WWVB Receiver...

[Rick C]

On Sunday, December 19, 2021 at 2:46:54 PM UTC-4, John Walliker wrote:

On Sunday, 19 December 2021 at 17:55:25 UTC, gnuarm.del...@gmail.com wrote:
There are diodes specifically designed for capacitance tuning. Although they are largely obsolete and hard to find, they still available. I don\'t recall the capacitance required to tune a ferrite loop to 60 kHz, but some hundreds of pF should suffice for fine tuning. No? I think the trick would be dealing with the temperature characteristic of the fixed capacitor already installed, but I think you said you would control the temperature. Even so, you might want to replace that with your own capacitor with a more stable temperature coefficient.

So is this feasible or am I missing something?


- Get 1,000 miles of free Supercharging
- Tesla referral code - https://ts.la/richard11209
Why does the ferrite rod need to be highly tuned if there is plenty of signal?
An alternative approach might be to use a broadly tuned ferrite antenna
with just enough selectivity to avoid overloading subsequent stages
with other signals and then digitise the 60kHz WWVB signal with a
wide-band 24-bit audio ADC, perhaps sampling at around 192ksa/s.
This would be fine for acquiring a 60kHz signal. You could then process
the ADC output with a linear-phase digital filter to get the final selectivity
you need.

The ferrite rod needs to be highly tuned because unless you are very close to the transmitter, the signal gets swamped by extraneous interferers. The tuning is actually how a loop antenna works. The effective height of a typical small loop antenna is small, but is multiplied by the Q to determine the output voltage. It is the high Q that makes a sensitive loop.

Even with the high Q of a tuned loop antenna additional filtering is important as interferers are often strong.

--

Rick C.

-- Get 1,000 miles of free Supercharging
-- Tesla referral code - https://ts.la/richard11209

 

[Rick C]

On Sunday, December 19, 2021 at 3:38:49 PM UTC-4, jla...@highlandsniptechnology.com wrote:

On Sun, 19 Dec 2021 07:39:10 GMT, Jan Panteltje
pNaOnSt...@yahoo.com> wrote:

On a sunny day (Sun, 19 Dec 2021 05:14:04 -0000 (UTC)) it happened Jan Frank
spa...@not.com> wrote in <XnsAE052628EB...@144.76.35.252>:

jla...@highlandsniptechnology.com wrote:

On Sat, 18 Dec 2021 18:39:31 -0000 (UTC), Jan Frank <nos...@not.com
wrote:

I am making a WWVB receiver to compare the diurnal phase shift with GPS.

Ferrite loop antennas are available on Amazon, but they are fixed
frequency and I need to find a way to tune them.

I wonder if a magnetic field will affect the permeability so I can shift
the frequency with a small magnet? It would be followed by a 60KHz xtal
filter to narrow the bandwidth, so a reduction in Q is acceptable.

The assembly would have to be in a temperature-controlled environment,
but that\'s fairly easy.

Any thoughts?

How about a giant loop antenna? Use multiconductor shielded cable.
Lots of pickup area.

Normally one would ground one end of the shield, but why not resonate
the shield?

Thanks for the suggestion, but I don\'t need a large pickup area. I have a
number of WWVB clocks, including two Casio Waveceptor wrist watches. They
have a tiny ferrite antenna to fit in the small case, but still have enough
signal to synchronize perfectly every night:

https://www.amazon.ca/s?k=casio+waveceptor+watches+men

The Amazon ferrite loop is much larger and should provide plenty of signal:

https://www.amazon.ca/gp/product/B01KH3VEGS/

In addition there is no way to trim the resonance of a large loop except by
changing the capacitor. This would require a huge capacitor and a
microstepping motor to turn the shaft. This would make a large and
cumbersome assembly.

I need to compare the phase of the WWVB signal with the GPS to analyze the
diurnal time shift affecting both signals. A GPS signal does not provide
phase information, so I need to use the 1 PPS signal and compare the time
with WWVB. A Ublox LEA-5T-0-003 50-channel 15ns 1pps Timing GPS module will
provide the 1 PPS:

https://www.ebay.com/itm/333297389966

The all I need is a way to offset the 10 MHz GPSDO clock to cancel the
dirurnal time shift of the GPS signal. Hopefully this will give better than
the typical 1e-12 error of the GPS.

