开云体育

On Sun, 30 Dec 2012 14:40:09 -0800 (PST), you wrote:

Yes.

The Tek DPO, based on the 7704 (or possibly the 7704A) was the first
AFAIK.

That would be the Tektronix 7D20 (as a combined timebase/dual channel
scope) which was intended to run at least in the 760x series.

No. The DPO uses an LSI-11 as a 'puter. I have one.

The 7D20 is much later. The DPO separated thye PI and display sections of
the scope with a digitizer unit, maybe the d7001.

10 MHz bandwidth for the 468, and about 70 Mhz bandwidth for the 7D20
(IIRC).

Are you sure about the 70 MHz on the 7D20. I doubt that it was that good
one shot.

There were also some Transient Data Digitizers that used specialized
double ended CRTs, I think.

That would be the 7912, IIRC, which used the 7B16P and 7B90P
programmable plugins as a minimum. Never had one, but it looks like
an interesting possibility to have.

There were actually several different ones.

-John

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

Harvey

Vladan-

I got the cover off without trouble. By testing I note that when the middle (40 dB) switch is triggered, the output turns off completely. As far as I can determine, the plunger is actuated correctly. Moving it manually also results in the loss of output. The other two switches (10 and 20 dB) work fine.

I suppose that either the switch contacts are failing or the 40 dB resistor is open. Apparently both switches and resistors are inside the two-piece machined steel block. All the RF path seems to be inside. It would appear that I need to get this apart, but I am not sure how to. There are quite a few screws attaching the solenoids to the steel block, but I am not seeing what is holding the steel block together.

This attenuator has a PC board with several chips on it. I have noticed before that HP seems to have adopted the philosophy that if 2 chips are good, then 20 must be better. It is a surprisingly complicated device.

Any advice??

Thanks!

Jim



At 08:45 PM 12/30/2012, you wrote:

Jim,

You will have to pull out the front panel. There are a few flat-head screws along the front frame (top and bottom, maybe even sides - don't remember any more). Once you remove them, the front panel pulls out of the front frame, though it still has a couple of umbilicals attached. From there, you will see how to get the attenuator out.

The most common failure has to do with the tiny rubber O-rings on the plungers. This is all delicate magnifier work. Do not succumb to the "magnetized housing" myth of repair. The attenuator depends on magnets for latching. It's just that when the O-ring fails and either splits or falls off, the mechanism shifts closer to the magnet and the force of magnetic attraction increases. At that point, the solenoid (which has nothing wrong with it) no longer can pull the parts away from the magnet (which also has nothing wrong with it). This is when some people decide that demagnetizing everything will fix things. If you have the patience, study the design, it's interesting. The goal was to use current only when switching a stage is required and then interrupt all current to the coils as soon as the task is accomplished.

You open the attenuator by removing one of the end caps and sliding the U-shaped steel cover off. It comes off relatively easily. Don't try to remove it by prying with screwdrivers.

There are two versions of the attenuators. The older ones are all electro-mechanical, the newer ones have a p.c. board with some ICs on them.

Good luck, they can be fixed, but the work requires some patience and you may have to order no O-rings from Agilent.

Vladan

--- In <mailto:hp_agilent_equipment%40yahoogroups.com>hp_agilent_equipment@..., Jim Schatzman <james.schatzman@...> wrote:

All-

O.k., I have an 8568B with a bad attenuator (40dB and below are dead).

My big question is - how do you get access to the attenuator to take it apart and clean/repair? It seems to be really buried under a maze of stuff...

Thanks!

Jim



At 01:44 PM 12/27/2012, you wrote:

Thanks a lot of this.

I just managed to get a 8568B that attenuator has some problems with the 20dB step.

These instructions encourage me to open the attenuator and clean the pads with isopropyl alcohol. Now it works fine.

I didn't made a testped for it i just cold moved carefully the solenoids and checked with dg8saq vnwa the attenuator pads. I'm not sure is the solenoid stuck or is it actually a contact failure.

The instructions how to disaasemble the attenuator was very accurate and easy to follow.

BR, Jarmo

[Non-text portions of this message have been removed]

[Non-text portions of this message have been removed]

Jim,

You will have to pull out the front panel. There are a few flat-head screws along the front frame (top and bottom, maybe even sides - don't remember any more). Once you remove them, the front panel pulls out of the front frame, though it still has a couple of umbilicals attached. From there, you will see how to get the attenuator out.

