Tektronix 1-3 GHz touch screen color DSO back in 1989 !

[snoopy]

Do people know that Tek had a touch screen, full color DSO with 10 bit ADC back in 1989 ? The 11403 had 1GHz bandwidth and the later 11403A had 3GHz bandwidth with 2GHz trigger bandwidth !! Anyone still using one of these old war horses ?



Check out this crusty old VHS video !!



Official datasheet

https://accusrc.com/uploads/datasheets/4351_11403A.pdf

http://w140.com/tekwiki/wiki/11403

[David Hess]

The real time sample rate was pretty low on these so they relied on equivalent time sampling and were not suitable for general purpose use.  There were other mainframes in this series which supported real time sampling at rates to support their 1 GHz bandwidth.

They are certainly impressive but this was the start of test instruments from Tektronix which lacked service documentation so relatively minor problems are major problems.

[bjcuizon]

The touchscreen is just a crt but with Multiplexed IR Leds and receivers on the sides of the crt. Basically, as you touch a certain point of the screen, you interrupt the IR light that is pointed at an IR receiver. Check this out:
Mr Carlson's Tek Scope

[Wuerstchenhund]

Do people know that Tek had a touch screen, full color DSO with 10 bit ADC back in 1989 ?

Yes, but these were sampling scopes, i.e. the real-time sample rate was abysmally low, so it was only really useful for  truly repetitive signals.

Also, touch screens weren't really new back then, computers with similar touch technology (IR LED scanners) have been available earlier than that.

The 11403 had 1GHz bandwidth and the later 11403A had 3GHz bandwidth with 2GHz trigger bandwidth !!

Yes, but so what? Back in 1989 HP had the 54120B which could go up to 50Ghz with the 54124A 4ch test set:



And Philips had the PM3340 2GHz 10bit 50kSa/s sampling scope back in 1988:

Anyone still using one of these old war horses ?

I doubt it. Sampling scopes are useful only in very few situations.

[David Hess]

Do people know that Tek had a touch screen, full color DSO with 10 bit ADC back in 1989 ?

Yes, but these were sampling scopes, i.e. the real-time sample rate was abysmally low, so it was only really useful for  truly repetitive signals.

The model being discussed is not a sampling oscilloscope.  There were 4 different series of these mainframes:

1. The 11400/11200 series are 20 MSamples/second and 10k with 10/9 bit resolution.
2. The DSA series are up to 2 GSamples/second and 32k with 8 bit resolution.
3. The 11300 series are 500 MHz MCP analog oscilloscopes with a lot of automatic amplitude and time measurement capability.  These did not use a touch screen.

These series all had at least dual independent timebase capability including dual delayed timebases.

Except for the "low cost" 11200, the above all took the same improved 7000 series vertical amplifier plug-ins.

4. The CSA/11800 series are sequential only sampling oscilloscopes operating at 200 kSamples/second and 8 bits with a bandwidth up to 40 or 50 GHz although the 20 GHz differential TDR/sampling head is probably the most interesting.
 
These used the SD series sampling heads.

[David Hess]

Anyone still using one of these old war horses ?

I considered getting one for their math capability.  These are one of the rare DSOs which can do real time vector network analysis using a step response input by differentiating the waveform and then displaying the magnitude and phase results from the FFT.  Modern DSOs which can do this are pretty rare and expensive; LeCroy has a bunch which can but I am not sure about anybody else and of course anything new will cost as much as a car.  Some of the old Tektronix TDS series can do it also.

I do not know of anybody still using these in a production or business environment.  Without service documentation they are difficult to maintain.

[Wuerstchenhund]

The model being discussed is not a sampling oscilloscope. There were 4 different series of these mainframes:

1. The 11400/11200 series are 20 MSamples/second and 10k with 10/9 bit resolution.

(emphasis mine)

So you're saying the 11403 which is shown on the picture isn't a sampling scope then? 

Although you're right re the DSA and CSA of course.

I do not know of anybody still using these in a production or business environment.  Without service documentation they are difficult to maintain.

Indeed, also by now pretty much all of them will have reached the end of their service lifespan.