Time-Nuts has numerous posts discussing the phase errors caused by resonant
circuits in the signal path, so I need a method of tuning the ferrite loop
and keeping it at resonance.

https://febo.com/pipermail/time-nuts_lists.febo.com/

Changing the capacitance is not feasible,

We have the same watch it seems.
Anyways large capacitance varicaps exist
(I have used those for very low frequencis)
or you could use some normal diode at low reverse bias for that.
Varicaps have ghastly tempcos that change with bias.

I think Jan is looking for high time resolution, nanoseconds maybe, so
varicaps and ferrites are out of the game.

If he\'s looking for a diurnal time shift, that could be overwhelmmed
by the diurnal temperature cycle whacking all the gear.

The ferrite or capacitance adjustment is to mitigate a source of error in the measurement. It may be within the project\'s error budget to have some error in the error compensation, a secondary impact on the measurement itself. We can only know what is an acceptable solution if we had detailed requirements. So we brainstorm and propose ideas that may well work depending on the details such as the quality of the temperature stabilization.

With temperature control of the electronics, this could provide a very stable reference.

--

Rick C.

-+ Get 1,000 miles of free Supercharging
-+ Tesla referral code - https://ts.la/richard11209

 

[Jan Frank]

piglet <erichpwagner@hotmail.com> wrote:

On 18/12/2021 6:39 pm, Jan Frank wrote:
I am making a WWVB receiver to compare the diurnal phase shift with
GPS.

Ferrite loop antennas are available on Amazon, but they are fixed
frequency and I need to find a way to tune them.

I wonder if a magnetic field will affect the permeability so I can
shift the frequency with a small magnet? It would be followed by a
60KHz xtal filter to narrow the bandwidth, so a reduction in Q is
acceptable.

The assembly would have to be in a temperature-controlled environment,
but that\'s fairly easy.

Any thoughts?

Yes driving a variable direct current thru second coil around the
ferrite rod will decrease the permeability. If you set the unsaturated
system to resonant slightly low then you could adjust the dc bias to hit
resonance. BTW how to you plan to know when the loop is resonant?

I wonder since you are comparing against the 1PS GPS timekeeping can you
simply go by the 1PS WWVB time code instead of measuring WWVB carrier
frequency?

piglet

Thanks for the info on permeability.

I need to compare the WWVB carrier phase against the GPS time signal. The
time code must be decoded which takes too long.

 

[Jan Frank]

jlarkin@highlandsniptechnology.com wrote:

On Sat, 18 Dec 2021 18:39:31 -0000 (UTC), Jan Frank <nospam@not.com
wrote:

[...]

What phase shift do you want to measure, the GPS or the WWVB?

You can get a cheap surplus rubidium standard that will drift
nanoseconds per day. That could be the reference that you measure the
other against.

It\'s fun to trigger a scope from one rubidium and zoom the rising edge
from another at ns per division. It looks internally triggered! Check
it occasionally and see it creep across the screen.

I want to compare the WWVB carrier phase against the GPS 1 pps time signal.

I have a number of Rb\'s to use in a N-corner hat to analyze the results. The
goal is to see if offsetting the GPS 1 pps by changes in the WWVB carrier
would help reduce the diurnal shift in GPS time. This could improve the
typical GPS accuracy from 1e-12 to possibly 0.1e-12.

 

[Jan Frank]

Martin Brown <\'\'\'newspam\'\'\'@nonad.co.uk> wrote:

On 19/12/2021 13:40, Jan Frank wrote:
Martin Brown <\'\'\'newspam\'\'\'@nonad.co.uk> wrote:

On 18/12/2021 18:39, Jan Frank wrote:
I am making a WWVB receiver to compare the diurnal phase shift with
GPS.

Ferrite loop antennas are available on Amazon, but they are fixed
frequency and I need to find a way to tune them.

You can get pre tuned ferrite loop antennas. You won\'t need to tune
them they are more or less spot on 60kHz +/- small fractions of a ppm.

I wonder if a magnetic field will affect the permeability so I can
shift the frequency with a small magnet? It would be followed by a
60KHz xtal filter to narrow the bandwidth, so a reduction in Q is
acceptable.

The assembly would have to be in a temperature-controlled
environment, but that\'s fairly easy.

You should look at the low frequency LBI studies by Duffet-Smith et al
in the 1980\'s. They were disciplining a local Rb clock with the MSF
Rugby signal so their local clock was way better frequency stability
than the instantaneous signal from Rugby.

