Empathy List Archives

VE

Volker Esper

Fri, Nov 2, 2012 3:32 PM

Dear fellows,

I'm searching for a counter that allows me to make automated measure
series. Of course, Allan Deviation is an important thing to determine
when experimenting with or developing GPSDOs (what I intend). So it
seemed to be a nice thing to measure phase deviation with a counter (HP
53132A or SR-620 or so).

When experimenting with that stuff I discovered, that all my equipment
would not be adequate to determine Allan Deviations beyond 10e-11 (or
so). The question is, can any counter deliver that precision / low noise
to make such measurements possible? Or does it have to be a more
sophisticated technique, such as cross spectrum analysis?

Thanks a lot in advance

Volker

(I guess you recognize, that I am just a newbie to that all)

Dear fellows, I'm searching for a counter that allows me to make automated measure series. Of course, Allan Deviation is an important thing to determine when experimenting with or developing GPSDOs (what I intend). So it seemed to be a nice thing to measure phase deviation with a counter (HP 53132A or SR-620 or so). When experimenting with that stuff I discovered, that all my equipment would not be adequate to determine Allan Deviations beyond 10e-11 (or so). The question is, can any counter deliver that precision / low noise to make such measurements possible? Or does it have to be a more sophisticated technique, such as cross spectrum analysis? Thanks a lot in advance Volker (I guess you recognize, that I am just a newbie to that all)

TV

Tom Van Baak

Fri, Nov 2, 2012 4:09 PM

Hi Volker,

Universal counters like 53132A or SR620 or CNT-91 do quite well for making measurements of GPSDO. Over minutes or hours they easily measure down to the 1e-12, 1e-13, and 1e-14 level. But yes, for short averaging times the noise of the counter is probably greater than the instability of your GPSDO so the measurements from the counter are not useful.

Similarly, for long averaging times beyond hours or days the instability of your frequency reference is probably greater than the instability of your GPSDO so the measurements are also not useful.

To make higher-resolution phase measurements use advanced commercial time interval instruments like Wavecrest DTS*, or Symmetricom TSC 5110/5120, or TimePod 5330A (http://www.miles.io/). On the cheap, you can also homebrew your own phase meter using several methods. For example, see:
http://tf.nist.gov/phase/Properties/main.htm
http://www.wriley.com/A%20Small%20DMTD%20System.pdf
http://www.ke5fx.com/tpll.htm

/tvb

----- Original Message -----
From: "Volker Esper" ailer2@t-online.de
To: "Discussion of precise time and frequency measurement" time-nuts@febo.com
Sent: Friday, November 02, 2012 8:32 AM
Subject: [time-nuts] A counter for phase measures

Dear fellows,

I'm searching for a counter that allows me to make automated measure
series. Of course, Allan Deviation is an important thing to determine
when experimenting with or developing GPSDOs (what I intend). So it
seemed to be a nice thing to measure phase deviation with a counter (HP
53132A or SR-620 or so).

When experimenting with that stuff I discovered, that all my equipment
would not be adequate to determine Allan Deviations beyond 10e-11 (or
so). The question is, can any counter deliver that precision / low noise
to make such measurements possible? Or does it have to be a more
sophisticated technique, such as cross spectrum analysis?

Thanks a lot in advance

Volker

(I guess you recognize, that I am just a newbie to that all)

Hi Volker, Universal counters like 53132A or SR620 or CNT-91 do quite well for making measurements of GPSDO. Over minutes or hours they easily measure down to the 1e-12, 1e-13, and 1e-14 level. But yes, for short averaging times the noise of the counter is probably greater than the instability of your GPSDO so the measurements from the counter are not useful. Similarly, for long averaging times beyond hours or days the instability of your frequency reference is probably greater than the instability of your GPSDO so the measurements are also not useful. To make higher-resolution phase measurements use advanced commercial time interval instruments like Wavecrest DTS*, or Symmetricom TSC 5110/5120, or TimePod 5330A (http://www.miles.io/). On the cheap, you can also homebrew your own phase meter using several methods. For example, see: http://tf.nist.gov/phase/Properties/main.htm http://www.wriley.com/A%20Small%20DMTD%20System.pdf http://www.ke5fx.com/tpll.htm /tvb ----- Original Message ----- From: "Volker Esper" <ailer2@t-online.de> To: "Discussion of precise time and frequency measurement" <time-nuts@febo.com> Sent: Friday, November 02, 2012 8:32 AM Subject: [time-nuts] A counter for phase measures Dear fellows, I'm searching for a counter that allows me to make automated measure series. Of course, Allan Deviation is an important thing to determine when experimenting with or developing GPSDOs (what I intend). So it seemed to be a nice thing to measure phase deviation with a counter (HP 53132A or SR-620 or so). When experimenting with that stuff I discovered, that all my equipment would not be adequate to determine Allan Deviations beyond 10e-11 (or so). The question is, can any counter deliver that precision / low noise to make such measurements possible? Or does it have to be a more sophisticated technique, such as cross spectrum analysis? Thanks a lot in advance Volker (I guess you recognize, that I am just a newbie to that all)

BC

Bob Camp

Fri, Nov 2, 2012 4:11 PM

Hi

If you want Adev of < 1.0 x 10^-11 at a tau of 1 second, then a counter
probably is not going to do it for you. To do a reasonable job on "time nut"
grade stuff you probably want something with a floor of < 1.0 x 10^-13 at a
tau of 1 second.

You have several choices:

Forget about 1 second and move out to 100 seconds. At that point your
53132's and the like will do ok. Their resolution goes up as the time gets
longer.
If you are measuring pps's go over to something like a Wavecrest.
If you are looking at 10 MHz either go to a DMTD or a TimePod.

In all cases, the gizmo it's self just takes data. The PC does the heavy
lifting of turning into pretty plots. You can get sort of plots on things
like a CNT-90 or a 5371, but the PC does a much better job.

Bob

-----Original Message-----
From: time-nuts-bounces@febo.com [mailto:time-nuts-bounces@febo.com] On
Behalf Of Volker Esper
Sent: Friday, November 02, 2012 11:32 AM
To: Discussion of precise time and frequency measurement
Subject: [time-nuts] A counter for phase measures

Dear fellows,

I'm searching for a counter that allows me to make automated measure
series. Of course, Allan Deviation is an important thing to determine
when experimenting with or developing GPSDOs (what I intend). So it
seemed to be a nice thing to measure phase deviation with a counter (HP
53132A or SR-620 or so).

When experimenting with that stuff I discovered, that all my equipment
would not be adequate to determine Allan Deviations beyond 10e-11 (or
so). The question is, can any counter deliver that precision / low noise
to make such measurements possible? Or does it have to be a more
sophisticated technique, such as cross spectrum analysis?

Thanks a lot in advance

Volker

(I guess you recognize, that I am just a newbie to that all)

Hi If you want Adev of < 1.0 x 10^-11 at a tau of 1 second, then a counter probably is not going to do it for you. To do a reasonable job on "time nut" grade stuff you probably want something with a floor of < 1.0 x 10^-13 at a tau of 1 second. You have several choices: 1) Forget about 1 second and move out to 100 seconds. At that point your 53132's and the like will do ok. Their resolution goes up as the time gets longer. 2) If you are measuring pps's go over to something like a Wavecrest. 3) If you are looking at 10 MHz either go to a DMTD or a TimePod. In all cases, the gizmo it's self just takes data. The PC does the heavy lifting of turning into pretty plots. You can get sort of plots on things like a CNT-90 or a 5371, but the PC does a much better job. Bob -----Original Message----- From: time-nuts-bounces@febo.com [mailto:time-nuts-bounces@febo.com] On Behalf Of Volker Esper Sent: Friday, November 02, 2012 11:32 AM To: Discussion of precise time and frequency measurement Subject: [time-nuts] A counter for phase measures Dear fellows, I'm searching for a counter that allows me to make automated measure series. Of course, Allan Deviation is an important thing to determine when experimenting with or developing GPSDOs (what I intend). So it seemed to be a nice thing to measure phase deviation with a counter (HP 53132A or SR-620 or so). When experimenting with that stuff I discovered, that all my equipment would not be adequate to determine Allan Deviations beyond 10e-11 (or so). The question is, can any counter deliver that precision / low noise to make such measurements possible? Or does it have to be a more sophisticated technique, such as cross spectrum analysis? Thanks a lot in advance Volker (I guess you recognize, that I am just a newbie to that all)

VE

Volker Esper

Sat, Nov 3, 2012 2:10 PM

Thank you for the interesting information. Now, the time has come to
look for an adequate counter - anyone who has experience with the HP
53132A and the SR620? If they both where at - say 1000 USD - what would
you prefer for the job of phase measurement? I've read about that
massive single shot capability of the SR, but - as being a newbie - is
there anything I overlook at this moment?

Thanks

Volker - DF9PL

Am 02.11.2012 16:32, schrieb Volker Esper:

Dear fellows,

I'm searching for a counter that allows me to make automated measure
series. Of course, Allan Deviation is an important thing to determine
when experimenting with or developing GPSDOs (what I intend). So it
seemed to be a nice thing to measure phase deviation with a counter
(HP 53132A or SR-620 or so).

When experimenting with that stuff I discovered, that all my equipment
would not be adequate to determine Allan Deviations beyond 10e-11 (or
so). The question is, can any counter deliver that precision / low
noise to make such measurements possible? Or does it have to be a more
sophisticated technique, such as cross spectrum analysis?

