On Thu, 07 Mar 2013 13:11:03 -0500, rickman wrote:

I understand. But this is intended to be *very* low power and I haven't
found an amp I can use that is in the low double digits uW power
consumption range. I plan to use no amp and go straight to digital.

I don't think that's possible. Unless your input A/D converter can
operate in the microvolt region, it's going to have a difficult time
dealing with the low signal levels. Fortunately, WWVB is on-off
keying with no amplitude component, so there's no incentive to add an
AGC controlled input amplifier in order to maximize the A/D converters
dynamic range. Still, you need to work with something more than a few
bits above the noise level. Incidentally, after midnight, you WWVB
delivers about 100 uV/meter or more to continental US.
http://tf.nist.gov/tf-cgi/wwvbmonitor_e.cgi (Java required)
I've seen it strong enough that I can see the waveform on an
oscilloscope after a 60Khz passive filter.

As for bandwidth, the code is sent at 1 baud (1 bit/sec) which
produces about a 2Hz occupied bandwidth. Therefore, the maximum Q of
the antenna would need to be:
60Khz/ 2Hz = 30,000
before the antenna bandwidth becomes a problem.

Incidentally, while Googling away merrily, I found this on SPICE
models for a loop antenna. It's not quite in your xformer format, but
it might be useful:
http://sidstation.loudet.org/antenna-theory-en.xhtml
I won't pretend to understand what the author is doing until I read it
more carefully.

Incidentally, I used a WWVB code simulator driving a signal generator
to test my receiver:
http://www.leapsecond.com/notes/wwvb2.htm

If you're seriously into this, I suggest asking questions on the
time-nuts mailing list:
https://www.febo.com/mailman/listinfo/time-nuts

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

On Thu, 07 Mar 2013 11:39:45 -0800, Jeff Liebermann
wrote:

Fortunately, WWVB is on-off
keying with no amplitude component, so there's no incentive to add an
AGC controlled input amplifier in order to maximize the A/D converters
dynamic range.

Oops. WWVB does have an amplitude component and is not quite on-off
keying (OOK). There's a -17dB drop in RF signal level at the
beginning of each 1 second marker pulse. It was -10dB prior to 2005.
http://en.wikipedia.org/wiki/WWVB#Modulation_depth




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

On 3/7/2013 2:39 PM, Jeff Liebermann wrote:
On Thu, 07 Mar 2013 13:11:03 -0500, wrote:

I understand. But this is intended to be *very* low power and I haven't
found an amp I can use that is in the low double digits uW power
consumption range. I plan to use no amp and go straight to digital.

I don't think that's possible. Unless your input A/D converter can
operate in the microvolt region, it's going to have a difficult time
dealing with the low signal levels. Fortunately, WWVB is on-off
keying with no amplitude component, so there's no incentive to add an
AGC controlled input amplifier in order to maximize the A/D converters
dynamic range. Still, you need to work with something more than a few
bits above the noise level. Incidentally, after midnight, you WWVB
delivers about 100 uV/meter or more to continental US.
http://tf.nist.gov/tf-cgi/wwvbmonitor_e.cgi (Java required)
I've seen it strong enough that I can see the waveform on an
oscilloscope after a 60Khz passive filter.

Yes, I have done my homework on the WWVB signal. I am at the fringe of
the 100 uV/m contour. I would very much like to see the signal on an
oscilloscope when I test this. They have a receiver not far from here
in Gaithersburg, MD and the signal is often strong during the day. So
much so that I don't follow why they say there is this day/night signal
strength fluctuation. It looks much more random to me.

The WWVB signal is not truly on-off keying. I believe they use a 10 dB
modulation factor for the AM signal. This is close to on-off I agree.
But they also phase modulate the signal and I will be demodulating both
to see which one works best in my design.