ISTR What they found was that the time delays depended critically on
the air humidity and amount of dew on the ground at the transmitter
site. This is hardly surprising as it affects the time constant of
their transmitting antenna. MSF Rugby were interested in the result.

GPS is essentially a better than Rb clock so you should be able to
reproduce their results (They were doing it to timestamp
interferometer data in a not quite VLBI method at low frequencies).

Looking for the white light fringe in LBI is still a nightmare even
when you have precision timestamps - particularly with the hardware of
that era.

This isn\'t the paper I remember but search terms into ADS elude me for
the moment. It is related but at 16kHz (submarine comms frequency).

https://ui.adsabs.harvard.edu/abs/1985RaF....28..574S/abstract

Thanks for the info. I am trying to cancel the diurnal time shift of
GPS signals by using phase information from WWVB signals to improve the
typical 1e-12 accuracy of the GPS. Even 100e-15 would be a significant
improvement.

I have a number of Rb clocks for use in a N-corner hat to compare the
GPS and WWVB signals. Ideally the reference should be a hydrogen maser
but that is far beyond most people\'s budgets. However, moving GPS
closer would be a valuable improvement at far lower cost.

OK. If you have an Rb clock to hand you definitely want the paper that
describes the MRAO portable Rb clock synchronisation to MSF & Loran.

Interesting comment about humidity. However, there are other signals in
the vlf region such as 100KHz that could also be used.

If memory serves there was a systematic ~2us lag in the early morning
signal compared to light travel time to the remote site which went away
as the sun dried the ground.

Not mentioned in the only paper about it I can find at 81.5MHz but the
references may well include one with more details of the hardware.

https://articles.adsabs.harvard.edu/pdf/1983MNRAS.205..625H

VLBI these days invariably use H masers as local clocks.

Thanks for the paper.

 

[Jan Frank]

jlarkin@highlandsniptechnology.com wrote:

On Sun, 19 Dec 2021 05:14:04 -0000 (UTC), Jan Frank <spamme@not.com
wrote:

[...]

Changing the capacitance is not feasible, so I am trying to find a way
to change the inductance. Hopefully, changing the permeability will do
the job.

Thanks for the help.

At 60 KHz, 1 ns resolution is 0.02 degrees. I don\'t think a ferrite
rod antenna will be stable to numbers like that. A loop or solenoid
air-core antenna might.

Maybe a solenoid in an oven! With tuning/amplifiers as needed. May as
well put your GPS in the oven too. (I\'m thinking ovens because we\'re
controlling some optical stuff to a few mK just now.)

https://www.dropbox.com/s/f6h8tfyq0xkqx1q/Oven_Cables_pub.jpg?raw=1

1 PPS doesn\'t provide much information. You could use a good GPS 10
MHz clock to sample the 60K waveform and average the heck out of that.
A cheap digital oscilloscope could digitize the 60K and 10M waveforms
for you, and you could math that. Channel-to-channel time drift is low
on most scopes, far better than trig-to-channel.

The 10 MHz GPS clock is derived from the 1 PPS. I will use the phase of the
WWVB carrier to compare against the GPS signal. I had already planned on
putting everything in an oven for stabilization.

 

[Rick C]

On Sunday, December 19, 2021 at 6:03:07 PM UTC-4, Jan Frank wrote:

piglet <erichp...@hotmail.com> wrote:

On 18/12/2021 6:39 pm, Jan Frank wrote:
I am making a WWVB receiver to compare the diurnal phase shift with
GPS.

Ferrite loop antennas are available on Amazon, but they are fixed
frequency and I need to find a way to tune them.

I wonder if a magnetic field will affect the permeability so I can
shift the frequency with a small magnet? It would be followed by a
60KHz xtal filter to narrow the bandwidth, so a reduction in Q is
acceptable.

The assembly would have to be in a temperature-controlled environment,
but that\'s fairly easy.

Any thoughts?

Yes driving a variable direct current thru second coil around the
ferrite rod will decrease the permeability. If you set the unsaturated
system to resonant slightly low then you could adjust the dc bias to hit
resonance. BTW how to you plan to know when the loop is resonant?

I wonder since you are comparing against the 1PS GPS timekeeping can you
simply go by the 1PS WWVB time code instead of measuring WWVB carrier
frequency?

piglet
Thanks for the info on permeability.