Thanks a lot in advance

Volker

(I guess you recognize, that I am just a newbie to that all)

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.

Thank you for the interesting information. Now, the time has come to look for an adequate counter - anyone who has experience with the HP 53132A and the SR620? If they both where at - say 1000 USD - what would you prefer for the job of phase measurement? I've read about that massive single shot capability of the SR, but - as being a newbie - is there anything I overlook at this moment? Thanks Volker - DF9PL Am 02.11.2012 16:32, schrieb Volker Esper: > Dear fellows, > > I'm searching for a counter that allows me to make automated measure > series. Of course, Allan Deviation is an important thing to determine > when experimenting with or developing GPSDOs (what I intend). So it > seemed to be a nice thing to measure phase deviation with a counter > (HP 53132A or SR-620 or so). > > When experimenting with that stuff I discovered, that all my equipment > would not be adequate to determine Allan Deviations beyond 10e-11 (or > so). The question is, can any counter deliver that precision / low > noise to make such measurements possible? Or does it have to be a more > sophisticated technique, such as cross spectrum analysis? > > Thanks a lot in advance > > Volker > > (I guess you recognize, that I am just a newbie to that all) > > > _______________________________________________ > 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.

MD

Magnus Danielson

Sat, Nov 3, 2012 3:28 PM

On 11/03/2012 03:10 PM, Volker Esper wrote:

Thank you for the interesting information. Now, the time has come to
look for an adequate counter - anyone who has experience with the HP
53132A and the SR620? If they both where at - say 1000 USD - what would
you prefer for the job of phase measurement? I've read about that
massive single shot capability of the SR, but - as being a newbie - is
there anything I overlook at this moment?

For short time-scales, single shot resolution and trigger jitter
dominates your measurement floor.

Single-shot resolution is the time resolution by which you make a single
measurement.

Trigger jitter is the noise at the trigger point. it's a combination of
thermal noise and the slew-rate at the trigger points. It is often that
trigger jitter is dominated by slew-rate, but there is also internal
sources of trigger jitter. The slope dependent trigger jitter follows
the formula:

t_jitter = v_noise / s_slew

t_jitter is the trigger jitter (s)
v_noise is the noise power (V)
s_slew is the slew rate (V/s)

When the time-span of a measurement is long, long-term stability comes
in as well as systematic drifts. Also, systematic noise such as hum also
becomes important.

To see how much you depend on slew-rate limitation, you can reduce the
amplitude, and as this reduces the slew-rate you can separate the
slew-rate dependent jitter from the intrinsic jitter of the input. It
also helps you to identify if you need to work on the slew-rate limit
rather than anything else.

So, it may not be the single-shot resolution which limits you, but a
combination of things.

I would recommend you to pick up a SR620. It has 4 ps single shot
resolution and about 25 ps jitter (but you can get less). That is
significantly better than the 53152A provides.

SR620 manual (one of many links):
http://ilrs.gsfc.nasa.gov/docs/timing/sr620_manual.pdf

Cheers,
Magnus

On 11/03/2012 03:10 PM, Volker Esper wrote: > > Thank you for the interesting information. Now, the time has come to > look for an adequate counter - anyone who has experience with the HP > 53132A and the SR620? If they both where at - say 1000 USD - what would > you prefer for the job of phase measurement? I've read about that > massive single shot capability of the SR, but - as being a newbie - is > there anything I overlook at this moment? For short time-scales, single shot resolution and trigger jitter dominates your measurement floor. Single-shot resolution is the time resolution by which you make a single measurement. Trigger jitter is the noise at the trigger point. it's a combination of thermal noise and the slew-rate at the trigger points. It is often that trigger jitter is dominated by slew-rate, but there is also internal sources of trigger jitter. The slope dependent trigger jitter follows the formula: t_jitter = v_noise / s_slew t_jitter is the trigger jitter (s) v_noise is the noise power (V) s_slew is the slew rate (V/s) When the time-span of a measurement is long, long-term stability comes in as well as systematic drifts. Also, systematic noise such as hum also becomes important. To see how much you depend on slew-rate limitation, you can reduce the amplitude, and as this reduces the slew-rate you can separate the slew-rate dependent jitter from the intrinsic jitter of the input. It also helps you to identify if you need to work on the slew-rate limit rather than anything else. So, it may not be the single-shot resolution which limits you, but a combination of things. I would recommend you to pick up a SR620. It has 4 ps single shot resolution and about 25 ps jitter (but you can get less). That is significantly better than the 53152A provides. SR620 manual (one of many links): http://ilrs.gsfc.nasa.gov/docs/timing/sr620_manual.pdf Cheers, Magnus

TK

Tom Knox

Sat, Nov 3, 2012 4:58 PM

Hi Magnus;
I hope this is not to far off thread. Has anyone in the group done real world measurements of single shot res, and jitter on the new Agilent and Tektronix/Pendulum counters compared to the SR620 and Agilent 53132A. I would imagine that counter designs would be an area that really benefits from ongoing advances in digital technology. I have also found that it is much easier to claim specs then meet specs.
Thanks;
Thomas Knox

Date: Sat, 3 Nov 2012 16:28:42 +0100
From: magnus@rubidium.dyndns.org
To: time-nuts@febo.com
Subject: Re: [time-nuts] A counter for phase measures

On 11/03/2012 03:10 PM, Volker Esper wrote:

Thank you for the interesting information. Now, the time has come to
look for an adequate counter - anyone who has experience with the HP
53132A and the SR620? If they both where at - say 1000 USD - what would
you prefer for the job of phase measurement? I've read about that
massive single shot capability of the SR, but - as being a newbie - is
there anything I overlook at this moment?

For short time-scales, single shot resolution and trigger jitter
dominates your measurement floor.

Single-shot resolution is the time resolution by which you make a single
measurement.

Trigger jitter is the noise at the trigger point. it's a combination of
thermal noise and the slew-rate at the trigger points. It is often that
trigger jitter is dominated by slew-rate, but there is also internal
sources of trigger jitter. The slope dependent trigger jitter follows
the formula:

t_jitter = v_noise / s_slew

t_jitter is the trigger jitter (s)
v_noise is the noise power (V)
s_slew is the slew rate (V/s)

When the time-span of a measurement is long, long-term stability comes
in as well as systematic drifts. Also, systematic noise such as hum also
becomes important.

To see how much you depend on slew-rate limitation, you can reduce the
amplitude, and as this reduces the slew-rate you can separate the
slew-rate dependent jitter from the intrinsic jitter of the input. It
also helps you to identify if you need to work on the slew-rate limit
rather than anything else.

So, it may not be the single-shot resolution which limits you, but a
combination of things.

I would recommend you to pick up a SR620. It has 4 ps single shot
resolution and about 25 ps jitter (but you can get less). That is
significantly better than the 53152A provides.

SR620 manual (one of many links):
http://ilrs.gsfc.nasa.gov/docs/timing/sr620_manual.pdf

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.

Hi Magnus; I hope this is not to far off thread. Has anyone in the group done real world measurements of single shot res, and jitter on the new Agilent and Tektronix/Pendulum counters compared to the SR620 and Agilent 53132A. I would imagine that counter designs would be an area that really benefits from ongoing advances in digital technology. I have also found that it is much easier to claim specs then meet specs. Thanks; Thomas Knox > Date: Sat, 3 Nov 2012 16:28:42 +0100 > From: magnus@rubidium.dyndns.org > To: time-nuts@febo.com > Subject: Re: [time-nuts] A counter for phase measures > > On 11/03/2012 03:10 PM, Volker Esper wrote: > > > > Thank you for the interesting information. Now, the time has come to > > look for an adequate counter - anyone who has experience with the HP > > 53132A and the SR620? If they both where at - say 1000 USD - what would > > you prefer for the job of phase measurement? I've read about that > > massive single shot capability of the SR, but - as being a newbie - is > > there anything I overlook at this moment? > > For short time-scales, single shot resolution and trigger jitter > dominates your measurement floor. > > Single-shot resolution is the time resolution by which you make a single > measurement. > > Trigger jitter is the noise at the trigger point. it's a combination of > thermal noise and the slew-rate at the trigger points. It is often that > trigger jitter is dominated by slew-rate, but there is also internal > sources of trigger jitter. The slope dependent trigger jitter follows > the formula: > > t_jitter = v_noise / s_slew > > t_jitter is the trigger jitter (s) > v_noise is the noise power (V) > s_slew is the slew rate (V/s) > > When the time-span of a measurement is long, long-term stability comes > in as well as systematic drifts. Also, systematic noise such as hum also > becomes important. > > To see how much you depend on slew-rate limitation, you can reduce the > amplitude, and as this reduces the slew-rate you can separate the > slew-rate dependent jitter from the intrinsic jitter of the input. It > also helps you to identify if you need to work on the slew-rate limit > rather than anything else. > > So, it may not be the single-shot resolution which limits you, but a > combination of things. > > I would recommend you to pick up a SR620. It has 4 ps single shot > resolution and about 25 ps jitter (but you can get less). That is > significantly better than the 53152A provides. > > SR620 manual (one of many links): > http://ilrs.gsfc.nasa.gov/docs/timing/sr620_manual.pdf > > 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.