The ADC in my design is truly one bit. It is an LVDS input on an FPGA.
I looked at delta-sigma (or is it sigma-delta? conversion and got
code from the chip vendor for a simplistic implementation. I don't
think I have the power budget for that and am using a much simpler 1 bit
ADC at 4x the carrier rate. The bit stream is multiplied by quadrature
carriers at 60 kHz and each stream summed for 1/30 of a second to
implement what can be considered a DFT bin, a decimated FIR filter or a
decimated down conversion; take your pick, they are all mathematically
the same in this case because the sampling is synchronous to the carrier
(or very close to synchronous).

What comes out the other end of this processing gains nearly 40 dB in
SNR. My simulations show a recoverable signal when it is more than 20
dB below the noise.

Of course, I have not tested this yet on a real signal. I want to run
some tests on the antenna and coupling transformer to verify the
simulation. Then I will start working with the FPGA to see if I can
make the LVDS input do what I want. I have ideas on how to bend digital
circuits to do my bidding. This LVDS input is why I want as large a
signal as possible from the antenna. With the high impedance input on
the chip I should be able to boost the signal pretty well with just
passive devices and signal processing.

The loop antenna is rather large. I would like to end up with something
smaller. Once I get this working with a shielded loop antenna I will
check out the ferrite core antennas. My understanding is that they
don't produce as much signal.


As for bandwidth, the code is sent at 1 baud (1 bit/sec) which
produces about a 2Hz occupied bandwidth. Therefore, the maximum Q of
the antenna would need to be:
60Khz/ 2Hz = 30,000
before the antenna bandwidth becomes a problem.

I'm not sure how you came up with 2 Hz for the bandwidth. In this case
the bandwidth is not just twice the bit rate. I believe the stated
"system" bandwidth is around 5 Hz (from a 1995 paper prior to addition
of the phase modulation). Regardless, I am sampling at 30 Hz and if I
expect to see significant changes in phase or amplitude within one
sample time, I need an appropriate bandwidth.

Even so, that is not the limiting factor. The limiting factor is the
difficulty in holding tune with drift in passive component values. The
Q can be raised by increasing the turns ratio on the transformer, but it
becomes so sensitive to the parasitic capacitance that the sensitivity
drops 10 dB with a 1 pF change.


Incidentally, while Googling away merrily, I found this on SPICE
models for a loop antenna. It's not quite in your xformer format, but
it might be useful:
http://sidstation.loudet.org/antenna-theory-en.xhtml
I won't pretend to understand what the author is doing until I read it
more carefully.

Thanks. I will take a look at that.


Incidentally, I used a WWVB code simulator driving a signal generator
to test my receiver:
http://www.leapsecond.com/notes/wwvb2.htm

I will be needing a time code simulator. I designed a commercial
product that works with the IRIG-B time code which is similar. The
functionality is not hard, it is just a matter of generating the data,
encoding it into the modulation pattern, then impressing the carrier
with the modulation. Working in an FPGA this sort of stuff is easy.

The trouble is if you make the same mistake in both the generator and
receiver they work just fine in simulation, but not with other
equipment. lol

I'll take a look at this link.


If you're seriously into this, I suggest asking questions on the
time-nuts mailing list:
https://www.febo.com/mailman/listinfo/time-nuts

I might look into that. Certainly it can't hurt to get more input.

--

Rick

On Fri, 08 Mar 2013 13:44:36 -0500, rickman wrote:

I am at the fringe of
the 100 uV/m contour. I would very much like to see the signal on an
oscilloscope when I test this.

I built a passive 60KHz bandpass filters out of a collection of
ferrite cores from an old modem front end. I left it at a previous
consulting job, but can resurrect the design if necessary.
Incidentally, during my limited testing at home, I found that the
biggest determent to decent reception was all the switching power
supply noise found around the house. I finally ended up using a
battery power oscilloscope
http://802.11junk.com/jeffl/pics/drivel/slides/tek213.html
a gel cell for powering the RF amp, and turning off the main power to
the house. Then, I could sorta see a signal.