The most common failure has to do with the tiny rubber O-rings on the plungers. This is all delicate magnifier work. Do not succumb to the "magnetized housing" myth of repair. The attenuator depends on magnets for latching. It's just that when the O-ring fails and either splits or falls off, the mechanism shifts closer to the magnet and the force of magnetic attraction increases. At that point, the solenoid (which has nothing wrong with it) no longer can pull the parts away from the magnet (which also has nothing wrong with it). This is when some people decide that demagnetizing everything will fix things. If you have the patience, study the design, it's interesting. The goal was to use current only when switching a stage is required and then interrupt all current to the coils as soon as the task is accomplished.

You open the attenuator by removing one of the end caps and sliding the U-shaped steel cover off. It comes off relatively easily. Don't try to remove it by prying with screwdrivers.

There are two versions of the attenuators. The older ones are all electro-mechanical, the newer ones have a p.c. board with some ICs on them.

Good luck, they can be fixed, but the work requires some patience and you may have to order no O-rings from Agilent.

Vladan

All-

When all else fails, read the manual! The 8568B service manual says to remove the front panel. That is easy and the attenuator is then right in front of you. I was trying to get at it from the interior....

Jim



At 01:44 PM 12/27/2012, you wrote:

Thanks a lot of this.

I just managed to get a 8568B that attenuator has some problems with the 20dB step.

These instructions encourage me to open the attenuator and clean the pads with isopropyl alcohol. Now it works fine.

I didn't made a testped for it i just cold moved carefully the solenoids and checked with dg8saq vnwa the attenuator pads. I'm not sure is the solenoid stuck or is it actually a contact failure.

The instructions how to disaasemble the attenuator was very accurate and easy to follow.

BR, Jarmo

[Non-text portions of this message have been removed]

Hi Christian,

There is very little information available about the firmware revisions for the 8970B. Certainly, if you have a standard unit (1600 MHz high end), you will not want any version of the firmware that was made for the H1800 or 020 options. I am pretty sure the roms for the units with extended range had different part numbers.

I have recently looked at the manual change pages for the 8970B and I didn't see any obvious signs that the hardware changed so much as to require new firmware. But ... I don't know that for a fact. If you are concerned about firmware compatibility, your odds of success are higher if you choose a firmware version that is close to your meter's vintage than a very late date code. If I were in your situation, I would get a pair of spare roms and try the newer firmware to see what happens. Just don't erase the originals and you can always go back.

I think the text file that is included with the 8005 roms gives the full serial no. of the donor instrument. Compare the last four digits of that unit's SN with yours and you will know how far apart they were.

Vladan

I have a Phillips PM3375 out in the workshop, and I'm using an HP54601A in the radio shack to monitor transmitter RF; with accurate directional couplers, I can calibrate watt meters. Not far away, on a very heavy-duty turntable, sits a 141T with 8552B and 8554B plugins. I don't actually use that very much, mainly to check harmonics.

I was using a Leader 15 MHz analog scope with a cracked graticule (now at our radio club) which served almost as well,at least letting me know if the sound card output has reset itself to 100 percent again, but the digitizer lets me capture peaks and know what I have.

The 141T has an analog storage CRT, writing to a charged grid then flooding the CRT with electrons. It's finicky, and the display is perishable, so my "digital" storage is a digital camera on a tripod.
Couldn't afford Polaroid film now even if I had the camera.


Cortland
KA5S

On Sun, 30 Dec 2012 21:34:36 -0500, Harvey White
<madyn@...> wrote:

On Sun, 30 Dec 2012 19:57:46 -0600, you wrote:

On Sun, 30 Dec 2012 20:03:41 -0500, Harvey White
<madyn@...> wrote:

On Sun, 30 Dec 2012 14:40:09 -0800 (PST), you wrote:

Yes.

The Tek DPO, based on the 7704 (or possibly the 7704A) was the first AFAIK.

That would be the Tektronix 7D20 (as a combined timebase/dual channel
scope) which was intended to run at least in the 760x series.

I think he means the Tektronix P7001 which attaches to a 7704A between
the display and mainframe:

That's a new one on me.