BC

Bob Camp

Sat, Nov 3, 2012 4:59 PM

Hi

Both counters have their weak points long term. On the balance I think the 620 should last longer. The 620 is the higher resolution of the two. The 620 normally comes with a bit better reference. Both are supported by various Time Nut software packages. Both do GPIB and serial i/o. The 620 is a bit more controllable over serial. The 53132 takes up less space on your bench.

For the same price - go for the 620. For the usual 53132 is $1300 and the 620 is $2400, not so clear.

Bob

On Nov 3, 2012, at 10:10 AM, Volker Esper ailer2@t-online.de wrote:

Thank you for the interesting information. Now, the time has come to look for an adequate counter - anyone who has experience with the HP 53132A and the SR620? If they both where at - say 1000 USD - what would you prefer for the job of phase measurement? I've read about that massive single shot capability of the SR, but - as being a newbie - is there anything I overlook at this moment?

Thanks

Volker - DF9PL

Am 02.11.2012 16:32, schrieb Volker Esper:

Dear fellows,

I'm searching for a counter that allows me to make automated measure series. Of course, Allan Deviation is an important thing to determine when experimenting with or developing GPSDOs (what I intend). So it seemed to be a nice thing to measure phase deviation with a counter (HP 53132A or SR-620 or so).

When experimenting with that stuff I discovered, that all my equipment would not be adequate to determine Allan Deviations beyond 10e-11 (or so). The question is, can any counter deliver that precision / low noise to make such measurements possible? Or does it have to be a more sophisticated technique, such as cross spectrum analysis?

Thanks a lot in advance

Volker

(I guess you recognize, that I am just a newbie to that all)

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.

Hi Both counters have their weak points long term. On the balance I think the 620 should last longer. The 620 is the higher resolution of the two. The 620 normally comes with a bit better reference. Both are supported by various Time Nut software packages. Both do GPIB and serial i/o. The 620 is a bit more controllable over serial. The 53132 takes up less space on your bench. For the same price - go for the 620. For the usual 53132 is $1300 and the 620 is $2400, not so clear. Bob On Nov 3, 2012, at 10:10 AM, Volker Esper <ailer2@t-online.de> wrote: > > Thank you for the interesting information. Now, the time has come to look for an adequate counter - anyone who has experience with the HP 53132A and the SR620? If they both where at - say 1000 USD - what would you prefer for the job of phase measurement? I've read about that massive single shot capability of the SR, but - as being a newbie - is there anything I overlook at this moment? > > Thanks > > Volker - DF9PL > > > Am 02.11.2012 16:32, schrieb Volker Esper: >> Dear fellows, >> >> I'm searching for a counter that allows me to make automated measure series. Of course, Allan Deviation is an important thing to determine when experimenting with or developing GPSDOs (what I intend). So it seemed to be a nice thing to measure phase deviation with a counter (HP 53132A or SR-620 or so). >> >> When experimenting with that stuff I discovered, that all my equipment would not be adequate to determine Allan Deviations beyond 10e-11 (or so). The question is, can any counter deliver that precision / low noise to make such measurements possible? Or does it have to be a more sophisticated technique, such as cross spectrum analysis? >> >> Thanks a lot in advance >> >> Volker >> >> (I guess you recognize, that I am just a newbie to that all) >> >> >> _______________________________________________ >> 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.

BC

Bob Camp

Sat, Nov 3, 2012 5:04 PM

Hi

The 620 is still pretty good (when tuned up). It certainly beats the Pendulum's on a single shot basis. The 53230 is spec'd to be as good as the 620. I suspect it meets or exceeds it's stated specs.

Bob

On Nov 3, 2012, at 12:58 PM, Tom Knox actast@hotmail.com wrote:

Hi Magnus;
I hope this is not to far off thread. Has anyone in the group done real world measurements of single shot res, and jitter on the new Agilent and Tektronix/Pendulum counters compared to the SR620 and Agilent 53132A. I would imagine that counter designs would be an area that really benefits from ongoing advances in digital technology. I have also found that it is much easier to claim specs then meet specs.
Thanks;
Thomas Knox

Date: Sat, 3 Nov 2012 16:28:42 +0100
From: magnus@rubidium.dyndns.org
To: time-nuts@febo.com
Subject: Re: [time-nuts] A counter for phase measures

On 11/03/2012 03:10 PM, Volker Esper wrote:

Thank you for the interesting information. Now, the time has come to
look for an adequate counter - anyone who has experience with the HP
53132A and the SR620? If they both where at - say 1000 USD - what would
you prefer for the job of phase measurement? I've read about that
massive single shot capability of the SR, but - as being a newbie - is
there anything I overlook at this moment?

For short time-scales, single shot resolution and trigger jitter
dominates your measurement floor.

Single-shot resolution is the time resolution by which you make a single
measurement.

Trigger jitter is the noise at the trigger point. it's a combination of
thermal noise and the slew-rate at the trigger points. It is often that
trigger jitter is dominated by slew-rate, but there is also internal
sources of trigger jitter. The slope dependent trigger jitter follows
the formula:

t_jitter = v_noise / s_slew

t_jitter is the trigger jitter (s)
v_noise is the noise power (V)
s_slew is the slew rate (V/s)

When the time-span of a measurement is long, long-term stability comes
in as well as systematic drifts. Also, systematic noise such as hum also
becomes important.

To see how much you depend on slew-rate limitation, you can reduce the
amplitude, and as this reduces the slew-rate you can separate the
slew-rate dependent jitter from the intrinsic jitter of the input. It
also helps you to identify if you need to work on the slew-rate limit
rather than anything else.

So, it may not be the single-shot resolution which limits you, but a
combination of things.

I would recommend you to pick up a SR620. It has 4 ps single shot
resolution and about 25 ps jitter (but you can get less). That is
significantly better than the 53152A provides.

SR620 manual (one of many links):
http://ilrs.gsfc.nasa.gov/docs/timing/sr620_manual.pdf

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.

Hi The 620 is still pretty good (when tuned up). It certainly beats the Pendulum's on a single shot basis. The 53230 is spec'd to be as good as the 620. I suspect it meets or exceeds it's stated specs. Bob On Nov 3, 2012, at 12:58 PM, Tom Knox <actast@hotmail.com> wrote: > > Hi Magnus; > I hope this is not to far off thread. Has anyone in the group done real world measurements of single shot res, and jitter on the new Agilent and Tektronix/Pendulum counters compared to the SR620 and Agilent 53132A. I would imagine that counter designs would be an area that really benefits from ongoing advances in digital technology. I have also found that it is much easier to claim specs then meet specs. > Thanks; > Thomas Knox > > > >> Date: Sat, 3 Nov 2012 16:28:42 +0100 >> From: magnus@rubidium.dyndns.org >> To: time-nuts@febo.com >> Subject: Re: [time-nuts] A counter for phase measures >> >> On 11/03/2012 03:10 PM, Volker Esper wrote: >>> >>> Thank you for the interesting information. Now, the time has come to >>> look for an adequate counter - anyone who has experience with the HP >>> 53132A and the SR620? If they both where at - say 1000 USD - what would >>> you prefer for the job of phase measurement? I've read about that >>> massive single shot capability of the SR, but - as being a newbie - is >>> there anything I overlook at this moment? >> >> For short time-scales, single shot resolution and trigger jitter >> dominates your measurement floor. >> >> Single-shot resolution is the time resolution by which you make a single >> measurement. >> >> Trigger jitter is the noise at the trigger point. it's a combination of >> thermal noise and the slew-rate at the trigger points. It is often that >> trigger jitter is dominated by slew-rate, but there is also internal >> sources of trigger jitter. The slope dependent trigger jitter follows >> the formula: >> >> t_jitter = v_noise / s_slew >> >> t_jitter is the trigger jitter (s) >> v_noise is the noise power (V) >> s_slew is the slew rate (V/s) >> >> When the time-span of a measurement is long, long-term stability comes >> in as well as systematic drifts. Also, systematic noise such as hum also >> becomes important. >> >> To see how much you depend on slew-rate limitation, you can reduce the >> amplitude, and as this reduces the slew-rate you can separate the >> slew-rate dependent jitter from the intrinsic jitter of the input. It >> also helps you to identify if you need to work on the slew-rate limit >> rather than anything else. >> >> So, it may not be the single-shot resolution which limits you, but a >> combination of things. >> >> I would recommend you to pick up a SR620. It has 4 ps single shot >> resolution and about 25 ps jitter (but you can get less). That is >> significantly better than the 53152A provides. >> >> SR620 manual (one of many links): >> http://ilrs.gsfc.nasa.gov/docs/timing/sr620_manual.pdf >> >> 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. > > _______________________________________________ > 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.

MD

Magnus Danielson

Sat, Nov 3, 2012 5:20 PM

Hi Tom,

On 11/03/2012 05:58 PM, Tom Knox wrote:

Hi Magnus;
I hope this is not to far off thread. Has anyone in the group done
real world measurements of single shot res, and jitter on the new
Agilent and Tektronix/Pendulum counters compared to the SR620 and
Agilent 53132A. I would imagine that counter designs would be an
area that really benefits from ongoing advances in digital technology.
I have also found that it is much easier to claim specs then meet specs.

I haven't tested the new 53230 counter, and not the CNT-91/PM-6691
counter either, but much of the others.

It should not be too hard to test this. I have made some tests with the
aggregate of single-shot and trigger noise on a sine signal, i.e. 5 MHz
out of a BVA. It gave the expected separation as a linear slope on the
ADEV. On the other hand just tossing a sine into counters isn't
necessary the most fair comparison, so in that sense it just gave a
rough image.