I need to compare the WWVB carrier phase against the GPS time signal. The
time code must be decoded which takes too long.

That\'s not how the time code is usually handled. The time code is decoded by looking at the bit timings and pulling out the start of each bit and decoding the value. This value is always late, reporting the time of the minute index mark the message started with. In your case you don\'t need to consider the data at all. You simply look for the starting edge of each bit on one second intervals. There could be an acquisition mode where the algorithm searches for valid bits without restriction on the timing details. After finding some number of successive bits it can then look for the start of bits only within the window near the 1 second boundaries to help minimize noise impacts. The rising edge time of the pulse will depend on a number of effects on the bandwidth of the entire path. So you can be the judge of the stability of these effects.

To sync to the carrier may have issues with receiving a clean carrier. I\'ve never bothered looking at that detail, so I can\'t advise on issues there. I suppose with the averaging available this can be mitigated. A simple high-Q resonant circuit may be adequate.

--

Rick C.

+- Get 1,000 miles of free Supercharging
+- Tesla referral code - https://ts.la/richard11209

 

[Jan Frank]

jlarkin@highlandsniptechnology.com wrote:

On Sun, 19 Dec 2021 07:39:10 GMT, Jan Panteltje
pNaOnStPeAlMtje@yahoo.com> wrote:

On a sunny day (Sun, 19 Dec 2021 05:14:04 -0000 (UTC)) it happened Jan
Frank <spamme@not.com> wrote in
XnsAE052628EBB7idtokenpost@144.76.35.252>:

jlarkin@highlandsniptechnology.com wrote:

On Sat, 18 Dec 2021 18:39:31 -0000 (UTC), Jan Frank <nospam@not.com
wrote:

I am making a WWVB receiver to compare the diurnal phase shift with
GPS.

[...]

The assembly would have to be in a temperature-controlled
environment, but that\'s fairly easy.

[...]

I need to compare the phase of the WWVB signal with the GPS to analyze
the diurnal time shift affecting both signals. A GPS signal does not
provide phase information, so I need to use the 1 PPS signal and
compare the time with WWVB. A Ublox LEA-5T-0-003 50-channel 15ns 1pps
Timing GPS module will provide the 1 PPS:

https://www.ebay.com/itm/333297389966

[...]

Changing the capacitance is not feasible,

We have the same watch it seems.
Anyways large capacitance varicaps exist
(I have used those for very low frequencis)
or you could use some normal diode at low reverse bias for that.

Varicaps have ghastly tempcos that change with bias.

I think Jan is looking for high time resolution, nanoseconds maybe, so
varicaps and ferrites are out of the game.

If he\'s looking for a diurnal time shift, that could be overwhelmmed
by the diurnal temperature cycle whacking all the gear.

The varicap tempco is not important. It is in a pll inside an oven.

 

[Jan Frank]

Jan Frank <nospam@not.com> wrote:

I am making a WWVB receiver to compare the diurnal phase shift with GPS.

Ferrite loop antennas are available on Amazon, but they are fixed
frequency and I need to find a way to tune them.

I wonder if a magnetic field will affect the permeability so I can shift
the frequency with a small magnet? It would be followed by a 60KHz xtal
filter to narrow the bandwidth, so a reduction in Q is acceptable.

The assembly would have to be in a temperature-controlled environment,
but that\'s fairly easy.

Any thoughts?

Breakthrough. Paul, W1VLF describes how to make a ferrite rod antenna using
around 500pf to resonate at 60KHz:

https://www.youtube.com/watch?v=ySWpaAXiJCQ WWVB Antenna Part 1
https://www.youtube.com/watch?v=2zrV60J52dc WWVB Antenna Part 2
https://www.youtube.com/watch?v=3zNG7L7fQf4 WWVB antenna build part 3

A MVAM109 will reach 500pf:

http://njsemi.com/datasheets/MVAM108%20-%20MVAM125.pdf MVAM108 -
MVAM125.pdf

This would also allow a 3:1 tuning range covering 60KHz to 180KHz. This
includes signals at 100KHz that might also be valuable.