The P7001 is not as common as the 7854 which is the same basic design.
I have seen one or two in the past couple of years for sale but would
prefer the 7854 if only because it is relatively common.

Then there was the magnificent 7854. IMO, this is the BEST lab 'scope
ever made. It is still my 'scope of choice.

and I'd like to have one, but that's still out there, somewhere.....

The downfall of both of these units was the relatively slow A/D by modern
standards.

10 MHz bandwidth for the 468, and about 70 Mhz bandwidth for the 7D20
(IIRC).

The 7D20 uses a CCD based sampler like the 2430 series of DSOs.

Yep, and they do mention that the CCD has absolutely no static
immunity.

Those CCD sampling DSOs occupy a weird place in history. Since
Tektronix had access to 100 MS/sec 8 bit ADCs like those used in the
2232, I am not sure why they used them for as long as they did. My
guess is that the CCDs were easier to interleave then the digital side
of the fast ADCs. The 2232 is limited to 50 MS/sec in chop mode
because of that.

All-

O.k., I have an 8568B with a bad attenuator (40dB and below are dead).

My big question is - how do you get access to the attenuator to take it apart and clean/repair? It seems to be really buried under a maze of stuff...

Thanks!

Jim

I will admit that there is not a complete and free 2230 service manual
available online but ArtekManuals and others have scanned service
manuals available inexpensively. Maintenance is made easier since the
2230 shares so many parts with the ubiquitous 2232, 2235A, 2221, and
others in that series.

The 2232 is easier to use and repair than the 2230 but generally at
least twice as expensive.

On Mon, 31 Dec 2012 13:20:51 +1100, Yuting Wan
<ywan03@...> wrote:

I've looked into the Tek 2230 before and found it's performance and ease of use is nowhere near the Fluke PM3390. It's also very hard to get detailed info of 2230. For Fluke PM3390, you can purchase a service manual for very little but some parts are very hard to get. Then you end up getting a few units and start cannibalise them.


On 31/12/2012, at 12:46 PM, David wrote:

Tektronix made a number of oscilloscopes like you describe including
some interesting hybrids. Phillips made some oscilloscopes
contemporary to the Tektronix 2232 series but supporting 4 channels,
200 MHz bandwidths, and with more features. I have avoided them
because of the lack of service information.

Here are the Tektronix analog oscilloscopes that include at least some
form of digital storage roughly in order:

The Tektronix P7001 attaches to a 7704A (200 MHz) between the display
and mainframe:




The Tektronix 7854 (400 MHz) works like a P7001 but is all one unit.
It uses the analog timebase for digital storage which has the side
effect of allowing equivalent time sampling without a clock delay
timer. With a 7B87 timebase, it also supports low speed real time
single shot captures with pretrigger. The user interface and
programming make a great companion to your HP RPN calculator.

The Tektronix 468 is a 465 (100 MHz) with a digitizer module (8 bits
and 25 MS/sec) mounted on top. It apparently supports some type of
equivalent time sampling called "jitter correction" in the manual but
is still limited to a 10 MHz bandwidth in storage mode.

The 2232 (100 MHz 100 MS/sec), 2224 (60 MHz and 100 MS/sec), and 2230
(100 MHz 20 MS/sec) support real time sampling, equivalent time
sampling, and peak detection. The other 22xx series oscilloscopes
operate with real time sampling only and no peak detection. The 2221A
(100 MHz 100 MS/sec) and 2221 (100 MHz 20 MS/sec) are single timebase
versions of the 2232 and 2230.

My 2230 is my go to oscilloscope although if I need to backup
waveforms for later, I use my 2232 instead because my 2230 does not
have the memory backup option.

The Tektronix 2252 is a 2247A (100 MHz 4 channels) with limited
digital storage using sequential sampling.

Some of the above oscilloscopes have much higher storage resolution
than you would expect. The 2252 is 12 bits vertical. The 7854 is 10
bits horizontal and vertical.

Of the above, the 2230 and 2232 are the most common and easiest to
maintain. The 2221 and 2221A are just as easy to maintain because
they are essentially the same as the 2230 and 2232. The 7854 is the
most versatile but lacks real time operation and peak detection.

