Ref out on Keysight 53230A
Bob Camp said:
The other rabbit hole involves the 5323x counters in general. Their ref out
ports have issues. One could be a bit more descriptive, but this is a public
list.
What did they do wrong? There must be an educational story behind that.
--
These are my opinions. I hate spam.
Hi
About as far as I can go on this one:
Early on TVB got ahold of an example of one of these. He did some testing and was
not real excited by the results. He posted the info here on the list and most of us avoided
the counters as a result.
Eventually ( being the insane person that you all know me to be ) I gave one a try. The
YouTube video roped me in :). What can I say? marketing works ….
I dug into this and that. I duplicated (and improved on) the data in the video. The big issue
seemed to be the Ref out. For me that was OK since I’m not going to daisy chain the units.
They all will run on an external standard of some sort. Yes’ it’s a shame you can’t use that
OCXO in the counter ( if you got that option) for anything. The caution from TVB is spot
on in this respect.
Best guess from looking at what’s going on:
When they get done with a series of readings, the power demand drops abruptly.
( = whatever it is does correlate with measurements ending). That load dump on
the power supply / ground creates a bump somehow ( there’s none obvious, but
I’d bet there’s one there …). That dump puts a phase hit on the ref out. ( = Watch
then ref out and you can see it has issues relative to ref in.)
So not very scientific. I dug in a bit, but didn’t find anything obvious. Since it was
not a big deal for me, I stopped digging.
Bob
On Jan 19, 2023, at 2:10 PM, Hal Murray via time-nuts time-nuts@lists.febo.com wrote:
Bob Camp said:
The other rabbit hole involves the 5323x counters in general. Their ref out
ports have issues. One could be a bit more descriptive, but this is a public
list.
What did they do wrong? There must be an educational story behind that.
--
These are my opinions. I hate spam.
time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe send an email to time-nuts-leave@lists.febo.com
I also have a 53230A. Maybe comparing the freq out to the source of a good
short and long term reference with a couple SRS620 would be worth doing?
Bill
On Thu, Jan 19, 2023, 5:08 PM Bob Camp via time-nuts <
time-nuts@lists.febo.com> wrote:
Hi
About as far as I can go on this one:
Early on TVB got ahold of an example of one of these. He did some testing
and was
not real excited by the results. He posted the info here on the list and
most of us avoided
the counters as a result.
Eventually ( being the insane person that you all know me to be ) I gave
one a try. The
YouTube video roped me in :). What can I say? marketing works ….
I dug into this and that. I duplicated (and improved on) the data in the
video. The big issue
seemed to be the Ref out. For me that was OK since I’m not going to daisy
chain the units.
They all will run on an external standard of some sort. Yes’ it’s a shame
you can’t use that
OCXO in the counter ( if you got that option) for anything. The caution
from TVB is spot
on in this respect.
Best guess from looking at what’s going on:
When they get done with a series of readings, the power demand drops
abruptly.
( = whatever it is does correlate with measurements ending). That load
dump on
the power supply / ground creates a bump somehow ( there’s none obvious,
but
I’d bet there’s one there …). That dump puts a phase hit on the ref out. (
= Watch
then ref out and you can see it has issues relative to ref in.)
So not very scientific. I dug in a bit, but didn’t find anything obvious.
Since it was
not a big deal for me, I stopped digging.
Bob
On Jan 19, 2023, at 2:10 PM, Hal Murray via time-nuts <
time-nuts@lists.febo.com> wrote:
Bob Camp said:
The other rabbit hole involves the 5323x counters in general. Their ref
out
ports have issues. One could be a bit more descriptive, but this is a
public
list.
What did they do wrong? There must be an educational story behind that.
--
These are my opinions. I hate spam.
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
If you put the input and output into a Phase Station, you can see glitches …..
Bob
On Jan 19, 2023, at 7:16 PM, Bill Notfaded notfaded1@gmail.com wrote:
I also have a 53230A. Maybe comparing the freq out to the source of a good short and long term reference with a couple SRS620 would be worth doing?
