Why can't I buy a good clock?
Robert Scott
This rant is about clocks - the ordinary AC-powered kind that you buy
at retail outlets and put next to your bed so that you know when to
get up and start your day. Why is it posted in a newsgroup about
embedded systems? Because I am a developer of embedded systems and I
know, as do most of you, how easy it would be to design a good clock.
The sad fact is that it is not being done well at all.
I start with battery back-up. Here in the midwest, power failures are
a fact of life. And I don't mean little one or two minute
interruptions. I mean big four or five hour outages, or perphaps 47
hours, like we had two years ago. Many clocks offer battery backup.
But from the sampling that I have done, it seems that most low-cost
clocks can be off as much as 15 minutes after an 8-hour power failure.
Even the cheapest wrist watches can be expected to be within 2 seconds
over an 8-hour period. What possible excuse is there for not
maintaining that accuracy in battery-backup mode?
I finally found a clock that had decent accuracy in battery-backup
mode. It was made by Timex. But then I noticed that it was drifting
by about 2 minutes per month during normal operation. Even though it
is AC-powered, the people at Timex seem to have totally ignored the
possibility of locking to the AC-line during normal operation. (For
those of you too young to remember when all AC-powered clocks were
mechanically driven by AC synchronous motors, the electric power
utilities ensure precise long-term timing of such clocks.)
Ideally, an AC-powered clock should use the AC line frequency when it
is available. If the clock has a little bit of non-volatile memory,
it could even develop a calibration correction during normal
operation, which would apply during battery-backup periods.
I don't know about your house, but over here, clocks seem to breed
faster than coat hangers. This includes not only simple clocks, but
clocks embedded in everything from the VCR to the telephone answering
machine. Twice a year (for daylight savings time), and after every
power failure, we have the ritual of "find all the clocks". It
usually takes several days to re-discover all of them.
-Robert Scott
Ypsilanti, Michigan
(Reply through newsgroups, not by direct e-mail, as automatic reply address is fake.)
Uwe Hercksen
Robert Scott schrieb:
>
> This rant is about clocks - the ordinary AC-powered kind that you buy
> at retail outlets and put next to your bed so that you know when to
> get up and start your day. Why is it posted in a newsgroup about
> embedded systems? Because I am a developer of embedded systems and I
> know, as do most of you, how easy it would be to design a good clock.
> The sad fact is that it is not being done well at all.
>
Hello,
you should buy a radio controlled clock.
I have 3 such clocks for the german dcf77 transmitter. Very
precise, always better than a second error.
Look on this page for NIST radio stations in the USA and for
receivers:
http://www.bldrdoc.gov/timefreq/index.html
Bye
Spehro Pefhany
That or move to some place with reliable power. ;-)
The backup battery that sort-of keeps time for a few minutes is
all I've ever needed (if I could only remember to replace the
battery). I think your requirement is oddball for
the first world. Alarm clocks are cheap products, mostly made
by cheapskate companies such as Cxxxxx in Hong Kong, and
all corners are subject to cutting. If they can use an RC
oscillator instead of a crystal for the backup, they'll do
it, and the accuracy aspect won't buy them much in the way
of additional sales; the backup will, and it probably adds
less than 10 cents to the cost. Further, these features are
mostly fixed when the cheap ASIC is designed, and they are
aiming at the highest volume markets.
Best regards,
Spehro Pefhany
--
"it's the network..." "The Journey is the reward"
sp...@interlog.com Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.co
Paul Burke
>
> This rant is about clocks - the ordinary AC-powered kind that you buy
> at retail outlets and put next to your bed so that you know when to
> get up and start your day. Why is it posted in a newsgroup about
> embedded systems? Because I am a developer of embedded systems and I
> know, as do most of you, how easy it would be to design a good clock.
> The sad fact is that it is not being done well at all.
>
This is a result of Gresham's Law. Originally stated as "Bad money
drives out good", it applies to almost everything in the market place.
People can sell a crap clock for $10 or a good one for $40. Customers
just look at the price tag, buy it, complain that it doesn't work, in
the meantime the good clock designer has gone out of business.
For clocks read PCs, operating systems, power tools, hand tools, etc.
etc.
Paul Burke
Alex Colvin
>drives out good", it applies to almost everything in the market place.
>People can sell a crap clock for $10 or a good one for $40. Customers
>just look at the price tag, buy it, complain that it doesn't work, in
>the meantime the good clock designer has gone out of business.
>For clocks read PCs, operating systems, power tools, hand tools, etc.
>etc.
except that with PC operating systems, you sell the crap for $40.
"you can shellac it and paint it colors, but if you poke at it
hard enough you'll find that it's still a turd".
--
mac the naïf
Richard Cochran
> This rant is about clocks - the ordinary AC-powered kind that you buy
> at retail outlets and put next to your bed so that you know when to
> get up and start your day. Why is it posted in a newsgroup about
> embedded systems? Because I am a developer of embedded systems and I
> know, as do most of you, how easy it would be to design a good clock.
> The sad fact is that it is not being done well at all.
>
>
> I don't know about your house, but over here, clocks seem to breed
> faster than coat hangers. This includes not only simple clocks, but
> clocks embedded in everything from the VCR to the telephone answering
> machine. Twice a year (for daylight savings time), and after every
> power failure, we have the ritual of "find all the clocks". It
> usually takes several days to re-discover all of them.
LOL! I agree with your complaints. In particular, for that last
item, I wish designers would realize that, just because you CAN
put a clock in a device for virtually no additional cost,
doesn't mean that you SHOULD. If a clock is not self-setting,
then it's a maintenance item, and I'd prefer not to have it in
my house. I wish I could find a modern microwave oven without
a clock.
I've got a couple of Oregon Scientific radio controlled clocks,
and they are wonderful. Always accurate to better than a second,
completely battery powered, so no power interruption problems,
self-setting, self-adjusting for DST, in short, almost ideal.
See
http://www.geocities.com/gregclep/clocks/reviews.html
for a review of lots of these (the guy who put together that
page is even more fanatical about them than I am).
But they're still not quite ideal. I wanted an alarm that would
sound on weekdays, but not on weekends. And I don't want to
have to remember to flip a switch depending on the day of the
week. I already had built a PIC-based device to control some
landscape lighting via the X-10 home automation protocol, so
I added an X-10 chime, and programmed the device to sound the
chime on weekday mornings. Finally, I wired the alarm signal
from a radio controlled clock into my homebuilt device, so that
my device syncs up every day with the precisely correct time.
Voila! My perfect alarm clock! It survived a 17-hour power
interruption earlier this month, and when power was restored,
it was still accurate to better than a second.
See http://www.lanset.com/rcochran/cm11a.html for details on
what I did.
--Rich
Charles Wood
"Robert Scott" <no-...@nowhere.com> wrote in message
news:3de36a3...@news.wwnet.net...
>
> This rant is about clocks - the ordinary AC-powered kind that you buy
> at retail outlets and put next to your bed so that you know when to
> get up and start your day.
Check out these features: (found it on google)
H.H. Scott SCR120 E-Z Set AM/FM Dual-Alarm Clock Radio
The E-Z Set clock radio could just as well be called the "Never-Set." That's
because its built-in computer chip will automatically update all time
changes for nearly a century. Plug it in and the E-Z Set instantaneously
adjusts to the correct year, month, date, day, and time. The only thing you
have to set is your time zone. After that, all your settings are held, even
if the AC power source is interrupted.
This unit has a nice look, manageable size, and a great display that
includes a dimmer with high/low settings. At a glance you'll know the time,
date, and weekday. Since the E-Z Set self-adjusts for Daylight Savings and
leap year, you'll never wake again wondering what the day, date, or time is.
Dual independent alarms offer flexible wake options. For each alarm, you can
choose from weekday, weekend, or everyday wake modes, as well as radio or
buzzer alarm options. The E-Z Set's snooze lets you slumber in nine-minute
increments for up to an hour, and its snooze bar is easy to access. Another
handy feature is the programmable sleep timer. Set to your preference (up to
90 minutes) and drift to sleep to the built-in radio.
