LED alarm clocks all lose accuracy over time

[JW]

On Sat, 19 May 2012 11:55:04 -0700 Jeff Liebermann <jeffl@cruzio.com>
wrote in Message id: <gkqfr7tifijlt1n4l7frdrb5pumccv7tad@4ax.com>:

On Sat, 19 May 2012 18:24:04 +0000 (UTC), "Geoffrey S. Mendelson"
gsm@mendelson.com> wrote:

It's because the US is trying to get rid of those broadcasts. With GPS they
are obsolete.

I beg to differ. The cost of a GPS disciplined oscillator or clock in
a consumer product is prohibitive. Running continuously, GPS is a
major power drain. GPS doesn't work well indoors. There are huge
number of products that currently use 60KHz time sync that will go
dark if the US pulls the plug on WWVH and WWVB. That's not going to
happen. Quite the contrary, there are plans to add a US east coast
transmitter. See:
http://en.wikipedia.org/wiki/WWVB
under "Service Improvement Plans".

Incidentally, I'm building my own 10MHz GPSDO for running my test
eqipment and ham junk. It's NOT a trivial or inexpensive exercise:
http://www.jrmiller.demon.co.uk/projects/ministd/frqstd.htm

Too expensive.
Get this while it lasts for $135:
http://www.ebay.com/itm/TRIMBLE-GPS-RECEIVER-10MHZ-CLOCK-THUNDERBOLT-/170845054315?pt=LH_DefaultDomain_0&hash=item27c728a56b

I almost went for one of those, but instead decided on this:
http://www.ebay.com/itm/280655233263?ru=http%3A%2F%2Fwww.ebay.com%2Fsch%2Fi.html%3F_sacat%3D0%26_nkw%3D280655233263%26_rdc%3D1

Not quite as accurate as the thunderbolt, but good enough for my bench.

 

On May 19, 7:13 pm, Jim Yanik <jya...@abuse.gov> wrote:

"Ian Field" <gangprobing.al...@ntlworld.com> wrote innews:9bTtr.549288$yJ4.364502@fx07.am4:











"gregz" <ze...@comcast.net> wrote in message
news:1124369627359136873.854687zekor-comcast.net@news.eternal-september
.org...
"N_Cook" <dive...@tcp.co.uk> wrote:
Bill Proms <n...@anywhere.us> wrote in message
news:jp850p$fk2$1@dont-email.me...
I have 3 Intelli-Time LED alarm clocks around the house, just like
the one
here:

http://www.acurite.com/clock/alarm-clock/intelli-time-digital-alarm-c
lock-13 027a2.html

I initially bought these due to them keeping time when the power
goes off
and auto resetting for DST.  There is a problem, however.  Each of
the clocks becomes inaccurate over time.  If I set them all
manually to the
same
time, within a few months, each one will be off by 3-5 minutes.

So I ask, what is the problem and is there any way to repair it?

Thanks in advance,
Bill

I've always put this down to hash on the mains being interpreted as
extra cycles by the clock monitoring input. The supply companies
contractually have to correct the mains frequency so an exact number
of cycles per day (50/60)x60x60x24, but at any instant can be above
or below the nominal frequency.

I doubt if any use power line for sync. Most have battery backup.
Crystals jump frequency from time to time.

Even one I bought recently uses mains sync, but it has a battery &
crystal divider backup to cover outages.

The mains frequency varies depending on peak demand/off peak, but long
term its average has less drift than the cheap crystal oscillator
they're going to put in a radio alarm clock.

radio alarm clock probably won't even use an xtal;like Jeff L. says,they
may use a cheap RC osc.

I note my MW oven clock that derives it's clock from line freq. has better
stability than other "digital" clocks.(like my PC clock....)

(of course,I use an internet program to keep the PC clock fairly close. I
used to use a program(Atomic Clock) that direct-dialed the Naval
Observatory,but the long distance calls cost too much.)

--
Jim Yanik
jyanik
at
localnet
dot com

Try PC Atomic Sync. You select your time server and the PC can sync
every hour or day or manual. I turn machines off at night and have the
sync program run once at boot AND set the PC 15 seconds fast. When
they start recording TV they never clip the head.

The OP referenced a clock that sets itself. How can it be wrong? WWVB
clocks usually have problems receiving the data during the day - too
much noise but work well at night. This implies you need a decent time
base between successful data reads. My 'goofy clock' uses the power
line as the time base but switches to the uProc clock during power
failures. The uProc clock can be 'calibrated' for reasonable accuracy.
What happens is the crystal frequency is divided down to make a fake
60Hz reference pulse when the power line is absent and the divisor is
'tweaked'. I have the WWVB receivers and the code is mostly written
for goofy.