Bill
On Thu, Jan 19, 2023, 5:08 PM Bob Camp via time-nuts <time-nuts@lists.febo.com mailto:time-nuts@lists.febo.com> wrote:
Hi
About as far as I can go on this one:
Early on TVB got ahold of an example of one of these. He did some testing and was
not real excited by the results. He posted the info here on the list and most of us avoided
the counters as a result.
Eventually ( being the insane person that you all know me to be ) I gave one a try. The
YouTube video roped me in :). What can I say? marketing works ….
I dug into this and that. I duplicated (and improved on) the data in the video. The big issue
seemed to be the Ref out. For me that was OK since I’m not going to daisy chain the units.
They all will run on an external standard of some sort. Yes’ it’s a shame you can’t use that
OCXO in the counter ( if you got that option) for anything. The caution from TVB is spot
on in this respect.
Best guess from looking at what’s going on:
When they get done with a series of readings, the power demand drops abruptly.
( = whatever it is does correlate with measurements ending). That load dump on
the power supply / ground creates a bump somehow ( there’s none obvious, but
I’d bet there’s one there …). That dump puts a phase hit on the ref out. ( = Watch
then ref out and you can see it has issues relative to ref in.)
So not very scientific. I dug in a bit, but didn’t find anything obvious. Since it was
not a big deal for me, I stopped digging.
Bob
On Jan 19, 2023, at 2:10 PM, Hal Murray via time-nuts <time-nuts@lists.febo.com mailto:time-nuts@lists.febo.com> wrote:
Bob Camp said:
The other rabbit hole involves the 5323x counters in general. Their ref out
ports have issues. One could be a bit more descriptive, but this is a public
list.
What did they do wrong? There must be an educational story behind that.
--
These are my opinions. I hate spam.
time-nuts mailing list -- time-nuts@lists.febo.com mailto:time-nuts@lists.febo.com
To unsubscribe send an email to time-nuts-leave@lists.febo.com mailto:time-nuts-leave@lists.febo.com
time-nuts mailing list -- time-nuts@lists.febo.com mailto:time-nuts@lists.febo.com
To unsubscribe send an email to time-nuts-leave@lists.febo.com mailto:time-nuts-leave@lists.febo.com
On Friday, 20 January 2023 01:54:25 CET Bob Camp via time-nuts wrote:
If you put the input and output into a Phase Station, you can see glitches
Never used the external output for something important before, but as the
devices were standing on the table, I wanted a record for the future:
Setup: 10 MHz Reference to both "Ext Ref In" and Phase station, "Int Ref Out"
to DUT input of Phase station: Phase drifts rather slowly. As soon as I
connect some ~100MHz input to the 53230A, the "Int Ref Out" shows periodic
phase jumps. Turning the knob to increase the Gating time 0.1s->1s temporarily
disturbs the phase even more; the Glitch period seems unchanged though.
Still like this counter though...
Cheers,
Jürgen
Hi
Yup, that’s the sort of thing it does to the Ref Out.
As long as you don’t use the Ref Out, it’s a very nice counter.
Bob
On Jan 24, 2023, at 8:32 AM, Jürgen Appel via time-nuts time-nuts@lists.febo.com wrote:
On Friday, 20 January 2023 01:54:25 CET Bob Camp via time-nuts wrote:
If you put the input and output into a Phase Station, you can see glitches
Never used the external output for something important before, but as the
devices were standing on the table, I wanted a record for the future:
Setup: 10 MHz Reference to both "Ext Ref In" and Phase station, "Int Ref Out"
to DUT input of Phase station: Phase drifts rather slowly. As soon as I
connect some ~100MHz input to the 53230A, the "Int Ref Out" shows periodic
phase jumps. Turning the knob to increase the Gating time 0.1s->1s temporarily
disturbs the phase even more; the Glitch period seems unchanged though.
Still like this counter though...
Cheers,
Jürgen
<53230A-intRef.png>_______________________________________________
time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe send an email to time-nuts-leave@lists.febo.com
Jürgen
Is the period of the phase jumps of the ref out related to the frequency
difference of the free running internal ref and the external supplied ref?