The AM/FM tuner includes a one-touch on/off button, tuning scale, band
selector, and thumbwheel controls for tuning and volume. It has one speaker,
positioned at the back, with monaural output. This unit is primarily a clock
radio--don't expect high-fidelity speaker performance here.
H.H. Scott designed the E-Z Set with an eye for ease, and indeed, most
programming buttons are intuitive and accessible. One exception is the dual
alarms' selector switch, which we found awkward to use and poorly spaced.
Additionally, the sleep timer button could be larger.
Despite these quibbles, however, this clock radio is notable for its ease of
use--you shouldn't find yourself reaching for the user's guide very often.
To complete the package, the E-Z Set comes with a great power-failure system
that relies on and includes a 3-volt lithium battery, which should last from
three to five years. And, as added backup, you can always manually set the
clock. H.H. Scott includes a limited one-year parts and labor warranty.
Uwe Hercksen
Spehro Pefhany schrieb:
>
> That or move to some place with reliable power. ;-)
Hello,
there are places with relaible power within the USA?
Sorry, could not resist.
Bye
Uwe Hercksen
Robert Scott schrieb:
>
> I finally found a clock that had decent accuracy in battery-backup
> mode. It was made by Timex. But then I noticed that it was drifting
> by about 2 minutes per month during normal operation. Even though it
> is AC-powered, the people at Timex seem to have totally ignored the
Hello,
2 minutes per month, this is 44.8 ppm. Not too bad for a
standard consumer quartz oscillator.
2.7 seconds per month would require a 1 ppm quartz.
Bye
Trevor Barton
> Dual independent alarms offer flexible wake options. For each alarm, you can
> choose from weekday, weekend, or everyday wake modes, as well as radio or
> buzzer alarm options. The E-Z Set's snooze lets you slumber in nine-minute
> increments for up to an hour, and its snooze bar is easy to access. Another
Now there's a thing - snooze buttons often seem to be 9 minutes (well, I've
noticed 9 minutes several times in the past). Why 9? Can anyone figure out
what's magical about 9? 32768 / 60 (secs/minute) / 60 (Hz) gives 9.1, but
I've seen 9 minutes on 50Hz clocks as well?
It's a mystery!
Trev
--
Trevor Barton
Isotek Electronics Ltd, 9 Clayton Wood Bank, Leeds, LS16 6QZ, UK.
Tel: +44 (113) 275 1339, Fax +44 (113) 224 9827
Remove X from t...@Xisotek.co.uk before replying.
Views expressed are my own and not necessarily those of Isotek Electronics Ltd.
rickman
>
> "Robert Scott" <no-...@nowhere.com> wrote in message
> news:3de36a3...@news.wwnet.net...
> >
> > This rant is about clocks - the ordinary AC-powered kind that you buy
> > at retail outlets and put next to your bed so that you know when to
> > get up and start your day.
>
> Check out these features: (found it on google)
>
> H.H. Scott SCR120 E-Z Set AM/FM Dual-Alarm Clock Radio
My clock is not quite that fancy since it does not even have an alarm.
But it will set itself once it is started and setup. It is an analog
display clock and you have to do a little dance with the one button on
the back to let it know where the hands are. After that it picks up the
WWV signal and synchronizes to it.
It even automatically adjusts itself for daylight savings. This last
change I happened to be up at 2 AM when my PC set itself back to 1 AM.
I noticed that my analog clock was running oddly. Seems it is not
capable of turning itself backwards, so it was advancing at eight times
the normal rate! It had to set itself forward by 11 hours and took some
hour and a half to do it. It was pretty weird watching it run so fast.
Made me feel like I was getting older by the minute. :)
--
Rick "rickman" Collins
rick.c...@XYarius.com
Ignore the reply address. To email me use the above address with the XY
removed.
Arius - A Signal Processing Solutions Company
Specializing in DSP and FPGA design URL http://www.arius.com
4 King Ave 301-682-7772 Voice
Frederick, MD 21701-3110 301-682-7666 FAX
rickman
>
> On Tue, 26 Nov 2002 22:22:03 GMT, Charles Wood wrote:
> > Dual independent alarms offer flexible wake options. For each alarm, you can
> > choose from weekday, weekend, or everyday wake modes, as well as radio or
> > buzzer alarm options. The E-Z Set's snooze lets you slumber in nine-minute
> > increments for up to an hour, and its snooze bar is easy to access. Another
>
> Now there's a thing - snooze buttons often seem to be 9 minutes (well, I've
> noticed 9 minutes several times in the past). Why 9? Can anyone figure out
> what's magical about 9? 32768 / 60 (secs/minute) / 60 (Hz) gives 9.1, but
> I've seen 9 minutes on 50Hz clocks as well?
>
> It's a mystery!
You are thinking too hard about this. Forget all the binary stuff, it
is just a single minute digit! To go to 10 minutes adds another digit.
rickman
45 ppm may sound good, but I believe watches are trimmed at the factory
to get closer to the correct frequency. I may be wrong, but I have
never had a watch that was that far off. I have a wall clock that is
within a minute a year. I have another wall clock that has a radio to
synchronize to WWV. It ran for a year without me realizing that it was
not picking up the signal! I had it in the middle of an office building
and it kept perfect time without ever synchronizing. I think I finally
realized it was not synching when I changed the time for daylight
savings and when I could not get it to synch, realized that if it were
synching it would have self corrected for the time change.
Hans-Bernhard Broeker
> 45 ppm may sound good, but I believe watches are trimmed at the factory
> to get closer to the correct frequency.
Decent ones: probably. I wouldn't bet on that in a 2 US$
trash-it-when-the-battery's-up type of watch.
> I may be wrong, but I have never had a watch that was that far off.
There's another detail that makes watches easier to calibrate than
stand-alone clocks: the owner's wrist makes for a rather nice thermal
regulator --- i.e. no drift by temperature variation to be taken care
of. Ever wondered why even watches with plastic housing all have at
least their back plate made out of metal? Thermal connection to the
skin may very well be the reason.
--
Hans-Bernhard Broeker (bro...@physik.rwth-aachen.de)
Even if all the snow were burnt, ashes would remain.
Robert Scott
<bro...@physik.rwth-aachen.de> wrote:
>
>There's another detail that makes watches easier to calibrate than
>stand-alone clocks: the owner's wrist makes for a rather nice thermal
>regulator --- i.e. no drift by temperature variation to be taken care
>of. Ever wondered why even watches with plastic housing all have at
>least their back plate made out of metal? Thermal connection to the
>skin may very well be the reason.
I've heard this argument before, but somehow, I just don't buy it.
Most wrist watches spend a considerable amount of time off the wrist,
lying in a drawer. My watch spends only about 8 hours a day on my
wrist. The rest of the time it is exposed to the same temperature
variations that would affect any other quartz-controlled clock in a
residential environment. And since the ambient temperature in my
house in considerably different from my skin temperature, the
uncertain duty cycle of wrist-use would make calibration a bit harder
than the calibration of a clock that is expected to be in a typical
residential environment 100% of the time.
Trevor Barton
> Trevor Barton wrote:
>>
>> On Tue, 26 Nov 2002 22:22:03 GMT, Charles Wood wrote:
>> > Dual independent alarms offer flexible wake options. For each alarm, you can
>> > choose from weekday, weekend, or everyday wake modes, as well as radio or
>> > buzzer alarm options. The E-Z Set's snooze lets you slumber in nine-minute
>> > increments for up to an hour, and its snooze bar is easy to access. Another
>>
>> Now there's a thing - snooze buttons often seem to be 9 minutes (well, I've
>> noticed 9 minutes several times in the past). Why 9? Can anyone figure out
>> what's magical about 9? 32768 / 60 (secs/minute) / 60 (Hz) gives 9.1, but
>> I've seen 9 minutes on 50Hz clocks as well?
>>
>> It's a mystery!
>
> You are thinking too hard about this. Forget all the binary stuff, it
> is just a single minute digit! To go to 10 minutes adds another digit.
But even that's hard, in that the first minute is spent on the same digit
you woke up on and so you have to stop at the start of (digit-1). I'd
believe that if it was 10 minutes, so you just stop the snooze when the digit
has not been, and then comes back to, the digit you started on. In the 10
minute case, you need only do logical comparisons, in the 9 minute case
you need to do maths on the original number.