G˛

 

[Geoffrey S. Mendelson]

isw wrote:

If you're building a clock, the noise and jitter aren't very important;
long-term drift is.

It's also part of the modern perception that precision is much greater than
it is. A clock that reads hours and minutes is accurate to around 30 seconds.
A clock that reads hours minutes and seconds is accurate to 1/2 second.

Neither are accurate to a millisecond.

I it started with airlines who would write 12:30 for "sometime before 1
o'clock" for departure and 2:00 for "around 1:30" for arrival. People
expect exact numbers where they are approximate, and it is easy to
arrive on time if you allow enough "slop" to commenpensate for anything.

Mousillini (pardon the spelling) "made the trains run on time" by adjusting
the schedules to reality.

I've also seen it in ham radio where the frequency really is around 14.200,
but someone logs it as 14.203154 because that's what their (inacurate)
receiver reads it as and in computers where someone thinks floating
point numbers are integers. :-(

For most people a clock that reads in minutes is ok, and for almost everyone
who needs more accuracy seconds is ok. Just about every clock made stays
within a second for a few minutes, and auto correcting via GPS, NTP or WWV
would do well enough.

I would expect that an HF receiver clock where you set the minutes and it
autocorrects to the minute pips on WWV or CHU would do fairly well, and
in most of the US and Canada do it without the interference problems
the VLF radios have.

Note that I don't have access to any of those sources, or the EU equivalents.
The best that I can do is to run real NTP clients on all my computers,
which sync to a main NTP server on my network.

The main NTP server syncs to a variety of sources, which confuses it because
they are all within a millisecond of each other except for two Apple EU ones
which are 5 seconds off. I put them in a long time ago and probably should
remove them.

The irony of all of this is the only thing that needs accurate timing is
catching a bus, which never comes at any exact time anyway and recording
programs off of the DBS system I susbscibe to.

Since we have their PVR, it gets its time and programing information
from the feed, but is set to start recording early and end late.
They don't even trust themselves.

Geoff

--
Geoffrey S. Mendelson, N3OWJ/4X1GM/KBUH7245/KBUW5379
In 1969 the US could put a man on the moon, now teenagers just howl at it. :-(

 

[William Sommerwerck]

"Cydrome Leader" <presence@MUNGEpanix.com> wrote in message
news:jpc2ij$8cq$4@reader1.panix.com...

William Sommerwerck <grizzledgeezer@comcast.net> wrote:

I think it was GE that had an alarm clock like that
in the late 70s. It was actually pretty cool. It had a
little keypad on it.

I mentioned that in a preceding post. It was the 7-4760,
I believe. I still have it. The keypad needed cleaning
every couple of years.

I once had the older vacuum flourescent version. They
changed to LEDs at one some point. I forgot how/why
mine broke.

The vacuum-fluorescent version did 'not have a digital tuner. (I have that
one, too. It works, but it could stand a bit of fixing-up.)

 

[William Sommerwerck]

It's also part of the modern perception that precision
is much greater than it is. A clock that reads hours
and minutes is accurate to around 30 seconds. A
clock that reads hours minutes and seconds is
accurate to 1/2 second.

You're confusing accuracy and resolution. You're also ignoring the fact that
the user can //see// when the minutes change. If you make sure the minutes
change at the same time your reference clock changes, the clock's accuracy
can be less than one second.

I've also seen it in ham radio where the frequency really
is around 14.200, but someone logs it as 14.203154
because that's what their (inacurate) receiver reads it as
and in computers where someone thinks floating
point numbers are integers. :-(

I don't think most digital communications receivers display the frequency
with a precision the LO is capable of.

 

[William Sommerwerck]

I've also seen it in ham radio where the frequency
really is around 14.200, but someone logs it as
14.203154 because that's what their (inaccurate)
receiver reads it as, and in computers, where
someone thinks floating point numbers are integers.

I don't think most digital communications receivers
display the frequency with a precision the LO is
capable of.

WHOOPS! That should have been "incapable of".

 

[William Sommerwerck]

My 1980's Kenwood TS-430 is capable of tuning in 10Hz
steps, and the modern high-priced rigs in 1Hz steps.

I'd like to make a joke about "high-priced [band]spread", but I will
refrain.

The former resolution would require the synthesizer's crystal to be accurate
to about 1 part in 100,000, which is not out of the question.

My Yaesu FTD-1000 is in storage. I don't remember its resolution, or the
tolerance of its crystals.

 

[Geoffrey S. Mendelson]

William Sommerwerck wrote:

WHOOPS! That should have been "incapable of".