[David Hess]

The model being discussed is not a sampling oscilloscope. There were 4 different series of these mainframes:

1. The 11400/11200 series are 20 MSamples/second and 10k with 10/9 bit resolution.

(emphasis mine)

So you're saying the 11403 which is shown on the picture isn't a sampling scope then? 

That is right; the 11403A shown is not a sampling oscilloscope.  Those vertical amplifier plug-ins are not sampling heads; they are an improved version of the 7000 series vertical amplifier plug-ins.

If that is not enough, then a 3 GHz bandwidth is pitiful by sampling oscilloscope standards of 20 years before the 11403A was produced and the 11403A's horizontal timing specifications are not consistent with a sampling oscilloscope.

[Hydron]

The User Reference: http://w140.com/tek_11402A_11403_user_ref.pdf
suggests that it's running at 20MSamples/s for one channel, dropping to 5 and 2.5 for 2/3 channels respectively. Also talks about "Equivalent Time Sampling", and that in single shot mode you only get one sample per 50ns.

Sorry, but it's not as exciting as it looks!

[Wuerstchenhund]

The model being discussed is not a sampling oscilloscope. There were 4 different series of these mainframes:

1. The 11400/11200 series are 20 MSamples/second and 10k with 10/9 bit resolution.

(emphasis mine)

So you're saying the 11403 which is shown on the picture isn't a sampling scope then? 

That is right; the 11403A shown is not a sampling oscilloscope.

Well, it doesn't look to be a true real-time scope either as 20 MSa/s is hardly enough to capture the 1Ghz BW of the 11A71 amplifier plugins on the photo.

The User Reference: http://w140.com/tek_11402A_11403_user_ref.pdf
suggests that it's running at 20MSamples/s for one channel, dropping to 5 and 2.5 for 2/3 channels respectively. Also talks about "Equivalent Time Sampling", and that in single shot mode you only get one sample per 50ns.

Ok, so that would give it a real-time BW of roughly 8Mhz in single channel mode, 2Mhz in two channel mode and 1Mhz in 3 channel mode, the full BW seems to be available in ETS mode only => looks a lot like a sampling scope.

[David Hess]

Well, it doesn't look to be a true real-time scope either as 20 MSa/s is hardly enough to capture the 1Ghz BW of the 11A71 amplifier plugins on the photo.

It is not a real time oscilloscope although the display is updated in real time for which 20 MSamples/second is plenty fast.  Tektronix never claimed it was a real time oscilloscope and DSOs with sample rates to support a Nyquist frequency greater than their bandwidth were the exception back then.

A 2230 is 100 MHz but only has a 20 MSample/second sampling rate but that does not make it a sampling oscilloscope.  A 7854 is 400 MHz but only has a 500 kSample/second sampling rate and might be considered the predecessor to the 11400 series but it is not a sampling oscilloscope either.

Ok, so that would give it a real-time BW of roughly 8Mhz in single channel mode, 2Mhz in two channel mode and 1Mhz in 3 channel mode, the full BW seems to be available in ETS mode only => looks a lot like a sampling scope.

So what distinguishes a sampling oscilloscope like the HP 54120B or Philips PM3340?

1. The sampling gate width is much less than the sampling gate time constant so the input sampling capacitance never has time to charge to the input voltage.  Another way to put this is that the sampling efficiency is below 1; usually it is a lot lower than 1.

2. As a consequence of 1 above, the input is roughly integrated over the sampling gate width producing a non-linear frequency response which follows the sin(x)/x curve with a first null at the reciprocal of the sampling gate width.  The familiar bandwidth = 0.35 / transition time rule does not apply.

3. To take advantage of the extremely high bandwidth this produces, there are no amplification stages between the sampler and the input.  (1) Usually there are no attenuation stages, delay lines, or other networks either.  Even a length of transmission line would compromise the potential bandwidth and often does.

(1) I know of one exception to this.  High input impedance samplers include a *low* frequency buffer before the sampling gate to remove accumulated charge (2) from any input capacitance which could not otherwise be discharged by the high input impedance.  High frequencies bypass this buffer.  As far as I know, Tektronix was the only one to make high input impedance samplers.