Meindert Sprang
news:slrnau9r1...@sheep.isotek.co.uk...
> >
> > You are thinking too hard about this. Forget all the binary stuff, it
> > is just a single minute digit! To go to 10 minutes adds another digit.
>
> But even that's hard, in that the first minute is spent on the same digit
> you woke up on and so you have to stop at the start of (digit-1). I'd
> believe that if it was 10 minutes, so you just stop the snooze when the
digit
> has not been, and then comes back to, the digit you started on. In the 10
> minute case, you need only do logical comparisons, in the 9 minute case
> you need to do maths on the original number.
>
I think I know why it is 9 minutes: if you hit the snooze button and the
clock remembers the current minute digit to compare to for the next alarm,
this alarm would occur immediately, since the current minute is not over. It
takes less logic just to compare to the previous digit than first having to
wait until the next minute and then starting to compare.
Meindert
Casey
> On Wed, 27 Nov 2002 10:14:35 -0500, rickman wrote:
> > Trevor Barton wrote:
> >>
> >> Now there's a thing - snooze buttons often seem to be 9 minutes (well, I've
> >> noticed 9 minutes several times in the past). Why 9? Can anyone figure out
> >> what's magical about 9? 32768 / 60 (secs/minute) / 60 (Hz) gives 9.1, but
> >> I've seen 9 minutes on 50Hz clocks as well?
> >
> > is just a single minute digit! To go to 10 minutes adds another digit.
>
> But even that's hard, in that the first minute is spent on the same digit
> you woke up on and so you have to stop at the start of (digit-1). I'd
> believe that if it was 10 minutes, so you just stop the snooze when the digit
> has not been, and then comes back to, the digit you started on. In the 10
> minute case, you need only do logical comparisons, in the 9 minute case
> you need to do maths on the original number.
This 9 minute thing has always bothered me as well.
I don't see the logic in it either, unless it just because common
and worked its way into the clock chips.
I once owned a GE clock radio that actually allowed you to set the
snooze duration.
At least a 10 minute snooze would be less confusing. I'll confess to
laying in bed, looking at the clock and trying to calculate 9 minutes
from the last time the @$&# clock went off...
Casey
"It is easier to get older than it is to get wiser."
Spehro Pefhany
Or it's actually ten minutes and the display is truncated rather than
rounded.
Meindert Sprang
news:vw8F9.7689$0b....@news2.central.cox.net...
>
> This 9 minute thing has always bothered me as well.
> I don't see the logic in it either, unless it just because common
> and worked its way into the clock chips.
>
> I once owned a GE clock radio that actually allowed you to set the
> snooze duration.
>
> At least a 10 minute snooze would be less confusing. I'll confess to
> laying in bed, looking at the clock and trying to calculate 9 minutes
> from the last time the @$&# clock went off...
That's it! Trying to make that calculation wakes you up sufficiently so
you'll never make it to the next alarm...
Meindert
Jim Granville
>
> On 27 Nov 2002 11:30:26 GMT, Trevor Barton <t...@Xisotek.co.uk> wrote:
>
> >Now there's a thing - snooze buttons often seem to be 9 minutes (well, I've
> >noticed 9 minutes several times in the past). Why 9? Can anyone figure out
> >what's magical about 9? 32768 / 60 (secs/minute) / 60 (Hz) gives 9.1, but
> >I've seen 9 minutes on 50Hz clocks as well?
>
A good example of how far back 'legacy' can go :)
From a frugal silicon perspective, (ignoring the legacy
electromechanical
precendent for a moment) :
With 9 minutes, you can enable and set snooze in a single 4 bit
RAM variable, and avoid second comparisons. The SW also does not
change the larger AlarmSet variable
something like
IF AlarmOFF THEN
Snooze := 0FH;
ELSIF SnoozeButtonEdge THEN
Snooze := (Minutes + 9) MOD 10; (* Keep in BCD *)
END;
and to run
IF Snooze < 10 THEN (* 0FH is off, 0..9 is active *)
IF Minutes = Snooze THEN (* Disables if Snooze pressed again *)
AlarmOn;
END;
END;
and in operation, because the seconds is not compared, the
exact sleep-quiet-time would vary depending on when the Snooze was
pressed.
( tho mostly it would be within 1..10 seconds :)
Advancing by 10 minutes would have to include LSB, and so use more
RAM space, and code space.
-jg
Tim
> Even the cheapest wrist watches can be expected to be within 2 seconds
> over an 8-hour period. What possible excuse is there for not
> maintaining that accuracy in battery-backup mode?
Wrist watches are nicely temperature stabilizer at 98F.
Are they still accurate when used as wall clocks, with a
magnifying glass :-)
Tim
> Even the cheapest wrist watches can be expected to be within 2 seconds
> over an 8-hour period. What possible excuse is there for not
> maintaining that accuracy in battery-backup mode?
Wrist watches are nicely temperature stabilizer at 98F.
Tim
> Even the cheapest wrist watches can be expected to be within 2 seconds
> over an 8-hour period. What possible excuse is there for not
> maintaining that accuracy in battery-backup mode?
Wrist watches are nicely temperature stabilizer at 98F.
Tim
> Even the cheapest wrist watches can be expected to be within 2 seconds
> over an 8-hour period. What possible excuse is there for not
> maintaining that accuracy in battery-backup mode?
Wrist watches are nicely temperature stabilizer at 98F.
Tim
Uwe Hercksen
> Robert Scott wrote:
>
> > Even the cheapest wrist watches can be expected to be within 2 seconds
> > over an 8-hour period. What possible excuse is there for not
> > maintaining that accuracy in battery-backup mode?
Hello,
2 seconds of an 8-hour period, that is 70 ppm error, or 3
minutes in a month.
Bye
Uwe Hercksen
rickman schrieb:
>
> 45 ppm may sound good, but I believe watches are trimmed at the factory
> to get closer to the correct frequency. I may be wrong, but I have
> never had a watch that was that far off. I have a wall clock that is
Hello,
PXO • Uncompensated Crystal Oscillators (Clocks, XO)
have a stability of +- 10 to 100 pm.
http://www.telequarz.de/pxo.htm
TCXO • Temperature Compensated Crystal Oscillators usually
have a stability of +- 2 to 5 ppm and a yearly aging of 1
ppm.
http://www.telequarz.de/tcxo.htm
OCXO • Oven Controlled Crystal Oscillators are better, but
not useful for a battery operated clock.
http://www.telequarz.de/ocxo.htm
Bye
Meindert Sprang
news:as3nah$qug$1$8300...@news.demon.co.uk...
> Robert Scott wrote:
>
> > Even the cheapest wrist watches can be expected to be within 2 seconds
> > over an 8-hour period. What possible excuse is there for not
> > maintaining that accuracy in battery-backup mode?
>
> Wrist watches are nicely temperature stabilizer at 98F.
Somehow I have trouble buying this stabilizer stories. I take my watch off
at night and when I'm riding my bicycle in wintertime I find it hard to
believe that the watch is still at a stable bodytemerature. Still it runs
with amaring precision. Never more than a few seconds off every six months I
change time.
Meindert
Spehro Pefhany
<mhsp...@NOcustomSPAMware.nl> wrote:
>Somehow I have trouble buying this stabilizer stories. I take my watch off
>at night and when I'm riding my bicycle in wintertime I find it hard to
>believe that the watch is still at a stable bodytemerature. Still it runs
>with amaring precision. Never more than a few seconds off every six months I
>change time.
Compare with an automobile clock that might have to deal with a -20'C
to 40'C range.
Best regards,
RoyalHeart
Yes, there ARE places in the USA with reliable power. I live in SC, and
the last time we had a power loss was, erm, ah, umm, let me see.....when
was it? Hmm....can't remember, it's been so long.....6 months, a year
maybe? Not sure.
RoyalHeart
>
>
> Now there's a thing - snooze buttons often seem to be 9 minutes (well, I've
> noticed 9 minutes several times in the past).
The snooze on my clock gives me 7 extra minutes of shut-eye. Never thought about
about it's 7 minutes. Any thoughts as to why?