Well, my 1980's Kenwood TS-430 is capable of tuning in 10Hz steps, and
the modern high priced rigs in 1Hz steps.

Geof.


--
Geoffrey S. Mendelson, N3OWJ/4X1GM/KBUH7245/KBUW5379
In 1969 the US could put a man on the moon, now teenagers just howl at it. :-(

 

[Jeff Liebermann]

On Mon, 21 May 2012 00:00:49 -0700, isw <isw@witzend.com> wrote:

I don't understand why there would ever be a missed once-per-second
pulse (assuming you know how to design digital stuff), but even so, the
next time a NMEA sentence comes along (or whenever you decide to read
the next one), you'd know you were off, and which way.

The missing pulse problem is with the 1pps clock implementation, not
the NEMA sentence reader. Clock slip with 1pps is very real. I can
do it with my weather station if I transmit on my VHF HT within about
2ft.

Maybe true, but I would have no easy way to determine if it needed to
be resynced with the GPS clock. If it drifted off frequency for some
reason, all my measurements would be off.

But if it ran at the same rate as the GPS clock, then it wouldn't be
drifting ...

Only if the clock didn't miss a pulse.

I think we have a problem here. You're mixing a NEMA sentence driven
clock, with a 1pps conventional counting clock. The NEMA sentence
device does NOT have a clock slip problem and does not need
recalibration. The 1pps counting clock can easily lose count and
offers no easy way to determine that the clock has slipped and that it
requires recalibration.

Setting the frequency to
1*10^-11 takes days to set, and the same time to reset. Might as well
leave it on all the time.

Well, of course it's on all the time; there's no other way to keep the
oscillator stable.

OK. Then forget about battery operation.

But if your local notion of time (what the NTP folks
call "epoch") tracks the data coming from the GPS, then it will
eventually be within that error band. And if it stays on longer, it'll
become ever more accurate (long-term).

Overkill (except for leap seconds). I don't think anyone cares about
sub-second accuracy in a home alarm clock. However, the last digit
(seconds) should be accurate.

Use whatever crystal you want/need to get the temperature/frequency
curve you need.

If I do that, the design becomes more difficult. For example, a
higher freq/temp slope for the crystal will require more insulation.
Crystals with such a slope tend to have large frequency drift during
aging characteristics. Reverse slopes in the operating area will
create some rather bizarre tuning characteristics.

Pick an oven temperature range that gives the slope you
need.

Well, if I use an AT cut crystal at 80C, the slope is about 2ppm/C. At
10MHz, that's 5Hz/C. You could probably adjust the temperature over a
20C range (to avoid the dip in the AT curve at 75C) yielding a 100Hz
range. Yep, that might work, but you'll need to reduce the oscillator
thermal mass, and greatly increase the insulation.

I don't see what's so hard about it.

Pretend you're on a large boat, which has about a 3 minute delay
between when you turn the rudder and when the vessel changes
direction. Newton's 2nd law and hysteresis delay at its best. To
speed up the turn, one tends to over shoot the rudder direction, and
then bring it back to the correct orientation. To anyone lacking
experience in piloting such an over-damped system, the path traveled
will be rather erratic. See control system damping calculations. your
manual oven corrections might be similarly erratic.

It's not even clear to me why
you even need a low-noise oscillator for a clock, assuming it's being
disciplined by a more accurate "master" (the GPS time system).

I don't need a low noise oscillator for an alarm clock. I mentioned
the low noise in reference to by building a GPSDO, which will be used
for everything from running my test equipment clock, to synchronizing
transmitters. I don't see any reason to build multiple versions when
one device will do it all.

The 1/f phase noise of an oscillator is directly related to the power
density of the device. That means using a moderately high current
device at about 10% of its rated current. That's also why nobody uses
low noise RF front end transistor for low noise oscillators. That
leaves flicker noise, which reduced with lots of negative feedback.

The short-term noise and the long-term stability are different. That's
why super-precision sources use rubidium oscillators (low noise, poor
drift), stabilized by cesium devices (higher noise, essentially zero
long-term drift).

True and thanks for recognizing that a rubidium source is not a
primary frequency standard. The issue with noise is that it has an
effect on accuracy in that one cannot measure the frequency to a
resolution less than the noise. For example, if there is 1Hz of FM
noise on the oscillator, the frequency cannot be measured to less than
1Hz. Of course, one could average the measurements, which only works
if the noise spectra is symmetrical. Also, 1Hz of FM noise at the
10MHz clock reference, multiplied up to a 10GHz transmitter, is
1000Hz, which is audible and quite fatal to data.

If you're building a clock, the noise and jitter aren't very important;
long-term drift is.