They have a receiver not far from here
in Gaithersburg, MD and the signal is often strong during the day. So
much so that I don't follow why they say there is this day/night signal
strength fluctuation. It looks much more random to me.

http://tf.nist.gov/tf-cgi/wwvbgraph_e.cgi?5636103007
Very random. Compare the above graph with Santa Clara which looks
less random:
http://tf.nist.gov/tf-cgi/wwvbgraph_e.cgi?5636105007

On the east coast, besides a weak signal, you also have the potential
for 60KHz interference from the UK:
http://en.wikipedia.org/wiki/MSF_time_signal

I had a 100KHz LORAN antenna on the roof of a former employer. The
signal was just fine, until someone turned on the mercury vapor arc
parking lot lamps at night. They were changed to low pressure sodium,
which made testing possible at night.

Incidentally, got any clue as to the vertical scale? My guess(tm) is
20 uv/meter signal strength per division, but I'm not sure.

The WWVB signal is not truly on-off keying. I believe they use a 10 dB
modulation factor for the AM signal. This is close to on-off I agree.

It's now 17dB drop at the beginning of each UTC second. The change
came in about 2008.

But they also phase modulate the signal and I will be demodulating both
to see which one works best in my design.

The BPSK signal is much better at rejecting interference and digging
the signal out of the noise. I don't know exactly how much, but I'm
sure it's in a NIST publication somewhere.

The ADC in my design is truly one bit. It is an LVDS input on an FPGA.
I looked at delta-sigma (or is it sigma-delta?

It's delta-sigma.

The loop antenna is rather large. I would like to end up with something
smaller. Once I get this working with a shielded loop antenna I will
check out the ferrite core antennas. My understanding is that they
don't produce as much signal.

Not exactly. Small loopsticks receive a proportional amount of noise.
The ratio of signal to atmospheric noise remains roughly the same
within a fixed bandwidth for any antenna. That's why tiny little
loopsticks, inside "atomic time" wris****ches work. The small
loopsticks also use the magnetic field instead of the electric field,
which is why they can be made so small.
http://en.wikipedia.org/wiki/Loop_antenna#Small_loops

As for bandwidth, the code is sent at 1 baud (1 bit/sec) which
produces about a 2Hz occupied bandwidth. Therefore, the maximum Q of
the antenna would need to be:
60Khz/ 2Hz = 30,000
before the antenna bandwidth becomes a problem.

I'm not sure how you came up with 2 Hz for the bandwidth. In this case
the bandwidth is not just twice the bit rate. I believe the stated
"system" bandwidth is around 5 Hz (from a 1995 paper prior to addition
of the phase modulation).

Ok, I made a bad guess(tm). Even at 5Hz BW, the maximum Q of
60KHz / 5Hz = 12,000
is not going to happen in a loop or loopstick antenna.

Even so, that is not the limiting factor. The limiting factor is the
difficulty in holding tune with drift in passive component values.

Agreed.


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

On 3/10/2013 1:32 AM, Jeff Liebermann wrote:
On Fri, 08 Mar 2013 13:44:36 -0500, wrote:

I am at the fringe of
the 100 uV/m contour. I would very much like to see the signal on an
oscilloscope when I test this.

I built a passive 60KHz bandpass filters out of a collection of
ferrite cores from an old modem front end. I left it at a previous
consulting job, but can resurrect the design if necessary.
Incidentally, during my limited testing at home, I found that the
biggest determent to decent reception was all the switching power
supply noise found around the house. I finally ended up using a
battery power oscilloscope
http://802.11junk.com/jeffl/pics/drivel/slides/tek213.html
a gel cell for powering the RF amp, and turning off the main power to
the house. Then, I could sorta see a signal.

Holy crap! That's a lot of trouble to see a signal. By "see" I assume
you mean on the scope. How large was the signal?

The place where I am working currently is not very close to much and
there isn't much in the house. I'm told the fridge is the biggest
source of noise. We'll see how the CFL lamps do.

Funny, last night my two RCC's both updated like they should. One is an
analog clock and runs at 8x speed to get the hour ahead. In the fall it
does this to go 11 hours ahead. Quite a sight! They both did the job,
but my PC didn't update until it had been on for awhile, without being
connected to the I'net.