Spehro Pefhany
wrote:
>I too don't understand why xtal controlled clocks (and all of them
>are) don't use the mains power frequency to do a correction.
I'll bet that I won't find a crystal in my AC-powered bedroom clock.
Uwe Hercksen
Peter schrieb:
>
> I too don't understand why xtal controlled clocks (and all of them
> even do an error detection over a long period, say 1 week, and do an
> internal trim of the crystal. Very simple.
Hello,
you need a very good filtering of the mains power to remove
all spikes.
Long ago I had a Perkin Elmer processor with onboard clock
derived from the main frequency.
The clock error was impressive, 5 minutes within an hour. I
soldered a capacitor to the right place and got a reduction
to only 5 seconds per hour.
Bye
Ian Okey
> On Sat, 30 Nov 2002 19:51:06 +0000, the renowned z...@ds1.com (Peter)
> wrote:
>
> >I too don't understand why xtal controlled clocks (and all of them
> >are) don't use the mains power frequency to do a correction.
>
> I'll bet that I won't find a crystal in my AC-powered bedroom clock.
>
> Best regards,
> Spehro Pefhany
I have a mains powered analogue clock that keeps excellent time,
correct
it for BST/GMT change twice a year. IIRC here in the UK the power
generators
are required by law to produce the correct number of 50Hz mains supply
oscillations each year, hence a synchronous motor makes an extremely
accurate clock.
Ian
rickman
>
> Uwe Hercksen <Herc...@mew.uni-erlangen.de> wrote
>
> >you need a very good filtering of the mains power to remove
> >all spikes.
>
> the mains.
Years ago I was building a digital clock and was trying to calibrate it
against the AC line not knowing that it is only long term accurate and
not short term. I saw drift of the AC line frequency of at least +- 1
Hz lasting for periods of several tens of seconds or even minutes. This
would require a *very* slow filter to keep the AC line frequency
deviations from dragging the PLL around. I guess since the AC line
doesn't vary enough to cause a noticable error in the clock display, the
PLL will have an even smaller error.
But I still like my Atomic Clocks, especially the analog one that runs
backwards for the DST change in the fall. :)
Kenneth Porter
news:22526900.02112...@posting.google.com:
> If a clock is not self-setting,
> then it's a maintenance item, and I'd prefer not to have it in
> my house. I wish I could find a modern microwave oven without
> a clock.
After suffering having to set a VCR's clock regularly, and disabling the
auto-set feature because the local PBS station had a lousy timebase, I'm
now addicted to my TiVo's sense of time, which sets itself each night with
an NTP connection while collecting TV schedule updates from TiVo's web-
based schedule service. Most people do this via dial-up, but you can also
use a DHCP-configured USB NIC in the TiVo Series 2 boxes.
That suggests that modern clock-equipped equipment should include WiFi or
Bluetooth capability to connect to the house's broadband connection and get
time via NTP. The hard part is of course creating a zero-config consumer
network client.
--
Kenneth Porter
http://www.sewingwitch.com/ken/
Kenneth Porter
news:fuSE9.32713$hi6....@nwrddc02.gnilink.net:
> The E-Z Set's snooze lets you slumber in nine-minute
> increments for up to an hour, and its snooze bar is easy to access.
> Another handy feature is the programmable sleep timer. Set to your
> preference (up to 90 minutes) and drift to sleep to the built-in
> radio.
Back in the early 80's I bought a GE clock/radio/phone that had
programmable *snooze*, settable from 0-59 minutes. Why don't they offer
that anymore?
Kenneth Porter
@mew.uni-erlangen.de:
> you should buy a radio controlled clock.
I did that for my wristwatch:
Alas, the reception isn't great indoors so pages are iffy.
Now if they'd just put GPS position, a databank, and PC upload/download
capability all together in one unit.... Timex' GPS unit just gives distance
and velocity, not position, and has no PC integration, so it's not really
useful for orienteering.
Uwe Hercksen
rickman schrieb:
>
> Years ago I was building a digital clock and was trying to calibrate it
> against the AC line not knowing that it is only long term accurate and
> not short term. I saw drift of the AC line frequency of at least +- 1
> Hz lasting for periods of several tens of seconds or even minutes. This
> would require a *very* slow filter to keep the AC line frequency
> deviations from dragging the PLL around. I guess since the AC line
> doesn't vary enough to cause a noticable error in the clock display, the
> PLL will have an even smaller error.
Hello,
+- 1 Hz is very unstable. The European common mains
frequency drifts only +- 50 mHz.
Bye
rickman
I could easily be wrong about the magnitude, it was a long time ago. I
do remember thinking I had a drift problem in the crystal and trying to
debug it for a day or two before I realized that it was the power line
that was drifting. It would go back and forth. I was trying to adjust
the crystal and every time I would get it right and leave it for 5
minutes, it would be off again when I came back.
It seems 50 mHz is very, very small considering that this is coming from
very large mechanical devices. Can they really control the speed of a
large generator that well?
Jim Granville
<snip>
> very large mechanical devices. Can they really control the speed of a
> large generator that well?
Thats one part in 1000, or 1 rpm per 1000 rpm, so is not a difficult
regulation.
Also to engage/disengage a generator, they must phase lock within
a fraction of a cycle, so 1Hz wobble on a grid must be a nightmare...
A quick test here showed the mains 35mHz above 50Hz
-jg
Robert Scott
wrote:
>
>It seems 50 mHz is very, very small considering that this is coming from
>very large mechanical devices. Can they really control the speed of a
>large generator that well?
>
A specification of frequency tolerance is meaningful only if an
averaging period is specified. 50 mHz would not be difficult to
maintain if the frequency were measured over a period of several
seconds.
For purposes of locking to the power grid, it is not the frequency but
the instantaneous phase that is important. If two power souces are
out of phase by a certain number of degrees, large currents would flow
between the two generators. The relationship between frequency and
phase is one of a function and its integral. 50 mHz frequency error,
if continued over a 10 second period, would result in a phase error of
180 degrees, which would be absolutely disasterous.
-Robert Scott
Ypsilanti, Michigan
(Reply through newsgroups, not by direct e-mail, as automatic reply address is fake.)
Ville Voipio
> For purposes of locking to the power grid, it is not the frequency but
> the instantaneous phase that is important. If two power souces are
> out of phase by a certain number of degrees, large currents would flow
> between the two generators.
If we talk about generators, they are usually synchronous generators
whose output power depends on the phase shift. So, if one generator
wants to run faster than the others, it advances slightly in phase and
takes larger part of the load. So, as long as you connect them to the
grid in the right phase, they'll keep in phase.
As the load varies, the frequency changes unavoidably.
However, the frequency can be adjusted higher by adding more power
production to the grid and slower by decreasing the amount of power
produced. Of course, the changes in production are slower
than changes in load, so it is quite impossible to keep the frequency
very stable.
A lot depends on the power grid size. If you have a large grid with
huge amounts of power production capacity, its frequency is
probably more stable than that of a small grid.
I am not aware of the systems used in the US, but at least here we
have a three-country grid (Finland, Norway, Sweden). IIRC, the
maximum deviation over long time is one cycle per day (i.e.
7.3 seconds per year). In the short run the deviation may be
larger. Right at the moment the deviation seems to be 8.45 s and
frequency 49.98 Hz. (In practice, the cumulative deviation is zero
in the long run, as the grid is synchronized to a primary atomic
clock, but the official specification is 7.3 seconds per year.)
(For real-time information, see:
http://www.fingrid.fi/engl/markkinainfo/markkinainfo_vaakajako.html )
So, at least here it is possible to make a decent clock by using
the power grid frequency. That is, the clock is good in the sense that
if you don't care about ten seconds' lead or lag, you won't need to
touch the clock ever. OTOH, if you need millisecond accuracy or good
frequency stability, the power grid is a lousy reference.
- Ville
--
Ville Voipio, Dr.Tech., M.Sc. (EE)
Rick Merrill
> So, at least here it is possible to make a decent clock by using
> the power grid frequency.
If you're going to *cheat* :-) and use an external source,
why not use the GPS signals? They use an atomic clock!
