke...@Quitt.net (Kevin D. Quitt) wrote:
>
> > The first battery-powered clock I had still went tick-tick-tick. It was a
> > wind-up clock, and when the spring unwound, the battery powered a motor that
> > wound it back up.
>
> The operating principle, so I was told at the time, of clocks in the
> better sort of car.
>
> Was there ever a device more notorious for failure (before personal
> computers, of course) than the pre-digital car clocks?
The one of this sort in my 1956 Pontiac was still working when I sold
the car in 1964 having bought it from its original owner 15 months
earlier for $200. As was everything else.
My present car has a quartz clock which still has hands thank
goodness.
--
Nick Spalding
Happy Dog
> Recently had an old clock in to fix: owner complained it ran backwards.
> Plugged in in and it ran forwards for me...
>
> A little investigation showed that with the (2-pin) plug in one way, it
> usually runs forwards, plugged in other way it usually runs backwards...
>
> It contains a little sealed motor: very odd??!! ;-)
Just by chance. Like flipping a coin. The motor can't tell which side the
hot is on.
hd
D.H. Kelly
>on the back labeled "start -->" that you twirl to get the clock
>started, You can start it backwards if you wish
Right- you jogged my memory- however, the shaded pole stator soon eliminated
that.(jus as the electric starter did wonders for cars).
Wow- get xxx- self starting electric clock! only $4.95
Dan Drake
wrote:
>...
>
> I believe my mother *still* uses a mains-powered synchronous motor clock
> as her primary timekeeper. It's about as simple as an electric clock
> can be - just a small motor, and some step-down gearing. I know for a fact
> that it is forty-five years old, and I believe it is at least seven years
> older than that.
>
> ...
> Provided there were no power cuts the clock never had to be adjusted.
> Accuracy to that level (let's say one minute a year, which works out
> to around two parts in a million) doesn't happen by chance. If the
> long-term frequency was not adjusted [to precisely 50Hz, in this case]
> the clock would have needed periodic corrections...
Here's a bit of folklore, which I believe (or believed) to be true because
I heard it only a few years after the event, from my brother, who was old
enough to remember it. In the late 1940s there was a severe drought in
California, and at times the generators (we had lots of hyrdoelectric
power then) were run at a reduced speed; the output was below 60 Hz, and
people had to reset their synchronous clocks at intervals.
The more I think about this, the less plausible it seems. Any comments
from people who actually know power generation? (And remember, this is
about pre-digital days when there was also no highly redundant national
power network to make it possible to plunge the entire coast into darkness
at once.)
M.J.Powell
>Here's a bit of folklore, which I believe (or believed) to be true because
>I heard it only a few years after the event, from my brother, who was old
>enough to remember it. In the late 1940s there was a severe drought in
>California, and at times the generators (we had lots of hyrdoelectric
>power then) were run at a reduced speed; the output was below 60 Hz, and
>people had to reset their synchronous clocks at intervals.
>
>The more I think about this, the less plausible it seems. Any comments
>from people who actually know power generation? (And remember, this is
>about pre-digital days when there was also no highly redundant national
>power network to make it possible to plunge the entire coast into darkness
>at once.)
Yes, I've heard of this. In UK where frequency is 50 c/s, the lower
limit of freq is about 58.5, after that it's candle-time again.
I don't know why.
--
Mike The life that I have
Is all that I have
And the life that I have
Is yours.
Dick Alvarez
because I heard it only a few years after the event, from my brother, who
was old enough to remember it. In the late 1940s there was a severe
drought in California, and at times the generators (we had lots of
hyrdoelectric power then) were run at a reduced speed; the output was
below 60 Hz, and people had to reset their synchronous clocks at
That is real. I am old enough to remember it. My recollection is
that the big drought that year was mainly in northern California, and did
not much include southern California. I think that the electric power
line frequency was reduced to 59 Hz. (Actually cycles per second; we did
not have Hertz then!) That slightly reduced the speed of induction
motors, and thus saved a little bit of energy. A side-effect was that
synchronous electric clocks lost 1 minute per hour, or 24 minutes per day.
We quickly became accustomed to that.
Incidentally, about synchronous clocks and sealed motor units and
self-starting and shaded poles: At least many years ago, a typical
synchronous electric clock motor had a C-shaped laminated magnetic core.
The electrical winding was on a spool which fitted around the magnetic
core. Each pole-piece was forked, and had a thick copper ring around one
tine of the fork. Those copper rings retarded the magnetic field in those
2 tines. That produced a rotating component of magnetic field to start
the clock. Inside of the sealed copper case, there was a little steel
ring with 2 spokes. The ring rotated synchronously with the applied
magnetic field, at 60 revolutions per second. The sealed copper case
contained a gear train. A shaft, with a gear, protruded from the sealed
copper case, and engaged a gear that drove the clock hands. The whole
clock consumed about 2 Watts.
In school, one of my electricial engineering class-mates was a
brilliant theoretician but not so much on hardware. In those days,
Telechron self-starting synchronous clock magnetic cores were made in
3 pieces, bolted together. Some of us opened the guy's clock (non-alarm
type), unbolted the magnetic core, interchanged the 2 pole-pieces, and
reassembled the clock. This caused the magnetic field, hence the motor
and clock hands, to rotate backward. Of course we conveniently just
happened to be there when the guy returned to his room. He looked at the
clock for a long time, and then announced that it seemed to be running
backward. We told him to unplug it, and put the plug in the wall the
other way. Still the clock ran backward. We told him that the electric
utility must have had a polarity reversal, and we would call them. After
the guy left, we restored his clock to normal.
