I just used a fully digital oscilloscope for the first time

[Fried Chicken]

A Tektronix 1072b made for the education market.  I have to preface this:  Tektronix did a fantastic job with the interface.  The whole thing has an intuition to it that's fantastic.  The different colors are presented beautifully.  The fact that the knob has a light that indicates it needs to be used to change the settings is really clever.  Seeing the instantaneous measured frequency onscreen is similarly fantastic, and having a live FFT is the meat and potatoes.  Seeing a slow signal populate across the screen;  fantastic.  Same with the silent design.  No fan, no heat, just signals.

I say all that, because.  Wow is it a pain in the ass to use compared to my analog/solid state 2230.  Just from startup: the 2230 is on in about 5 seconds, the tektronix.... sit back, relax, there's a full boot sequence waiting to happen.  The interface on this digital thing is also a total disaster.  Manually setting a scale hunting a signal; it easily takes 5 seconds for the scale to update.  It's so easy to fly past the scale setting you want.  It's the same story with every. single. function.  There's no obvious indicator that the trigger is set to channel 2.  Trying to decipher that takes forever.  For a lot of things (AC vs Ground vs DC) requires going into a submenu that again, 5 seconds every time you do anything.  There is no guess and check, there is know, push, and wait.  Getting the beam centered around zero, I couldn't find a quick way to do it, and the ridiculously slow update meant trying to set it back where you wanted could put an autistic person over the edge and into the facilities.  There's no fast way of knowing what the current settings on the scope are.  There's no fast way to do anything with this scope.  On the 2230, it's a matter of seconds between starting the scope, click click click, boom I have the signal onscreen.  The buttons are all physical, the screen updates at the literal speed of light, if I get completely lost, I can quickly try each and every setting to see what/where it's at and go from there.  It's easier MUCH easier to pick up small nuances in a signal or what you're measuring.

Unreal, and this doesn't even get into the whole feel of using the thing.  I know that doesn't matter in this day and age, but it just didn't feel right.

Idk.  I was expecting better.  There's very little information on which digital scopes are actually good.  Can I hook my 2230 up to a computer and quickly/easily do any of the math I might want?

[ataradov]

No fan, no heat, just signals.

Unless those signals are digital and not periodic, then you are screwed.

You got a new tool and could not figure out how to use it in 5 minutes, so the tool must be bad?

[alm]

Idk.  I was expecting better.  There's very little information on which digital scopes are actually good.  Can I hook my 2230 up to a computer and quickly/easily do any of the math I might want?

Sure, if you have the RS-232 or GPIB option in the scope and a way to interface with them on your computer. You'll probably have to write your own software since I doubt there's anything that works on modern PCs that supports this scope. And you'll be limited to the 20 MS/s sample rate (about 4 MHz usable analog bandwidth), compared to 1 GS/s. Let's not talk about record length: record length is bare minimum on both the 2230 and TDS1000B series (this says more about the TBS1000B series than about how competitive the 2230 is, though).

There's no denying that the old front panel with one control per function was easier to use, until the number of features start exceeding what you can reasonably cram into a front panel. Have a look at the Tek 7000-series dual trace and dual timebase plugins for examples of that. That's why modern scopes (starting in the late eighties with the Tek 11000 series) use a different UI with a smaller set of controls and soft menus. Imagine what the front panel would look like with buttons for all 30 or so automated measurements, FFT window, average mode, trigger conditions, trend plotting, etc.

It certainly takes some getting used to to figure out which option is in which soft menu and which icon indicates what, but in the older Tek TDS scopes I used, which had a pretty similar UI, this was fairly easy after spending some time playing with them. Something like which channel the trigger is set on is generally clearly indicated. For example looking at the screenshot on page 1 of this datasheet, I'd say trigger is on channel 1 rising edge at a level of 2.62V (bottom right), and the displayed waveform is at 2V/div vertical and 1 ms/div horizontal. You're looking at the edges of the screen rather than at the knobs and enunciators on older scopes. But yes, changing trigger coupling takes more button presses than on an older scope with a physical knob/switch to do this. Especially if you are not still in the trigger soft menu because you switched to doing something else. But after a while you'll develop a muscle memory for the frequently used functions like press trigger button, then third soft button from the top, then first. Just like with any new tool.

[bdunham7]

On the 2230, it's a matter of seconds between starting the scope, click click click, boom I have the signal onscreen. 