Rick
Merrill
Kenneth Porter
news:88rpuugu4e7s579qs...@4ax.com:
>>Back in the early 80's I bought a GE clock/radio/phone that had
>>programmable *snooze*, settable from 0-59 minutes. Why don't they offer
>>that anymore?
>
> Because no one could figure out how to program it?
This one was pretty easy to use: Hit Snooze, then Hit up/down and fast to
set it to what you want, the same way you'd adjust the sleep timer.
> the majority of people out there buying consumer electronics can't
> set the time without the manual, and even then, its painful.
Why do you suppose that is? Fundamental problem in most people's way of
thinking (ie. are we the ones that are weird?) or problems in our
educational system, or something else?
(I'm reminded of a gag VCR "repair kit" that includes a short strip of
black electrical tape to put over the blinking "12:00".)
Mark Zenier
I have a much more recent GE clock radio with that feature. It uses
the Sanyo LM8560 clock IC.
Mark Zenier mze...@eskimo.com Washington State resident
Jim Granville
>
> Alex Pavloff <BLAHap...@BLAHeason.com> wrote in
> news:88rpuugu4e7s579qs...@4ax.com:
>
> >>Back in the early 80's I bought a GE clock/radio/phone that had
> >>programmable *snooze*, settable from 0-59 minutes. Why don't they offer
> >>that anymore?
> >
> > Because no one could figure out how to program it?
>
> This one was pretty easy to use: Hit Snooze, then Hit up/down and fast to
> set it to what you want, the same way you'd adjust the sleep timer.
You mean it had a DOWN button as well as UP ?!
That's be the reason it failed. Something truly intuitive doesn't
seem like technology... :)
-jg
Casey
> Kenneth Porter wrote:
> > This one was pretty easy to use: Hit Snooze, then Hit up/down and fast to
> > set it to what you want, the same way you'd adjust the sleep timer.
>
> You mean it had a DOWN button as well as UP ?!
> That's be the reason it failed. Something truly intuitive doesn't
> seem like technology... :)
Yeah, next thing you know, they'll make it easy to set clocks
*back* one hour... sheesh.
Casey
"It is easier to get older than it is to get wiser."
Casey
> On Wed, 04 Dec 2002 19:24:21 GMT, Casey <c...@cox.netremove> wrote:
>
> >Jim Granville said...
> >> Kenneth Porter wrote:
> >
> >> > This one was pretty easy to use: Hit Snooze, then Hit up/down and fast to
> >> > set it to what you want, the same way you'd adjust the sleep timer.
> >>
> >> You mean it had a DOWN button as well as UP ?!
> >> That's be the reason it failed. Something truly intuitive doesn't
> >> seem like technology... :)
> >
> >Yeah, next thing you know, they'll make it easy to set clocks
> >*back* one hour... sheesh.
>
> while the minute hand sweeps around.... it only takes a minute or so
> of leaning over the steering wheel.
Strangely enough my Ford Expedition has the easiest clock to set back
out of all the ones I own. A couple of clicks to set it back an
hour.
Now, my microwave is pure torture. Just to set the clock on the
thing (when the power fails, which is frequently) I have to set the
hour, the minutes, select AM or PM (who cares), and the month, day,
and year! I have yet to find any reason why it cares, but it forces
me to set it.
Kelly Hall
"Alex Pavloff" <BLAHap...@BLAHeason.com> wrote in message
news:4bssuugihpmnr4n7u...@4ax.com...
> I love the clock in my car. $35,000 and I have to hold the knob out
> while the minute hand sweeps around.... it only takes a minute or so
> of leaning over the steering wheel.
1999 BMW F650 motorcycle - rack the forks to the right, reach up underneath
the wind fairing (from the front), feel around for the small metal arm on
the side of the clock case, press the arm to the casing, release when analog
clock spins to correct-ish time. Note - clock is only visible from the back
(ie, seat-side). This job is best accomplished by two people. Other than
that, I really like the bike ;)
Kelly
Ian Okey
The best one that I have is on my video recorder. It reads the
time broadcast by the TV company as part of the teletext service
and resete itself whenever the timezone changes, it loses power
or it just drifts off too far.
Now if I could just network.........
Ian
Kenneth Porter
@posting.google.com:
> The best one that I have is on my video recorder. It reads the
> time broadcast by the TV company as part of the teletext service
> and resete itself whenever the timezone changes, it loses power
> or it just drifts off too far.
Assuming your local station has decent time. Mine was off by several
minutes, making the feature useless. (The time in the blanking interval
apparently didn't use the same timebase used to time the start of
programs.)
Kenneth Porter
@eskinews.eskimo.com:
> I have a much more recent GE clock radio with that feature. It uses
> the Sanyo LM8560 clock IC.
What's the model? I'd love to find another one like that!
Kenneth Porter
@news2.central.cox.net:
> Now, my microwave is pure torture. Just to set the clock on the
> thing (when the power fails, which is frequently) I have to set the
> hour, the minutes, select AM or PM (who cares), and the month, day,
> and year! I have yet to find any reason why it cares, but it forces
> me to set it.
Put it on a UPS? :D
Casey
I was disappointed when my Sony VCR seemed to lock to the time
signal, but was a couple of minutes off.
Casey
> Casey <c...@cox.netremove> wrote in news:eqwH9.2626$xU1.1268
> @news2.central.cox.net:
>
> > Now, my microwave is pure torture. Just to set the clock on the
> > thing (when the power fails, which is frequently) I have to set the
> > hour, the minutes, select AM or PM (who cares), and the month, day,
> > and year! I have yet to find any reason why it cares, but it forces
> > me to set it.
>
> Put it on a UPS? :D
Good idea!
Then I could make some popcorn while sitting in the
dark with nothing else to do...
Matthew S. Staben
> This rant is about clocks - the ordinary AC-powered kind that you buy
> at retail outlets and put next to your bed so that you know when to
> get up and start your day. Why is it posted in a newsgroup about
> embedded systems? Because I am a developer of embedded systems and I
> know, as do most of you, how easy it would be to design a good clock.
> The sad fact is that it is not being done well at all.
I'll second that! Take for instance a Samsung microwave oven we
received as a wedding present. When we powered up the oven about 8
months later (bought a house, moved, needed it) I noticed it couldn't
keep the time. I'd set it, and lo, the next day it would be ahead by
more than an hour ...
Deciding my Seiko Chronograph which manges to keep the NIST time
pretty closely ought to be a reliable benchmark, I timed the minute
change on the microwave by starting the chronograph when the minute
changed, and stopping it when the minute changed again. I was a
little surprised by the results: The microwave ticked "one minute"
for each 56 seconds!
This means that at five bona-fide minutes, the microwave was 20
seconds ahead; one minute gained for every fifteen minutes; four
minutes gained every hour; 96 minutes (1 hour, 36 minutes) gained
every day! I consequently judged the clock on this microwave utterly
useless, if not dangerous. Fortunately, however, I found that if the
microwave is re-powered and the clock is never set, the display shows
a " :0" representing the time remaining for cooking, much better than
a mysterious number that coincides with the time once every 15 days
...
Matt
Eric Smith
> Back in the early 80's I bought a GE clock/radio/phone that had
> programmable *snooze*, settable from 0-59 minutes. Why don't they offer
> that anymore?
The GE models I liked (3880 or 3885) didn't have a phone, but they had
a numeric keypad. You could enter the time, alarm, or radio frequency
directly. They still had up and down buttons for the frequency as well.
I don't recall whether they had programmable snooze.
I *despise* setting most digital clocks. The alarm clock I use now,
a Sony, at least has both forward and reverse buttons so if you overshoot
you don't have to go all the way around again.
Unfortunately I no longer have my GE 3880, but I'm looking to buy one.
Otherwise I may have to design and build my own. If so, I'll do a
lot of things differently than the commercially available ones, for
instance:
* separate alarm times for weekdays and weekends - many clock
radios are sold claiming this, but they really just have two
separate alarms, and you have to remember to arm the right one
* gradually increasing alarm volume
* X10 control of my room lighting, to slowly ramp the intensity
for artificial sunrise
Casey
> * gradually increasing alarm volume
I won't even buy one without this any more.
