开云体育

Chuck,

The random sampling is intentional and is done to reduce the artifacts of
precisely spaced samples.

I think the 3T77A was the first of Tek's attempts. I clearly remember that
one of the 3 Series sampling sweeps was labeled that way.

-John

================

David wrote:

There are many ways to get a waveform using sampling. All of those

that sample waveforms that are higher bandwidth than the sampling rate
are storing only small parts of many, many, repetitions of the signal
under test. In the case of the 7D20, and the 7854, you may be looking
at snippets of hundreds of repetitions of that signal, just to get a
look
at a single copy. In the days of old, these were called sampling
oscilloscopes.

My old Tektronix catalogs always refer to them as digitizers or
digital storage oscilloscopes. The term sampling was always
associated with instruments that had actual sampling front ends.

It doesn't matter what they call it, if it can't store the whole
waveform in one shot, it is a sampling scope, just as sure as the
old N, 1SXX, 7SXX, etc. plugins were. The prime difference is the
old type N, 1SXX, and 7Sxx plugins used the screen's phosphor, in
combination with the refresh rate, to "store" the sampled bits long
enough for you to see the full waveform. The 7D20, and 7854 use
digital storage bins, filled in a fairly chaotic way, to store the
sampled bits for view. If you have ever watched a 7854 store a
400MHz sine wave using its 50K sampling rate, you know what I mean.
If you have it set to display the stored bits as they come in, you
will see dots randomly appear on the screen (like snowflakes) as
the waveform is generated in the digital memory.

I never picked up a 7D20 because it lacks peak detection but the slow
waveform regeneration rate of my 2230 has only rarely been a problem.

I use a 7854, or a 7D20, quite a lot... but only in the single shot
mode. I usually only need storage to handle things that are slower
than my visual refresh rate.

-Chuck Harris

It was almost certainly a 2230 (100 MHz and 20 MS/sec) or 2220 (60 MHz
and 20 MS/sec) which came out in 1986 or at least first showed up in
that year's catalog.

The sample clock is not dithered but instead the difference between
the trigger and sample clock is measured to within about 500ps which
allows the acquired samples to be positioned within the waveform
record. In order to gain anything from that process, the signal being
measured and the sample clock have to be asynchronous.

It is my most used oscilloscope although the updated version in the
form of the 2232 is superior in almost every way.

On Mon, 31 Dec 2012 17:40:05 -0000, "erich_schlecht"
<schlechtca@...> wrote:

Speaking of old scopes, the first digital scope I got circa 1986 was a Tek with a sample rate around 20 or 50 Msps, bandwidth 100 or 200 MHz. For repetitive signals it dithered the sample clock to reconstruct signals well above the Nyquist frequency over many cycles. It couldn't see fast single event signals, of course.

It also had a pure analog mode. For the time, it was a pretty decent instrument.Unfortunately, I've long forgotten the model number, but it looked like a 24xx series.

Erich

--- In hp_agilent_equipment@..., Chuck Harris <cfharris@> wrote:

Hi Peter,

As I said, "any competently designed DSO". An analog scope gives
you the full vertical bandwidth regardless of the timebase setting.
A competently designed DSO should also.

You can be a bit flexible about that requirement, though. If the
aliasing effects are too fast to see at a particular timebase
setting, it would be ok to slow the sample rate until they are
only marginally too fast to see.

-Chuck Harris

Peter Gottlieb wrote:

But it's not just filtering above the Nyquist. There are other ways a sampling
digital scope can give you a wrong picture of reality. If all of these scopes
ran their digitizers constantly at full rate, watched for envelope effects and
so forth they would go a long way towards eliminating these unwanted erroneous
displays.

Peter


On 12/31/2012 10:56 AM, Chuck Harris wrote:

If by "trust" you mean see things faster than the Nyquist
limit, I fully agree.

We introduced the 2230 world-wide in the fall of 1985, so 1986 was the big year for it. It has equivalent-time sampling for display of repetitive signals up to 100 MHz, and peak detection for anti-aliasing envelope display of undersampled signals, and it operates in regular analog scope mode - equivalent to the 2235, which it was based on. The 2220 was the same, but lesser BW to fill the lower market spot. The 2232 followed, with many improvements - especially 100 Ms/sec sampling rate.

Ed

--- In hp_agilent_equipment@..., David <davidwhess@> wrote:

It was almost certainly a 2230 (100 MHz and 20 MS/sec) or 2220 (60 MHz
and 20 MS/sec) which came out in 1986 or at least first showed up in
that year's catalog.

The sample clock is not dithered but instead the difference between
the trigger and sample clock is measured to within about 500ps which
allows the acquired samples to be positioned within the waveform
record. In order to gain anything from that process, the signal being
measured and the sample clock have to be asynchronous.

It is my most used oscilloscope although the updated version in the
form of the 2232 is superior in almost every way.

On Mon, 31 Dec 2012 17:40:05 -0000, "erich_schlecht"
<schlechtca@> wrote:

Speaking of old scopes, the first digital scope I got circa 1986 was a Tek with a sample rate around 20 or 50 Msps, bandwidth 100 or 200 MHz. For repetitive signals it dithered the sample clock to reconstruct signals well above the Nyquist frequency over many cycles. It couldn't see fast single event signals, of course.