Amidon sells MnZn type 33 material rods for $60 each. Two are required:

http://www.amidoncorp.com/33-material-ferrite-rods/

Amazon sells type 33 ferrite rods that combine to make a total of 27.5
inches:

5pcs 10mm*140mm Ferrite Rod Bar Loopstick for Radio Antenna Aerial Crystal
- Materials Metal Rod - 1 x 53pcs Red Heat Shrink Tubing, 1 x Storage Box
Price: $24.75

https://www.amazon.ca/gp/product/B07FYDN5P6/

I have the Ublox LEA-5T-0-003 50-channel 15ns 1pps Timing GPS on order from
eBay. Should arrive sometime in January. I will update when I have more
information.

 

[Rick C]

On Sunday, December 19, 2021 at 6:52:40 PM UTC-4, Jan Frank wrote:

Jan Frank <nos...@not.com> wrote:

I am making a WWVB receiver to compare the diurnal phase shift with GPS..

Ferrite loop antennas are available on Amazon, but they are fixed
frequency and I need to find a way to tune them.

I wonder if a magnetic field will affect the permeability so I can shift
the frequency with a small magnet? It would be followed by a 60KHz xtal
filter to narrow the bandwidth, so a reduction in Q is acceptable.

The assembly would have to be in a temperature-controlled environment,
but that\'s fairly easy.

Any thoughts?

Breakthrough. Paul, W1VLF describes how to make a ferrite rod antenna using
around 500pf to resonate at 60KHz:

https://www.youtube.com/watch?v=ySWpaAXiJCQ WWVB Antenna Part 1
https://www.youtube.com/watch?v=2zrV60J52dc WWVB Antenna Part 2
https://www.youtube.com/watch?v=3zNG7L7fQf4 WWVB antenna build part 3

A MVAM109 will reach 500pf:

http://njsemi.com/datasheets/MVAM108%20-%20MVAM125.pdf MVAM108 -
MVAM125.pdf

This would also allow a 3:1 tuning range covering 60KHz to 180KHz. This
includes signals at 100KHz that might also be valuable.

Amidon sells MnZn type 33 material rods for $60 each. Two are required:

http://www.amidoncorp.com/33-material-ferrite-rods/

Amazon sells type 33 ferrite rods that combine to make a total of 27.5
inches:

5pcs 10mm*140mm Ferrite Rod Bar Loopstick for Radio Antenna Aerial Crystal
- Materials Metal Rod - 1 x 53pcs Red Heat Shrink Tubing, 1 x Storage Box
Price: $24.75

https://www.amazon.ca/gp/product/B07FYDN5P6/

I have the Ublox LEA-5T-0-003 50-channel 15ns 1pps Timing GPS on order from
eBay. Should arrive sometime in January. I will update when I have more
information.

You are trying to fine tune the resonance at 60 kHz only, no? So wouldn\'t it be better to use a low tempco fixed cap at say, 470 pF and a tuning cap to trim the final ~30 pF? I believe the tuning rate drops off quickly as the value drops, so a 500 pF unit would require a higher voltage to get to 30 pF and at that point the sensitivity is low making it less sensitive to noise in the control voltage and the signal on the circuit.

--

Rick C.

++ Get 1,000 miles of free Supercharging
++ Tesla referral code - https://ts.la/richard11209

 

[server]

On Sun, 19 Dec 2021 22:39:03 -0000 (UTC), Jan Frank <spamme@not.com>
wrote:

jlarkin@highlandsniptechnology.com wrote:

On Sun, 19 Dec 2021 07:39:10 GMT, Jan Panteltje
pNaOnStPeAlMtje@yahoo.com> wrote:

On a sunny day (Sun, 19 Dec 2021 05:14:04 -0000 (UTC)) it happened Jan
Frank <spamme@not.com> wrote in
XnsAE052628EBB7idtokenpost@144.76.35.252>:

jlarkin@highlandsniptechnology.com wrote:

On Sat, 18 Dec 2021 18:39:31 -0000 (UTC), Jan Frank <nospam@not.com
wrote:

I am making a WWVB receiver to compare the diurnal phase shift with
GPS.

[...]

The assembly would have to be in a temperature-controlled
environment, but that\'s fairly easy.


[...]

I need to compare the phase of the WWVB signal with the GPS to analyze
the diurnal time shift affecting both signals. A GPS signal does not
provide phase information, so I need to use the 1 PPS signal and
compare the time with WWVB. A Ublox LEA-5T-0-003 50-channel 15ns 1pps
Timing GPS module will provide the 1 PPS:

https://www.ebay.com/itm/333297389966

[...]