This thread have again had me consider reviving the testing aspect, and
I decided to get a 53132A counter finally, now that prices have gone down.

Cheers,
Magnus

Hi Tom, On 11/03/2012 05:58 PM, Tom Knox wrote: > > Hi Magnus; > I hope this is not to far off thread. Has anyone in the group done > real world measurements of single shot res, and jitter on the new > Agilent and Tektronix/Pendulum counters compared to the SR620 and > Agilent 53132A. I would imagine that counter designs would be an > area that really benefits from ongoing advances in digital technology. > I have also found that it is much easier to claim specs then meet specs. I haven't tested the new 53230 counter, and not the CNT-91/PM-6691 counter either, but much of the others. It should not be too hard to test this. I have made some tests with the aggregate of single-shot and trigger noise on a sine signal, i.e. 5 MHz out of a BVA. It gave the expected separation as a linear slope on the ADEV. On the other hand just tossing a sine into counters isn't necessary the most fair comparison, so in that sense it just gave a rough image. This thread have again had me consider reviving the testing aspect, and I decided to get a 53132A counter finally, now that prices have gone down. Cheers, Magnus

VE

Volker Esper

Sat, Nov 3, 2012 7:42 PM

Hi Bob

I didn't expect something to be tunable in the counter (except for the
oscillator) - what is it that has to be calibrated?

Thanks

Volker

Am 03.11.2012 18:04, schrieb Bob Camp:

Hi

The 620 is still pretty good (when tuned up). It certainly beats the Pendulum's on a single shot basis. The 53230 is spec'd to be as good as the 620. I suspect it meets or exceeds it's stated specs.

Bob

On Nov 3, 2012, at 12:58 PM, Tom Knoxactast@hotmail.com wrote:

Hi Magnus;
I hope this is not to far off thread. Has anyone in the group done real world measurements of single shot res, and jitter on the new Agilent and Tektronix/Pendulum counters compared to the SR620 and Agilent 53132A. I would imagine that counter designs would be an area that really benefits from ongoing advances in digital technology. I have also found that it is much easier to claim specs then meet specs.
Thanks;
Thomas Knox

Date: Sat, 3 Nov 2012 16:28:42 +0100
From: magnus@rubidium.dyndns.org
To: time-nuts@febo.com
Subject: Re: [time-nuts] A counter for phase measures

On 11/03/2012 03:10 PM, Volker Esper wrote:

Thank you for the interesting information. Now, the time has come to
look for an adequate counter - anyone who has experience with the HP
53132A and the SR620? If they both where at - say 1000 USD - what would
you prefer for the job of phase measurement? I've read about that
massive single shot capability of the SR, but - as being a newbie - is
there anything I overlook at this moment?

For short time-scales, single shot resolution and trigger jitter
dominates your measurement floor.

Single-shot resolution is the time resolution by which you make a single
measurement.

Trigger jitter is the noise at the trigger point. it's a combination of
thermal noise and the slew-rate at the trigger points. It is often that
trigger jitter is dominated by slew-rate, but there is also internal
sources of trigger jitter. The slope dependent trigger jitter follows
the formula:

t_jitter = v_noise / s_slew

t_jitter is the trigger jitter (s)
v_noise is the noise power (V)
s_slew is the slew rate (V/s)

When the time-span of a measurement is long, long-term stability comes
in as well as systematic drifts. Also, systematic noise such as hum also
becomes important.

To see how much you depend on slew-rate limitation, you can reduce the
amplitude, and as this reduces the slew-rate you can separate the
slew-rate dependent jitter from the intrinsic jitter of the input. It
also helps you to identify if you need to work on the slew-rate limit
rather than anything else.

So, it may not be the single-shot resolution which limits you, but a
combination of things.

I would recommend you to pick up a SR620. It has 4 ps single shot
resolution and about 25 ps jitter (but you can get less). That is
significantly better than the 53152A provides.

SR620 manual (one of many links):
http://ilrs.gsfc.nasa.gov/docs/timing/sr620_manual.pdf

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.

Hi Bob I didn't expect something to be tunable in the counter (except for the oscillator) - what is it that has to be calibrated? Thanks Volker Am 03.11.2012 18:04, schrieb Bob Camp: > Hi > > The 620 is still pretty good (when tuned up). It certainly beats the Pendulum's on a single shot basis. The 53230 is spec'd to be as good as the 620. I suspect it meets or exceeds it's stated specs. > > Bob > > On Nov 3, 2012, at 12:58 PM, Tom Knox<actast@hotmail.com> wrote: > > >> Hi Magnus; >> I hope this is not to far off thread. Has anyone in the group done real world measurements of single shot res, and jitter on the new Agilent and Tektronix/Pendulum counters compared to the SR620 and Agilent 53132A. I would imagine that counter designs would be an area that really benefits from ongoing advances in digital technology. I have also found that it is much easier to claim specs then meet specs. >> Thanks; >> Thomas Knox >> >> >> >> >>> Date: Sat, 3 Nov 2012 16:28:42 +0100 >>> From: magnus@rubidium.dyndns.org >>> To: time-nuts@febo.com >>> Subject: Re: [time-nuts] A counter for phase measures >>> >>> On 11/03/2012 03:10 PM, Volker Esper wrote: >>> >>>> Thank you for the interesting information. Now, the time has come to >>>> look for an adequate counter - anyone who has experience with the HP >>>> 53132A and the SR620? If they both where at - say 1000 USD - what would >>>> you prefer for the job of phase measurement? I've read about that >>>> massive single shot capability of the SR, but - as being a newbie - is >>>> there anything I overlook at this moment? >>>> >>> For short time-scales, single shot resolution and trigger jitter >>> dominates your measurement floor. >>> >>> Single-shot resolution is the time resolution by which you make a single >>> measurement. >>> >>> Trigger jitter is the noise at the trigger point. it's a combination of >>> thermal noise and the slew-rate at the trigger points. It is often that >>> trigger jitter is dominated by slew-rate, but there is also internal >>> sources of trigger jitter. The slope dependent trigger jitter follows >>> the formula: >>> >>> t_jitter = v_noise / s_slew >>> >>> t_jitter is the trigger jitter (s) >>> v_noise is the noise power (V) >>> s_slew is the slew rate (V/s) >>> >>> When the time-span of a measurement is long, long-term stability comes >>> in as well as systematic drifts. Also, systematic noise such as hum also >>> becomes important. >>> >>> To see how much you depend on slew-rate limitation, you can reduce the >>> amplitude, and as this reduces the slew-rate you can separate the >>> slew-rate dependent jitter from the intrinsic jitter of the input. It >>> also helps you to identify if you need to work on the slew-rate limit >>> rather than anything else. >>> >>> So, it may not be the single-shot resolution which limits you, but a >>> combination of things. >>> >>> I would recommend you to pick up a SR620. It has 4 ps single shot >>> resolution and about 25 ps jitter (but you can get less). That is >>> significantly better than the 53152A provides. >>> >>> SR620 manual (one of many links): >>> http://ilrs.gsfc.nasa.gov/docs/timing/sr620_manual.pdf >>> >>> 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. >>> >> >> _______________________________________________ >> 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. > > >

BC

Bob Camp

Sat, Nov 3, 2012 8:08 PM

Hi

There is a fairly elaborate alignment procedure for the 620. It's been reported in great detail here on the list. The counter definitely does better after you go through the full process.

Bob

On Nov 3, 2012, at 3:42 PM, Volker Esper ailer2@t-online.de wrote:

Hi Bob

I didn't expect something to be tunable in the counter (except for the oscillator) - what is it that has to be calibrated?

Thanks

Volker

Am 03.11.2012 18:04, schrieb Bob Camp:

Hi

The 620 is still pretty good (when tuned up). It certainly beats the Pendulum's on a single shot basis. The 53230 is spec'd to be as good as the 620. I suspect it meets or exceeds it's stated specs.

Bob

On Nov 3, 2012, at 12:58 PM, Tom Knoxactast@hotmail.com wrote:

Hi Magnus;
I hope this is not to far off thread. Has anyone in the group done real world measurements of single shot res, and jitter on the new Agilent and Tektronix/Pendulum counters compared to the SR620 and Agilent 53132A. I would imagine that counter designs would be an area that really benefits from ongoing advances in digital technology. I have also found that it is much easier to claim specs then meet specs.
Thanks;
Thomas Knox

Date: Sat, 3 Nov 2012 16:28:42 +0100
From: magnus@rubidium.dyndns.org
To: time-nuts@febo.com
Subject: Re: [time-nuts] A counter for phase measures

On 11/03/2012 03:10 PM, Volker Esper wrote:

Thank you for the interesting information. Now, the time has come to
look for an adequate counter - anyone who has experience with the HP
53132A and the SR620? If they both where at - say 1000 USD - what would
you prefer for the job of phase measurement? I've read about that
massive single shot capability of the SR, but - as being a newbie - is
there anything I overlook at this moment?

For short time-scales, single shot resolution and trigger jitter
dominates your measurement floor.

Single-shot resolution is the time resolution by which you make a single
measurement.