On Sun, 30 Dec 2012 17:09:00 -0500, Bryce Schroeder
<bryce.schroeder@...> wrote:

Did HP (or Tek, or anyone for that matter) make any oscilloscopes that were
basically analog scopes, but with the ability to digitize what is seen on
the display (at some lesser resolution than what you might be able to see
on the actual display, presumably)? As I understand it - which may
admittedly be incorrect - some of the older spectrum analyzers are sort of
like this.

So, this hypothetical oscilloscope type would basically have the analog
display "in parallel" with an ADC, such that you could look at the screen
and see the waveform displayed like an analog scope, or use the ADC to see
a digital approximation. (Presumably the digital stuff could also be used
to capture and store one-shot things like a regular digital oscilloscope.)
This would be opposed to a purely digital oscilloscope where the ADC and
memory is always between the signal and the display.

The reason for doing this, of course, would be that the analog display
would be "higher resolution" than the ability of the ADC, but the digital
stuff would enable automation and storage ability. This would stop being
important once digital was high-enough resolution to look more or less
indistinguishable to humans.

Did such "hybrid" oscilloscopes ever exist? If so, what are some model
numbers?

Thanks for your time - sorry if that was a bit lengthy.

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

Yahoo! Groups Links

Back in the mid 80's I remember using the Tek 7D20 digitizer plug-in. It only had a 20MHz sampling rate with an over sampling bandwidth of 70MHz. It was OK for sampling in the audio range but not go at TTL rates.

On Sun, 30 Dec 2012 19:57:46 -0600, you wrote:

On Sun, 30 Dec 2012 20:03:41 -0500, Harvey White
<madyn@...> wrote:

On Sun, 30 Dec 2012 14:40:09 -0800 (PST), you wrote:

Yes.

The Tek DPO, based on the 7704 (or possibly the 7704A) was the first AFAIK.

That would be the Tektronix 7D20 (as a combined timebase/dual channel
scope) which was intended to run at least in the 760x series.

I think he means the Tektronix P7001 which attaches to a 7704A between
the display and mainframe:

That's a new one on me.

Then there was the magnificent 7854. IMO, this is the BEST lab 'scope
ever made. It is still my 'scope of choice.

and I'd like to have one, but that's still out there, somewhere.....

The downfall of both of these units was the relatively slow A/D by modern
standards.

10 MHz bandwidth for the 468, and about 70 Mhz bandwidth for the 7D20
(IIRC).

The 7D20 uses a CCD based sampler like the 2430 series of DSOs.

Yep, and they do mention that the CCD has absolutely no static
immunity.

Harvey

The A3 Directional Bridge (Port 1) has a pair of toroids on each side. The pair with brown "wire" is wound with coax. Is the grey "wire" on the other pair also coax?

I have an 85046A S parameter test set with identical faults on both directional couplers (I did not believe it either) and looking at the inside of one seems suspicious as the "Bridge" does not appear balanced in construction. Each end of the grey "wire" out of the toroid pair ends in a solder blob. I cannot see if it is/was a coax.

I looked at the coupler on the VNA from input to output after doing a B/R thru calibration and it starts at -20dB and climbs exponentially to -2dB at 1GHz, stays there then drops off a bit at the 3GHz end. I used a Transmission/Reflection Test Set with "Test" feeding the Transfer Switch end and with the Port 1 end on "B" on the 8753B. Coupler port terminated but it does not change anything.

When connected to the 8753B S11 and S22 with the ports open have identical shapes with two humps starting at -70dB rising to -15dB,and separated by a notch going to -55dB around 2GHz. When terminated they both rise exponentially from -70dB to -20dB at around 2GHz then flatten out. (this has nothing to do with the other thread I started, this is a different 85046A)

I have eliminated the splitter, attenuator, and Transfer Switch because I thought it had to be something in common to both Directional Bridges. I am now looking at the A3 as a separate component.

I can send a full description .docx file with pictures if I have an email.

Main question: is the grey "wire" on the toroids a coax?

I've looked into the Tek 2230 before and found it's performance and ease of use is nowhere near the Fluke PM3390. It's also very hard to get detailed info of 2230. For Fluke PM3390, you can purchase a service manual for very little but some parts are very hard to get. Then you end up getting a few units and start cannibalise them.