Changing the capacitance is not feasible,

We have the same watch it seems.
Anyways large capacitance varicaps exist
(I have used those for very low frequencis)
or you could use some normal diode at low reverse bias for that.

Varicaps have ghastly tempcos that change with bias.

I think Jan is looking for high time resolution, nanoseconds maybe, so
varicaps and ferrites are out of the game.

If he\'s looking for a diurnal time shift, that could be overwhelmmed
by the diurnal temperature cycle whacking all the gear.

The varicap tempco is not important. It is in a pll inside an oven.

OK. Someone was talking about a varicap to tune the antenna.



--

I yam what I yam - Popeye

 

[Jan Panteltje]

On a sunny day (Sun, 19 Dec 2021 13:00:38 -0800 (PST)) it happened Rick C
<gnuarm.deletethisbit@gmail.com> wrote in
<933eed2a-2922-4efe-a8c7-1ca9c16d1db2n@googlegroups.com>:

The ferrite rod needs to be highly tuned because unless you are very close to
the transmitter, the signal gets swamped by extraneous interferers. The
tuning is actually how a loop antenna works. The effective height of a typical
small loop antenna is small, but is multiplied by the Q to determine the
output voltage. It is the high Q that makes a sensitive loop.

Even with the high Q of a tuned loop antenna additional filtering is important
as interferers are often strong.

All true, even so it is amazing that my Casio Wavecepter watch gets the time / date here in minutes
while it is indoors and surrounded by .. 12 wallwarts, most of those switchers, many computers and a big UPS, what not.
But indeed if you want only the carrier you need high Q precise tuning.
But then the slightest tuning change will cause a phase shift I\'d expect.
Phase shift at such a low frequency is a lot of time...
Temperature stabilization is perhaps a must then.

 

[Martin Brown]

On 19/12/2021 22:35, Rick C wrote:

On Sunday, December 19, 2021 at 6:03:07 PM UTC-4, Jan Frank wrote:
piglet <erichp...@hotmail.com> wrote:

On 18/12/2021 6:39 pm, Jan Frank wrote:
I am making a WWVB receiver to compare the diurnal phase shift
with GPS.

Ferrite loop antennas are available on Amazon, but they are
fixed frequency and I need to find a way to tune them.

I wonder if a magnetic field will affect the permeability so I
can shift the frequency with a small magnet? It would be
followed by a 60KHz xtal filter to narrow the bandwidth, so a
reduction in Q is acceptable.

The assembly would have to be in a temperature-controlled
environment, but that\'s fairly easy.

Any thoughts?

Yes driving a variable direct current thru second coil around
the ferrite rod will decrease the permeability. If you set the
unsaturated system to resonant slightly low then you could adjust
the dc bias to hit resonance. BTW how to you plan to know when
the loop is resonant?

I wonder since you are comparing against the 1PS GPS timekeeping
can you simply go by the 1PS WWVB time code instead of measuring
WWVB carrier frequency?

piglet
Thanks for the info on permeability.

I need to compare the WWVB carrier phase against the GPS time
signal. The time code must be decoded which takes too long.

That\'s not how the time code is usually handled. The time code is
decoded by looking at the bit timings and pulling out the start of
each bit and decoding the value. This value is always late,
reporting the time of the minute index mark the message started with.
In your case you don\'t need to consider the data at all. You simply
look for the starting edge of each bit on one second intervals.
There could be an acquisition mode where the algorithm searches for
valid bits without restriction on the timing details. After finding
some number of successive bits it can then look for the start of bits
only within the window near the 1 second boundaries to help minimize
noise impacts. The rising edge time of the pulse will depend on a
number of effects on the bandwidth of the entire path. So you can be
the judge of the stability of these effects.

The trick is to generate a local oscillator derived from a Rb clock and
correlate that with the incoming MSF signal over the long term. Provided
you do it right and can prevent the LO blinding the MSF receiver you can
get a stable local clock that is exactly derived from the mean reference
signal at the remote site plus light travel time (which varies with air
pressure and humidity enough to be a problem if you are looking for
ultimate precision). ISTR it took them a lot of iterations before they
were happy that device worked exactly as intended.

As I said they had to defend it against being tricked into drifting when
there was dew on the ground plane at the transmitter site - but that
only became obvious after they had been using it for a while.