If so, this could imply the internal ref couples into the pll locking to
the external ref and creates a periodically pull that causes the
internal pll to jump.
If this is the case then the measurements done with the 53230A should
also have such a phase jump when an external reference is supplied. Not
sure if such a small phase jump would be visible on the 53230A.
On one of my counters the jump is only visible when I use a rather short
(10 ms or lower) gate time but then it is very visible. With a gate time
of 1 second the jump is totally invisible
If the measurements do not have such a phase jump there is no harm done.
Erik.
On 24-1-2023 18:37, Bob Camp via time-nuts wrote:
Hi
Yup, that’s the sort of thing it does to the Ref Out.
As long as you don’t use the Ref Out, it’s a very nice counter.
Bob
On Jan 24, 2023, at 8:32 AM, Jürgen Appel via time-nuts time-nuts@lists.febo.com wrote:
On Friday, 20 January 2023 01:54:25 CET Bob Camp via time-nuts wrote:
If you put the input and output into a Phase Station, you can see glitches
Never used the external output for something important before, but as the
devices were standing on the table, I wanted a record for the future:
Setup: 10 MHz Reference to both "Ext Ref In" and Phase station, "Int Ref Out"
to DUT input of Phase station: Phase drifts rather slowly. As soon as I
connect some ~100MHz input to the 53230A, the "Int Ref Out" shows periodic
phase jumps. Turning the knob to increase the Gating time 0.1s->1s temporarily
disturbs the phase even more; the Glitch period seems unchanged though.
Still like this counter though...
Cheers,
Jürgen
Hi
The jump period does not seem to be related to the frequency offset of
the internal oscillator. If the counter is idle, the system locks up and runs
fine. The Ref Out is pretty clean (= they never are 100% clean on any
Ref Out).
When the problem happens, it is not subtle at all. It will mess up just about
any downstream device to some degree. If you dig into the pops / crud, they
look worse and worse as you zoom in the time scale. Something pretty
significant is going on.
The jumps appear to be related to the counter finishing a measurement
or group of measurements. They appear to happen when the counter is
(more or less) idling. As far as the internals of the counter are concerned,
the jumps have no impact.
The counter is very accurate and the jumps are quite big. This tends to
confirm that they must happen between measurements / readings. If
they happened during a reading, it seems unlikely that the accuracy
would not be impacted.
Bob
On Jan 25, 2023, at 1:44 AM, Erik Kaashoek via time-nuts time-nuts@lists.febo.com wrote:
Jürgen
Is the period of the phase jumps of the ref out related to the frequency difference of the free running internal ref and the external supplied ref?
If so, this could imply the internal ref couples into the pll locking to the external ref and creates a periodically pull that causes the internal pll to jump.
If this is the case then the measurements done with the 53230A should also have such a phase jump when an external reference is supplied. Not sure if such a small phase jump would be visible on the 53230A.
On one of my counters the jump is only visible when I use a rather short (10 ms or lower) gate time but then it is very visible. With a gate time of 1 second the jump is totally invisible
If the measurements do not have such a phase jump there is no harm done.
Erik.
On 24-1-2023 18:37, Bob Camp via time-nuts wrote:
Hi
Yup, that’s the sort of thing it does to the Ref Out.
As long as you don’t use the Ref Out, it’s a very nice counter.
Bob
On Jan 24, 2023, at 8:32 AM, Jürgen Appel via time-nuts time-nuts@lists.febo.com wrote:
On Friday, 20 January 2023 01:54:25 CET Bob Camp via time-nuts wrote:
If you put the input and output into a Phase Station, you can see glitches
Never used the external output for something important before, but as the
devices were standing on the table, I wanted a record for the future:
Setup: 10 MHz Reference to both "Ext Ref In" and Phase station, "Int Ref Out"
to DUT input of Phase station: Phase drifts rather slowly. As soon as I
connect some ~100MHz input to the 53230A, the "Int Ref Out" shows periodic
phase jumps. Turning the knob to increase the Gating time 0.1s->1s temporarily
disturbs the phase even more; the Glitch period seems unchanged though.