Unfortunately, this narrows the field down a lot.
Greg Deuerling
says...
Ahh, a true engineer !!! At least I'm not the only one out
there who knows what chips are in his clock radio :)
I drive my wife nuts, "Why are you taking that apart ?", Me: "To see
whats in it !"
--
gad
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
=-=
= Greg Deuerling, Fermi National Accelerator Laboratory
=
= P.O.Box 500 MS368 Batavia, IL 60510 (630)840-4629, FAX (630)840-
5406 =
= Electronic Systems Engineering Group
=
= Work: eg...@fnal.gov Personal: g...@elnet.com
=
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
=-=
Hans-Bernhard Broeker
[...]
> I drive my wife nuts, "Why are you taking that apart ?", Me: "To see
> whats in it !"
As a matter of fact, that's what I think really is the best possible
description of someone who, at least in their heart if not by
profession, is an engineer: someone who wouldn't even _consider_
throwing away a broken gadget without opening it and having a look
inside --- to check if it can possibly be repaired, but at least as
much out of sheer curiosity.
Yet more engineer-ish people will open a device as soon as the
warranty period is over (or whatever other obstacles are cleared,
including wild opposition by other family members ;-). Truly
hard-core engineers on the day they get it, no matter what.
--
Hans-Bernhard Broeker (bro...@physik.rwth-aachen.de)
Even if all the snow were burnt, ashes would remain.
Mark Zenier
Kenneth Porter <k...@sewingwitch.com> wrote:
>mze...@eskimo.com (Mark Zenier) wrote in news:aslg5u$qhf$1
>@eskinews.eskimo.com:
>
>> I have a much more recent GE clock radio with that feature. It uses
>> the Sanyo LM8560 clock IC.
>
>What's the model? I'd love to find another one like that!
Hmmm, looks like the programmable time is the "Sleep to Music", the
wakeup "Snooze" feature is still 9 minutes.
Anyway, 7-4816 or 7-4817.
Thomson Consumer Electronics, made in China.
rickman
That is so you can put a turkey in the oven at Easter and set the oven
to cook it for Thanksgiving. :)
--
Rick "rickman" Collins
rick.c...@XYarius.com
Ignore the reply address. To email me use the above address with the XY
removed.
Arius - A Signal Processing Solutions Company
Specializing in DSP and FPGA design URL http://www.arius.com
4 King Ave 301-682-7772 Voice
Frederick, MD 21701-3110 301-682-7666 FAX
Zefram Cochrane
"RoyalHeart" <royal...@yahoo.com> wrote in message
news:3DE6F3C3...@yahoo.com...
> Yes, there ARE places in the USA with reliable power. I live in SC, and
> the last time we had a power loss was, erm, ah, umm, let me see.....when
> was it? Hmm....can't remember, it's been so long.....6 months, a year
> maybe? Not sure.
Are you serious ? 6 months between power outages is "so long" ? Here, I've
had about 1 power outage in 10 years (due to severe wind).
Richard [in WR14]
Zefram Cochrane
"Rick Merrill" <RickM...@attbi.com> wrote in message
news:3DED0E45...@attbi.com...
Might as well just use one of the public-domain time frequency
standards ? These are a heck of a lot easier to receive (esp. indoors)
than GPS signals ?
Richard [in WR14]
Zefram Cochrane
"Kenneth Porter" <ken.bl...@sewingwitch.com> wrote in message
> Back in the early 80's I bought a GE clock/radio/phone that had
> programmable *snooze*, settable from 0-59 minutes. Why don't they offer
> that anymore?
Well, if I were the sort of persun who didn't want to wake up when
I was supposed to (I'm not), I'd have better things to do than thinking
"hm.. now how long do I want to snooze today ? 7 minutes 6 ? 11 ?"
and the attendant button pushing. If I were sufficiently alert to do all
that, I might as well get up anyway !
Most people don't colloquially speak of time in units of less than
10 (15? 5?) minutes anyway. No-one says "I'm at the bus stop;
I'll be home in 53 minutes". It's "an hour" or "45 mins".
Richard [in WR14]
Zefram Cochrane
"Ville Voipio" <vvo...@kosh.hut.fi> wrote in message
news:i3kel8y...@kosh.hut.fi...
[picking a post at random]
All this talk about generators and frequency tolerances: Why is it necessary
to
have the frequency locked to such a high degree of accuracy ? Is this done
solely
for the benefit of AC-derived clocks ?
Will my kettle, microwave, computer, lamps, all start to behave erratically
if they are presented with 47Hz or 53Hz over a long period of time ?
For industrial grade motors, surely they use some form of
microprocessor-governed
speed control these days, rather than being tied synchronously to the mains
?
What's the advantage of synch motors over DC motors ? Just the lack of a
commutator ?
Richard [in WR14]
Zefram Cochrane
"rickman" <spamgo...@yahoo.com> wrote in message
news:3DF4F541...@yahoo.com...
> That is so you can put a turkey in the oven at Easter and set the oven
> to cook it for Thanksgiving. :)
You missed the point that he had to set the "YEAR" as well. Even if I
do come from the GB, I know that Thanksgiving is AFTER Easter
in the same year (but before xmas), so you wouldn't need the year
functionality
to do that:
No, no, you want to put the poultry in the uwave oven and xmas in year
YYYY, so that it's ready for thanksgiving in year YYYY+1. THEN, you'd
need the year function.
Similar things that get my goat are:
1. central heating timers that are incapable letting me putt the heating/hw
on
FOR one hour. On all the ones I've seen, I either have to put it into
manual mode (on all the time) and then switch it off after one hour,
or I have to set start-time and finish-time (not start-time and duration, oh
no).
2. Car (and other radios) which advertise (e.g.) 21 preset stations, when
what
they mean is "7 preset buttons, plus a waveband switch LW/MW/FM". This
was OK in the age of mechanical station-selection buttons, but with modern
logic, it is trivial to store the WAVEBAND along with the frequency "in" the
button, so the user could mix/match what channels they wanted on a mix of
wavebands.
Richard [in WR14]
Richard [in WR14]
Zefram Cochrane
"Kenneth Porter" <ken.bl...@sewingwitch.com> wrote in message
> (I'm reminded of a gag VCR "repair kit" that includes a short strip of
> black electrical tape to put over the blinking "12:00".)
I wasn't aware these were available commercially ! Yes, I've stuck
insulting tape over a VCR's display in my time. I've done this for
several reasons:
One VCR I had wouldn't display the time in other than the 12h AM/PM
format, which I can do without.
I had several other clocks on surrounding equipment, all of which gave
slightly different times, and I wanted to have one definitive one.
It (still doesn't?) didn't occur to any VCR manufacturer then that it
might be a good idea to provide battery backup for the clock in
case it was moved from one room to another !
Richard [in WR14]
Zefram Cochrane
> Yeah, next thing you know, they'll make it easy to set clocks
> *back* one hour... sheesh.
Oh. Come on. That's being far too sensible. Better to have just a single
button
that *advances* the minutes (only) and which accelerates to 20min/second
after holding it down for 1.4 seconds, so that you consistently overshoot
even when adjusting forwards. This is a hell of a pig more difficult than
the relative ease of adjusting a clock backwards by pushing "+" 23 times!
CBFalconer
> > SC, and the last time we had a power loss was, erm, ah, umm,
> > let me see.....when was it? Hmm....can't remember, it's been so
> > long.....6 months, a year maybe? Not sure.
>
> Are you serious ? 6 months between power outages is "so long" ?
> Here, I've had about 1 power outage in 10 years (due to severe wind).
I suspect his 'on' period has been seriously truncated, since the
recent ice storms in NC and SC. In this part of the world most
power travels by wires strung on poles, which tend to come apart
when weighted down with heavy ice, not to mention tree branches
falling on them from the same overload.
--
Chuck F (cbfal...@yahoo.com) (cbfal...@worldnet.att.net)
Available for consulting/temporary embedded and systems.
<http://cbfalconer.home.att.net> USE worldnet address!