True. However, I'm building GPSDO, while you're building a GPS alarm
clock. I agree that the requirements are quite different.

Isaac

--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

 

[Jeff Liebermann]

On Mon, 21 May 2012 05:31:07 -0400, JW <none@dev.null> wrote:

Incidentally, I'm building my own 10MHz GPSDO for running my test
eqipment and ham junk. It's NOT a trivial or inexpensive exercise:
http://www.jrmiller.demon.co.uk/projects/ministd/frqstd.htm

Too expensive.
Get this while it lasts for $135:
http://www.ebay.com/itm/TRIMBLE-GPS-RECEIVER-10MHZ-CLOCK-THUNDERBOLT-/170845054315?pt=LH_DefaultDomain_0&hash=item27c728a56b

I almost went for one of those, but instead decided on this:
http://www.ebay.com/itm/280655233263?ru=http%3A%2F%2Fwww.ebay.com%2Fsch%2Fi.html%3F_sacat%3D0%26_nkw%3D280655233263%26_rdc%3D1

Try to find out which lamp they're using. Rubidium depletion is a
major cause of failure. I had an old TFT reference osc crap out on
me. The stupid lamp cost as much as the entire (used) unit. Some of
the cheap lamps have as little as 0.1mg inside, while the one's that
last almost forever have about 1mg.

Not quite as accurate as the thunderbolt, but good enough for my bench.

Yet another temptation to spend money...

I would have no problem with using either of these as a reference.
However, I have a unique problem. I have approximately 200ea Novatel
Allstar 12 boards which are looking for a home. The grand plan is to
build various projects (APRS tracker, GPSDO, GPS logger, test board,
etc) around these boards and possibly sell the kits.

Thanks.

--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

 

[jeff_wisnia]

Jim Yanik wrote:

"Ian Field" <gangprobing.alien@ntlworld.com> wrote in
news:9bTtr.549288$yJ4.364502@fx07.am4:


"gregz" <zekor@comcast.net> wrote in message
news:1124369627359136873.854687zekor-comcast.net@news.eternal-september
.org...

"N_Cook" <diverse@tcp.co.uk> wrote:

Bill Proms <none@anywhere.us> wrote in message
news:jp850p$fk2$1@dont-email.me...

I have 3 Intelli-Time LED alarm clocks around the house, just like
the one
here:



http://www.acurite.com/clock/alarm-clock/intelli-time-digital-alarm-c
lock-13 027a2.html

I initially bought these due to them keeping time when the power
goes off
and auto resetting for DST. There is a problem, however. Each of
the clocks becomes inaccurate over time. If I set them all
manually to the

same

time, within a few months, each one will be off by 3-5 minutes.

So I ask, what is the problem and is there any way to repair it?

Thanks in advance,
Bill



I've always put this down to hash on the mains being interpreted as
extra cycles by the clock monitoring input. The supply companies
contractually have to correct the mains frequency so an exact number
of cycles per day (50/60)x60x60x24, but at any instant can be above
or below the nominal frequency.

I doubt if any use power line for sync. Most have battery backup.
Crystals jump frequency from time to time.


Even one I bought recently uses mains sync, but it has a battery &
crystal divider backup to cover outages.

The mains frequency varies depending on peak demand/off peak, but long
term its average has less drift than the cheap crystal oscillator
they're going to put in a radio alarm clock.





radio alarm clock probably won't even use an xtal;like Jeff L. says,they
may use a cheap RC osc.

I note my MW oven clock that derives it's clock from line freq. has better
stability than other "digital" clocks.(like my PC clock....)

I wish the most recent MW oven I bought for our office worked like that.
Its clock gains about one minute a day and I doubt that there's any
syncing to the line frequency built in. It keeps far worse time than the
$10 battery powered wall clocks we use around the office.

Could be the problem is that the temperature of the electronics in that
MW oven moves around quite a bit more than those in the clock on the
wall next to it due to the high power loads when the oven is heating
stuff. That could screw up a simple RC or even xtal oscillator, eh?

OTOH the MW oven in our home's kitchen is always dead nuts on time,
barring power failures, after which it needs resetting.

Jeff

Jeffry Wisnia
(W1BSV + Brass Rat '57 EE)
The speed of light is 1.8*10e12 furlongs per fortnight.

(of course,I use an internet program to keep the PC clock fairly close. I
used to use a program(Atomic Clock) that direct-dialed the Naval
Observatory,but the long distance calls cost too much.)