To sync to the carrier may have issues with receiving a clean
carrier. I\'ve never bothered looking at that detail, so I can\'t
advise on issues there. I suppose with the averaging available this
can be mitigated. A simple high-Q resonant circuit may be adequate.

You can be pretty much assured that the carrier will be dirty. Only the
long term average of its correlation with a precision tuned local
oscillator is any good. The latter being used to discipline the Rb
clock. If he can find the hardware paper or one for VLBI synch protocols
that ought to provide enough clues on how to do it optimally today.

It relies on the weak law of large numbers to work. More signal is
always better but you have to accept that the VLF bands have more than
their fair share of noise in them from man made and natural events.

--
Regards,
Martin Brown

 

[Jan Panteltje]

On a sunny day (Mon, 20 Dec 2021 11:28:41 +0000) it happened Martin Brown
<\'\'\'newspam\'\'\'@nonad.co.uk> wrote in <spppda$99r$2@gioia.aioe.org>:

It relies on the weak law of large numbers to work. More signal is
always better but you have to accept that the VLF bands have more than
their fair share of noise in them from man made and natural events.

Right, I even did build a lightning detector with a ferrite rod tuned to some VLF frequency.
In case of lighting you are going to get periods with no signal (saturation) and all sort
of transients, also at other frequencies than the one you are looking for.
Lightning happens perhaps more often than you think, on an AM radio just causing cracking noises.

 

[Clive Arthur]

On 19/12/2021 05:14, Jan Frank wrote:

<snipped>

The Amazon ferrite loop is much larger and should provide plenty of signal:

https://www.amazon.ca/gp/product/B01KH3VEGS/

In addition there is no way to trim the resonance of a large loop except by
changing the capacitor. This would require a huge capacitor and a
microstepping motor to turn the shaft. This would make a large and
cumbersome assembly.

I\'m guessing you don\'t need much adjustment. Sliding a ferrite toroid
over the rod may well increase inductance a little, a brass ring may do
the reverse. Worth a try anyway.

--
Cheers
Clive

 

[Martin Brown]

On 20/12/2021 12:06, Jan Panteltje wrote:

On a sunny day (Mon, 20 Dec 2021 11:28:41 +0000) it happened Martin Brown
\'\'\'newspam\'\'\'@nonad.co.uk> wrote in <spppda$99r$2@gioia.aioe.org>:

It relies on the weak law of large numbers to work. More signal is
always better but you have to accept that the VLF bands have more than
their fair share of noise in them from man made and natural events.

Right, I even did build a lightning detector with a ferrite rod tuned to some VLF frequency.
In case of lighting you are going to get periods with no signal (saturation) and all sort
of transients, also at other frequencies than the one you are looking for.
Lightning happens perhaps more often than you think, on an AM radio just causing cracking noises.

Whistlers are really cool if you build a suitable wideband LF antenna
with low noise preamp and feed it into a waterfall display realtime FFT.

--
Regards,
Martin Brown

 

[Clifford Heath]

On 20/12/21 9:33 am, Jan Frank wrote:

jlarkin@highlandsniptechnology.com wrote:
1 PPS doesn\'t provide much information

The 10 MHz GPS clock is derived from the 1 PPS. I will use the phase of the
WWVB carrier to compare against the GPS signal.

The 1PPS is also derived by a PLL locked to the GPS signals, so you\'re
still dependent on the stability of the underlying VCO for low phase
noise. The GPS signal is well below the noise floor, so the PLL loop
filter is necessarily slow - meaning the VCO phase noise will probably
extend to sub-Hz frequencies before the GPS locking signal starts to
pull it in. The 10MHz output is no better, of course.

Folk who are enthusiastic about their GPS disciplined 10MHz reference
source seem to mostly disregard these issues.

Clifford Heath.

 

[Gerhard Hoffmann]

Am 03.01.22 um 23:58 schrieb Clifford Heath:

On 20/12/21 9:33 am, Jan Frank wrote:
jlarkin@highlandsniptechnology.com wrote:
1 PPS doesn\'t provide much information

The 10 MHz GPS clock is derived from the 1 PPS. I will use the phase
of the
WWVB carrier to compare against the GPS signal.