Trigger jitter is the noise at the trigger point. it's a combination of
thermal noise and the slew-rate at the trigger points. It is often that
trigger jitter is dominated by slew-rate, but there is also internal
sources of trigger jitter. The slope dependent trigger jitter follows
the formula:

t_jitter = v_noise / s_slew

t_jitter is the trigger jitter (s)
v_noise is the noise power (V)
s_slew is the slew rate (V/s)

When the time-span of a measurement is long, long-term stability comes
in as well as systematic drifts. Also, systematic noise such as hum also
becomes important.

To see how much you depend on slew-rate limitation, you can reduce the
amplitude, and as this reduces the slew-rate you can separate the
slew-rate dependent jitter from the intrinsic jitter of the input. It
also helps you to identify if you need to work on the slew-rate limit
rather than anything else.

So, it may not be the single-shot resolution which limits you, but a
combination of things.

I would recommend you to pick up a SR620. It has 4 ps single shot
resolution and about 25 ps jitter (but you can get less). That is
significantly better than the 53152A provides.

SR620 manual (one of many links):
http://ilrs.gsfc.nasa.gov/docs/timing/sr620_manual.pdf

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.

Hi There is a fairly elaborate alignment procedure for the 620. It's been reported in great detail here on the list. The counter definitely does better after you go through the full process. Bob On Nov 3, 2012, at 3:42 PM, Volker Esper <ailer2@t-online.de> wrote: > Hi Bob > > I didn't expect something to be tunable in the counter (except for the oscillator) - what is it that has to be calibrated? > > Thanks > > Volker > > > Am 03.11.2012 18:04, schrieb Bob Camp: >> Hi >> >> The 620 is still pretty good (when tuned up). It certainly beats the Pendulum's on a single shot basis. The 53230 is spec'd to be as good as the 620. I suspect it meets or exceeds it's stated specs. >> >> Bob >> >> On Nov 3, 2012, at 12:58 PM, Tom Knox<actast@hotmail.com> wrote: >> >> >>> Hi Magnus; >>> I hope this is not to far off thread. Has anyone in the group done real world measurements of single shot res, and jitter on the new Agilent and Tektronix/Pendulum counters compared to the SR620 and Agilent 53132A. I would imagine that counter designs would be an area that really benefits from ongoing advances in digital technology. I have also found that it is much easier to claim specs then meet specs. >>> Thanks; >>> Thomas Knox >>> >>> >>> >>> >>>> Date: Sat, 3 Nov 2012 16:28:42 +0100 >>>> From: magnus@rubidium.dyndns.org >>>> To: time-nuts@febo.com >>>> Subject: Re: [time-nuts] A counter for phase measures >>>> >>>> On 11/03/2012 03:10 PM, Volker Esper wrote: >>>> >>>>> Thank you for the interesting information. Now, the time has come to >>>>> look for an adequate counter - anyone who has experience with the HP >>>>> 53132A and the SR620? If they both where at - say 1000 USD - what would >>>>> you prefer for the job of phase measurement? I've read about that >>>>> massive single shot capability of the SR, but - as being a newbie - is >>>>> there anything I overlook at this moment? >>>>> >>>> For short time-scales, single shot resolution and trigger jitter >>>> dominates your measurement floor. >>>> >>>> Single-shot resolution is the time resolution by which you make a single >>>> measurement. >>>> >>>> Trigger jitter is the noise at the trigger point. it's a combination of >>>> thermal noise and the slew-rate at the trigger points. It is often that >>>> trigger jitter is dominated by slew-rate, but there is also internal >>>> sources of trigger jitter. The slope dependent trigger jitter follows >>>> the formula: >>>> >>>> t_jitter = v_noise / s_slew >>>> >>>> t_jitter is the trigger jitter (s) >>>> v_noise is the noise power (V) >>>> s_slew is the slew rate (V/s) >>>> >>>> When the time-span of a measurement is long, long-term stability comes >>>> in as well as systematic drifts. Also, systematic noise such as hum also >>>> becomes important. >>>> >>>> To see how much you depend on slew-rate limitation, you can reduce the >>>> amplitude, and as this reduces the slew-rate you can separate the >>>> slew-rate dependent jitter from the intrinsic jitter of the input. It >>>> also helps you to identify if you need to work on the slew-rate limit >>>> rather than anything else. >>>> >>>> So, it may not be the single-shot resolution which limits you, but a >>>> combination of things. >>>> >>>> I would recommend you to pick up a SR620. It has 4 ps single shot >>>> resolution and about 25 ps jitter (but you can get less). That is >>>> significantly better than the 53152A provides. >>>> >>>> SR620 manual (one of many links): >>>> http://ilrs.gsfc.nasa.gov/docs/timing/sr620_manual.pdf >>>> >>>> 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. >>>> >>> >>> _______________________________________________ >>> 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. >> >> >> > > > _______________________________________________ > 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.

VE

Volker Esper

Sat, Nov 3, 2012 8:10 PM

Hi Bob

What is it, that limits a counters life, are you speaking of typical
counter specific failures or do you just mean the common wearout?

Sorry, I know those are no smart questions - but my heart is thumping
when I think of the price and the long way it has to go over the sea...

Thanks

Volker

Am 03.11.2012 17:59, schrieb Bob Camp:

Hi

Both counters have their weak points long term. On the balance I think the 620 should last longer. The 620 is the higher resolution of the two. The 620 normally comes with a bit better reference. Both are supported by various Time Nut software packages. Both do GPIB and serial i/o. The 620 is a bit more controllable over serial. The 53132 takes up less space on your bench.

For the same price - go for the 620. For the usual 53132 is $1300 and the 620 is $2400, not so clear.

Bob

On Nov 3, 2012, at 10:10 AM, Volker Esperailer2@t-online.de wrote:

Thank you for the interesting information. Now, the time has come to look for an adequate counter - anyone who has experience with the HP 53132A and the SR620? If they both where at - say 1000 USD - what would you prefer for the job of phase measurement? I've read about that massive single shot capability of the SR, but - as being a newbie - is there anything I overlook at this moment?

Thanks

Volker - DF9PL

Am 02.11.2012 16:32, schrieb Volker Esper:

Dear fellows,

I'm searching for a counter that allows me to make automated measure series. Of course, Allan Deviation is an important thing to determine when experimenting with or developing GPSDOs (what I intend). So it seemed to be a nice thing to measure phase deviation with a counter (HP 53132A or SR-620 or so).

When experimenting with that stuff I discovered, that all my equipment would not be adequate to determine Allan Deviations beyond 10e-11 (or so). The question is, can any counter deliver that precision / low noise to make such measurements possible? Or does it have to be a more sophisticated technique, such as cross spectrum analysis?

Thanks a lot in advance

Volker

(I guess you recognize, that I am just a newbie to that all)

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.

Hi Bob What is it, that limits a counters life, are you speaking of typical counter specific failures or do you just mean the common wearout? Sorry, I know those are no smart questions - but my heart is thumping when I think of the price and the long way it has to go over the sea... Thanks Volker Am 03.11.2012 17:59, schrieb Bob Camp: > Hi > > Both counters have their weak points long term. On the balance I think the 620 should last longer. The 620 is the higher resolution of the two. The 620 normally comes with a bit better reference. Both are supported by various Time Nut software packages. Both do GPIB and serial i/o. The 620 is a bit more controllable over serial. The 53132 takes up less space on your bench. > > For the same price - go for the 620. For the usual 53132 is $1300 and the 620 is $2400, not so clear. > > Bob > > > On Nov 3, 2012, at 10:10 AM, Volker Esper<ailer2@t-online.de> wrote: > > >> Thank you for the interesting information. Now, the time has come to look for an adequate counter - anyone who has experience with the HP 53132A and the SR620? If they both where at - say 1000 USD - what would you prefer for the job of phase measurement? I've read about that massive single shot capability of the SR, but - as being a newbie - is there anything I overlook at this moment? >> >> Thanks >> >> Volker - DF9PL >> >> >> Am 02.11.2012 16:32, schrieb Volker Esper: >> >>> Dear fellows, >>> >>> I'm searching for a counter that allows me to make automated measure series. Of course, Allan Deviation is an important thing to determine when experimenting with or developing GPSDOs (what I intend). So it seemed to be a nice thing to measure phase deviation with a counter (HP 53132A or SR-620 or so). >>> >>> When experimenting with that stuff I discovered, that all my equipment would not be adequate to determine Allan Deviations beyond 10e-11 (or so). The question is, can any counter deliver that precision / low noise to make such measurements possible? Or does it have to be a more sophisticated technique, such as cross spectrum analysis? >>> >>> Thanks a lot in advance >>> >>> Volker >>> >>> (I guess you recognize, that I am just a newbie to that all) >>> >>> >>> _______________________________________________ >>> 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. >> > > _______________________________________________ > 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. > > >

BC

Bob Camp

Sat, Nov 3, 2012 8:36 PM

Hi

In the case of the 53132, the power supply seems to be the weak link. Out of maybe a hundred or so in the fleet, we see maybe one or two die each month. On the SR620 the power supply also seems to go from time to time. Both have the normal keyboard and display issues, but those can be fixed. If you go back to things like the 5335, the weak point is the input amp, it blows if you get +5 on it. Like power transformers - not a replaceable item...

Bob

On Nov 3, 2012, at 4:10 PM, Volker Esper ailer2@t-online.de wrote:

Hi Bob

What is it, that limits a counters life, are you speaking of typical counter specific failures or do you just mean the common wearout?

Sorry, I know those are no smart questions - but my heart is thumping when I think of the price and the long way it has to go over the sea...