It also had a pure analog mode. For the time, it was a pretty decent instrument.Unfortunately, I've long forgotten the model number, but it looked like a 24xx series.

Erich

--- In hp_agilent_equipment@..., Chuck Harris <cfharris@> wrote:

Hi Peter,

As I said, "any competently designed DSO". An analog scope gives
you the full vertical bandwidth regardless of the timebase setting.
A competently designed DSO should also.

You can be a bit flexible about that requirement, though. If the
aliasing effects are too fast to see at a particular timebase
setting, it would be ok to slow the sample rate until they are
only marginally too fast to see.

-Chuck Harris

Peter Gottlieb wrote:

But it's not just filtering above the Nyquist. There are other ways a sampling
digital scope can give you a wrong picture of reality. If all of these scopes
ran their digitizers constantly at full rate, watched for envelope effects and
so forth they would go a long way towards eliminating these unwanted erroneous
displays.

Peter


On 12/31/2012 10:56 AM, Chuck Harris wrote:

If by "trust" you mean see things faster than the Nyquist
limit, I fully agree.

Chuck,

The random sampling is intentional and is done to reduce the artifacts of
precisely spaced samples.

I think the 3T77A was the first of Tek's attempts. I clearly remember that
one of the 3 Series sampling sweeps was labeled that way.

-John

================

David wrote:

There are many ways to get a waveform using sampling. All of those

that sample waveforms that are higher bandwidth than the sampling rate
are storing only small parts of many, many, repetitions of the signal
under test. In the case of the 7D20, and the 7854, you may be looking
at snippets of hundreds of repetitions of that signal, just to get a
look
at a single copy. In the days of old, these were called sampling
oscilloscopes.

My old Tektronix catalogs always refer to them as digitizers or
digital storage oscilloscopes. The term sampling was always
associated with instruments that had actual sampling front ends.

It doesn't matter what they call it, if it can't store the whole
waveform in one shot, it is a sampling scope, just as sure as the
old N, 1SXX, 7SXX, etc. plugins were. The prime difference is the
old type N, 1SXX, and 7Sxx plugins used the screen's phosphor, in
combination with the refresh rate, to "store" the sampled bits long
enough for you to see the full waveform. The 7D20, and 7854 use
digital storage bins, filled in a fairly chaotic way, to store the
sampled bits for view. If you have ever watched a 7854 store a
400MHz sine wave using its 50K sampling rate, you know what I mean.
If you have it set to display the stored bits as they come in, you
will see dots randomly appear on the screen (like snowflakes) as
the waveform is generated in the digital memory.

I never picked up a 7D20 because it lacks peak detection but the slow
waveform regeneration rate of my 2230 has only rarely been a problem.

I use a 7854, or a 7D20, quite a lot... but only in the single shot
mode. I usually only need storage to handle things that are slower
than my visual refresh rate.

-Chuck Harris

Yeah !
IMHO this instrument was THE great successor of the legendary 465: easy to use (except menus) and exceptional & reliable readings due to the top performance "peak-det" feature implemented in this very instrument. I used most the digital mode because it was so trustable! After many years I sold it for more recent models. And finally missed it and bought a used 2232 which featured many nice improvements over the former. This one is now my prefered general purpose workhorse among three others (2440, 2467BHD & HP 54542A). A friend of mine which I adviced to buy a 2232 said that this was really what he long awaited for...

Congrats for the great job! I suspect later scopes were not designed by the same team... Right ?

Christian F1GWR


Le 1 janv. 2013 à 04:39, "Ed Breya" <edbreya@...> a écrit :

We introduced the 2230 world-wide in the fall of 1985, so 1986 was the big year for it. It has equivalent-time sampling for display of repetitive signals up to 100 MHz, and peak detection for anti-aliasing envelope display of undersampled signals, and it operates in regular analog scope mode - equivalent to the 2235, which it was based on. The 2220 was the same, but lesser BW to fill the lower market spot. The 2232 followed, with many improvements - especially 100 Ms/sec sampling rate.

Ed

In order to use a HP 8757A SNA as a display, the noise figure meter 8970B

My datecode is only 2725 yelding E36 error. Bad luck!
To display firmware release enter 99.9 SP on the 8970B.
If someone could provide me with related ROM dump, would be great!

David wrote:

There are many ways to get a waveform using sampling. All of those

that sample waveforms that are higher bandwidth than the sampling rate
are storing only small parts of many, many, repetitions of the signal
under test. In the case of the 7D20, and the 7854, you may be looking
at snippets of hundreds of repetitions of that signal, just to get a
look
at a single copy. In the days of old, these were called sampling
oscilloscopes.

My old Tektronix catalogs always refer to them as digitizers or
digital storage oscilloscopes. The term sampling was always
associated with instruments that had actual sampling front ends.