Robert Scott
>
>All this talk about generators and frequency tolerances: Why is it necessary
>to
>have the frequency locked to such a high degree of accuracy ? Is this done
>solely
>for the benefit of AC-derived clocks ?
A pre-requisite to combining power from various and sundry generators
is that these generators be phase locked. Once you have the ability
to do that, the additional effort to make that phase lock produce good
long-term frequency stability is small. And at once time, almost all
electric clocks were AC synchronous motors, so the advantage to having
these clocks track over the long term was great.
-Robert Scott
Ypsilanti, Michigan
(Reply through newsgroups, not by direct e-mail, as automatic reply address is fake.)
Robert Scott
>
>Are you serious ? 6 months between power outages is "so long" ? Here, I've
>had about 1 power outage in 10 years (due to severe wind).
>
You must live very close to a substation.
RoyalHeart
Wrong. I'm on the GA side. NC got hit hard, but we didn't lose power.
RoyalHeart
Overworked and underpaid. I'll help others when I have the time...and can
afford it.
Hans-Bernhard Broeker
> It (still doesn't?) didn't occur to any VCR manufacturer then that it
> might be a good idea to provide battery backup for the clock in
> case it was moved from one room to another !
At least one got it right. I suspect that was by accident rather than
by design, though ;-)
That VCR I saw keeps its master clock and a copy of all pre-programmed
recordings in the *remote control*. Which is battery-driven anyway,
so it's exactly the right place to put anything that needs batter
backup.
This means you can disconnect the VCR from mains power, either
purposefully or by accident, and just press the "send programming to
VCR" button on the remote after powering it back on and all's well.
Stef
>
>> It (still doesn't?) didn't occur to any VCR manufacturer then that it
>> might be a good idea to provide battery backup for the clock in
>> case it was moved from one room to another !
My new VCR just uses teletext on programmed channel 1 to get the current
time and date. No need for backup and automatic daylight savings time
adjustment. :-)
The old one lost it's time on a 1 second power interruption and needed
resetting, very annoying indeed.
Stef
RP Henry
"Zefram Cochrane" <b@b.b> wrote in message
news:3dfcc42f$1...@news.teranews.com...
I've had two in the last year, both due to auto accidents against power
poles. The first was a minivan which veered off the road and was flipped on
its back by a support guy wire. The pole and guy were undamaged, but the
power lines whipped together and caused the pole-mounted breaker to trip.
The second was a pickup truck whose driver dropped a cigarette and drove off
the road while looking for it. He struck the pole just north of the one
mentioned above and left it completely broken. Before that was about 8
years of unbroken power, which dates back to the time we moved here.
Zefram Cochrane
news:3dfd2e4...@news.wwnet.net...
> >Are you serious ? 6 months between power outages is "so long" ? Here,
I've
> >had about 1 power outage in 10 years (due to severe wind).
> >
>
> You must live very close to a substation.
I am referring to the most recent 10 years, wherein I have moved
house 7 times. During all those 10 years, I've only suffered one power
outage. None of those houses was within about 10kM of a SS.
Thank you for the explication about the need to have multiple generators
locked in freq and phase.
Actually, maybe we'd all be better off with DC. What's the advantage
of AC, anyway (just that it's difficult to transform ?) Is the usage of AC
just a legacy from the Edison Electric Chair wars.
Richard [in WR14]
Zefram Cochrane
news:atkpe7
> That VCR I saw keeps its master clock and a copy of all pre-programmed
> recordings in the *remote control*. Which is battery-driven anyway,
> so it's exactly the right place to put anything that needs batter
> backup.
That's exactly what my Hitachi VCR does: it keeps the time/date in the
RC. This is so that it "knows" the current time date, and can deal with
concepts like "today","tomorrow" and cross-midnight timings and translate
these to real times and dates when it xmits to the VCR.
However, the VCR also automatically syncs its clock from the teletext
signal. I asked the guy in the shop what happened if there was a conflict
(obviously it can't xmit back to the RC) and he found it a strange request.
Richard [in WR14]
rickman
I think you have it backwards. DC is very difficult to efficiently
distribute since it is hard to ship at low voltage and dangerous and
difficult to use at high voltages. AC on the other hand can be
transformed between the two quite easily and efficiently.
Zefram Cochrane
"Alex Pavloff" <BLAHap...@BLAHeason.com> wrote in message
news:phesvus6k5c4iejps...@4ax.com...
> I find myself when dealing with electronics far too often screaming
> "WHO DESIGNS THIS SH....."
Yep. Join the club. Just about every piece of electronics that
I possess, from my digital watch, to my computer, to my VCR,
to my washing machine, all have user-interface issues that make
me think "why the Feck did they do it like that ?"!
> and then remembering that it was the poor
> add-on embedded guy given the cheapest part they could find and
> absolutely no direction on what to do.
Oh, I've been in that situation (crappy parts, no time, no
specification), but I almost always take the care and time
to finesse things properly (even in my own time). Sometimes
I deliberately don't ask for a specification if they don't
give me one up-front; that way, I have the leeway to design
the thing properly.
Richard [in WR14]
Zefram Cochrane
> I think you have it backwards. DC is very difficult to efficiently
> distribute since it is hard to ship at low voltage and dangerous and
> difficult to use at high voltages. AC on the other hand can be
> transformed between the two quite easily and efficiently.
Yes. I got this backwards, of course. My mistake. Transformers
can't transform DC.
Why is DC more dangerous though, and why is it more difficult
to transport at low voltages than AC ? I know low voltages
per se are easier to transport, but why is AC easier to transport
than DC, other than the ability to xform.
Richard [in WR14]
Jim Granville
>
> Zefram Cochrane wrote:
<snip>
> > Actually, maybe we'd all be better off with DC. What's the advantage
> > of AC, anyway (just that it's difficult to transform ?) Is the usage of AC
> > just a legacy from the Edison Electric Chair wars.
>
> I think you have it backwards. DC is very difficult to efficiently
> distribute since it is hard to ship at low voltage and dangerous and
> difficult to use at high voltages. AC on the other hand can be
> transformed between the two quite easily and efficiently.
They both have their place.
AC allows simple power transformers : 50Hz / 60Hz / 400Hz
to move anywhere in the Volt-Amp curve.
Above a certain power level, the skin effect of AC ( even at 50Hz )
causes problems.
Undersea HiPower cables are typically DC, with large DC-AC converters
at either end.
Super conducting cables may have shifted this - anyone know if they
are deployed under-sea yet ?
-jg
Anthony Marchini
> Actually, maybe we'd all be better off with DC. What's the advantage
> of AC, anyway (just that it's difficult to transform ?) Is the usage of AC
> just a legacy from the Edison Electric Chair wars.
>
Tony
Geoff McCaughan
The reason AC is in such widespread use, is that there are two conflicting
requirements for electrical power distribution.
1) Power needs to be at a reasonably low voltage in order to be safely used
around the home. Imagine the insulation required if your toaster operated at
100KV.
2) Power loss on an electrical transmission line is proportional to the
*square* of the current, so in order to minimise wastage, long distance
transmission needs to have the lowest possible current, [and therefore very
high voltages].
With AC, both these requirements can be easily and cheaply met by using
transformers to reduce the high transmission voltages to the 110 or 240
volts used at the consumption point. If you used DC, you'd either have to
put up with high transmission losses, or unacceptably high [and thus
dangerous] voltages in your home wiring, or use expensive and inefficient
DC-DC converters everywhere to reduce the voltage.
CBFalconer
>
... snip ...
>
> Why is DC more dangerous though, and why is it more difficult
> to transport at low voltages than AC ? I know low voltages
> per se are easier to transport, but why is AC easier to transport
> than DC, other than the ability to xform.
Because higher voltages mean less current for the same power
level. Less current means less ohmic losses. DC transmission at
household voltage levels restricts the power source to being a few
miles (or km) from the destination. AC can be transmitted for
many hundreds of miles with respectable losses. P = E*I. Double
the voltage and halve the current, and the same cable has 1/4 the
IsquareR loss.
Everett M. Greene
> "Ville Voipio" <vvo...@kosh.hut.fi> wrote
>
> [picking a post at random]
>
> All this talk about generators and frequency tolerances: Why is it necessary
> to
> have the frequency locked to such a high degree of accuracy ? Is this done
> solely
> for the benefit of AC-derived clocks ?
