Topics

Need help with 11801 delay jitter #photo-notice


lccavalheiro@...
 

Hi everyone,

I recently purchased a 11801B with an SD-26 module. Because I haven't used one before, I'm trying to understand if the behavior I'm seeing is expected or if the equipment is compromised. I'm using the Bodnar edge generator for this, and when I acquire as close as I can to the trigger (delay ~ 49ns), the edge looks fairly good, but when I add a bit more delay (delay ~ 11us), it seems that the jitter noise makes the measurement very unreliable and noisy. I've uploaded 2 images of what I'm referring to, and I'm hoping someone that also has this scope can tell me if this is the normal behavior or whether I mine has a problem.
The link to the pictures is: https://groups.io/g/TekScopes/album?id=250925

PS: This looks the same if I use the 11801 internal clock output.

Looking forward to getting any feedback.

Cheer!


Clark Foley
 

It has been a long time since I have used the 11801, but I remember well that the jitter accumulation as a function of delay is profound. What you are seeing is not unusual from what I recall. The scope was optimized for short delay times. The timing hardware is not locked to an internal timing standard; instead, it is done with analog linear ramps. Check the jitter specifications and you will see a term something like "4ppm of delay setting." I think that this is an RMS and typical. The display will capture close to the peak-to-peak jitter which is many times the RMS jitter.
Use the scope's RMS jitter measurement and compare to the number that you calculate from the specification.

Clark


Reginald Beardsley
 

Use the calibrator output. Leo's pulsers have significant jitter.

"Initialize" the scope under "Utilities", set the time base to "internal" and connect the calibrator output to the SD-26. The calibrator is ~19 ps rise time, so significantly better than Leo's BNC unit and slightly better than the 3.5 and 2.48 mm units.

Have Fun!
Reg


lccavalheiro@...
 

Hi Clark,

You made a great point about the jitter spec and I've take the measurement you suggested. Without smoothing, the jitter is within specs for delays lower than few hundreds of ns, but if I get to more than a few us, it seems to fall out of the spec, even the 2ps + 4ppm of the delay. Either way, I think the equipment is working just fine, and I don't think I'll need to use large delays.
Thanks again for the feedback.

Lucas


lccavalheiro@...
 

Hi Reginald,

I got pretty excited with the idea of getting a 19ps rise time from the calibrator, but when I connect it to the sampling head I'm only measuring hundreds of ps rise time. The waveform is synchronizing and looks fine, but the rise time is definitely much slower than 19ps. Do you think this could be an issue with my calibration signal, or is there any configuration that can change that?

Thanks,

Lucas


Chuck Harris
 

It is important that you remember your 11801 is not an
ordinary digital storage scope: It is a sampling scope.

Sampling scopes do their magic by taking small samples of
hundreds, or even thousands of repetitions of a given signal.

Each time they trigger, they grab another single voltage
measurement, and each time they trigger, they offset to a
slightly different timing point in the waveform.

Only after they have gathered a thousand, or so samples from a
thousand or so triggers of the waveform, do they have enough data
to reconstruct one full picture of the waveform.

If... each trigger isn't exactly like every other... because jitter
is happening, then you will get nothing but a garbage representation
of the waveform.

-Chuck Harris

lccavalheiro@... wrote:

Hi Clark,

You made a great point about the jitter spec and I've take the measurement you suggested. Without smoothing, the jitter is within specs for delays lower than few hundreds of ns, but if I get to more than a few us, it seems to fall out of the spec, even the 2ps + 4ppm of the delay. Either way, I think the equipment is working just fine, and I don't think I'll need to use large delays.
Thanks again for the feedback.

Lucas




John Gord
 

Lucas,
From what I have been able to find, the 11801 has a 20ps rise time calibrator, but the 11801C has a 250ps rise time calibrator. I don't know about the A and B versions.
--John Gord

On Thu, Jul 23, 2020 at 04:59 PM, <lccavalheiro@...> wrote:


Hi Reginald,

I got pretty excited with the idea of getting a 19ps rise time from the
calibrator, but when I connect it to the sampling head I'm only measuring
hundreds of ps rise time. The waveform is synchronizing and looks fine, but
the rise time is definitely much slower than 19ps. Do you think this could be
an issue with my calibration signal, or is there any configuration that can
change that?

Thanks,

Lucas


Albert Otten
 

I thought it was well known by now that 2 types of calibrator units exist. Regrettably I only have a CSA803 and CSA803A, hence the slower type.
See also https://groups.io/g/TekScopes/message/155394?p=,,,20,0,0,0::Created,,11801+calibrator+connector,20,2,0,30465052 .

Albert


Chuck Harris
 

From what I remember, one special option of the 11801
came out with a calibrator that was half of the S26 TDR
head. It has a 20ps calibrator. It can be recognized
by its gold APC-3.5 connector... where as the common
calibrator had a stainless APC-3.5 connector.

The fancy calibrator was done so you could roll your own
TDR system using the other sampling heads.