--

 

[Spehro Pefhany]

On Sun, 20 May 2012 17:25:54 +0000 (UTC), the renowned Cydrome Leader
<presence@MUNGEpanix.com> wrote:

gregz <zekor@comcast.net> wrote:
"William Sommerwerck" <grizzledgeezer@comcast.net> wrote:
Many years ago I reviewed Heath's "Most-Accurate Clock" for one of Ed Dell's
magazines. It used the Bureau of Standards' shortwave time signals. Sync was
a bit touchy (I eventually replaced the carbon calibration pots with
ceramic), but it otherwise worked very well. It even had an interface that
allowed your computer to reset its clock each time the machine restarted.

When I needed money a few years back, I sold it on eBay for something like
$400, without anyone questioning the price.

If Heath wants to come back as a kit company, it needs to design products
that have no commercial equivalents.

Heath had an fm tuner with direct frequency entry push button. I never saw
another for home use, but there may have been another or commercial use.

I think it was GE that had an alarm clock like that in the late 70s. It
was actually pretty cool. It had a little keypad on it.

I had one of those- nice. IIRC, the keys failed over time though.


Best regards,
Spehro Pefhany
--
"it's the network..." "The Journey is the reward"
speff@interlog.com Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.com

 

[Geoffrey S. Mendelson]

jeff_wisnia wrote:

I wish the most recent MW oven I bought for our office worked like that.
Its clock gains about one minute a day and I doubt that there's any
syncing to the line frequency built in. It keeps far worse time than the
$10 battery powered wall clocks we use around the office.

More likely it has one of those chips that detects if there is power line
signal or battery backup and uses a built in crystal to keep time if the
power line goes off.

Unfortunately, at least one of the cheap Chinese versions is broken and uses
the crystal all of the time. I have a clock like that.

The crystals are spec'ed to run a microprocessor in the chip and not keep
time. :-(

Geoff.

--
Geoffrey S. Mendelson, N3OWJ/4X1GM/KBUH7245/KBUW5379
In 1969 the US could put a man on the moon, now teenagers just howl at it. :-(

 

[Michael A. Terrell]

"Geoffrey S. Mendelson" wrote:

Bill Proms wrote:

As a possible replacement, I have considered an atomic LED clock, but these
appear to be next to impossible to come by for some reason. I see LCD
atomic clocks everywhere, but most or all have to have the backlight pressed
to see the time in dim conditions.

It's because the US is trying to get rid of those broadcasts. With GPS they
are obsolete.

Then why did they replace the WWVB towers & transmitters a couple
years ago? They built a better antenna array, and raised the
transmitter power so that it can be received in Florida on a $20 'Atomic
clock'.

Not only are they not looking to discontinue the service, but they
are looking at a new modulation method to improve noise immunity. This
web page from NIST says that there are about 50,000,000 radio controlled
clocks using WWVB in the United States:

http://www.nist.gov/pml/newsletter/radio.cfm

--
You can't have a sense of humor, if you have no sense.

 

[Michael A. Terrell]

Jeff Liebermann wrote:

On Sun, 20 May 2012 00:41:44 -0700, isw <isw@witzend.com> wrote:

The 1pps output can be used to run a digital clock, but I would hate
to see the final cost.

Why use the 1 pps? Any cheap GPS you get on eBay will output NMEA
"sentences" in ASCII that tell you the precise time. Just use those.

The NEMA sentences are not synchronized to GPS time and add delays in
the decoding process. The time might be off a fraction of a secod.
However, using the NEMA sentence is probably adequate for a consumer
alarm clock. $GPZDA give the time as:
$--ZDA,hhmmss.ss,xx,xx,xxxx,xx,xx
hhmmss.ss = UTC
xx = Day, 01 to 31
xx = Month, 01 to 12
xxxx = Year
xx = Local zone description, 00 to +/- 13 hours
xx = Local zone minutes description (same sign as hours)

The 1pps output has the advantage of simplicity and easy of
integration with existing digital clock designs.

Say, $15 for the GPS, another $15 for an Arduino, $5 for an LCD, and
whatever crystal you have on hand, stuck in a home-made oven. Maybe
another $20 for all the "glue", and the rest, as they say, is just
software.

The cheapest GPS board or module that I could find is about $40.
http://www.sparkfun.com/categories/4?sort_by=price_asc&per_page=50
Chips seem to run about $8/1000. Using your prices, a commercial
digital alarm clock product would retail about $175-$200. Meanwhile,
WWVB controlled "atomic" alarm clocks are selling for $15.