They have a receiver not far from here
in Gaithersburg, MD and the signal is often strong during the day. So
much so that I don't follow why they say there is this day/night signal
strength fluctuation. It looks much more random to me.

http://tf.nist.gov/tf-cgi/wwvbgraph_e.cgi?5636103007
Very random. Compare the above graph with Santa Clara which looks
less random:
http://tf.nist.gov/tf-cgi/wwvbgraph_e.cgi?5636105007

On the east coast, besides a weak signal, you also have the potential
for 60KHz interference from the UK:
http://en.wikipedia.org/wiki/MSF_time_signal

Loop antennas have a null that can be steered toward the source of
interference. I expect that will solve that problem...


I had a 100KHz LORAN antenna on the roof of a former employer. The
signal was just fine, until someone turned on the mercury vapor arc
parking lot lamps at night. They were changed to low pressure sodium,
which made testing possible at night.

Incidentally, got any clue as to the vertical scale? My guess(tm) is
20 uv/meter signal strength per division, but I'm not sure.

The WWVB signal is not truly on-off keying. I believe they use a 10 dB
modulation factor for the AM signal. This is close to on-off I agree.

It's now 17dB drop at the beginning of each UTC second. The change
came in about 2008.

But they also phase modulate the signal and I will be demodulating both
to see which one works best in my design.

The BPSK signal is much better at rejecting interference and digging
the signal out of the noise. I don't know exactly how much, but I'm
sure it's in a NIST publication somewhere.

That's for an ideal receiver. I have my limitations and I have no idea
how that will impact the reception.


The ADC in my design is truly one bit. It is an LVDS input on an FPGA.
I looked at delta-sigma (or is it sigma-delta?

It's delta-sigma.

Actually I always say that with a smiley as it can be either.


The loop antenna is rather large. I would like to end up with something
smaller. Once I get this working with a shielded loop antenna I will
check out the ferrite core antennas. My understanding is that they
don't produce as much signal.

Not exactly. Small loopsticks receive a proportional amount of noise.
The ratio of signal to atmospheric noise remains roughly the same
within a fixed bandwidth for any antenna. That's why tiny little
loopsticks, inside "atomic time" wris****ches work. The small
loopsticks also use the magnetic field instead of the electric field,
which is why they can be made so small.
http://en.wikipedia.org/wiki/Loop_antenna#Small_loops

In my case I am not worried that the SNR isn't better, I just need a
strong enough signal to drive the LVDS input. I will be providing
feedback to eliminate any DC bias, but even that will only be so good.
The input is claimed to have no hysteresis, but even a tiny amount can
ruin this design. I will only know if this will work when I try it.


As for bandwidth, the code is sent at 1 baud (1 bit/sec) which
produces about a 2Hz occupied bandwidth. Therefore, the maximum Q of
the antenna would need to be:
60Khz/ 2Hz = 30,000
before the antenna bandwidth becomes a problem.

I'm not sure how you came up with 2 Hz for the bandwidth. In this case
the bandwidth is not just twice the bit rate. I believe the stated
"system" bandwidth is around 5 Hz (from a 1995 paper prior to addition
of the phase modulation).

Ok, I made a bad guess(tm). Even at 5Hz BW, the maximum Q of
60KHz / 5Hz = 12,000
is not going to happen in a loop or loopstick antenna.

Even so, that is not the limiting factor. The limiting factor is the
difficulty in holding tune with drift in passive component values.

Agreed.




--

Rick

On Sun, 10 Mar 2013 17:09:45 -0400, rickman wrote:

On 3/10/2013 1:32 AM, Jeff Liebermann wrote:

Holy crap! That's a lot of trouble to see a signal. By "see" I assume
you mean on the scope. How large was the signal?

Turning off the house was easier than finding the multiple sources of
noise at 60KHz. What drove me nuts for about an hour was that much of
the noise was coming from my bench oscilloscope. Argh.

