While evaluating a Trimble Mini-T GPSDO it appears the 10 MHz output
frequency,  with loop control enabled, is jumping up/down every second
or so in the order of 5e-10  (green trace in attached image) resulting
in a 0.1 s ADEV of 8e-11.
In holdover mode (no GPS input and no PPS output) the output frequency
has a much better short term stability  (purple trace in attached
image) resulting in a 0.1 s ADEV of 5e-12
The time constant was set to 100s and the damping factor to 1. The
Mini-T was powered by a linear bench supply.
The DOCXO used has reference has a 0.1s - 10 s ADEV of at most 5e-12.
Is there anything in the tuning that can reduce the jumps or is this
to be expected from a Mini-T?
Erik.

time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe send an email to time-nuts-leave@lists.febo.com

Hi Magnus,
Thanks for telling me to check everything.
I turned out to be leakage from another 10MHz source with an output
coax running close to the coax from the reference.
To ensure the PLL worked correctly I first validated in open loop mode
the control range of the DAC, which turned out to have 10MHz correctly
centered in the middle of the DAC range.
Then I enabled the PLL and after about 10 minutes the PLL was stable
and within 1e-10 of a Rubidium reference.
The 0.1s ADEV without PLL active is just below 0.8e-11. With PLL
active the 01. - 100 s ADEV is at about 3e-11 which is good enough for
me. The active PLL ADEV can be further improved by making the loop
time constant longer if needed (it is now at 100s)
Erik.

On 5-11-2023 0:35, Magnus Danielson via time-nuts wrote:

Erik,

That looks to me as the beating from being unable to lock. One
possible reason is that the oscillator drifted too far so it no
longer can really reach 10 MHz. You keep getting cycle-slips.

It might be better if you measured it in free-running state, so you
start the Mini-T up with no antenna connected, but measure the 10 MHz
output. Then you can observe the oscillators behaviour without the
PLL pulling it here and there.

The beat-frequency indicates the unmodulated oscillators frequency
offset. Usually should this provide a component for the PI loop to
pull it in, but if the steering range is not working, then that will
stop it.

You should be able to diagnoze more with Lady Heather or other
analysis of PLL state.

Cheers,
Magnus

On 2023-11-04 12:49, Erik Kaashoek via time-nuts wrote:

While evaluating a Trimble Mini-T GPSDO it appears the 10 MHz output
frequency,  with loop control enabled, is jumping up/down every
second or so in the order of 5e-10 (green trace in attached image)
resulting in a 0.1 s ADEV of 8e-11.
In holdover mode (no GPS input and no PPS output) the output
frequency has a much better short term stability  (purple trace in
attached image) resulting in a 0.1 s ADEV of 5e-12
The time constant was set to 100s and the damping factor to 1. The
Mini-T was powered by a linear bench supply.
The DOCXO used has reference has a 0.1s - 10 s ADEV of at most 5e-12.
Is there anything in the tuning that can reduce the jumps or is this
to be expected from a Mini-T?
Erik.

---

time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe send an email to time-nuts-leave@lists.febo.com

---

time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe send an email to time-nuts-leave@lists.febo.com

time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe send an email to time-nuts-leave@lists.febo.com

Hi Erik,

Really happy to hear you got hold of the issue, and happy that my
little suggestions got you in the right track quick enough.

With ADEV, do you mean the Lady Heather self-reports or that of a
separate reference measurement?

Cheers,
Magnus

On 2023-11-06 16:47, Erik Kaashoek via time-nuts wrote:

Hi Magnus,
Thanks for telling me to check everything.
I turned out to be leakage from another 10MHz source with an output
coax running close to the coax from the reference.
To ensure the PLL worked correctly I first validated in open loop
mode the control range of the DAC, which turned out to have 10MHz
correctly centered in the middle of the DAC range.
Then I enabled the PLL and after about 10 minutes the PLL was stable
and within 1e-10 of a Rubidium reference.
The 0.1s ADEV without PLL active is just below 0.8e-11. With PLL
active the 01. - 100 s ADEV is at about 3e-11 which is good enough
for me. The active PLL ADEV can be further improved by making the
loop time constant longer if needed (it is now at 100s)
Erik.

Hi Magnus
The ADEV was a separate measurement using the 10MHz output of the
Mini-T against my most stable DOCXO using a tinyPFA
Erik.

On 6-11-2023 20:12, Magnus Danielson via time-nuts wrote:

Hi Erik,

Really happy to hear you got hold of the issue, and happy that my
little suggestions got you in the right track quick enough.

With ADEV, do you mean the Lady Heather self-reports or that of a
separate reference measurement?

Cheers,
Magnus

On 2023-11-06 16:47, Erik Kaashoek via time-nuts wrote:

Hi Magnus,
Thanks for telling me to check everything.
I turned out to be leakage from another 10MHz source with an output
coax running close to the coax from the reference.
To ensure the PLL worked correctly I first validated in open loop
mode the control range of the DAC, which turned out to have 10MHz
correctly centered in the middle of the DAC range.
Then I enabled the PLL and after about 10 minutes the PLL was stable
and within 1e-10 of a Rubidium reference.
The 0.1s ADEV without PLL active is just below 0.8e-11. With PLL
active the 01. - 100 s ADEV is at about 3e-11 which is good enough
for me. The active PLL ADEV can be further improved by making the
loop time constant longer if needed (it is now at 100s)
Erik.

time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe send an email to time-nuts-leave@lists.febo.com