On 31/12/2012, at 12:46 PM, David wrote:

Tektronix made a number of oscilloscopes like you describe including
some interesting hybrids. Phillips made some oscilloscopes
contemporary to the Tektronix 2232 series but supporting 4 channels,
200 MHz bandwidths, and with more features. I have avoided them
because of the lack of service information.

Here are the Tektronix analog oscilloscopes that include at least some
form of digital storage roughly in order:

The Tektronix P7001 attaches to a 7704A (200 MHz) between the display
and mainframe:




The Tektronix 7854 (400 MHz) works like a P7001 but is all one unit.
It uses the analog timebase for digital storage which has the side
effect of allowing equivalent time sampling without a clock delay
timer. With a 7B87 timebase, it also supports low speed real time
single shot captures with pretrigger. The user interface and
programming make a great companion to your HP RPN calculator.

The Tektronix 468 is a 465 (100 MHz) with a digitizer module (8 bits
and 25 MS/sec) mounted on top. It apparently supports some type of
equivalent time sampling called "jitter correction" in the manual but
is still limited to a 10 MHz bandwidth in storage mode.

The 2232 (100 MHz 100 MS/sec), 2224 (60 MHz and 100 MS/sec), and 2230
(100 MHz 20 MS/sec) support real time sampling, equivalent time
sampling, and peak detection. The other 22xx series oscilloscopes
operate with real time sampling only and no peak detection. The 2221A
(100 MHz 100 MS/sec) and 2221 (100 MHz 20 MS/sec) are single timebase
versions of the 2232 and 2230.

My 2230 is my go to oscilloscope although if I need to backup
waveforms for later, I use my 2232 instead because my 2230 does not
have the memory backup option.

The Tektronix 2252 is a 2247A (100 MHz 4 channels) with limited
digital storage using sequential sampling.

Some of the above oscilloscopes have much higher storage resolution
than you would expect. The 2252 is 12 bits vertical. The 7854 is 10
bits horizontal and vertical.

Of the above, the 2230 and 2232 are the most common and easiest to
maintain. The 2221 and 2221A are just as easy to maintain because
they are essentially the same as the 2230 and 2232. The 7854 is the
most versatile but lacks real time operation and peak detection.

On Sun, 30 Dec 2012 17:09:00 -0500, Bryce Schroeder
<bryce.schroeder@...> wrote:

Did HP (or Tek, or anyone for that matter) make any oscilloscopes that were
basically analog scopes, but with the ability to digitize what is seen on
the display (at some lesser resolution than what you might be able to see
on the actual display, presumably)? As I understand it - which may
admittedly be incorrect - some of the older spectrum analyzers are sort of
like this.

So, this hypothetical oscilloscope type would basically have the analog
display "in parallel" with an ADC, such that you could look at the screen
and see the waveform displayed like an analog scope, or use the ADC to see
a digital approximation. (Presumably the digital stuff could also be used
to capture and store one-shot things like a regular digital oscilloscope.)
This would be opposed to a purely digital oscilloscope where the ADC and
memory is always between the signal and the display.

The reason for doing this, of course, would be that the analog display
would be "higher resolution" than the ability of the ADC, but the digital
stuff would enable automation and storage ability. This would stop being
important once digital was high-enough resolution to look more or less
indistinguishable to humans.

Did such "hybrid" oscilloscopes ever exist? If so, what are some model
numbers?

Thanks for your time - sorry if that was a bit lengthy.

[Non-text portions of this message have been removed]

Hi Vlandan,
Thank you for your answer.
Do you think version 8005x is compatible with all 8970B's? (including mine: S/N 2721Axxxx)
Seems strange that this number is a five-digit where mine has a four-digit version...
Christian

Le 31 d閏. 2012 � 02:48, pianovt a 閏rit :

Hi Christian,

I also see a newer version (standard, not Opt 020) there:

)_ROM_Images_and_Drivers/HP_8970B/HP_8970B_08970_8005X.zip

Vladan

--- In hp_agilent_equipment@..., f1gwr <f1gwr@...> wrote:
...

Please note K04BB's site only holds 2705 version of the firmware ROM, see:
)_ROM_Images_and_Drivers/HP_8970B

So I'm looking for 2800 release or later.

Thanks for your help,
Christian
...