This is typical. WWVH through an active preamp showing the effect of
power line noise (probably from attached switching power supplies).
http://www.prc68.com/I/Images/AMRAD110.GIF
and after adding some better line filtering:
http://www.prc68.com/I/Images/AMRAD_BT.GIF
Main page:
http://www.prc68.com/I/LF-Ant.shtml

I didn't log the setup or take pictures. So, let's do the math and
guesswork.
http://vk1od.net/calc/FS2RPCalc.htm
I plugged in some guesses and recollections as to what the antenna
(Q=30) and amp (+20dB gain) were doing and got:
http://802.11junk.com/jeffl/crud/WWVH-rx-signal-estimate.jpg
-15.8dBm or about 36mv into 50 ohms. I amplified this about 20dB with
two or three U310 JFET's (I forgot what I did) to about 3V rms on the
scope. I didn't bother with the 50 ohm to scope input Z conversion.
Most of what I saw was noise, noise, and more noise. However, if I
was patient, I could see the data fade in an out. As I vaguely
recall, it was less than 1 division or about 0.1v change.

The place where I am working currently is not very close to much and
there isn't much in the house. I'm told the fridge is the biggest
source of noise. We'll see how the CFL lamps do.

Sigh. Most of what I found at 60KHz was coming from lightning storms
over Florida. The local sources were all switching power supplies,
including those in my test equipment. I didn't have an CFL or LED
room lights at the time. I've recently found them to be a rather
nasty noise source. Also, the switching power supply wall warts were
rather awful. My standard test is to fire up my antique IC-735 HF
xceiver, attach a long length of RG-58c/u to the antenna with a
resonant loop at the end, tune it to 100KHz (as low as it will go),
and sniff around the house.

What you'll see on a spectrum analyzer.
http://www.prc68.com/I/Spec_0002.shtml
If you're thinking of removing all that junk with a 5Hz wide digital
filter in software, please note that you'll need to have the input A/D
handle the total power of almost all that junk. Also, the amplifier
that you're trying to avoid between the antenna and A/D will also need
to be rather linear, and therefore rather high power, in order to
avoid producing more spurious junk via intermodulation products.

Funny, last night my two RCC's both updated like they should. One is an
analog clock and runs at 8x speed to get the hour ahead. In the fall it
does this to go 11 hours ahead. Quite a sight! They both did the job,
but my PC didn't update until it had been on for awhile, without being
connected to the I'net.

So that's how they change daylight savings time. If I had known, I
would have stayed and watched. Thanks for the tip.

On the east coast, besides a weak signal, you also have the potential
for 60KHz interference from the UK:
http://en.wikipedia.org/wiki/MSF_time_signal

Loop antennas have a null that can be steered toward the source of
interference. I expect that will solve that problem...

The depth of the notch seems to be less as the antenna shrinks in
size. I'm not sure about this as I haven't attempted to recently
model a 60KHz magnetic loop with 4NEC2, but that's what my tinkering
shows. If there were a deep notch, most of the home "atomic clock"
receivers would be orientation sensitive and I would expect warnings
in the docs.

The BPSK signal is much better at rejecting interference and digging
the signal out of the noise. I don't know exactly how much, but I'm
sure it's in a NIST publication somewhere.

That's for an ideal receiver. I have my limitations and I have no idea
how that will impact the reception.

Well, you have to start somewhere, and an ideal receiver is a good
place to start. The advantage is that reality only makes everything
worse, never better. You should be able to build the BPSK
demodulator, and then use a PC to decode the data. I've seen several
such programs that do not require I/Q outputs. Here's one based on
FreeBSD intended to sync the system clock to WWV/WWVH:
http://docs.freebsd.org/doc/4.0-RELEASE/usr/share/doc/ntp/driver36.htm
I'm sure there are others.

In my case I am not worried that the SNR isn't better, I just need a
strong enough signal to drive the LVDS input.

Gain at 60KHz is very cheap. Watch out for overload issues. If you
design it to work at full scale with whatever you get at 50uV/m, and
the signal climbs to 100uV/m, your input A/D isn't going to be very
happy. AGC will help, but I don't think it will be needed if you
calculate your signal levels so that the A/D input amp isn't clipping.