(2) I think the problem here was kickout from the sampling strobe charging the input capacitance causing the DC level to shift.  The Tektronix 50 kilohm samplers (Tektronix S-3) did not need a low frequency buffer but their 1 megohm samplers (Tektronix S-5) had one.

[snoopy]

I think the correct term is sub-sampling scope or equivalent time sampling which requires repetitive waveforms for equivalent sampling rates greater than the ADC sampling rate !

[David Hess]

I think the correct term is sub-sampling scope or equivalent time sampling which requires repetitive waveforms for equivalent sampling rates greater than the ADC sampling rate!

The correct term in place of what?

There is nothing to preclude DSOs (digital *storage* oscilloscopes) with real time sampling from using equivalent time sampling and many do to achieve even higher sampling rates so the term equivalent time sampling is ambiguous in distinguishing them.

[Johnny10]

I use the DSA602 and 602A as my everyday scopes.

[Wuerstchenhund]

Well, it doesn't look to be a true real-time scope either as 20 MSa/s is hardly enough to capture the 1Ghz BW of the 11A71 amplifier plugins on the photo.

It is not a real time oscilloscope although the display is updated in real time for which 20 MSamples/second is plenty fast.

A digital scope is either a real-time scope (aquires the waveform in real-time) or a sampling scope (i.e. it acquires the waveform over multiple acquisition passes).

A 2230 is 100 MHz but only has a 20 MSample/second sampling rate but that does not make it a sampling oscilloscope.  A 7854 is 400 MHz but only has a 500 kSample/second sampling rate and might be considered the predecessor to the 11400 series but it is not a sampling oscilloscope either.

Both the 2230 and the 7854 are are analog/digital combi-scopes which both reach their analog BW in repetitive sampling or analog mode only. You're right that none of them are true sampling scopes as they are combi scopes with an analog and a digital mode.

So what distinguishes a sampling oscilloscope like the HP 54120B or Philips PM3340?

[....]

Back in the old days, Tek used to define¹ a sampling scope as a scope that put the sampler in front of the amplifiers and a DSO as a scope that has amplifiers in front of the sampling stage. But this was never an universally accepted definition and wasn't even correct as not all sampling scopes followed Tek's definition (i.e. LeCroy WaveExpert and I believe also the Agilent/Keysight DCA-J/-X). HP also had some sampling scopes (i.e. the 54501A) which didn't follow Tek's rule as well.

Today, a sampling scope is generally considered a fully digital scope that acquires its waveform over subsequent acquisitions to reach its full analog BW. It doesn't matter what sampling technique is used (random or sequential), or how it is actually implemented.



¹ http://www.tek.com/sites/tek.com/files/media/media/resources/Sampling%20Oscilloscope%20Techniques.pdf

[David Hess]

A digital scope is either a real-time scope (aquires the waveform in real-time) or a sampling scope (i.e. it acquires the waveform over multiple acquisition passes).

Great!  So depending on the time/div setting and the input frequency content or bandwidth limiting, a DSO could be either.  Or maybe all of those DSOs which can operate either way are a myth.  That sure unambiguously simplifies things.

Both the 2230 and the 7854 are are analog/digital combi-scopes which both reach their analog BW in repetitive sampling or analog mode only. You're right that none of them are true sampling scopes as they are combi scopes with an analog and a digital mode.

The analog capability of these oscilloscope is irrelevant to the discussion.  Why bring it up?

And you are flat out wrong.  Neither of these oscilloscope are bandwidth limited below Nyquist.  The bandwidth for both is independent of sample rate.  This is easy to demonstrate by making an RMS noise measurement at different sample rates; lower sample rates which cause aliasing do not change the measured amplitude because the input bandwidth does not change.  The same measurement made of a sine wave or any other signal will produce the same result.  Some DSOs take advantage of this for their external calibration.

So what distinguishes a sampling oscilloscope like the HP 54120B or Philips PM3340?

Back in the old days, Tek used to define¹ a sampling scope as a scope that put the sampler in front of the amplifiers and a DSO as a scope that has amplifiers in front of the sampling stage. But this was never an universally accepted definition and wasn't even correct as not all sampling scopes followed Tek's definition (i.e. LeCroy WaveExpert and I believe also the Agilent/Keysight DCA-J/-X). HP also had some sampling scopes (i.e. the 54501A) which didn't follow Tek's rule as well.