Idk.  I was expecting better.  There's very little information on which digital scopes are actually good.  Can I hook my 2230 up to a computer and quickly/easily do any of the math I might want?

Yes, if you just want to simply look at a signal for some reason, the 2230 will be faster as well as easier if you are used to it.  I believe there was an available parallel interface that would allow you to download data in one form or another, but it won't be very much data nor very fast compared to a reasonable modern DSO. 

There's all sorts of threads here about entry level DSOs that you can read.  I doubt the TBS1000 series would be recommended in any of them due to a lack of any advanced features and very small memory.  I haven't used that model specifically, but on every modern DSO I've seen (even the bad ones) the information you want should be on the screen and it shouldn't take 5 seconds for scale changes or anything else.  And there's always the autoset button and a few scopes even have dual (YT) continuous autoranging.

[nctnico]

I have some hands-on experience with the TBS1000 series. Nothing wrong with it to look at signals. Nice crisp display and it works well. The only problem is that Tektronix should have put a touchscreen on it. In most cases it is reversed, but the UI on the TBS1000 would work better as a touch UI instead of using function buttons / knobs. Having owned the 2230 as well, I'd certainly choose the TBS1000 over the 2230.

edit: I meant the TBS2000 series which is nicer. The TBS1000 series is older (still not bad but rather dated).

[2N3055]

I have some hands-on experience with the TBS1000 series. Nothing wrong with it to look at signals. Nice crisp display and it works well. The only problem is that Tektronix should have put a touchscreen on it. In most cases it is reversed, but the UI on the TBS1000 would work better as a touch UI instead of using function buttons / knobs. Having owned the 2230 as well, I'd certainly choose the TBS1000 over the 2230.

TBS1000B series have only 2.5kPts. That is very small buffer that makes it very limited.

[tggzzz]

No fan, no heat, just signals.

Unless those signals are digital and not periodic, then you are screwed.

The digital-vs-analogue signal is irrelevant in that context.

The waveforms seen by a scope must be analogue. They only become digital signals after time/voltage thresholding - and that raises the issue of whether the instrument's thresholding is equivalent to the signal receiver's thresholding.

Signals can frequently be made periodic, one way or another. However, true one-off events always have been the USP of storage scopes, analogue or digitising.

Signals don't have to be periodic to be visualised on a scope; the classic example is eye diagrams.

[nctnico]

I have some hands-on experience with the TBS1000 series. Nothing wrong with it to look at signals. Nice crisp display and it works well. The only problem is that Tektronix should have put a touchscreen on it. In most cases it is reversed, but the UI on the TBS1000 would work better as a touch UI instead of using function buttons / knobs. Having owned the 2230 as well, I'd certainly choose the TBS1000 over the 2230.

TBS1000B series have only 2.5kPts. That is very small buffer that makes it very limited.

You are right, I was thinking about the TBS2000 series which has way more memory. The TBS1000 series is older.

[2N3055]

I have some hands-on experience with the TBS1000 series. Nothing wrong with it to look at signals. Nice crisp display and it works well. The only problem is that Tektronix should have put a touchscreen on it. In most cases it is reversed, but the UI on the TBS1000 would work better as a touch UI instead of using function buttons / knobs. Having owned the 2230 as well, I'd certainly choose the TBS1000 over the 2230.

TBS1000B series have only 2.5kPts. That is very small buffer that makes it very limited.

You are right, I was thinking about the TBS2000 series which has way more memory. The TBS1000 series is older.

I agree. New TBS2000  is quite decent.  Simple but does the job of basic scope well.

[tautech]

You got a new tool and could not figure out how to use it in 5 minutes, so the tool must be bad?

It does take time for some to embrace new technologies especially when their attention times are limited or when teaching an old dog new tricks can be a challenge.
However these early DSO's were designed to mimic analog scope performance best they could so to not stress new users grappling with new ways of doing the same thing.

Less switches, levers and knobs can really f**k with ones mind initially but the true test is learning to work with such to get a sensible and correct result.

Best advice for the new DSO user is to KISS and systematically move to more advanced usage. < don't try to run while still in crawling mode.

[bdunham7]

You are right, I was thinking about the TBS2000 series which has way more memory. The TBS1000 series is older.