This one decodes NEMA sentences, and has some compromises due to
running on battery power:
http://www.siliconchip.com.au/cms/A_111192/article.html
To conserve the battery, the GPS module is only used
to synchronise the clock every 44 hours and following
synchronisation, the clock will either skip seconds
or double-step to reach the correct time. After
synchronisation the microcontroller is also able to
calculate the inherent inaccuracy of its crystal oscillator
and will compensate by occasionally skipping or
double-stepping a second. This process can also compensate
for aging of the crystal and will keep the clock accurate
between synchronisations.
http://geoffg.net/GPS_Synchronised_Clock.html
http://www.siliconchip.com.au/cms/gallery/article.html?a=111709

$7.79 Trimble GPS module:

http://www.ebay.com/itm/190626494449

$5.99 GPS antenna with MMCX connector:

http://www.ebay.com/itm/170496886891

$32.00 for a GPS receiver & antenna module:

http://www.ebay.com/itm/180815989220


Be VERY careful if you buy any Rockwell GPS receiver boards on Ebay.
At least one seller is advertising a board with 10 KHz out then ship a
different board, even though they have been sent the information by
people who have tried to use them.

--
You can't have a sense of humor, if you have no sense.

 

[Michael A. Terrell]

N_Cook wrote:

Confucius, he say, but even a stopped clock tells the right time twice a
day

Confucius was confused.


--
You can't have a sense of humor, if you have no sense.

 

[isw]

In article <90okr7d0ikv2jt1td8gop835i1ia0jbmdo@4ax.com>,
Jeff Liebermann <jeffl@cruzio.com> wrote:

On Mon, 21 May 2012 00:00:49 -0700, isw <isw@witzend.com> wrote:

I don't understand why there would ever be a missed once-per-second
pulse (assuming you know how to design digital stuff), but even so, the
next time a NMEA sentence comes along (or whenever you decide to read
the next one), you'd know you were off, and which way.

The missing pulse problem is with the 1pps clock implementation, not
the NEMA sentence reader. Clock slip with 1pps is very real. I can
do it with my weather station if I transmit on my VHF HT within about
2ft.

Maybe true, but I would have no easy way to determine if it needed to
be resynced with the GPS clock. If it drifted off frequency for some
reason, all my measurements would be off.

But if it ran at the same rate as the GPS clock, then it wouldn't be
drifting ...

Only if the clock didn't miss a pulse.

I think we have a problem here. You're mixing a NEMA sentence driven
clock, with a 1pps conventional counting clock. The NEMA sentence
device does NOT have a clock slip problem and does not need
recalibration. The 1pps counting clock can easily lose count and
offers no easy way to determine that the clock has slipped and that it
requires recalibration.

I'm proposing to build a time-of-the-year clock that uses a (say) 10 MHz
oscillator for a timebase, with dividers (hardware or software) to
provide all the various outputs. The time-of-year output is available to
the frequency-controlling loop, which compares its value with what it
gets from the NMEA sentences -- if the local clock is ahead, slow down
the 10 MHz, and vice-versa.

After a while, that oscillator is going to be dead on.

Setting the frequency to
1*10^-11 takes days to set, and the same time to reset. Might as well
leave it on all the time.

Well, of course it's on all the time; there's no other way to keep the
oscillator stable.

OK. Then forget about battery operation.

Using temperature to control frequency, probably; using other means, not
much of a problem. A decent GPS chip set, a little CMOS computer, and an
LCD for readout will come in at a small fraction of a watt. If you
picked a better-than-average crystal, stability-wise, you could turn off
everything but the oscillator for most of the time and still get
exceptional long-term accuracy. I expect that some clever design could
produce an "oven" that could be heated by a quarter-watt resistor glued
to the crystal case, so even using the temperature-control-of-frequency
method could still give you something that could operate all day on a
battery.

But if your local notion of time (what the NTP folks
call "epoch") tracks the data coming from the GPS, then it will
eventually be within that error band. And if it stays on longer, it'll
become ever more accurate (long-term).

Overkill (except for leap seconds). I don't think anyone cares about
sub-second accuracy in a home alarm clock. However, the last digit
(seconds) should be accurate.

Right. But what they might care about is that the clock never drifts
away from "real" time, no matter what the power line does. That, of
course, is the appeal of all "radio clocks".

Use whatever crystal you want/need to get the temperature/frequency
curve you need.

If I do that, the design becomes more difficult. For example, a
higher freq/temp slope for the crystal will require more insulation.

Not really. Remember, it's in a temperature-controlled oven. The
temperature control loop will keep the crystal at whatever temperature
is called for to keep the frequency at precisely 10.000... MHz.

Crystals with such a slope tend to have large frequency drift during
aging characteristics.

Which will be compensated for continuously by the loop.

Reverse slopes in the operating area will
create some rather bizarre tuning characteristics.

Well, don't operate it in that regime ...

Pick an oven temperature range that gives the slope you
need.