> Will my kettle, microwave, computer, lamps, all start to behave erratically
> if they are presented with 47Hz or 53Hz over a long period of time ?
>
> For industrial grade motors, surely they use some form of
> microprocessor-governed speed control these days,
> rather than being tied synchronously to the mains?
Have you priced speed controllers for large motors?
> What's the advantage of synch motors over DC motors?
> Just the lack of a commutator?
That'll do. Efficiency is another.
> Richard [in WR14]
Paul Keinanen
<geo...@spam.fqdn.com> wrote:
>With AC, both these requirements can be easily and cheaply met by using
>transformers to reduce the high transmission voltages to the 110 or 240
>volts used at the consumption point. If you used DC, you'd either have to
>put up with high transmission losses, or unacceptably high [and thus
>dangerous] voltages in your home wiring, or use expensive and inefficient
>DC-DC converters everywhere to reduce the voltage.
While his has been the situation for nearly a century, however, the
situation may change. Unfortunately, the mains frequency is quite low,
thus requiring large and heavy transformers. The maximum RPM you can
get out of an AC motor is limited and thus the AC motors have to be
heavy and bulky.
More and more AC motors are driven by inverters (to get higher RPM or
variable speed) and more and more electronic appliances use switching
mode power supplies. In these, the low frequency AC is first
rectified, then switched at a high frequency and in power supplies
transformed with a small high frequency transformer and then rectified
again.
I think it is time to define a low voltage DC distribution voltage. In
this case, the first rectifier in the inverters and SMPSs can be
eliminated as well as the power factor problems.
Since a heavy transformer is still needed between the medium voltage
(5-40 kV) AC distribution and the low voltage system, it would be more
economical to just add some secondary windings to generate phase
shifts and thus a large number of phases (6-12), which simplifies the
rectification at the distribution site and reduces the DC hum and also
reduces the harmonics on the medium voltage side.
While an additional DC distribution network might be impractical for
residential areas, it might be quite usable in large industrial sites,
which have their own medium/low voltage transformers, provided that
the DC distribution voltage could be standardised.
Paul
Hans-Bernhard Broeker
Actually, he's not kidding on this Electric Chair war stuff, either.
He just has it exactly backwards, too. Edison was dogmatically
*against* AC. Edison mounted a massive FUD campaign stating that AC
was prohibitively dangerous to use. Lots and lots of stray cats and
dogs dragged in off the streets are said to have been electrocuted by
Edison Electric just to prove this point. He also pressed the first
electric chair to be built on AC to drive home this sime gripe of his.
Fact is, he was a mile away from the truth, on almost all counts. DC
is actually *more* dangerous than AC, in the practicable household
voltage range. Grabbing a 230 V DC line you would be unable to drop
it again because of massive cramps, and the line voltage would
steadily electrolytize you. You'll have to stay under close
surveillance in a hospital for upwards of a week just to _assess_ the
damage.
With an AC line, you at least stand a chance of tearing yourself
loose, and you don't suffer almost unpredictable medium-term
electrolytic effects. The key danger is cardiac arrest caused by
irritation of the heart's internal clock --- this is very serious, but
you stand a chance of being rescued by paramedics.
Jim Stewart
From a fire and burn standpoint, AC is safer as well. Strike an arc on
230 VAC or below and it will probably extinguish at the next zero
crossing. Not so with DC.
Anthony Marchini
"Jim Stewart" <jste...@jkmicro.com> wrote in message
news:DDFCCE553023E531.0D6A0BEE...@lp.airnews.net...
>
>
> Hans-Bernhard Broeker wrote:
> > Anthony Marchini <amarchini@am_eri_c.anl0ckerdot.com> wrote:
> >
> >>>Actually, maybe we'd all be better off with DC. What's the advantage
> >>>of AC, anyway (just that it's difficult to transform ?) Is the usage of
AC
> >>>just a legacy from the Edison Electric Chair wars.
> >>>
> >>
> >>You better be kidding.
> >
> >
> > Actually, he's not kidding on this Electric Chair war stuff, either.
> > He just has it exactly backwards, too. Edison was dogmatically
> > *against* AC. Edison mounted a massive FUD campaign stating that AC
> > was prohibitively dangerous to use. Lots and lots of stray cats and
> > dogs dragged in off the streets are said to have been electrocuted by
> > Edison Electric just to prove this point. He also pressed the first
> > electric chair to be built on AC to drive home this sime gripe of his.
No, I meant was the person kidding on the question of "What's the advantage
of AC, anyway ?". The answer is academic to anyone with an electricity
background.
The response was condescending and maligns those who are looking for
answers here but are not necessarily electricity experts. I appologise.
On a positive note, the advantage in AC is its ability to be easily
"transformed" from
a low to high voltage and back (with a "little" loss). When you kick it up
to high
voltage, its current drops. When you run this over miles of wire the lower
current means
less energy loss from heat. Less loss means more energy getting to the other
end of the wire
which is a good thing.
Tony
Zefram Cochrane
"Anthony Marchini" <amarchini@am_eri_c.anl0ckerdot.com> wrote in message
news:iD3M9.1341
> No, I meant was the person kidding on the question of "What's the
advantage
> of AC, anyway ?". The answer is academic to anyone with an electricity
> background.
I don't have an electricity-type background. That's why I asked. I can see
by observation that AC is more widely used for distribution. I know that
only AC can be xformed, and am familiar with the "Higher voltages give less
I squared R losses" business.
But what I was getting at was "(why) is it better to transport say 11Kv
at AC" rather than (at the equivalent RMS voltage) DC". Does AC have
lower I^2R losses than the equivalent RMS DC ?
Is the whole advantage in the ability to use transformers to xform from
132Kv(?) all the way down to 240/120V AC via several stages of
transformering.
> The response was condescending and maligns those who are looking for
It was ? I never read it as such.
I didn't know much about the Electric Chair business, only that Edison
was promoting the usage of electricity as a means of "electromort" to show
just how dangerous it was (which it probably wasn't) compared to "his". Now,
I can't remember whether TAE was pro-AC or pro-DC.
Wasn't Westinghouse involved in this somehow ? I seem to remember that there
was a similar AC/DC hoo-hah when they were about to electrify the
Metropolitan
District Railway (forms part of today's District/Circle Line of London
Underground)
Geoff McCaughan
> But what I was getting at was "(why) is it better to transport say 11Kv
> at AC" rather than (at the equivalent RMS voltage) DC". Does AC have
> lower I^2R losses than the equivalent RMS DC ?
The differences in actually transporting AC vs DC are negligible. DC has the
advantage thaqt you don't have to closely sync widely separated generation
plants. DC has a considerable advantage when transporting via undersea
cables, as the conductive nature of seawater results in inductive losses
when using AC.
> Is the whole advantage in the ability to use transformers to xform from
> 132Kv(?) all the way down to 240/120V AC via several stages of
> transformering.
That's pretty much it. Transformers are comparitively cheap, efficient and
low-maintenance. When the electrical infrastructure was first being built,
the only way to do DC-DC conversion was with motor-generator sets.
> I didn't know much about the Electric Chair business, only that Edison
> was promoting the usage of electricity as a means of "electromort" to show
> just how dangerous it was (which it probably wasn't) compared to "his". Now,
> I can't remember whether TAE was pro-AC or pro-DC.
> Wasn't Westinghouse involved in this somehow ?
Westinghouse was pro AC, Edison was pro DC.
Zefram Cochrane
> The differences in actually transporting AC vs DC are negligible.
Yes. That's what I thought.
>[advantages of DC: no need to sync; better behavior in submarine cables]
>[AC can be transformed]
> That's pretty much it. Transformers are comparitively cheap, efficient and
OK. Thank you all. I got it now.
> Westinghouse was pro AC, Edison was pro DC.
So Edison developed the AC electric chair to show how dangerous
AC allegedly was: is that correct ?
In those states that use electric chairs, do they use AC or DC nowadays ?
Isn't there one town where usage of the electric chair lowers the
voltage available for the rest of the town ?