Out of service for a day. It seems that about 30 years of chemistry
experiments has finally destroyed much of the kitchen sink plumbing. I
hate plumbing.

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

On 3/10/2013 11:42 PM, Jeff Liebermann wrote:
On Sun, 10 Mar 2013 17:09:45 -0400, wrote:

On 3/10/2013 1:32 AM, Jeff Liebermann wrote:

Holy crap! That's a lot of trouble to see a signal. By "see" I assume
you mean on the scope. How large was the signal?

Turning off the house was easier than finding the multiple sources of
noise at 60KHz. What drove me nuts for about an hour was that much of
the noise was coming from my bench oscilloscope. Argh.

Yes, like I said, not much in this house and there is not much near it.
I have a laptop and my roommate (when he is here) uses one along with
an iPhone. I suppose they might generate some noise, but he turns off
his laptop at night I'm sure. Otherwise, there just isn't much in the
house that isn't 10 or 15 years old. I have a car radio on a linear
regulator and an electric shaver that sits charging (part of the time).
Otherwise it should be pretty quiet electrically here.


This is typical. WWVH through an active preamp showing the effect of
power line noise (probably from attached switching power supplies).
http://www.prc68.com/I/Images/AMRAD110.GIF
and after adding some better line filtering:
http://www.prc68.com/I/Images/AMRAD_BT.GIF
Main page:
http://www.prc68.com/I/LF-Ant.shtml

I didn't log the setup or take pictures. So, let's do the math and
guesswork.
http://vk1od.net/calc/FS2RPCalc.htm
I plugged in some guesses and recollections as to what the antenna
(Q=30) and amp (+20dB gain) were doing and got:
http://802.11junk.com/jeffl/crud/WWVH-rx-signal-estimate.jpg
-15.8dBm or about 36mv into 50 ohms. I amplified this about 20dB with
two or three U310 JFET's (I forgot what I did) to about 3V rms on the
scope. I didn't bother with the 50 ohm to scope input Z conversion.
Most of what I saw was noise, noise, and more noise. However, if I
was patient, I could see the data fade in an out. As I vaguely
recall, it was less than 1 division or about 0.1v change.

I'm not sure why you used 1900 meters for the distance. I also don't
get why you used 5 ohms for the receiver input impedance.


The place where I am working currently is not very close to much and
there isn't much in the house. I'm told the fridge is the biggest
source of noise. We'll see how the CFL lamps do.

Sigh. Most of what I found at 60KHz was coming from lightning storms
over Florida. The local sources were all switching power supplies,
including those in my test equipment. I didn't have an CFL or LED
room lights at the time. I've recently found them to be a rather
nasty noise source. Also, the switching power supply wall warts were
rather awful. My standard test is to fire up my antique IC-735 HF
xceiver, attach a long length of RG-58c/u to the antenna with a
resonant loop at the end, tune it to 100KHz (as low as it will go),
and sniff around the house.

There's just not much of that in this house.


What you'll see on a spectrum analyzer.
http://www.prc68.com/I/Spec_0002.shtml
If you're thinking of removing all that junk with a 5Hz wide digital
filter in software, please note that you'll need to have the input A/D
handle the total power of almost all that junk. Also, the amplifier
that you're trying to avoid between the antenna and A/D will also need
to be rather linear, and therefore rather high power, in order to
avoid producing more spurious junk via intermodulation products.

What A/D? Oh, you mean the LVDS input. How do you saturate a 1 bit ADC?


Funny, last night my two RCC's both updated like they should. One is an
analog clock and runs at 8x speed to get the hour ahead. In the fall it
does this to go 11 hours ahead. Quite a sight! They both did the job,
but my PC didn't update until it had been on for awhile, without being
connected to the I'net.

So that's how they change daylight savings time. If I had known, I
would have stayed and watched. Thanks for the tip.

On the east coast, besides a weak signal, you also have the potential
for 60KHz interference from the UK:
http://en.wikipedia.org/wiki/MSF_time_signal

Loop antennas have a null that can be steered toward the source of
interference. I expect that will solve that problem...