Time-keeping gradually has become much more sophisticated during the
past many decades. We want our clocks to be synchronous with the apparent
solar day. But for several reasons, that varies considerably, with both
annual periodicity and with long-term irregular drift. We also want the
second to be of constant length. So the second is defined internationally
in terms of an atomic transition frequency, which is constant as well as
anybody presently knows. To reconcile the constant second and Earth's
irregular motion, we insert or delete a leap-second in our standard time
when and if required at the end of December 31 and June 30, Coordinated
Universal Time (not local standard time). Commercial electric power line
frequency generally is carefully kept long-term synchronous with atomic
time. So even without power failures, synchronous clocks can accumulate
1-second errors because of those leap-seconds.
Dick Alvarez
alvarez at best dot com
David Scheidt
: Kevin D. Quitt wrote:
: You are absolutely right. I dredged the wrong word out of my memory.
: It is nearly 45 years since pendulum clocks were part of my every day
: life. There is a word, or phrase, for the type of mechanism where the
: going train is driven by a low power spring or weight which is
: automatically rewound at intervals but I cannot for the life of me
: remember it.
Self-winding?
David ""Scheidt
David Scheidt <dsch...@enteract.com>
The presumption of the flamers is, I assume, that folks barging in to AFU with
tired old stories and off-topic drivel have already ignored the polite and
subtle clues, and require a thwack with the clue-by-four. -- Andrew Reid
Usenet Reader
In article <36D9F61B...@cs.uoregon.edu> on Sun, 28 Feb 1999
18:06:19 -0800, Bret Wood <bret...@cs.uoregon.edu> writes:
> "William L. Bahn" wrote:
>
> <Excellent post on clocks snipped except for the following line>
>
> > or the clocks used by
> > airlines to schedule and track aircraft flying between different cities
>
> Does anyone have definitive proof that airlines actually use clocks?
> I'm kind of skeptical myself. :)
I have observed that the times shown on airport monitors can be
wrong by several minutes, either way.
Before these flight monitors were common, it could be quite hard
to find the official time at an airport. This is a problem if
you're not certain of the current time zone. I heard that clocks
were deliberately not installed in public view in case some clock
somewhere got wrong, wasn't noticed and corrected, and made
someone miss a flight.
Several decades ago clocks in institutions ran on a master-slave
principle. A master clock generated pulses that made the slave
clocks tick. This allowed all the clocks to be changed together,
say for daylight savings time. I think that in large cities
Western Union had a subscription service to distribute clock
pulses throughout the central business district.
comp.dcom.telecom described this a few years ago.
--- Wm. Randolph U Franklin, WRFUSE at MAB.ECSE.RPI.DELETETHIS.EDU
Robert M Best
> going train is driven by a low power spring or weight which is
> automatically rewound at intervals but I cannot for the life of me
> remember it.
Escapement mechanism?
Jim
in my Buick, Olds, or Chevy ((I had one of each at the time)) but it
was a GM car), had a form of self-regulation. If you set it forward, it
assumed that it had been running slow and so it speeded up a bit,
and vice versa. Kept excellent time, except that daylight savings time
meant resetting it every couple of days for a month or so.
J.Michael
e-ply to: <capu...@mcmsys.com>
- Response via DejaNews because my server didn't pick up the post -
- DejaNews e-ddress is a black hole -
-----------== Posted via Deja News, The Discussion Network ==----------
http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
Foobar T. Clown
>
> [My old car's clock], had a form of self-regulation. If you set it
> forward, it assumed that it had been running slow and so it speeded
> up a bit, and vice versa. Kept excellent time, except that daylight
> savings time meant resetting it every couple of days for a month or
> so.
Is it just my faulty memory, or did the high end watchmakers figure out
a way to cancel the self-reg action if the time was changed by more than
a few minutes?
Bret Wood
Dunedin wrote:
>
> Several decades ago? They are current products and big
> business around the world today. The reason for having
> synchronised clocks of any system (and there are many
> different types) is to reduce conflict in the workplace and
> thereby maximise productivity.
They also use them here at the U of O. When you have classes which
are supposed to be meeting on a synchronized schedule, in buildings
spread over a relatively large campus, it is important that the
classes all start and stop at about the same time. Unfortunately,
there's been a problem with the clocks lately, and it screws up
every clock on campus.
-Bret Wood
-bret...@cs.uoregon.edu
Dunedin
Usenet Reader wrote in message
<7bk3k2$fqm$1...@newsfeeds.rpi.edu>...
>
>Several decades ago clocks in institutions ran on a
master-slave
>principle. A master clock generated pulses that made the
slave
>clocks tick. This allowed all the clocks to be changed
together,
>say for daylight savings time. I think that in large
cities
>Western Union had a subscription service to distribute
clock
>pulses throughout the central business district.
>comp.dcom.telecom described this a few years ago.
>
Several decades ago? They are current products and big
business around the world today. The reason for having
synchronised clocks of any system (and there are many
different types) is to reduce conflict in the workplace and
thereby maximise productivity. A good clock system which
keeps all the clocks showing the same time will play a big
part in achieving this. I have seen stop-work meetings from
boiling resentment to complain that x-department was getting
off one minute ahead of y-department. The classic would have
to be the complaint that x-department were getting to the
cafeteria a minute ahead of everyone else and scoring the
freshest pies. After investigation is was confirmed that
x-department had a big ladder which reached their clock. In
this day of beeping digital watches and radios playing in
the factory it is essential to have accurate industrial time
displays to at least the second. That's why synchronised
clock systems which auto correct and self correct are such
big business.
The US has several dozen current clock manufacturers
supplying world markets but got behind in technology
compared to Asia and Europe.
Jim
With my luck, they thought to do that starting with the next car off the
assembly line after mine . . .