Well, if I use an AT cut crystal at 80C, the slope is about 2ppm/C. At
10MHz, that's 5Hz/C. You could probably adjust the temperature over a
20C range (to avoid the dip in the AT curve at 75C) yielding a 100Hz
range. Yep, that might work, but you'll need to reduce the oscillator
thermal mass, and greatly increase the insulation.

Nope. Very slow response times are no problem. The loop will eventually
arrange things so that precisely ten million cycles will elapse per one
second elapsed, over the long term. Still, small thermal mass is easy --
just pick a small crystal package, and heat that package directly with a
small resistor or two. Insulation is cheap and a lot of it is not a
problem. I'd use fairly long, very thin, stainless steel wires to
connect the crystal (and those resistors) to the rest of the circuit,
too. One major source of problems on high-stability oscillators is
thermal or acoustical shock running right up the connecting wires
directly to the quartz itself. I've seen ovenized oscillators that were
disrupted every time the thermostat clicked, and not because of
electrical transients.

I don't see what's so hard about it.

Pretend you're on a large boat, which has about a 3 minute delay
between when you turn the rudder and when the vessel changes
direction. Newton's 2nd law and hysteresis delay at its best. To
speed up the turn, one tends to over shoot the rudder direction, and
then bring it back to the correct orientation. To anyone lacking
experience in piloting such an over-damped system, the path traveled
will be rather erratic. See control system damping calculations. your
manual oven corrections might be similarly erratic.

I've done a few control loops, some of them a bit "peculiar". If you
insist on having the fastest possible transition to the new heading to
within some specified error band, then you're correct. It you're willing
to do things a lot more slowly while still (eventually) arriving at the
new heading, then the loop gets very simple. For the above problem, just
put a really slow motor on the power steering for the rudder.

It's not even clear to me why
you even need a low-noise oscillator for a clock, assuming it's being
disciplined by a more accurate "master" (the GPS time system).

I don't need a low noise oscillator for an alarm clock. I mentioned
the low noise in reference to by building a GPSDO, which will be used
for everything from running my test equipment clock, to synchronizing
transmitters. I don't see any reason to build multiple versions when
one device will do it all.

Consider the 10 MHz output from this sort of clock as the frequency
reference for whatever sort of synthesizer you want. If the clock runs
at the proper rate, the oscillator is at the specified frequency. After
it's run for a while, the frequency will always be very, very close to
exactly ten megahertz.

The 1/f phase noise of an oscillator is directly related to the power
density of the device. That means using a moderately high current
device at about 10% of its rated current. That's also why nobody uses
low noise RF front end transistor for low noise oscillators. That
leaves flicker noise, which reduced with lots of negative feedback.

The short-term noise and the long-term stability are different. That's
why super-precision sources use rubidium oscillators (low noise, poor
drift), stabilized by cesium devices (higher noise, essentially zero
long-term drift).

True and thanks for recognizing that a rubidium source is not a
primary frequency standard. The issue with noise is that it has an
effect on accuracy in that one cannot measure the frequency to a
resolution less than the noise. For example, if there is 1Hz of FM
noise on the oscillator, the frequency cannot be measured to less than
1Hz. Of course, one could average the measurements, which only works
if the noise spectra is symmetrical. Also, 1Hz of FM noise at the
10MHz clock reference, multiplied up to a 10GHz transmitter, is
1000Hz, which is audible and quite fatal to data.

One of the advantages of this sort of loop (thermally controlled) is
that there is literally nothing in the circuit that prevents the noise
from being as low as it can possibly be for whatever sort of oscillator
you choose, while still constraining it to being very close to the
design frequency.

Isaac

 

[Geoffrey S. Mendelson]

Michael A. Terrell wrote:

"Geoffrey S. Mendelson" wrote:
It's because the US is trying to get rid of those broadcasts. With GPS they
are obsolete.


Then why did they replace the WWVB towers & transmitters a couple
years ago? They built a better antenna array, and raised the
transmitter power so that it can be received in Florida on a $20 'Atomic
clock'.

They are playing "cheap catch up". The original plans were to build an
east coast station, but they were unable to get any government installation
to "host" it, (NIMBY) and lost the funding.

Not only are they not looking to discontinue the service, but they
are looking at a new modulation method to improve noise immunity. This
web page from NIST says that there are about 50,000,000 radio controlled
clocks using WWVB in the United States:

The are using the improved modulation to keep relevant. In most large cities,
the noise from computer and home electronic equipment, BPL (still very much
in use but not for internet to customers), aDSL, etc has made it next
to impossible to receive a signal.