The 1PPS is also derived by a PLL locked to the GPS signals, so you\'re
still dependent on the stability of the underlying VCO for low phase
noise. The GPS signal is well below the noise floor, so the PLL loop
filter is necessarily slow - meaning the VCO phase noise will probably
extend to sub-Hz frequencies before the GPS locking signal starts to
pull it in. The 10MHz output is no better, of course.

Folk who are enthusiastic about their GPS disciplined 10MHz reference
source seem to mostly disregard these issues.

You seem to disregard multipath shortwave propagation and
effects of the ionosphere on WWV. And the SNR of GPS is
positive when the signal is folded back to its information
bandwidth. In the phase noise/short term and mid term you
can only rely on your crystal oscillator. Only in the long
term the GPS system lends a helping hand. The MTI-260 oven
in my GPS receiver is alone by itself not much worse than
a rubidium ( that must rely on its crystal osc for the
short term, too).

< http://www.hoffmann-hochfrequenz.de/downloads/DoubDist.pdf >
for a glimpse into a qualified GPS receiver.

Cheers, Gerhard

 

[whit3rd]

On Monday, January 3, 2022 at 4:18:53 PM UTC-8, Gerhard Hoffmann wrote:

Am 03.01.22 um 23:58 schrieb Clifford Heath:

You seem to disregard multipath shortwave propagation and
effects of the ionosphere on WWV.

The longwave WWVB ( 60 kHz) is dominated by ionospheric
conduction, so multipath is a non-issue for that
source (other WWV stations at 5, 10, 15 MHz are less
predictable). GPS ought to be less dependent on atmosphere
effects, of course \'cuz in the UP direction there\'s only a few miles of atmosphere.

For best results, you might just want to make an atomic fountain of your own.

<https://www.atlasobscura.com/places/nist-f1-cesium-fountain-atomic-clock>

 

[Jan Frank]

Clifford Heath <no.spam@please.net> wrote:

On 20/12/21 9:33 am, Jan Frank wrote:
jlarkin@highlandsniptechnology.com wrote:
1 PPS doesn\'t provide much information

The 10 MHz GPS clock is derived from the 1 PPS. I will use the phase of
the WWVB carrier to compare against the GPS signal.

The 1PPS is also derived by a PLL locked to the GPS signals, so you\'re
still dependent on the stability of the underlying VCO for low phase
noise. The GPS signal is well below the noise floor, so the PLL loop
filter is necessarily slow - meaning the VCO phase noise will probably
extend to sub-Hz frequencies before the GPS locking signal starts to
pull it in. The 10MHz output is no better, of course.

Folk who are enthusiastic about their GPS disciplined 10MHz reference
source seem to mostly disregard these issues.

Clifford Heath.

You have got it backwards. The 1PPS is from the GPS signal. The 10MHz is
from an OCXO locked to the 1PPS.

This is where most people go wrong. There is 20ns to 50ns jitter in the
1PPS signal. This is from the internal clock in the GPS receiver that
decodes the GPS signal and issues the 1PPS pulse. It knows how much error
there is between the GPS time and when it is going to issue the 1PPS pulse.
You can extract the error and use it to correct the 1PPS pulse, but this is
complicated and adds its own jitter.

A conventional phase detector using an RC loop filter will drift during the
period between samples, and pass the jitter to the OCXO. Duty cycle, or PWM
phase detectors, have huge jitter and suffer from false lock where you can
get a non-integer divide ratio. Then you need to filter the error signal
with a very long time constant, perhaps days. With care, you can get about
1e-12 error in the 10MHz clock. This makes it ideal for an in-house
standard, which is why a GPSDO is so popular.

I have designed a zdncpfd: zero deadband noise cancelling phase frequency
detector that locks two clocks together but omits the jitter of either
clock, such as the jitter in the 1PPS. It bypasses the 20*log(N) of
conventional multipliers and dividers. It has many uses outside of GPS and
WWVB.

It is a full PFD so it guarantees lock if the locked oscillator has enough
range. It is a combination of the first zero deadband PFD patent that I got
around 1970, and more recent inventions on noise cancelling high frequency
samplers. Yes, such a thing is possible.

It should be noted that Tom Van Baak of Time-Nuts has designed PIC dividers
with extemely low jitter that I will use in the countdown. He has also
published curves of diurnal phase shift of the WWVB carrier compared to a
Hydrogen Maser (how he can afford such a thing is beyound me) and shows the
severe shift at sunrise and sunset. Measuring Time is a very absorbing and
addictive hobby.

HNY