Thanks

Volker

Am 03.11.2012 17:59, schrieb Bob Camp:

Hi

Both counters have their weak points long term. On the balance I think the 620 should last longer. The 620 is the higher resolution of the two. The 620 normally comes with a bit better reference. Both are supported by various Time Nut software packages. Both do GPIB and serial i/o. The 620 is a bit more controllable over serial. The 53132 takes up less space on your bench.

For the same price - go for the 620. For the usual 53132 is $1300 and the 620 is $2400, not so clear.

Bob

On Nov 3, 2012, at 10:10 AM, Volker Esperailer2@t-online.de wrote:

Thank you for the interesting information. Now, the time has come to look for an adequate counter - anyone who has experience with the HP 53132A and the SR620? If they both where at - say 1000 USD - what would you prefer for the job of phase measurement? I've read about that massive single shot capability of the SR, but - as being a newbie - is there anything I overlook at this moment?

Thanks

Volker - DF9PL

Am 02.11.2012 16:32, schrieb Volker Esper:

Dear fellows,

I'm searching for a counter that allows me to make automated measure series. Of course, Allan Deviation is an important thing to determine when experimenting with or developing GPSDOs (what I intend). So it seemed to be a nice thing to measure phase deviation with a counter (HP 53132A or SR-620 or so).

When experimenting with that stuff I discovered, that all my equipment would not be adequate to determine Allan Deviations beyond 10e-11 (or so). The question is, can any counter deliver that precision / low noise to make such measurements possible? Or does it have to be a more sophisticated technique, such as cross spectrum analysis?

Thanks a lot in advance

Volker

(I guess you recognize, that I am just a newbie to that all)

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.

Hi In the case of the 53132, the power supply seems to be the weak link. Out of maybe a hundred or so in the fleet, we see maybe one or two die each month. On the SR620 the power supply also seems to go from time to time. Both have the normal keyboard and display issues, but those can be fixed. If you go back to things like the 5335, the weak point is the input amp, it blows if you get +5 on it. Like power transformers - not a replaceable item... Bob On Nov 3, 2012, at 4:10 PM, Volker Esper <ailer2@t-online.de> wrote: > > Hi Bob > > What is it, that limits a counters life, are you speaking of typical counter specific failures or do you just mean the common wearout? > > Sorry, I know those are no smart questions - but my heart is thumping when I think of the price and the long way it has to go over the sea... > > Thanks > > Volker > > > > > > Am 03.11.2012 17:59, schrieb Bob Camp: >> Hi >> >> Both counters have their weak points long term. On the balance I think the 620 should last longer. The 620 is the higher resolution of the two. The 620 normally comes with a bit better reference. Both are supported by various Time Nut software packages. Both do GPIB and serial i/o. The 620 is a bit more controllable over serial. The 53132 takes up less space on your bench. >> >> For the same price - go for the 620. For the usual 53132 is $1300 and the 620 is $2400, not so clear. >> >> Bob >> >> >> On Nov 3, 2012, at 10:10 AM, Volker Esper<ailer2@t-online.de> wrote: >> >> >>> Thank you for the interesting information. Now, the time has come to look for an adequate counter - anyone who has experience with the HP 53132A and the SR620? If they both where at - say 1000 USD - what would you prefer for the job of phase measurement? I've read about that massive single shot capability of the SR, but - as being a newbie - is there anything I overlook at this moment? >>> >>> Thanks >>> >>> Volker - DF9PL >>> >>> >>> Am 02.11.2012 16:32, schrieb Volker Esper: >>> >>>> Dear fellows, >>>> >>>> I'm searching for a counter that allows me to make automated measure series. Of course, Allan Deviation is an important thing to determine when experimenting with or developing GPSDOs (what I intend). So it seemed to be a nice thing to measure phase deviation with a counter (HP 53132A or SR-620 or so). >>>> >>>> When experimenting with that stuff I discovered, that all my equipment would not be adequate to determine Allan Deviations beyond 10e-11 (or so). The question is, can any counter deliver that precision / low noise to make such measurements possible? Or does it have to be a more sophisticated technique, such as cross spectrum analysis? >>>> >>>> Thanks a lot in advance >>>> >>>> Volker >>>> >>>> (I guess you recognize, that I am just a newbie to that all) >>>> >>>> >>>> _______________________________________________ >>>> 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. >>> >> >> _______________________________________________ >> 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.

VE

Volker Esper

Sat, Nov 3, 2012 9:27 PM

Ah, ok, that helps. Meanwhile, I've found the manual and the schematics,
and it seems to be possible, to get a died power supply back running.

Thanks a lot!

Volker

Am 03.11.2012 21:36, schrieb Bob Camp:

Hi

In the case of the 53132, the power supply seems to be the weak link. Out of maybe a hundred or so in the fleet, we see maybe one or two die each month. On the SR620 the power supply also seems to go from time to time. Both have the normal keyboard and display issues, but those can be fixed. If you go back to things like the 5335, the weak point is the input amp, it blows if you get +5 on it. Like power transformers - not a replaceable item...

Bob

On Nov 3, 2012, at 4:10 PM, Volker Esperailer2@t-online.de wrote:

Hi Bob

What is it, that limits a counters life, are you speaking of typical counter specific failures or do you just mean the common wearout?

Sorry, I know those are no smart questions - but my heart is thumping when I think of the price and the long way it has to go over the sea...

Thanks

Volker

Am 03.11.2012 17:59, schrieb Bob Camp:

Hi

Both counters have their weak points long term. On the balance I think the 620 should last longer. The 620 is the higher resolution of the two. The 620 normally comes with a bit better reference. Both are supported by various Time Nut software packages. Both do GPIB and serial i/o. The 620 is a bit more controllable over serial. The 53132 takes up less space on your bench.

For the same price - go for the 620. For the usual 53132 is $1300 and the 620 is $2400, not so clear.

Bob

On Nov 3, 2012, at 10:10 AM, Volker Esperailer2@t-online.de wrote:

Thank you for the interesting information. Now, the time has come to look for an adequate counter - anyone who has experience with the HP 53132A and the SR620? If they both where at - say 1000 USD - what would you prefer for the job of phase measurement? I've read about that massive single shot capability of the SR, but - as being a newbie - is there anything I overlook at this moment?

Thanks

Volker - DF9PL

Am 02.11.2012 16:32, schrieb Volker Esper:

Dear fellows,

I'm searching for a counter that allows me to make automated measure series. Of course, Allan Deviation is an important thing to determine when experimenting with or developing GPSDOs (what I intend). So it seemed to be a nice thing to measure phase deviation with a counter (HP 53132A or SR-620 or so).

When experimenting with that stuff I discovered, that all my equipment would not be adequate to determine Allan Deviations beyond 10e-11 (or so). The question is, can any counter deliver that precision / low noise to make such measurements possible? Or does it have to be a more sophisticated technique, such as cross spectrum analysis?

Thanks a lot in advance

Volker

(I guess you recognize, that I am just a newbie to that all)

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.

Ah, ok, that helps. Meanwhile, I've found the manual and the schematics, and it seems to be possible, to get a died power supply back running. Thanks a lot! Volker Am 03.11.2012 21:36, schrieb Bob Camp: > Hi > > In the case of the 53132, the power supply seems to be the weak link. Out of maybe a hundred or so in the fleet, we see maybe one or two die each month. On the SR620 the power supply also seems to go from time to time. Both have the normal keyboard and display issues, but those can be fixed. If you go back to things like the 5335, the weak point is the input amp, it blows if you get +5 on it. Like power transformers - not a replaceable item... > > Bob > > On Nov 3, 2012, at 4:10 PM, Volker Esper<ailer2@t-online.de> wrote: > > >> Hi Bob >> >> What is it, that limits a counters life, are you speaking of typical counter specific failures or do you just mean the common wearout? >> >> Sorry, I know those are no smart questions - but my heart is thumping when I think of the price and the long way it has to go over the sea... >> >> Thanks >> >> Volker >> >> >> >> >> >> Am 03.11.2012 17:59, schrieb Bob Camp: >> >>> Hi >>> >>> Both counters have their weak points long term. On the balance I think the 620 should last longer. The 620 is the higher resolution of the two. The 620 normally comes with a bit better reference. Both are supported by various Time Nut software packages. Both do GPIB and serial i/o. The 620 is a bit more controllable over serial. The 53132 takes up less space on your bench. >>> >>> For the same price - go for the 620. For the usual 53132 is $1300 and the 620 is $2400, not so clear. >>> >>> Bob >>> >>> >>> On Nov 3, 2012, at 10:10 AM, Volker Esper<ailer2@t-online.de> wrote: >>> >>> >>> >>>> Thank you for the interesting information. Now, the time has come to look for an adequate counter - anyone who has experience with the HP 53132A and the SR620? If they both where at - say 1000 USD - what would you prefer for the job of phase measurement? I've read about that massive single shot capability of the SR, but - as being a newbie - is there anything I overlook at this moment? >>>> >>>> Thanks >>>> >>>> Volker - DF9PL >>>> >>>> >>>> Am 02.11.2012 16:32, schrieb Volker Esper: >>>> >>>> >>>>> Dear fellows, >>>>> >>>>> I'm searching for a counter that allows me to make automated measure series. Of course, Allan Deviation is an important thing to determine when experimenting with or developing GPSDOs (what I intend). So it seemed to be a nice thing to measure phase deviation with a counter (HP 53132A or SR-620 or so). >>>>> >>>>> When experimenting with that stuff I discovered, that all my equipment would not be adequate to determine Allan Deviations beyond 10e-11 (or so). The question is, can any counter deliver that precision / low noise to make such measurements possible? Or does it have to be a more sophisticated technique, such as cross spectrum analysis? >>>>> >>>>> Thanks a lot in advance >>>>> >>>>> Volker >>>>> >>>>> (I guess you recognize, that I am just a newbie to that all) >>>>> >>>>> >>>>> _______________________________________________ >>>>> 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. >>>> >>>> >>> _______________________________________________ >>> 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. >> > > _______________________________________________ > 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. > > >

VE

Volker Esper

Sat, Nov 3, 2012 9:29 PM

Great, your answere gives me hope, that the calibration procedure can be
done at my home :-)

Thank you!