They exist today because people are willing to accept the poor service they
get as it is the only game in town at that price tag. Most users never
pay attention to how often they get sync, if ever.

If they have to pay $100-$150 for a BPSK decoding clock, GPS or Wifi NTP clocks
will seem a lot better deal.

I'd love to know how they came up with the number of clocks in use. Anyone
have an idea?

Geoff.



--
Geoffrey S. Mendelson, N3OWJ/4X1GM/KBUH7245/KBUW5379
In 1969 the US could put a man on the moon, now teenagers just howl at it. :-(

 

[Jeff Liebermann]

On Tue, 22 May 2012 11:49:04 +0000 (UTC), "Geoffrey S. Mendelson"
<gsm@mendelson.com> wrote:

The are using the improved modulation to keep relevant. In most large cities,
the noise from computer and home electronic equipment, BPL (still very much
in use but not for internet to customers), aDSL, etc has made it next
to impossible to receive a signal.

Many switching power supplies run at about 60KHz. I have one
somewhere around my computer/TV pile, which kills WWVB reception if I
get anywhere near it.

They exist today because people are willing to accept the poor service they
get as it is the only game in town at that price tag. Most users never
pay attention to how often they get sync, if ever.

Most WWVB devices have an indicator on the LCD display to show that
the clock was recently synced with WWVB time. My weather stations and
assorted digital clocks all have this feature.

If they have to pay $100-$150 for a BPSK decoding clock, GPS or Wifi NTP clocks
will seem a lot better deal.

It's my understanding that only the modulation scheme will change, not
the encoded data. A universal chip that works using both system
should be possible without a major price jump.

I'm a bit mystified with the "new type of PM receiving antenna"
mentioned in:
<http://www.nist.gov/pml/newsletter/radio.cfm>
I didn't know that antennas were modulation specific.

I'd love to know how they came up with the number of clocks in use. Anyone
have an idea?

Same as how they get wi-fi device numbers. Marketing research firms,
that specialize in selling industry statistics and predictions, survey
the chip manufacturers for how many chips they've sold. In this case,
the leading manufactory is C-Max:
<http://www.c-max-time.com>
I suspect the largest numbers are in "atomic time" wristwatches.

The accuracy of sales statistics are always questionable, but are
usually accurate within an order of magnitude. The problem is while
sales statistics are fairly simple to generate, devices in use are
not. One could add up all the sales from the last 10 years, assume
that few of the devices were trashed, and produce a very large number
in use. Whether it has any value is very doubtful. Besides, sales
statistics is what companies are willing to pay for, and that's what
the marketing research firms tend to produce.


--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

 

[Geoffrey S. Mendelson]

Jeff Liebermann wrote:

Many switching power supplies run at about 60KHz. I have one
somewhere around my computer/TV pile, which kills WWVB reception if I
get anywhere near it.

Exactly, how many people have one?

Most WWVB devices have an indicator on the LCD display to show that
the clock was recently synced with WWVB time. My weather stations and
assorted digital clocks all have this feature.

What's recent? 1 minute? 1 Hour? 1 Day? a Week?

It's my understanding that only the modulation scheme will change, not
the encoded data. A universal chip that works using both system
should be possible without a major price jump.

Well no. The data is the same, but a new receiver needs to be used.
The old one just did on/off for an AM pulse, the new one uses BPSK,
which is two tone modulation. So not only does it have to decode
the carrier being there at all, it has to decode two different tones.

Then you have to decode the BPSK stream to get the data. This not a big
deal, you could do it with a sound card and a microprocessor, but it's
a different receiver design, and reprograming the microprocessor.

The kind of thing that if you really were going to sell 50 million of them
you could do it for a few dollars a chipset/board, which is probably what the
current ones cost, but if you want to break even with 10,000 you have to
sell them for at least $100, maybe more.

It's like I saw an article about an Israeli startup that had sold 200,000 of
their product. The article was entitled "sales of xxx disappointing".
I guess they planned on selling a million of them. :-(

I'm a bit mystified with the "new type of PM receiving antenna"
mentioned in:
http://www.nist.gov/pml/newsletter/radio.cfm
I didn't know that antennas were modulation specific.

Where have you been the last five years? I surprised that you have not
been swamped with HDTV antennas.

I expect it's another gimick to say you need to buy a higher gain antenna,
or that's why your device can't sync. I expect that everyone will need
to buy 1/2 wavelength end fed wires.

(for the humor impared, that's a joke, a wavelength is 5 kilometers).


Geoff.
--
Geoffrey S. Mendelson, N3OWJ/4X1GM/KBUH7245/KBUW5379
In 1969 the US could put a man on the moon, now teenagers just howl at it. :-(