The TBS2000 is way nicer.  I looked at the manual for the TBS1000B and it appears to be very similar to my TPS2000 except for the isolation feature.  There are currently 3 scopes living on my bench--a Tek 2221A (w/ A6909 isolator), a Sigent SDS2354X+ and the TPS2024.  I'd say for most simple tasks, the TPS2024 and the 2221A are roughly equivalent in many ways.  The persistence/DPO feature on the TPS is weak whereas the 2221A obviously has an analog mode.  The TPS does have a basic but functional FFT and autoset/autorange (like the TBS1000B) as well as basic digital features and automatic measurements. 

https://www.testunlimited.com/pdf/an/Tektronix_TBS1000B_usermanual.pdf

[csuhi17]

I couldn't figure out the point of this post.
Considering the price of the Tektronix 1072b, I don't think it's reasonable.
Was it released in 2015?!
Based on its specifications, you can get much better for less.
Almost any Siglent or maybe a Rigol or two seems like a better choice.
Maybe even an Owon is better than this Tektronix..

"There's no obvious indicator that the trigger is set to channel 2. Trying to decipher that takes forever"
From the few videos I found about it, its use seems clear to me, I don't understand what your problem is with the trigger. In the lower right part of the display, when you switch the trigger to CH2, does the color of the text not change, or does "CH1 / 0.00V" not change to "CH2 / 0.00V"?

"Getting the beam centered around zero, I couldn't find a quick way to do it,"
Can you press the vertical positioning button? On my Micsig and Rigol, I can use them to center the zero point of the channel in the middle of the screen.

I would like to look at a digital scope where all the functions have their own buttons and knobs, would a surface of half a square meter be enough for a modern scope?

[Fungus]

I would like to look at a digital scope where all the functions have their own buttons and knobs, would a surface of half a square meter be enough for a modern scope?

No.

[Fried Chicken]

Yes, if you just want to simply look at a signal for some reason, the 2230 will be faster as well as easier if you are used to it.  I believe there was an available parallel interface that would allow you to download data in one form or another, but it won't be very much data nor very fast compared to a reasonable modern DSO. 

The Analog scope is orders of magnitude faster.  I knew the digital scope might be a bit slower, but not to that ridiculous extent.  Even the autoset ranging was slower than simply dicking with the buttons.  The digital scope was so slow I'm genuinely surprised analog scopes were discontinued in the mid 90s and didn't continue production/development/refinement to this day!

If I were working in an environment and someone took my 2230 away and gave me something as slow and clunky as that digital scope, I would have revolted.

I'm agnostic on the interfaces, and I understand different paradigms fundamentally work differently, but holy hell sometimes you shouldn't throw the baby out with the bathwater when embracing new technologies.

Of course the example I used might have been especially slow, meant for the education market. I looked at a review of a newer keysight scope, and it looks way way better, but it's also three decades newer.

Much of the speed will depend on whether they have competent programmers who use C, or whether they use fairytale programming languages.  Unfortunately, that that doesn't seem to be the case!.

[lugaw]

Having used a DS1054z I wouldn't use my analog scopes anymore unless I am feeling nostalgic.  I wish I could afford the modern high bandwidth oscilloscope so I wouldn't use my TDS744.

[alm]

The Analog scope is orders of magnitude faster.  I knew the digital scope might be a bit slower, but not to that ridiculous extent.  Even the autoset ranging was slower than simply dicking with the buttons.  The digital scope was so slow I'm genuinely surprised analog scopes were discontinued in the mid 90s and didn't continue production/development/refinement to this day!

Are you maybe at slow time base settings? It may be that the TBS only updates things like vertical settings after a full sweep, so if you have the timebase set to say 100 ms/div, this could add to the perceived latency. The waveform update rate (number of sweeps per second) will likely be a lot slower for this scope than for an analog scope. Although more modern scopes like the Keysight scope are a lot better here, and limiting writing speed of the CRT would also limit the effective waveform update rate on analog scopes.

[Fried Chicken]

The Analog scope is orders of magnitude faster.  I knew the digital scope might be a bit slower, but not to that ridiculous extent.  Even the autoset ranging was slower than simply dicking with the buttons.  The digital scope was so slow I'm genuinely surprised analog scopes were discontinued in the mid 90s and didn't continue production/development/refinement to this day!