Today, a sampling scope is generally considered a fully digital scope that acquires its waveform over subsequent acquisitions to reach its full analog BW. It doesn't matter what sampling technique is used (random or sequential), or how it is actually implemented.

If the marketing departments of other companies want to confuse the issue, that is up to them.  One reason they did so is because sampling oscilloscopes were traditionally difficult to use which is why the Tektronix 7S14/5S14 were designed with a built in delay line which is the primary thing limiting their performance.  The delay line was either external or optionally built into later Tektronix sampling oscilloscopes.  The Philips PM3340 has a delay line as well limiting its bandwidth to 2 GHz but making it much easier to use.

The HP 54501A is just another digital storage oscilloscope with analog triggering, a bandwidth of 100 MHz, and a sample rate of 10 MSamples/second making it comparable to a 2230 or the 11403A being discussed except for the exceptional bandwidth of the 11403A.  The comparison with the 11403A is apt because both HP and Tektronix fell into the "one big knob" user interface trap at that time.  In the future, one big knob will control everything!

The definition that Tektronix used has the virtue of being less confusing but even they later limited it to the product descriptions instead of the product names leading to things like "communications signal analyser", "digital signal analyser", and "digitizing oscilloscope" which could mean anything including a 7104 with a digital camera or a scan converter.

So what distinguishes the performance of a sampling oscilloscope from your examples?  Overload recovery time is one.  Sampling oscilloscope have no overload recovery time which is very useful or even indispensable for some measurements.  If I want the best overload recovery time, then I want a sampling oscilloscope no matter what they are calling it today.  Picotech makes some and they even call them sampling oscilloscopes.  Everything else they make is unambiguously real time but some of their Nyquist bandwidths are not much higher than their input bandwidths; I much prefer ETS support when that is the case.

[shteii01]

I have 11403 (not A) with 300 MHz plugin for personal use.  I have not done much with mine simply because of being busy.

Interesting side note.  I know from Tek Wiki about 1989 date.  But I also looked at the Tek catalogs and 11403 does not show in 1989... If I recall correctly it first shows in 1991 or '92.

[edavid]

That is right; the 11403A shown is not a sampling oscilloscope.  Those vertical amplifier plug-ins are not sampling heads; they are an improved version of the 7000 series vertical amplifier plug-ins.

Except for the 11A81 3GHz "presampler" vertical plugin.

(Does anyone know the presampling rate?)

[Jay_Diddy_B]

Hi,

I have a couple of DSA602A in my collection. They were probably the peak of this series of scopes.

Here is a screen shot:



At some point I wrote a C program to convert the image format to TIFF.

The 'A' version had a floppy, the non A didn't.

There were four 500M s/s digitizer, that gave you up to 2Gs/s with interleaving on a signal channel. Memory depth was 32k samples.

I also has 11801 with SD24 sampling heads. This is 20 GHz bandwidth, but 20 M sps.



There were later versions of the 11801 with color displays etc.

Regards,

Jay_Diddy_B

[Wuerstchenhund]

A digital scope is either a real-time scope (aquires the waveform in real-time) or a sampling scope (i.e. it acquires the waveform over multiple acquisition passes).

Great!  So depending on the time/div setting and the input frequency content or bandwidth limiting, a DSO could be either.  Or maybe all of those DSOs which can operate either way are a myth.  That sure unambiguously simplifies things.

Nope, that has nothing to do with settings. If a scope's max sample rate is sufficient to fully capture a signal at its rated BW in a single acquisition then it'sa real-time scope, if the max sample rate is insufficient then it's a sampling scope.

Both the 2230 and the 7854 are are analog/digital combi-scopes which both reach their analog BW in repetitive sampling or analog mode only. You're right that none of them are true sampling scopes as they are combi scopes with an analog and a digital mode.

The analog capability of these oscilloscope is irrelevant to the discussion.  Why bring it up?