Volker

Am 03.11.2012 21:08, schrieb Bob Camp:

Hi

There is a fairly elaborate alignment procedure for the 620. It's been reported in great detail here on the list. The counter definitely does better after you go through the full process.

Bob

On Nov 3, 2012, at 3:42 PM, Volker Esperailer2@t-online.de wrote:

Hi Bob

I didn't expect something to be tunable in the counter (except for the oscillator) - what is it that has to be calibrated?

Thanks

Volker

Am 03.11.2012 18:04, schrieb Bob Camp:

Hi

The 620 is still pretty good (when tuned up). It certainly beats the Pendulum's on a single shot basis. The 53230 is spec'd to be as good as the 620. I suspect it meets or exceeds it's stated specs.

Bob

On Nov 3, 2012, at 12:58 PM, Tom Knoxactast@hotmail.com wrote:

Hi Magnus;
I hope this is not to far off thread. Has anyone in the group done real world measurements of single shot res, and jitter on the new Agilent and Tektronix/Pendulum counters compared to the SR620 and Agilent 53132A. I would imagine that counter designs would be an area that really benefits from ongoing advances in digital technology. I have also found that it is much easier to claim specs then meet specs.
Thanks;
Thomas Knox

Date: Sat, 3 Nov 2012 16:28:42 +0100
From: magnus@rubidium.dyndns.org
To: time-nuts@febo.com
Subject: Re: [time-nuts] A counter for phase measures

On 11/03/2012 03:10 PM, Volker Esper wrote:

Thank you for the interesting information. Now, the time has come to
look for an adequate counter - anyone who has experience with the HP
53132A and the SR620? If they both where at - say 1000 USD - what would
you prefer for the job of phase measurement? I've read about that
massive single shot capability of the SR, but - as being a newbie - is
there anything I overlook at this moment?

For short time-scales, single shot resolution and trigger jitter
dominates your measurement floor.

Single-shot resolution is the time resolution by which you make a single
measurement.

Trigger jitter is the noise at the trigger point. it's a combination of
thermal noise and the slew-rate at the trigger points. It is often that
trigger jitter is dominated by slew-rate, but there is also internal
sources of trigger jitter. The slope dependent trigger jitter follows
the formula:

t_jitter = v_noise / s_slew

t_jitter is the trigger jitter (s)
v_noise is the noise power (V)
s_slew is the slew rate (V/s)

When the time-span of a measurement is long, long-term stability comes
in as well as systematic drifts. Also, systematic noise such as hum also
becomes important.

To see how much you depend on slew-rate limitation, you can reduce the
amplitude, and as this reduces the slew-rate you can separate the
slew-rate dependent jitter from the intrinsic jitter of the input. It
also helps you to identify if you need to work on the slew-rate limit
rather than anything else.

So, it may not be the single-shot resolution which limits you, but a
combination of things.

I would recommend you to pick up a SR620. It has 4 ps single shot
resolution and about 25 ps jitter (but you can get less). That is
significantly better than the 53152A provides.

SR620 manual (one of many links):
http://ilrs.gsfc.nasa.gov/docs/timing/sr620_manual.pdf

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.

Great, your answere gives me hope, that the calibration procedure can be done at my home :-) Thank you! Volker Am 03.11.2012 21:08, schrieb Bob Camp: > Hi > > There is a fairly elaborate alignment procedure for the 620. It's been reported in great detail here on the list. The counter definitely does better after you go through the full process. > > Bob > > On Nov 3, 2012, at 3:42 PM, Volker Esper<ailer2@t-online.de> wrote: > > >> Hi Bob >> >> I didn't expect something to be tunable in the counter (except for the oscillator) - what is it that has to be calibrated? >> >> Thanks >> >> Volker >> >> >> Am 03.11.2012 18:04, schrieb Bob Camp: >> >>> Hi >>> >>> The 620 is still pretty good (when tuned up). It certainly beats the Pendulum's on a single shot basis. The 53230 is spec'd to be as good as the 620. I suspect it meets or exceeds it's stated specs. >>> >>> Bob >>> >>> On Nov 3, 2012, at 12:58 PM, Tom Knox<actast@hotmail.com> wrote: >>> >>> >>> >>>> Hi Magnus; >>>> I hope this is not to far off thread. Has anyone in the group done real world measurements of single shot res, and jitter on the new Agilent and Tektronix/Pendulum counters compared to the SR620 and Agilent 53132A. I would imagine that counter designs would be an area that really benefits from ongoing advances in digital technology. I have also found that it is much easier to claim specs then meet specs. >>>> Thanks; >>>> Thomas Knox >>>> >>>> >>>> >>>> >>>> >>>>> Date: Sat, 3 Nov 2012 16:28:42 +0100 >>>>> From: magnus@rubidium.dyndns.org >>>>> To: time-nuts@febo.com >>>>> Subject: Re: [time-nuts] A counter for phase measures >>>>> >>>>> On 11/03/2012 03:10 PM, Volker Esper wrote: >>>>> >>>>> >>>>>> Thank you for the interesting information. Now, the time has come to >>>>>> look for an adequate counter - anyone who has experience with the HP >>>>>> 53132A and the SR620? If they both where at - say 1000 USD - what would >>>>>> you prefer for the job of phase measurement? I've read about that >>>>>> massive single shot capability of the SR, but - as being a newbie - is >>>>>> there anything I overlook at this moment? >>>>>> >>>>>> >>>>> For short time-scales, single shot resolution and trigger jitter >>>>> dominates your measurement floor. >>>>> >>>>> Single-shot resolution is the time resolution by which you make a single >>>>> measurement. >>>>> >>>>> Trigger jitter is the noise at the trigger point. it's a combination of >>>>> thermal noise and the slew-rate at the trigger points. It is often that >>>>> trigger jitter is dominated by slew-rate, but there is also internal >>>>> sources of trigger jitter. The slope dependent trigger jitter follows >>>>> the formula: >>>>> >>>>> t_jitter = v_noise / s_slew >>>>> >>>>> t_jitter is the trigger jitter (s) >>>>> v_noise is the noise power (V) >>>>> s_slew is the slew rate (V/s) >>>>> >>>>> When the time-span of a measurement is long, long-term stability comes >>>>> in as well as systematic drifts. Also, systematic noise such as hum also >>>>> becomes important. >>>>> >>>>> To see how much you depend on slew-rate limitation, you can reduce the >>>>> amplitude, and as this reduces the slew-rate you can separate the >>>>> slew-rate dependent jitter from the intrinsic jitter of the input. It >>>>> also helps you to identify if you need to work on the slew-rate limit >>>>> rather than anything else. >>>>> >>>>> So, it may not be the single-shot resolution which limits you, but a >>>>> combination of things. >>>>> >>>>> I would recommend you to pick up a SR620. It has 4 ps single shot >>>>> resolution and about 25 ps jitter (but you can get less). That is >>>>> significantly better than the 53152A provides. >>>>> >>>>> SR620 manual (one of many links): >>>>> http://ilrs.gsfc.nasa.gov/docs/timing/sr620_manual.pdf >>>>> >>>>> 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. >>>>> >>>>> >>>> >>>> _______________________________________________ >>>> 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. >>> >>> >>> >>> >> >> _______________________________________________ >> 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. > > >

DH

David Hooke

Sat, Nov 3, 2012 9:44 PM

Folks,

Given that slew rate is so critical, why do we distribute sine waves and
perform the zero-crossing detection at every target instrument?

david

Trigger jitter is the noise at the trigger point. it's a combination
of thermal noise and the slew-rate at the trigger points. It is often
that trigger jitter is dominated by slew-rate, but there is also
internal sources of trigger jitter. The slope dependent trigger jitter
follows the formula:

t_jitter = v_noise / s_slew

t_jitter is the trigger jitter (s)
v_noise is the noise power (V)
s_slew is the slew rate (V/s)

When the time-span of a measurement is long, long-term stability comes
in as well as systematic drifts. Also, systematic noise such as hum
also becomes important.

To see how much you depend on slew-rate limitation, you can reduce the
amplitude, and as this reduces the slew-rate you can separate the
slew-rate dependent jitter from the intrinsic jitter of the input. It
also helps you to identify if you need to work on the slew-rate limit
rather than anything else.