Are you maybe at slow time base settings? It may be that the TBS only updates things like vertical settings after a full sweep, so if you have the timebase set to say 100 ms/div, this could add to the perceived latency. The waveform update rate (number of sweeps per second) will likely be a lot slower for this scope than for an analog scope. Although more modern scopes like the Keysight scope are a lot better here, and limiting writing speed of the CRT would also limit the effective waveform update rate on analog scopes.

No, the whole thing was just slow.  Pushing a menu button, it took like 5 seconds to display a menu, changing a menu was decently fast, but that's after 10 seconds of waiting for it to display a menu.  Changing the timebase was disastrously slow, whether in the MHz or Hz range.  It was almost impossible to find the signal without overshooting first.

This isn't a scope I bought, I was just testing it to make sure it's working for someone and comparing it to my decently calibrated 2230.

[tggzzz]

I'm agnostic on the interfaces, and I understand different paradigms fundamentally work differently, but holy hell sometimes you shouldn't throw the baby out with the bathwater when embracing new technologies.

It is irritating when people do that. Sometimes it is beneficial to realise that if you know how someone earns their money, you can predict what they will say.

Much of the speed will depend on whether they have competent programmers who use C, or whether they use fairytale programming languages.

Both HP and Tektronix made instruments based on Smalltalk, in the late 80s early 90s. Those programmers knew what they were doing.

[bdunham7]

No, the whole thing was just slow.  Pushing a menu button, it took like 5 seconds to display a menu, changing a menu was decently fast, but that's after 10 seconds of waiting for it to display a menu.  Changing the timebase was disastrously slow, whether in the MHz or Hz range.  It was almost impossible to find the signal without overshooting first.

This isn't a scope I bought, I was just testing it to make sure it's working for someone and comparing it to my decently calibrated 2230.

I made a video of my TPS2024 booting up and looking at a 1kHz 1V square wave with some basic operations such as channel coupling, cursors, zoom, vertical scale, timebase, FFT and measurements.  Is the TBS1072B comparable or slower?

https://youtu.be/toOrmcHtwws

[ebastler]

I just used a fully digital oscilloscope for the first time  [...]
Wow is it a pain in the ass to use compared to my analog/solid state 2230.

I just used a bicycle for the first time. And boy does it suck, compared to walking. So wobbly! You can't even stand still without falling over! And don't get me started about walking backwards... I'm giving up on this thing, I have certainly had enough.

 

[joeqsmith]

 

I remember my mom trying to learn to ride a bicycle in her 50s.  She grew up during the great depression and I doubt had access.  She never learned to ride but she skied into her 70s.   

[Fried Chicken]

No, the whole thing was just slow.  Pushing a menu button, it took like 5 seconds to display a menu, changing a menu was decently fast, but that's after 10 seconds of waiting for it to display a menu.  Changing the timebase was disastrously slow, whether in the MHz or Hz range.  It was almost impossible to find the signal without overshooting first.

This isn't a scope I bought, I was just testing it to make sure it's working for someone and comparing it to my decently calibrated 2230.

I made a video of my TPS2024 booting up and looking at a 1kHz 1V square wave with some basic operations such as channel coupling, cursors, zoom, vertical scale, timebase, FFT and measurements.  Is the TBS1072B comparable or slower?

https://youtu.be/toOrmcHtwws

The TSP2024 is about twice as fast, but I would still consider it super slow.  Unless I needed the digital functionality, I would much prefer an analog scope for looking at that signal.

I'm agnostic on the interfaces, and I understand different paradigms fundamentally work differently, but holy hell sometimes you shouldn't throw the baby out with the bathwater when embracing new technologies.

It is irritating when people do that. Sometimes it is beneficial to realise that if you know how someone earns their money, you can predict what they will say.

I'm not sure what you're trying to say here

Much of the speed will depend on whether they have competent programmers who use C, or whether they use fairytale programming languages.

Both HP and Tektronix made instruments based on Smalltalk, in the late 80s early 90s. Those programmers knew what they were doing.

It's possible those late 80s early 90s digital scopes were faster.

[tggzzz]

I'm agnostic on the interfaces, and I understand different paradigms fundamentally work differently, but holy hell sometimes you shouldn't throw the baby out with the bathwater when embracing new technologies.

It is irritating when people do that. Sometimes it is beneficial to realise that if you know how someone earns their money, you can predict what they will say.

I'm not sure what you're trying to say here.

That there is validity to your point.
That sometimes people have a financial interest in ignoring/downplaying the point.