-Chuck Harris

John Gord via groups.io wrote:

Lucas,
From what I have been able to find, the 11801 has a 20ps rise time calibrator, but the 11801C has a 250ps rise time calibrator. I don't know about the A and B versions.
--John Gord

On Thu, Jul 23, 2020 at 04:59 PM, <lccavalheiro@...> wrote:


Hi Reginald,

I got pretty excited with the idea of getting a 19ps rise time from the
calibrator, but when I connect it to the sampling head I'm only measuring
hundreds of ps rise time. The waveform is synchronizing and looks fine, but
the rise time is definitely much slower than 19ps. Do you think this could be
an issue with my calibration signal, or is there any configuration that can
change that?

Thanks,

Lucas



lccavalheiro@...
 

I was actually thinking about that. I'm wondering what would be the worst-case jitter on Leo's pulser. The laser driver use claims typical 4 ps pk-pk deterministic and 0.55 ps RMS random jitter. I guess that would be the major source of jitter there.


Reginald Beardsley
 

The 11801 calibrator is the pulser from an SD-24. The SD-26 is the same head without a pulser. for a long time all I had was an SD-26 and the calibrator.

I'm stunned! My desire for an 11801C probably just vanished unless it's *real* cheap. 250 ps rise time calibrator? Oh, how the mighty have fallen! Clear evidence of an ascendancy of accountants.

FYI A group of birds is a flock. A group of accountants is an "ascendancy".

Reg


Harvey White
 

the proper name for a group of accountants can't be repeated in polite society.

Wonder if you could make something that had a better rise time?

Harvey

On 7/26/2020 6:08 PM, Reginald Beardsley via groups.io wrote:
The 11801 calibrator is the pulser from an SD-24. The SD-26 is the same head without a pulser. for a long time all I had was an SD-26 and the calibrator.

I'm stunned! My desire for an 11801C probably just vanished unless it's *real* cheap. 250 ps rise time calibrator? Oh, how the mighty have fallen! Clear evidence of an ascendancy of accountants.

FYI A group of birds is a flock. A group of accountants is an "ascendancy".

Reg



Leon Robinson
 

BEANCOUNTERS, the downfall of many good company's.

Leon Robinson    K5JLR

Political Correctness is a Political Disease.

Politicians and Diapers should be changed
often and for the same reasons.

On Sunday, July 26, 2020, 5:26:40 PM CDT, Harvey White <madyn@...> wrote:

the proper name for a group of accountants can't be repeated in polite
society.

Wonder if you could make something that had a better rise time?

Harvey


On 7/26/2020 6:08 PM, Reginald Beardsley via groups.io wrote:
The 11801 calibrator is the pulser from an SD-24.  The SD-26 is the same head without a pulser.  for a long time all I had was an SD-26 and the calibrator.

I'm stunned!  My desire for an 11801C probably just vanished unless it's *real* cheap.  250 ps rise time calibrator?  Oh, how the mighty have fallen!  Clear evidence of an ascendancy of accountants.

FYI A group of birds is a flock.  A group of accountants is an "ascendancy".

Reg




Chuck Harris
 

I'm pretty sure that the 20ps calibrator is an option available
on the entire series, and standard to none.

-Chuck Harris

Reginald Beardsley via groups.io wrote:

The 11801 calibrator is the pulser from an SD-24. The SD-26 is the same head without a pulser. for a long time all I had was an SD-26 and the calibrator.

I'm stunned! My desire for an 11801C probably just vanished unless it's *real* cheap. 250 ps rise time calibrator? Oh, how the mighty have fallen! Clear evidence of an ascendancy of accountants.

FYI A group of birds is a flock. A group of accountants is an "ascendancy".

Reg




John Gord
 

Chuck,
The 11801 Service Reference Manual has a test (page 2-44) calling for the 11801 to measure its own calibrator at 35ps or better for the sampler/calibrator combination.
No mention is made of any optional calibrator versions. I don't know about the options that might have been available for the A/B/C versions.
--John Gord

On Sun, Jul 26, 2020 at 05:58 PM, Chuck Harris wrote:


I'm pretty sure that the 20ps calibrator is an option available
on the entire series, and standard to none.

-Chuck Harris


Chuck Harris
 

The 11801C specifically calls on the calibrator to be 250ps risetime.

I'm sure that I got the information about the 25ps risetime calibrator
being an option from one of you on this group. Maybe Leo?

It was during a show and tell session, and someone pointed out that
the calibrators with gold APC3.5 connectors were fast TDR compliant
calibrators, and were optional.

The 11801C manual I have doesn't discuss that option, though.

-Chuck Harris

John Gord via groups.io wrote:

Chuck,
The 11801 Service Reference Manual has a test (page 2-44) calling for the 11801 to measure its own calibrator at 35ps or better for the sampler/calibrator combination.
No mention is made of any optional calibrator versions. I don't know about the options that might have been available for the A/B/C versions.
--John Gord


On Sun, Jul 26, 2020 at 05:58 PM, Chuck Harris wrote:


I'm pretty sure that the 20ps calibrator is an option available
on the entire series, and standard to none.

-Chuck Harris