Re: [time-nuts] 4046 experiment for gpsdo
Another method would be to measure the phase-detector beat-note frequency
(most have mixer-like behavior), which you should be able to measure with
quite good precision, then set the EFC accordingly and then close the loop.
How do you get the sign out of a beat note?
--
These are my opinions. I hate spam.
Hi Hal,
On 09/26/2015 11:47 PM, Hal Murray wrote:
Another method would be to measure the phase-detector beat-note frequency
(most have mixer-like behavior), which you should be able to measure with
quite good precision, then set the EFC accordingly and then close the loop.
How do you get the sign out of a beat note?
You don't directly. You need to augment it, and there is several ways of
doing that.
For instance, you can use the PFD detector of the 4046 to know if you
are above or below in frequency. The normal low-pass filter on the
output of the charge-pump should do it.
Another method is directly on the beat-note. You simply try either
direction. The beat-note will either be twice the rate or be very low
frequency. Since you already established the rate, if you haven't seen
it for the period of the rate you had (if it doubles, it will be at
least once in that period) then you guessed right, otherwise you guessed
wrong and just need to flip the sign and re-verify. To make sure you can
even let the verification period be multiple cycles of the detected
beat-note rate.
The beat-note method does not work very well or quickly if you have a
steered source which has a very narrow range to begin with, say a
rubidium. In that case looking at the beat-note is just a monitoring
solution to know when to start attempting lock.
There is so many ways to do this. I just wanted to illustrate another
method, and that relatively simple approaches, some minor logic applied
to a CPU and a handful lines of code can create a relatively robust design.
Remember, in the end you want a good phase-detector with a PI-loop, at
least. How you get it to lock up smoothly and relatively robustly there
is myriad of tricks to use. The beat-note approach may suffice for some
hobbyist work while being relative simple to understand.
The beat-note of the unlocked phase-detector is also how phase-detector
lock is often detected in analog designs, at least those that I've seen.
Essentially they detect the lack of AC signal in the phase-detector
response, which is due to the cycle-slip. Each cycle-slip is a beat-note
event. The rate of it is the difference in frequency, and as it is
relatively low frequency, using some 48 MHz counter to count the period
time provides a good precision. A rubidium with say maximum of 1E-9
error, even the quickest beat-note would be 100 s, so it is clearly not
a useful solution for quick-lock.
Cheers,
Magnus
Hi
If you digitize the beat note it’s fairly simple:
The beat note is not really a sine wave. It’s periodic, but not a pure sine. The
reason is fairly simple. The frequency changes as the beat note changes the EFC.
You have a lower frequency as it gets closer to the “zero frequency”.
The net result is a signal that is a bit flat on the side that is towards the zero. It’s a
bit sharper away from the zero. Fit it with a sine wave and you will get a residual. From that
you can work out which side is which.
Bob
On Sep 26, 2015, at 5:47 PM, Hal Murray hmurray@megapathdsl.net wrote:
Another method would be to measure the phase-detector beat-note frequency
(most have mixer-like behavior), which you should be able to measure with
quite good precision, then set the EFC accordingly and then close the loop.
How do you get the sign out of a beat note?
--
These are my opinions. I hate spam.
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Hello All,
Thank you for all the answers. I have been experimenting relentlessly with
phase detectors, and finally I came up with a circuit that will isolate
lagging and forward pulses.
So, if signal is lagging the compare, it will send one pulse from the
output of the first flip flop. If the reverse, it will send pulse from the
output of second ff.
I have also built the PC2 of 4046 with nand gates, it does encode direction
of phase difference, but I have not particularly found it useful. The
circuit above works in simulation, but I am to test it tomorrow in my
workshop.
It turns out the heart of this operation is the phase comparator, and there
are many phase-frequency comparators. I am expecting the circuit above work
with picosecond precision.
There are also many chips that does phase-frequency comparison. Maybe one
would be suitable for this better?
Best Regards,
C.A.
On Sun, Sep 27, 2015 at 1:12 AM, Magnus Danielson magnus@rubidium.se
wrote:
Hi Hal,
On 09/26/2015 11:47 PM, Hal Murray wrote:
Another method would be to measure the phase-detector beat-note
frequency
(most have mixer-like behavior), which you should be able to measure
with
quite good precision, then set the EFC accordingly and then close the
loop.