Still like this counter though...
Cheers,
Jürgen
time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe send an email to time-nuts-leave@lists.febo.com
From Bob-
"If they happened during a reading, it seems unlikely that the accuracy
would not be impacted."
This is true of ref out... If you're using it. The normal function of the
counter outputting over GPIB or USB using an external reference isn't
affected. It's only if you are using the ref out for something. I
personally never use the ref out on this counter. I suppose just daisy
chaining counters may be some use case but, speaking for myself, I think
most of us would use a better reference split to multiple counters ref in
instead. At least that's how I've usually done it. I really like this
counter and generally would recommend it especially after upgrading the
firmware to current.
Jealous of the phase station btw...
Best Regards,
Bill
On Wed, Jan 25, 2023, 9:40 AM Bob Camp via time-nuts <
time-nuts@lists.febo.com> wrote:
Hi
The jump period does not seem to be related to the frequency offset of
the internal oscillator. If the counter is idle, the system locks up and
runs
fine. The Ref Out is pretty clean (= they never are 100% clean on any
Ref Out).
When the problem happens, it is not subtle at all. It will mess up just
about
any downstream device to some degree. If you dig into the pops / crud, they
look worse and worse as you zoom in the time scale. Something pretty
significant is going on.
The jumps appear to be related to the counter finishing a measurement
or group of measurements. They appear to happen when the counter is
(more or less) idling. As far as the internals of the counter are
concerned,
the jumps have no impact.
The counter is very accurate and the jumps are quite big. This tends to
confirm that they must happen between measurements / readings. If
they happened during a reading, it seems unlikely that the accuracy
would not be impacted.
Bob
On Jan 25, 2023, at 1:44 AM, Erik Kaashoek via time-nuts <
time-nuts@lists.febo.com> wrote:
Jürgen
Is the period of the phase jumps of the ref out related to the frequency
difference of the free running internal ref and the external supplied ref?
If so, this could imply the internal ref couples into the pll locking to
the external ref and creates a periodically pull that causes the internal
pll to jump.
If this is the case then the measurements done with the 53230A should
also have such a phase jump when an external reference is supplied. Not
sure if such a small phase jump would be visible on the 53230A.
On one of my counters the jump is only visible when I use a rather short
(10 ms or lower) gate time but then it is very visible. With a gate time of
1 second the jump is totally invisible
If the measurements do not have such a phase jump there is no harm done.
Erik.
On 24-1-2023 18:37, Bob Camp via time-nuts wrote:
Hi
Yup, that’s the sort of thing it does to the Ref Out.
As long as you don’t use the Ref Out, it’s a very nice counter.
Bob
On Jan 24, 2023, at 8:32 AM, Jürgen Appel via time-nuts <
time-nuts@lists.febo.com> wrote:
On Friday, 20 January 2023 01:54:25 CET Bob Camp via time-nuts wrote:
If you put the input and output into a Phase Station, you can see
glitches
Never used the external output for something important before, but as
the
devices were standing on the table, I wanted a record for the future:
Setup: 10 MHz Reference to both "Ext Ref In" and Phase station, "Int
Ref Out"
to DUT input of Phase station: Phase drifts rather slowly. As soon as I
connect some ~100MHz input to the 53230A, the "Int Ref Out" shows
periodic
phase jumps. Turning the knob to increase the Gating time 0.1s->1s
temporarily
disturbs the phase even more; the Glitch period seems unchanged though.
Still like this counter though...
Cheers,
Jürgen
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
I very much doubt that whatever it is only impacts the ref out. That might be
the case, but it would take a deep dive to prove it. All we really know is that
something in the chain from the OCXO to the REF OUT takes a hit. It is unclear
just where that hit is or how it occurs.
I took a quick look at power supplies. They don’t seem to have any massive
spikes on them. Even there, I would not. rule them out, it was a quick look.
Since there are no schematics (that I know of) for the device, tracing anything
is going to be exciting. Even with schematics, you would get to “goes into unknown
IC U302 and then to IC U689 …. “. Multilayer board and maybe IC’s with
connection on the underside. Could be a lot of fun.