[Non-text portions of this message have been removed]

On Sun, 30 Dec 2012 20:03:41 -0500, Harvey White
<madyn@...> wrote:

On Sun, 30 Dec 2012 14:40:09 -0800 (PST), you wrote:

Yes.

The Tek DPO, based on the 7704 (or possibly the 7704A) was the first AFAIK.

That would be the Tektronix 7D20 (as a combined timebase/dual channel
scope) which was intended to run at least in the 760x series.

I think he means the Tektronix P7001 which attaches to a 7704A between
the display and mainframe:

Then there was the magnificent 7854. IMO, this is the BEST lab 'scope
ever made. It is still my 'scope of choice.

and I'd like to have one, but that's still out there, somewhere.....

The downfall of both of these units was the relatively slow A/D by modern
standards.

10 MHz bandwidth for the 468, and about 70 Mhz bandwidth for the 7D20
(IIRC).

The 7D20 uses a CCD based sampler like the 2430 series of DSOs.

Thanks for all this information. Certainly there are more options to
consider than I thought there would be if it existed at all.

Hi Christian,

I also see a newer version (standard, not Opt 020) there:

)_ROM_Images_and_Drivers/HP_8970B/HP_8970B_08970_8005X.zip

Vladan

Tektronix made a number of oscilloscopes like you describe including
some interesting hybrids. Phillips made some oscilloscopes
contemporary to the Tektronix 2232 series but supporting 4 channels,
200 MHz bandwidths, and with more features. I have avoided them
because of the lack of service information.

Here are the Tektronix analog oscilloscopes that include at least some
form of digital storage roughly in order:

The Tektronix P7001 attaches to a 7704A (200 MHz) between the display
and mainframe:




The Tektronix 7854 (400 MHz) works like a P7001 but is all one unit.
It uses the analog timebase for digital storage which has the side
effect of allowing equivalent time sampling without a clock delay
timer. With a 7B87 timebase, it also supports low speed real time
single shot captures with pretrigger. The user interface and
programming make a great companion to your HP RPN calculator.

The Tektronix 468 is a 465 (100 MHz) with a digitizer module (8 bits
and 25 MS/sec) mounted on top. It apparently supports some type of
equivalent time sampling called "jitter correction" in the manual but
is still limited to a 10 MHz bandwidth in storage mode.

The 2232 (100 MHz 100 MS/sec), 2224 (60 MHz and 100 MS/sec), and 2230
(100 MHz 20 MS/sec) support real time sampling, equivalent time
sampling, and peak detection. The other 22xx series oscilloscopes
operate with real time sampling only and no peak detection. The 2221A
(100 MHz 100 MS/sec) and 2221 (100 MHz 20 MS/sec) are single timebase
versions of the 2232 and 2230.

My 2230 is my go to oscilloscope although if I need to backup
waveforms for later, I use my 2232 instead because my 2230 does not
have the memory backup option.

The Tektronix 2252 is a 2247A (100 MHz 4 channels) with limited
digital storage using sequential sampling.

Some of the above oscilloscopes have much higher storage resolution
than you would expect. The 2252 is 12 bits vertical. The 7854 is 10
bits horizontal and vertical.

Of the above, the 2230 and 2232 are the most common and easiest to
maintain. The 2221 and 2221A are just as easy to maintain because
they are essentially the same as the 2230 and 2232. The 7854 is the
most versatile but lacks real time operation and peak detection.

On Sun, 30 Dec 2012 17:09:00 -0500, Bryce Schroeder
<bryce.schroeder@...> wrote:

Did HP (or Tek, or anyone for that matter) make any oscilloscopes that were
basically analog scopes, but with the ability to digitize what is seen on
the display (at some lesser resolution than what you might be able to see
on the actual display, presumably)? As I understand it - which may
admittedly be incorrect - some of the older spectrum analyzers are sort of
like this.

So, this hypothetical oscilloscope type would basically have the analog
display "in parallel" with an ADC, such that you could look at the screen
and see the waveform displayed like an analog scope, or use the ADC to see
a digital approximation. (Presumably the digital stuff could also be used
to capture and store one-shot things like a regular digital oscilloscope.)
This would be opposed to a purely digital oscilloscope where the ADC and
memory is always between the signal and the display.