It doesn't matter what they call it, if it can't store the whole
waveform in one shot, it is a sampling scope, just as sure as the
old N, 1SXX, 7SXX, etc. plugins were. The prime difference is the
old type N, 1SXX, and 7Sxx plugins used the screen's phosphor, in
combination with the refresh rate, to "store" the sampled bits long
enough for you to see the full waveform. The 7D20, and 7854 use
digital storage bins, filled in a fairly chaotic way, to store the
sampled bits for view. If you have ever watched a 7854 store a
400MHz sine wave using its 50K sampling rate, you know what I mean.
If you have it set to display the stored bits as they come in, you
will see dots randomly appear on the screen (like snowflakes) as
the waveform is generated in the digital memory.

I never picked up a 7D20 because it lacks peak detection but the slow
waveform regeneration rate of my 2230 has only rarely been a problem.

I use a 7854, or a 7D20, quite a lot... but only in the single shot
mode. I usually only need storage to handle things that are slower
than my visual refresh rate.

-Chuck Harris

You're probably thinking of a 2232. The 24xx digital series have no analog
mode.

--- In hp_agilent_equipment@..., Yuting Wan <ywan03@...>
wrote:

Modern DSO has much higher sampling rate, called over sampling. I

remember I read this in the 90s oversampling can greatly eliminate the
phase distortion caused by a stiff anti-aliasing filter, instead a more
relaxed filter is used under oversampling.

Tim
On 01/01/2013, at 4:40 AM, erich_schlecht wrote:

Speaking of old scopes, the first digital scope I got circa 1986 was a

Tek with a sample rate around 20 or 50 Msps, bandwidth 100 or 200 MHz. For
repetitive signals it dithered the sample clock to reconstruct signals well
above the Nyquist frequency over many cycles. It couldn't see fast single
event signals, of course.

It also had a pure analog mode. For the time, it was a pretty decent

instrument.Unfortunately, I've long forgotten the model number, but it
looked like a 24xx series.

Erich

------------------------------------

Yahoo! Groups Links

Modern DSO has much higher sampling rate, called over sampling. I remember I read this in the 90s oversampling can greatly eliminate the phase distortion caused by a stiff anti-aliasing filter, instead a more relaxed filter is used under oversampling.

Tim
On 01/01/2013, at 4:40 AM, erich_schlecht wrote:

Speaking of old scopes, the first digital scope I got circa 1986 was a Tek with a sample rate around 20 or 50 Msps, bandwidth 100 or 200 MHz. For repetitive signals it dithered the sample clock to reconstruct signals well above the Nyquist frequency over many cycles. It couldn't see fast single event signals, of course.

It also had a pure analog mode. For the time, it was a pretty decent instrument.Unfortunately, I've long forgotten the model number, but it looked like a 24xx series.

Erich

Speaking of old scopes, the first digital scope I got circa 1986 was a Tek with a sample rate around 20 or 50 Msps, bandwidth 100 or 200 MHz. For repetitive signals it dithered the sample clock to reconstruct signals well above the Nyquist frequency over many cycles. It couldn't see fast single event signals, of course.

It also had a pure analog mode. For the time, it was a pretty decent instrument.Unfortunately, I've long forgotten the model number, but it looked like a 24xx series.

Erich

--- In hp_agilent_equipment@..., Chuck Harris <cfharris@...> wrote:

Hi Peter,

As I said, "any competently designed DSO". An analog scope gives
you the full vertical bandwidth regardless of the timebase setting.
A competently designed DSO should also.

You can be a bit flexible about that requirement, though. If the
aliasing effects are too fast to see at a particular timebase
setting, it would be ok to slow the sample rate until they are
only marginally too fast to see.

-Chuck Harris

Peter Gottlieb wrote:

But it's not just filtering above the Nyquist. There are other ways a sampling
digital scope can give you a wrong picture of reality. If all of these scopes
ran their digitizers constantly at full rate, watched for envelope effects and
so forth they would go a long way towards eliminating these unwanted erroneous
displays.

Peter


On 12/31/2012 10:56 AM, Chuck Harris wrote:

If by "trust" you mean see things faster than the Nyquist
limit, I fully agree.

I recall Lothar telling me several years ago that these O-rings were available for watch repair. -Eric

--- In hp_agilent_equipment@..., Steve Krull <Steve-Krull@> wrote:

I've got an attenuator for an 8672A that suffers from failed O-rings. In
this case it appears someone used something nasty to try to clean the
spring contacts and attenuator pads and whatever was used dripped down
the nylon plungers and turned all the O-rings to a mushy black paste.
Not a single survivor. Agilent no longer stocks them but provides the
dimensions and material type so I just need to research a source.

Steve

I've got an attenuator for an 8672A that suffers from failed O-rings. In
this case it appears someone used something nasty to try to clean the
spring contacts and attenuator pads and whatever was used dripped down
the nylon plungers and turned all the O-rings to a mushy black paste.
Not a single survivor. Agilent no longer stocks them but provides the
dimensions and material type so I just need to research a source.

Steve

Hi Folks

Anybody have any advice on getting a decade divider to go in front of a 5385
counter so I can measure frequencies around X band. Divide by 8 is not
acceptable. Clugey is acceptable as long as it is cheap.
Eric