Folks, Given that slew rate is so critical, why do we distribute sine waves and perform the zero-crossing detection at every target instrument? david > Trigger jitter is the noise at the trigger point. it's a combination > of thermal noise and the slew-rate at the trigger points. It is often > that trigger jitter is dominated by slew-rate, but there is also > internal sources of trigger jitter. The slope dependent trigger jitter > follows the formula: > > t_jitter = v_noise / s_slew > > t_jitter is the trigger jitter (s) > v_noise is the noise power (V) > s_slew is the slew rate (V/s) > > When the time-span of a measurement is long, long-term stability comes > in as well as systematic drifts. Also, systematic noise such as hum > also becomes important. > > To see how much you depend on slew-rate limitation, you can reduce the > amplitude, and as this reduces the slew-rate you can separate the > slew-rate dependent jitter from the intrinsic jitter of the input. It > also helps you to identify if you need to work on the slew-rate limit > rather than anything else. >

MD

Magnus Danielson

Sat, Nov 3, 2012 10:14 PM

On 11/03/2012 09:36 PM, Bob Camp wrote:

Hi

In the case of the 53132, the power supply seems to be the weak link. Out of maybe a hundred or so in the fleet, we see maybe one or two die each month. On the SR620 the power supply also seems to go from time to time. Both have the normal keyboard and display issues, but those can be fixed. If you go back to things like the 5335, the weak point is the input amp, it blows if you get +5 on it. Like power transformers - not a replaceable item...

Another weak point on the 5335 is the relay. We had to replace it and
the relay-holder, but once that was done, it was back up operational.
The 5335 ticks in as the most human-friendly of the counters at work,
while the 53132 is competing with the 5372 as being the most
human-unfriendly, where the 5372 has more capabilities to present, so it
gets used more.

Cheers,
Magnus

On 11/03/2012 09:36 PM, Bob Camp wrote: > Hi > > In the case of the 53132, the power supply seems to be the weak link. Out of maybe a hundred or so in the fleet, we see maybe one or two die each month. On the SR620 the power supply also seems to go from time to time. Both have the normal keyboard and display issues, but those can be fixed. If you go back to things like the 5335, the weak point is the input amp, it blows if you get +5 on it. Like power transformers - not a replaceable item... Another weak point on the 5335 is the relay. We had to replace it and the relay-holder, but once that was done, it was back up operational. The 5335 ticks in as the most human-friendly of the counters at work, while the 53132 is competing with the 5372 as being the most human-unfriendly, where the 5372 has more capabilities to present, so it gets used more. Cheers, Magnus

MD

Magnus Danielson

Sat, Nov 3, 2012 10:27 PM

David,

On 11/03/2012 10:44 PM, David Hooke wrote:

Folks,

Given that slew rate is so critical, why do we distribute sine waves and
perform the zero-crossing detection at every target instrument?

... or rather, why do we design our input stages so they are so
slew-rate sensitive?

Sine isn't necessary a bad choice, the benefit of a sine is that you
would not have to be as wide-band as to handle a whole number of
overtones. That translates into lower amount of noise.

There isn't really one right way of doing it, you can go about it in
several ways, but you need to do it consistently.

I've modified my TADD-2:s such that I use the input treatment to drive
one of the outputs, so that they will square up sines for me. For some
signals this have lowered my trigger jitter and hence improved my
ability to see more of the actual signal I want to see.

Just as much as you can get a counter with very high single shot
resolution, it doesn't help if you do not treat your signals properly to
get the most of that counter.

When doing DMTD tricks, the mixer is the easy part, squaring the signal
up to get good trigger jitter for the total is what takes a lot of effort.

Cheers,
Magnus

David, On 11/03/2012 10:44 PM, David Hooke wrote: > Folks, > > Given that slew rate is so critical, why do we distribute sine waves and > perform the zero-crossing detection at every target instrument? ... or rather, why do we design our input stages so they are so slew-rate sensitive? Sine isn't necessary a bad choice, the benefit of a sine is that you would not have to be as wide-band as to handle a whole number of overtones. That translates into lower amount of noise. There isn't really one right way of doing it, you can go about it in several ways, but you need to do it consistently. I've modified my TADD-2:s such that I use the input treatment to drive one of the outputs, so that they will square up sines for me. For some signals this have lowered my trigger jitter and hence improved my ability to see more of the actual signal I want to see. Just as much as you can get a counter with very high single shot resolution, it doesn't help if you do not treat your signals properly to get the most of that counter. When doing DMTD tricks, the mixer is the easy part, squaring the signal up to get good trigger jitter for the total is what takes a lot of effort. Cheers, Magnus

AB

Azelio Boriani

Sat, Nov 3, 2012 10:32 PM

Usually you don't need a BVA to test the single-shot capability of a
counter: a length (say 50nS) of good RF coaxial cable and your preferred
OCXO/Rb/GPSDO should be enough.

On Sat, Nov 3, 2012 at 11:14 PM, Magnus Danielson <
magnus@rubidium.dyndns.org> wrote:

On 11/03/2012 09:36 PM, Bob Camp wrote:

Hi

In the case of the 53132, the power supply seems to be the weak link. Out
of maybe a hundred or so in the fleet, we see maybe one or two die each
month. On the SR620 the power supply also seems to go from time to time.
Both have the normal keyboard and display issues, but those can be fixed.
If you go back to things like the 5335, the weak point is the input amp, it
blows if you get +5 on it. Like power transformers - not a replaceable
item...

Another weak point on the 5335 is the relay. We had to replace it and the
relay-holder, but once that was done, it was back up operational.
The 5335 ticks in as the most human-friendly of the counters at work,
while the 53132 is competing with the 5372 as being the most
human-unfriendly, where the 5372 has more capabilities to present, so it
gets used more.

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.

Usually you don't need a BVA to test the single-shot capability of a counter: a length (say 50nS) of good RF coaxial cable and your preferred OCXO/Rb/GPSDO should be enough. On Sat, Nov 3, 2012 at 11:14 PM, Magnus Danielson < magnus@rubidium.dyndns.org> wrote: > On 11/03/2012 09:36 PM, Bob Camp wrote: > >> Hi >> >> In the case of the 53132, the power supply seems to be the weak link. Out >> of maybe a hundred or so in the fleet, we see maybe one or two die each >> month. On the SR620 the power supply also seems to go from time to time. >> Both have the normal keyboard and display issues, but those can be fixed. >> If you go back to things like the 5335, the weak point is the input amp, it >> blows if you get +5 on it. Like power transformers - not a replaceable >> item... >> > > Another weak point on the 5335 is the relay. We had to replace it and the > relay-holder, but once that was done, it was back up operational. > The 5335 ticks in as the most human-friendly of the counters at work, > while the 53132 is competing with the 5372 as being the most > human-unfriendly, where the 5372 has more capabilities to present, so it > gets used more. > > 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. >

BC

Bob Camp

Sat, Nov 3, 2012 11:05 PM

Hi

If you have a sine wave, it gets into everything. You can identify it and take it out of your data.

If you have a broad band uber-fast high level pules, it gets into everything. Identifying it's impact and taking it out of the data - not so easy.

That may sound a bit crazy. I've actually worked in a place that went over to a square wave based system. It lasted for about a day. Got into all sorts of things, total nightmare.

Bob

On Nov 3, 2012, at 5:44 PM, David Hooke dhooke@gmail.com wrote:

Folks,

Given that slew rate is so critical, why do we distribute sine waves and perform the zero-crossing detection at every target instrument?

david

Trigger jitter is the noise at the trigger point. it's a combination of thermal noise and the slew-rate at the trigger points. It is often that trigger jitter is dominated by slew-rate, but there is also internal sources of trigger jitter. The slope dependent trigger jitter follows the formula:

t_jitter = v_noise / s_slew

t_jitter is the trigger jitter (s)
v_noise is the noise power (V)
s_slew is the slew rate (V/s)

When the time-span of a measurement is long, long-term stability comes in as well as systematic drifts. Also, systematic noise such as hum also becomes important.

To see how much you depend on slew-rate limitation, you can reduce the amplitude, and as this reduces the slew-rate you can separate the slew-rate dependent jitter from the intrinsic jitter of the input. It also helps you to identify if you need to work on the slew-rate limit rather than anything else.

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.

Hi If you have a sine wave, it gets into everything. You can identify it and take it out of your data. If you have a broad band uber-fast high level pules, it gets into everything. Identifying it's impact and taking it out of the data - not so easy. That may sound a bit crazy. I've actually worked in a place that went over to a square wave based system. It lasted for about a day. Got into all sorts of things, total nightmare. Bob On Nov 3, 2012, at 5:44 PM, David Hooke <dhooke@gmail.com> wrote: > > > > Folks, > > Given that slew rate is so critical, why do we distribute sine waves and perform the zero-crossing detection at every target instrument? > > david > >> Trigger jitter is the noise at the trigger point. it's a combination of thermal noise and the slew-rate at the trigger points. It is often that trigger jitter is dominated by slew-rate, but there is also internal sources of trigger jitter. The slope dependent trigger jitter follows the formula: >> >> t_jitter = v_noise / s_slew >> >> t_jitter is the trigger jitter (s) >> v_noise is the noise power (V) >> s_slew is the slew rate (V/s) >> >> When the time-span of a measurement is long, long-term stability comes in as well as systematic drifts. Also, systematic noise such as hum also becomes important. >> >> To see how much you depend on slew-rate limitation, you can reduce the amplitude, and as this reduces the slew-rate you can separate the slew-rate dependent jitter from the intrinsic jitter of the input. It also helps you to identify if you need to work on the slew-rate limit rather than anything else. >> > > > > > _______________________________________________ > 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@lists.febo.com

A counter for phase measures