The depth of the notch seems to be less as the antenna shrinks in
size. I'm not sure about this as I haven't attempted to recently
model a 60KHz magnetic loop with 4NEC2, but that's what my tinkering
shows. If there were a deep notch, most of the home "atomic clock"
receivers would be orientation sensitive and I would expect warnings
in the docs.

I wouldn't say the *depth* of the null depends on the size of the loop.
I think it is a null with a Q, much like a resonance peak, but a null
of course. The smaller the loop, the sharper the null like a high Q
resonance, so the orientation becomes very critical. In theory at
least, the null is perfect, 0 signal.


The BPSK signal is much better at rejecting interference and digging
the signal out of the noise. I don't know exactly how much, but I'm
sure it's in a NIST publication somewhere.

That's for an ideal receiver. I have my limitations and I have no idea
how that will impact the reception.

Well, you have to start somewhere, and an ideal receiver is a good
place to start. The advantage is that reality only makes everything
worse, never better. You should be able to build the BPSK
demodulator, and then use a PC to decode the data. I've seen several
such programs that do not require I/Q outputs. Here's one based on
FreeBSD intended to sync the system clock to WWV/WWVH:
http://docs.freebsd.org/doc/4.0-RELEASE/usr/share/doc/ntp/driver36.htm
I'm sure there are others.

PC?!!! We don't need no stinking PCs! The demodulator is simple. The
signal is beat with a quadrature reference which will bring it down to 0
Hz. This gives two values, a sin and a cos signal. Take the ratio and
do an arcTan. This is a simple table lookup made simpler by some
convenient math relations. For example, the table only needs to cover
0° to 45° since the ratio can be swapped for 45° to 90° and the other
three quadrants distinguished by the sign bits.

Some folks would like you to think this has to be done like a high
fidelity receiver, but it only has to pull the signal out of the noise.


In my case I am not worried that the SNR isn't better, I just need a
strong enough signal to drive the LVDS input.

Gain at 60KHz is very cheap. Watch out for overload issues. If you
design it to work at full scale with whatever you get at 50uV/m, and
the signal climbs to 100uV/m, your input A/D isn't going to be very
happy. AGC will help, but I don't think it will be needed if you
calculate your signal levels so that the A/D input amp isn't clipping.

If this is *signal* strength then it won't matter. If this is noise you
are talking about, I'm not sure it will be a problem, the signal will
still be able to be dug out with enough processing gain.


Out of service for a day. It seems that about 30 years of chemistry
experiments has finally destroyed much of the kitchen sink plumbing. I
hate plumbing.

Not much fun, but then what it if you have to work on your knees and get
dirty? I don't enjoy working on my car anymore either.

--

Rick

On Mon, 11 Mar 2013 18:10:23 -0400, rickman wrote:

So, let's do the math and
guesswork.
http://vk1od.net/calc/FS2RPCalc.htm
I plugged in some guesses and recollections as to what the antenna
(Q=30) and amp (+20dB gain) were doing and got:
http://802.11junk.com/jeffl/crud/WWVH-rx-signal-estimate.jpg
-15.8dBm or about 36mv into 50 ohms. I amplified this about 20dB with
two or three U310 JFET's (I forgot what I did) to about 3V rms on the
scope. I didn't bother with the 50 ohm to scope input Z conversion.
Most of what I saw was noise, noise, and more noise. However, if I
was patient, I could see the data fade in an out. As I vaguely
recall, it was less than 1 division or about 0.1v change.

I'm not sure why you used 1900 meters for the distance. I also don't
get why you used 5 ohms for the receiver input impedance.

(Quick reply... still working on my expanding plumbing problem).