Because these scopes are not fully digital scopes, they are combi-scopes, and on combi-scopes the digital part is usually kept simple compared to a contemporary digital scope. BTW, these were your examples, not mine.

And you are flat out wrong.  Neither of these oscilloscope are bandwidth limited below Nyquist.  The bandwidth for both is independent of sample rate.  This is easy to demonstrate by making an RMS noise measurement at different sample rates; lower sample rates which cause aliasing do not change the measured amplitude because the input bandwidth does not change.  The same measurement made of a sine wave or any other signal will produce the same result.  Some DSOs take advantage of this for their external calibration.

Yes, for something simple like noise measurement you're right, but then you don't need a scope for that. But that doesn't change the fact than neither of these boat anchors can (digitally) capture a true waveform of a signal at BW limit in a single acquisition. And for a scope, this is what counts.

If the marketing departments of other companies want to confuse the issue, that is up to them.

It's not confusing the issue, and it's not even driven by marketing.

This is what Keysight has to say:

Oscilloscopes fall into two groups, real-time oscilloscopes and sampling oscilloscopes (also called equivalent-time oscilloscopes) and it is important to understand the difference between the two types. Real-time oscilloscopes digitize a signal in real-time. Imagine a repetitive AC signal - the real-time oscilloscope acts like a camera, taking a series of frames of the signal during each cycle. The amount of frames the real-time oscilloscope captures depends upon the bandwidth, memory depth, and other attributes that we will soon discuss. A sampling oscilloscope, on the other hand, takes only one shot of the signal per cycle. By repeating this one shot, but at slightly different time frames, the sampling oscilloscope can reconstruct the signal with a high degree of accuracy.


I know that it's difficult to understand if you only know Tek because that's what you grew up with but they don't dictate the rules. Tek's definition back then was just that, something used by Tek based on the products they had at that time. Others however used a wider definition and for what it's worth this is what has been well established in the T&M field.

If I want the best overload recovery time, then I want a sampling oscilloscope no matter what they are calling it today.  Picotech makes some and they even call them sampling oscilloscopes.  Everything else they make is unambiguously real time but some of their Nyquist bandwidths are not much higher than their input bandwidths; I much prefer ETS support when that is the case.

Funny you mention PicoScope, as it seems they, too, disagree with your definition of Sampling Scope:

https://www.picotech.com/download/manuals/picoscope-9000-series-questions-and-answers.pdf:

A sampling oscilloscope is a special type of oscilloscope that uses a technique called sequential time sampling. This type of sampling is best suited to repetitive waveforms or those that are derived from a regular clock, such as serial data streams, clock waveforms and pulses in digital circuits, semiconductor test patterns, and amplifier pulse-response and rise-time tests. Signals like these tend to have very high bandwidths or high data rates. A sampling scope captures just one sample from one trigger event, typically a single cycle of the waveform or clock, and then repeats the process over a large number of cycles, varying the timing of the sample by a small increment from one sample to the next. The resulting collection of samples is then assembled into a picture of a typical cycle.

The advantage of a sampling scope is that even with input signals in the gigahertz range, the output of the sampler is at a much lower frequency, typically in the audio frequency band. This allows high-fidelity, low-frequency amplifiers and ADCs to be used to capture the signal. For example, the PicoScope 9300 Series can capture waveforms up to 20 GHz with high precision (16-bit resolution before scaling and processing). A real-time DSO that could capture a single cycle of the same 20 GHz waveform would be prohibitively expensive. For example the 20 GHz Agilent DSOX92004A, with a real-time sampling rate of 80 GS/s, has a base price of $177,000 — almost 12 times the price of the PicoScope 9301 while yielding only 4 samples per cycle. The PicoScope 9300 Series, sampling at an effective rate of 15 THz, would yield 750 samples per cycle of the same waveform.

A further benefit is that every sample point in the sequential sampling process is timed from the trigger event, and therefore from the original signal, rather than derived from an internal oscillator. Sampling jitter is therefore vastly reduced compared to the singletrigger, clock-derived sampling of the real-time DSO counterpart.


As I said, I understand where you're coming from but there's no point to fixate of a some 30 years old definition by a single manufacturer while constantly refering to ancient kit that has long been obsolete when the rest of the world has moved on.