Don’t use the REF OUT and just move on ….
Bob
On Jan 26, 2023, at 12:42 AM, Bill Notfaded notfaded1@gmail.com wrote:
From Bob-
"If they happened during a reading, it seems unlikely that the accuracy would not be impacted."
This is true of ref out... If you're using it. The normal function of the counter outputting over GPIB or USB using an external reference isn't affected. It's only if you are using the ref out for something. I personally never use the ref out on this counter. I suppose just daisy chaining counters may be some use case but, speaking for myself, I think most of us would use a better reference split to multiple counters ref in instead. At least that's how I've usually done it. I really like this counter and generally would recommend it especially after upgrading the firmware to current.
Jealous of the phase station btw...
Best Regards,
Bill
On Wed, Jan 25, 2023, 9:40 AM Bob Camp via time-nuts <time-nuts@lists.febo.com mailto:time-nuts@lists.febo.com> wrote:
Hi
The jump period does not seem to be related to the frequency offset of
the internal oscillator. If the counter is idle, the system locks up and runs
fine. The Ref Out is pretty clean (= they never are 100% clean on any
Ref Out).
When the problem happens, it is not subtle at all. It will mess up just about
any downstream device to some degree. If you dig into the pops / crud, they
look worse and worse as you zoom in the time scale. Something pretty
significant is going on.
The jumps appear to be related to the counter finishing a measurement
or group of measurements. They appear to happen when the counter is
(more or less) idling. As far as the internals of the counter are concerned,
the jumps have no impact.
The counter is very accurate and the jumps are quite big. This tends to
confirm that they must happen between measurements / readings. If
they happened during a reading, it seems unlikely that the accuracy
would not be impacted.
Bob
On Jan 25, 2023, at 1:44 AM, Erik Kaashoek via time-nuts <time-nuts@lists.febo.com mailto:time-nuts@lists.febo.com> wrote:
Jürgen
Is the period of the phase jumps of the ref out related to the frequency difference of the free running internal ref and the external supplied ref?
If so, this could imply the internal ref couples into the pll locking to the external ref and creates a periodically pull that causes the internal pll to jump.
If this is the case then the measurements done with the 53230A should also have such a phase jump when an external reference is supplied. Not sure if such a small phase jump would be visible on the 53230A.
On one of my counters the jump is only visible when I use a rather short (10 ms or lower) gate time but then it is very visible. With a gate time of 1 second the jump is totally invisible
If the measurements do not have such a phase jump there is no harm done.
Erik.
On 24-1-2023 18:37, Bob Camp via time-nuts wrote:
Hi
Yup, that’s the sort of thing it does to the Ref Out.
As long as you don’t use the Ref Out, it’s a very nice counter.
Bob
On Jan 24, 2023, at 8:32 AM, Jürgen Appel via time-nuts <time-nuts@lists.febo.com mailto:time-nuts@lists.febo.com> wrote:
On Friday, 20 January 2023 01:54:25 CET Bob Camp via time-nuts wrote:
If you put the input and output into a Phase Station, you can see glitches
Never used the external output for something important before, but as the
devices were standing on the table, I wanted a record for the future:
Setup: 10 MHz Reference to both "Ext Ref In" and Phase station, "Int Ref Out"
to DUT input of Phase station: Phase drifts rather slowly. As soon as I
connect some ~100MHz input to the 53230A, the "Int Ref Out" shows periodic
phase jumps. Turning the knob to increase the Gating time 0.1s->1s temporarily
disturbs the phase even more; the Glitch period seems unchanged though.
Still like this counter though...
Cheers,
Jürgen
time-nuts mailing list -- time-nuts@lists.febo.com mailto:time-nuts@lists.febo.com
To unsubscribe send an email to time-nuts-leave@lists.febo.com mailto:time-nuts-leave@lists.febo.com
time-nuts mailing list -- time-nuts@lists.febo.com mailto:time-nuts@lists.febo.com
To unsubscribe send an email to time-nuts-leave@lists.febo.com mailto:time-nuts-leave@lists.febo.com