How do you get the sign out of a beat note?
You don't directly. You need to augment it, and there is several ways of
doing that.
For instance, you can use the PFD detector of the 4046 to know if you are
above or below in frequency. The normal low-pass filter on the output of
the charge-pump should do it.
Another method is directly on the beat-note. You simply try either
direction. The beat-note will either be twice the rate or be very low
frequency. Since you already established the rate, if you haven't seen it
for the period of the rate you had (if it doubles, it will be at least once
in that period) then you guessed right, otherwise you guessed wrong and
just need to flip the sign and re-verify. To make sure you can even let the
verification period be multiple cycles of the detected beat-note rate.
The beat-note method does not work very well or quickly if you have a
steered source which has a very narrow range to begin with, say a rubidium.
In that case looking at the beat-note is just a monitoring solution to know
when to start attempting lock.
There is so many ways to do this. I just wanted to illustrate another
method, and that relatively simple approaches, some minor logic applied to
a CPU and a handful lines of code can create a relatively robust design.
Remember, in the end you want a good phase-detector with a PI-loop, at
least. How you get it to lock up smoothly and relatively robustly there is
myriad of tricks to use. The beat-note approach may suffice for some
hobbyist work while being relative simple to understand.
The beat-note of the unlocked phase-detector is also how phase-detector
lock is often detected in analog designs, at least those that I've seen.
Essentially they detect the lack of AC signal in the phase-detector
response, which is due to the cycle-slip. Each cycle-slip is a beat-note
event. The rate of it is the difference in frequency, and as it is
relatively low frequency, using some 48 MHz counter to count the period
time provides a good precision. A rubidium with say maximum of 1E-9 error,
even the quickest beat-note would be 100 s, so it is clearly not a useful
solution for quick-lock.
Cheers,
Magnus
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Hello All,
Of course I could not wait for tomorrow, so built and tested the phase
comparator that I mentioned in my previous email. (with a change)
I hooked up the output (labeled 4 and 5) to my scope, and introduced a very
small phase difference between clocks,
and I get tiny tiny pulses from 4 and 5 depending on whichever is lagging.
So it works. There is a little DC bias on the flip flop outputs, but I
guess the signal diode will take care of it.
So here is my plan, in summary: hookup GPS 1PPS output to one input, and
1PPS output from my RTC to other.
The changes in phase introduce tiny pulses from outputs of flip-flops 4 or
5 whether depending on whichever one is leading.
I get those pulses, and feed it into 2 seperate RC networks, (like
described at the 1ns GPSDO email) - But instead of one I will have two
networks, and 2 adc's sampling, one for lagging other for trailing pulses,
which then I will be able to adjust the speed of my XTAL or RTC.
Currently I am using 74LS versions of the chips, since thats what I had in
my junkbox, but monday I will buy 74HC or 74HCT versions. I would love to
hear about your ideas on which family to use, for this application.
Once I am done constructing a proper circuit, and finish the project I
decided to open source the code on github. (It is written in energia, but
it will be fully compatible with arduino, it might need some changes for
interrupts since on arduino you can not use any pin for interrupt like in
texas instruments launchpads)
I am also very curious about what people use in metrology labs for time
comparison? If we were chasing pico or femto seconds syncing, how would we
compare phases?
Best regards, and thank you very much for your input.
-Can Altineller
On Sun, Sep 27, 2015 at 4:30 AM, Can Altineller altineller@gmail.com
wrote:
Hello All,
Thank you for all the answers. I have been experimenting relentlessly with
phase detectors, and finally I came up with a circuit that will isolate
lagging and forward pulses.
So, if signal is lagging the compare, it will send one pulse from the
output of the first flip flop. If the reverse, it will send pulse from the
output of second ff.
I have also built the PC2 of 4046 with nand gates, it does encode
direction of phase difference, but I have not particularly found it useful.
The circuit above works in simulation, but I am to test it tomorrow in my
workshop.
It turns out the heart of this operation is the phase comparator, and
there are many phase-frequency comparators. I am expecting the circuit
above work with picosecond precision.
There are also many chips that does phase-frequency comparison. Maybe one
would be suitable for this better?
Best Regards,
C.A.