[Leo Bodnar]

I regularly use CSA803A with SD-30 sampling head (end of 1980's - beginning of 1990's tech) and I have to say that 11800/CSA803 has the worst ever user interface I have experienced - hands down.  It's so experimental, random, confusing and eclectic.

Many operations include pressing a dedicated function button (often twice or more), then using the touch screen - or rather un-touch as it activates on finger withdrawal rather than touch (it does not detect push) and then rotating one of the large knobs.  It's just awful.  And yes, there are popup menus that blend with main screen and don't look like pop-up menus.

When i have tried to use it for the first time without warning, I spent about 30 minutes trying to modify timebase scale (what else do you do with the scope?) and in the end have given up and had to read the manual.  It turned out you had to press one of the function buttons twice to get to the right touchscreen and knob roll.

User interfaces with state are evil - you need to concentrate on what the screen is telling you and you can't just engage muscle memory to get you to the right place.

Oh, and the required channel is turned on/off by a tiny button on the head that has no tactile feedback, beep or any other way of knowing the state.  And yes, active channel has a dim blinking LED when it's on.  And yes, channel traces randomly change colour - the first one you happen to turn on is green, the second is red - irrespective of their physical location.

And it looks even worse than it feels.

Other than that it's awesome - its sampling technology is genius, frontend hybrids are beautifully laid out and you can get a 20GHz dual channel capability for $500.  You often get an instrument that has been hardly used - I guess due to reasons above.

Leo

[snoopy]

I think the correct term is sub-sampling scope or equivalent time sampling which requires repetitive waveforms for equivalent sampling rates greater than the ADC sampling rate!

The correct term in place of what?

There is nothing to preclude DSOs (digital *storage* oscilloscopes) with real time sampling from using equivalent time sampling and many do to achieve even higher sampling rates so the term equivalent time sampling is ambiguous in distinguishing them.

Hi David

I have a Tek TDS784 scope with a max ADC sampling rate of 4Gs/s. Anything above that it jumps into equivalent time sampling upto an effective sampling rate of 250 Gs/s. I'm fine with that As long as the signal is repetitive then equivalent time works perfectly

cheers

[Leo Bodnar]

I have posted few 11800/CSA803A frontend photos elsewhere but they are so beautiful they need to be here too. They don't care how they are called.
Leo

2 channel 20GHz sampling head


1 channel 40GHz sampling hybrid


Amplifier/offset/gain hybrid

[snoopy]

As I said, I understand where you're coming from but there's no point to fixate of a some 30 years old definition by a single manufacturer while constantly refering to ancient kit that has long been obsolete when the rest of the world has moved on.

I'm not sure if the world has moved on that much but according to Teks current lineup this 80GHz scope only has a 300kS/S sampling rate ! Perhaps that is a misprint but it does say "sampling scope" !

DSA8300 Digital Sampling Oscilloscope

The DSA8300 is a state-of-the-art Equivalent Time Sampling Oscilloscope that provides the highest fidelity measurement and analysis capabilities for Communications Signal Analysis, Serial Data Network Analysis, and Serial Data Link Analysis applications

With intrinsic jitter of less than 100 femtoseconds for extremely accurate device characterization, the DSA8300 Series provides comprehensive support for Optical Communications Standards, Time Domain Reflectometry and S-parameters. The DSA8300 Digital Sampling Oscilloscope is a complete high-speed PHY Layer testing platform for data communications from 155 Mb/sec to 400G PAM4.

http://www.tek.com/oscilloscope/dsa8300-sampling-oscilloscope

[Wuerstchenhund]

I'm not sure if the world has moved on that much but according to Teks current lineup this 80GHz scope only has a 300kS/S sampling rate ! Perhaps that is a misprint but it does say "sampling scope" !

Well, it is a sampling scope (i.e. a scope that uses multiple acquisitions to capture a waveform at analog BW), so why should it be a misprint?

The DSA8300 is basically a rehash of the DSA8000, which is based on the old TDS7000, so it's not exactly a new scope.

The real advantage of these scopes is in their low price compared to a real-time scope with the same BW.