The 1900 meters is because I screwed up. It should be about 1900Km
from San Francisco to Denver. However, any distance greater than zero
will suffice for this calculation. The controlling numbers are the
100uV/m field strength, the -3dB antenna gain, and the receiver
bandwidth (5Hz). All of the other numbers can change without having
any effect on the recovered power. The 5 ohms rx input Z was because
the original antenna that I used, was a base loaded 100ft "whip"
antenna with a rather impedance. I couldn't decide if the field
strength to receive power form wanted the antenna impedance before the
50 ohm matching network, or if it treated the matching as part of the
antenna. I flipped a coin and chose 5 ohms. I guess for a loop,
100-200 ohms would be more appropriate. Again, the value makes no
difference in the calculations.

(Back to plumbing and fixing the 48" farm jack that my neighbor
borrowed and returned looking like a pretzel).




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

rickman wrote:
Funny, last night my two RCC's both updated like they should. One is an
analog clock and runs at 8x speed to get the hour ahead. In the fall it
does this to go 11 hours ahead. Quite a sight! They both did the job,
but my PC didn't update until it had been on for awhile, without being
connected to the I'net.

In Europe we have DCF-77 which is at 77.5 kHz.
The trouble receiving it is similar to WWVB.
I have several clocks in the house but some of them have only very
weak sync. Also, to save battery they only sync once every 12 hours
or so. At DST change, they may display the wrong time for a couple
of days, especially the one in the kitchen. I need to relocate it
to a place where I know there is better signal.

The problem is (harmonics of) switching power supplies here as well.
Once I had a big open-frame SMPS that I used to power my radio equipment
and that switched around 25 kHz. Under the right circumstances, the
3rd harmonic wiped away all DCF-77 receiving within 5 meters or so.
Old CRT computer monitors also were problematic.

I presume you have some specific needs, low power being among them,
to stay focussed on WWVB for your clock sync. Most computer users
would use GPS now, or simply sync via the internet. GPS has a different
receiving conditions problem, but at least it isn't so much affected
by prominently present local interference.

Of course a GPS receiver requires a lot more power than a WWVB receiver,
although this has come down over the years. Using some sort of on/off
switching (syncing with the received time and then having it run free
for some time) may help a bit, the battery powered radio clocks do
that as well.

On 3/11/2013 5:54 AM, Rob wrote:
wrote:
Funny, last night my two RCC's both updated like they should. One is an
analog clock and runs at 8x speed to get the hour ahead. In the fall it
does this to go 11 hours ahead. Quite a sight! They both did the job,
but my PC didn't update until it had been on for awhile, without being
connected to the I'net.

In Europe we have DCF-77 which is at 77.5 kHz.
The trouble receiving it is similar to WWVB.
I have several clocks in the house but some of them have only very
weak sync. Also, to save battery they only sync once every 12 hours
or so. At DST change, they may display the wrong time for a couple
of days, especially the one in the kitchen. I need to relocate it
to a place where I know there is better signal.

Yes, noise can be a problem I understand. I am hoping to get the
bandwidth down much more than most receivers so the noise won't be so
big a factor. With a signal bandwidth of a handful of Hz, it should be
possible.


The problem is (harmonics of) switching power supplies here as well.
Once I had a big open-frame SMPS that I used to power my radio equipment
and that switched around 25 kHz. Under the right circumstances, the
3rd harmonic wiped away all DCF-77 receiving within 5 meters or so.
Old CRT computer monitors also were problematic.

I presume you have some specific needs, low power being among them,
to stay focussed on WWVB for your clock sync. Most computer users
would use GPS now, or simply sync via the internet. GPS has a different
receiving conditions problem, but at least it isn't so much affected
by prominently present local interference.

Yes, this is actually a demo to illustrate how low power an FPGA can be.
An FPGA will run both the clock and the receiver and use power from
the environment rather than batteries.


Of course a GPS receiver requires a lot more power than a WWVB receiver,
although this has come down over the years. Using some sort of on/off
switching (syncing with the received time and then having it run free
for some time) may help a bit, the battery powered radio clocks do
that as well.

Yes, the receiver itself only has to run part of the time, 10% perhaps.
The clock has to run 100% obviously. Interesting enough, the FPGA has
a base power consumption (0 Hz) of nearly 50% of the power budget and I
am confident it will still make the goal.

--

Rick