On Sun, Sep 27, 2015 at 1:12 AM, Magnus Danielson magnus@rubidium.se
wrote:
Hi Hal,
On 09/26/2015 11:47 PM, Hal Murray wrote:
Another method would be to measure the phase-detector beat-note
frequency
(most have mixer-like behavior), which you should be able to measure
with
quite good precision, then set the EFC accordingly and then close the
loop.
How do you get the sign out of a beat note?
You don't directly. You need to augment it, and there is several ways of
doing that.
For instance, you can use the PFD detector of the 4046 to know if you are
above or below in frequency. The normal low-pass filter on the output of
the charge-pump should do it.
Another method is directly on the beat-note. You simply try either
direction. The beat-note will either be twice the rate or be very low
frequency. Since you already established the rate, if you haven't seen it
for the period of the rate you had (if it doubles, it will be at least once
in that period) then you guessed right, otherwise you guessed wrong and
just need to flip the sign and re-verify. To make sure you can even let the
verification period be multiple cycles of the detected beat-note rate.
The beat-note method does not work very well or quickly if you have a
steered source which has a very narrow range to begin with, say a rubidium.
In that case looking at the beat-note is just a monitoring solution to know
when to start attempting lock.
There is so many ways to do this. I just wanted to illustrate another
method, and that relatively simple approaches, some minor logic applied to
a CPU and a handful lines of code can create a relatively robust design.
Remember, in the end you want a good phase-detector with a PI-loop, at
least. How you get it to lock up smoothly and relatively robustly there is
myriad of tricks to use. The beat-note approach may suffice for some
hobbyist work while being relative simple to understand.
The beat-note of the unlocked phase-detector is also how phase-detector
lock is often detected in analog designs, at least those that I've seen.
Essentially they detect the lack of AC signal in the phase-detector
response, which is due to the cycle-slip. Each cycle-slip is a beat-note
event. The rate of it is the difference in frequency, and as it is
relatively low frequency, using some 48 MHz counter to count the period
time provides a good precision. A rubidium with say maximum of 1E-9 error,
even the quickest beat-note would be 100 s, so it is clearly not a useful
solution for quick-lock.
Cheers,
Magnus
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Can Altineller wrote:
There are also many chips that does phase-frequency comparison. Maybe one
would be suitable for this better?
You may enjoy reading the op/svc manual for the HP linear phase comparator. A copy is on Had's site:
http://www.to-way.com/tf/k34599a.pdf
/tvb
Hi,
On 09/27/2015 02:04 AM, Bob Camp wrote:
Hi
If you digitize the beat note it’s fairly simple:
The beat note is not really a sine wave. It’s periodic, but not a pure sine. The
reason is fairly simple. The frequency changes as the beat note changes the EFC.
You have a lower frequency as it gets closer to the “zero frequency”.
In closed-loop yes. I actually proposed to start in open-loop mode,
estimate the frequency and then close the loop after a coarse frequency
adjustment have been done.
In closed loop, the beat-note shape, which comes from non-linear
modulation, clearly indicates the direction. You can use this knowledge
to learn the polarity, as one of many methods.
A semi-closed loop will get you the shape, i.e. a loop where you only
close the proportional part, but not the integration part. Looking at
the shape (PWM) of the beat-note you can learn the polarity and the rate
the amplitude of the frequency error. Once you know the polarity and
error, just initiate the integrating part accordingly and close the
integration part of the loop (by setting a non-zero integrating constant).
Another method is naturally to do a quadrature detection. As I said,
there is so many ways to do it, and I wanted to illustrate a different
approach. If you can get a 360 degree phase measure, frequency measure
is almost trivial to derive from that.
The net result is a signal that is a bit flat on the side that is towards the zero. It’s a
bit sharper away from the zero. Fit it with a sine wave and you will get a residual. From that
you can work out which side is which.
You can also measure the PWM if you like.
PWM < 50% => positive frequency
PWM > 50% => negative frequency
Cheers,
Magnus
Bob
On Sep 26, 2015, at 5:47 PM, Hal Murray hmurray@megapathdsl.net wrote:
Another method would be to measure the phase-detector beat-note frequency
(most have mixer-like behavior), which you should be able to measure with
quite good precision, then set the EFC accordingly and then close the loop.
How do you get the sign out of a beat note?
--
These are my opinions. I hate spam.
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.