[vk6zgo]

The Analog scope is orders of magnitude faster.  I knew the digital scope might be a bit slower, but not to that ridiculous extent.  Even the autoset ranging was slower than simply dicking with the buttons.  The digital scope was so slow I'm genuinely surprised analog scopes were discontinued in the mid 90s and didn't continue production/development/refinement to this day!

If I were working in an environment and someone took my 2230 away and gave me something as slow and clunky as that digital scope, I would have revolted.

I'm agnostic on the interfaces, and I understand different paradigms fundamentally work differently, but holy hell sometimes you shouldn't throw the baby out with the bathwater when embracing new technologies.

Back in the 1990s, at the TV studio where I worked, both Tek & HP would send reps to show us the "latest & greatest" DSOs.
They would fiddle with them for a while & show us a few waveforms, then someone would spoil everything by saying "let's look at a video waveform".

The very earliest ones were starting to get a bit marginal at displaying a line rate signal (a whole line of analog TV occupies around 64 us) but would "sort of" show it, although some would have problems with aliasing against the colour subcarrier frequency, but they all came to grief if asked to display a Field rate signal (some 20ms wide).

At the long time/div rate required, because of the minimal memory in "those distant days", the sample rate was reduced radically, so that most of the frequency components of the signal were lost in a sea of aliasing.

They were patently useless for the task.

The "gurus" would try to argue that the mess was still a valid, if different type of display, with the inference that we were "dinosaurs".
That was far from the case, as by & large, we were all quite favourably disposed towards doing things digitally.
The important part of that is actually "doing things", which those early DSO's could not!

A few years later, we eventually got a Tek DSO which was usable for Field rate signals, although it did have an overlay of a "beat note" between the sample rate & the colour subcarrier.

When we bought a new 'scope for the transmitter site it was a 200MHz Iwatsu analogue!

[tggzzz]

The "gurus" would try to argue that the mess was still a valid, if different type of display, with the inference that we were "dinosaurs".
That was far from the case, as by & large, we were all quite favourably disposed towards doing things digitally.
The important part of that is actually "doing things", which those early DSO's could not!

That "blame the user" and "you don't count" attitude is an extremely annoying practice.

As someone that pushed the bounds of technology for most of my professional life, I became adept at distinguishing "new" from "better"...

Most new technologies were "new" but not "better". Some weren't even "new". 
Too many required advances that would have won Nobel/Turing Prizes - if the advances actually existed!
Some had useful advantages in limited circumstances.
A small proportion were a significant advance over the competition.

Frequently the smaller the advance or difference, the more heated the discussions/arguments.

[PwrElectronics]

First DSO I ever used was a Tek 2430A in around 1989-1990.  A couple years later a TDS420 and then TDS520 as well as THS720A and TDS220 by the end of the 90s.  Since then, various other Tek DSOs as well as Agilent and Lecroy.  I've liked the 2 Lecroys (mid 2000s era ones) the best.  Some of the Tek models use a 1-2-4 scale instead of the traditional 1-2-5 and that means things not aligned with the graticals.  I still like to use the graticals for quick eyeball measurments even though yes there are on screen measurements and cursors.

The only DSO I've used that really annoys me is the TPS2024 mentioned in this topic.  We have 2 of these at work and one of them is currently in my home lab (checked out during the pandemic).  For whatever reason, these (both do it) will "hang" and be unresponsive to user controls at times as if the processor is off doing who knows what.  I have not had that happen with any other model or brand of scope.

[bdunham7]

The TSP2024 is about twice as fast, but I would still consider it super slow.  Unless I needed the digital functionality, I would much prefer an analog scope for looking at that signal.

Once it boots (20 seconds or so) it is reasonably quick and responsive, so if that would still bug you it may be difficult for you to adapt to a DSO in general. 

[switchabl]

The Keysight DSOX/MSOX series are probably still fastest when it comes to UI responsiveness. They do take a while to boot but the controls are almost instant, they have a very "analog" feel. It's a shame Keysight has been neglecting that market segment for a while now. Some aspects aren't really competitive anymore, especially the memory depth. But I'm not getting rid of mine anytime soon, I still like the way it low-key "just works" for a lot of things.

From memory, the older Agilent 54600 models had a similar feel as well.

I don't think I've used a TBS1000 but Tek has made some horribly slow DSOs over the years. I'm not judging anyone who finds that annoying.