The reason for doing this, of course, would be that the analog display
would be "higher resolution" than the ability of the ADC, but the digital
stuff would enable automation and storage ability. This would stop being
important once digital was high-enough resolution to look more or less
indistinguishable to humans.

Did such "hybrid" oscilloscopes ever exist? If so, what are some model
numbers?

Thanks for your time - sorry if that was a bit lengthy.

On Sun, 30 Dec 2012 14:40:09 -0800 (PST), you wrote:

Yes.

The Tek DPO, based on the 7704 (or possibly the 7704A) was the first AFAIK.

That would be the Tektronix 7D20 (as a combined timebase/dual channel
scope) which was intended to run at least in the 760x series.

Then there was the magnificent 7854. IMO, this is the BEST lab 'scope
ever made. It is still my 'scope of choice.

and I'd like to have one, but that's still out there, somewhere.....

The downfall of both of these units was the relatively slow A/D by modern
standards.

10 MHz bandwidth for the 468, and about 70 Mhz bandwidth for the 7D20
(IIRC).

There were also some Transient Data Digitizers that used specialized
double ended CRTs, I think.

That would be the 7912, IIRC, which used the 7B16P and 7B90P
programmable plugins as a minimum. Never had one, but it looks like
an interesting possibility to have.

Harvey

-John

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

Did HP (or Tek, or anyone for that matter) make any oscilloscopes that
were
basically analog scopes, but with the ability to digitize what is seen on
the display (at some lesser resolution than what you might be able to see
on the actual display, presumably)? As I understand it - which may
admittedly be incorrect - some of the older spectrum analyzers are sort of
like this.

So, this hypothetical oscilloscope type would basically have the analog
display "in parallel" with an ADC, such that you could look at the screen
and see the waveform displayed like an analog scope, or use the ADC to see
a digital approximation. (Presumably the digital stuff could also be used
to capture and store one-shot things like a regular digital oscilloscope.)
This would be opposed to a purely digital oscilloscope where the ADC and
memory is always between the signal and the display.

The reason for doing this, of course, would be that the analog display
would be "higher resolution" than the ability of the ADC, but the digital
stuff would enable automation and storage ability. This would stop being
important once digital was high-enough resolution to look more or less
indistinguishable to humans.

Did such "hybrid" oscilloscopes ever exist? If so, what are some model
numbers?

Thanks for your time - sorry if that was a bit lengthy.

On Sun, 30 Dec 2012 17:09:00 -0500, you wrote:

Did HP (or Tek, or anyone for that matter) make any oscilloscopes that were
basically analog scopes, but with the ability to digitize what is seen on
the display (at some lesser resolution than what you might be able to see
on the actual display, presumably)? As I understand it - which may
admittedly be incorrect - some of the older spectrum analyzers are sort of
like this.

Look at the Tek 468 scope. It had a digital backpack, capable of
about 10 Mhz bandwidth. The processor was (IIRC) and 8051 style, with
limited memory, etc.

The basic scope was a 465, a 100 Mhz dual channel scope. The
circuitry was an add-on backpack that could take over the display
(without the processor running, the scope won't work). The amount of
digital controls available were minimal, but allowed you to measure
waveform extensions (amplitude) and timing conditions with several
manually adjustable cursors. IIRC there was one memory.

A more modern processor would allow much greater capability, and one
of these days, I might just make a plugin to make that happen.

If you look at the scope, there's a digital backpack with an add-on
hood for the display, and an extension with the processor boards. The
scope was capable of running the HPIB interface as a talker only.

Other scopes, say in the 24xx series for Tektronix, are completely
digital.

Harvey

So, this hypothetical oscilloscope type would basically have the analog
display "in parallel" with an ADC, such that you could look at the screen
and see the waveform displayed like an analog scope, or use the ADC to see
a digital approximation. (Presumably the digital stuff could also be used
to capture and store one-shot things like a regular digital oscilloscope.)
This would be opposed to a purely digital oscilloscope where the ADC and
memory is always between the signal and the display.

The reason for doing this, of course, would be that the analog display
would be "higher resolution" than the ability of the ADC, but the digital
stuff would enable automation and storage ability. This would stop being
important once digital was high-enough resolution to look more or less
indistinguishable to humans.

Did such "hybrid" oscilloscopes ever exist? If so, what are some model
numbers?

Thanks for your time - sorry if that was a bit lengthy.