Author Topic: Do You still use analog oscilloscopes in 2019?  (Read 13467 times)

0 Members and 1 Guest are viewing this topic.

Offline Mr. Scram

  • Super Contributor
  • ***
  • Posts: 8741
  • Country: 00
  • Display aficionado
Re: Do You still use analog oscilloscopes in 2019?
« Reply #175 on: August 13, 2019, 09:49:33 am »
Re: the Keithley....

I'm honestly not sure why that exists. Say I'm looking at PA transistor current. As the transistor gets hotter, current goes up. Difficult to read trends over a 5 second period with any kind of digital metering. The real time trend graph is good for sure but you can see the same thing watching a needle creep up and it doesn't cost $1750 which is a bad ROI unless you spend all day looking at it. If I want to log something over time then use a generic bench meter with a PC interface and do it there. Very limited in what you can do with the data on a device when there's Excel and Mathematica available on something larger and more powerful.

Some analogue action from me to demonstrate the current metering response...


Have you seen videos about the DMM6500 and DAQ6510? They're insanely flexible and powerful machines and a lot of processing can be done on the machine itself. Of course you can hook up a computer and do it that way but it's all about expediency. It's essentially the COTS versus rolling your own discussion. I don't think it's a stretch to say Keithley set a new standard. Of course the model pictured can scan multiple channels so doing more involved measurements can be easily done.
« Last Edit: August 13, 2019, 09:51:24 am by Mr. Scram »
 

Offline BU508A

  • Super Contributor
  • ***
  • Posts: 1394
  • Country: de
“Chaos is found in greatest abundance wherever order is being sought. It always defeats order, because it is better organized.”            - Terry Pratchett -
 

Offline bd139

  • Super Contributor
  • ***
  • Posts: 13697
  • Country: gb
Re: Do You still use analog oscilloscopes in 2019?
« Reply #177 on: August 13, 2019, 10:07:18 am »
Not sure I get the use cases for it. I built similar test platforms years ago using GPIB and composition. Composing instruments is always more powerful than having a black box. You can reuse the instruments for other cases later once your production run is over.

Also as always with these things, they tout 6.5 digits, 1MS sample rate but look at the datasheet. To sample all those channels you need a channel switch with a 3ms rate change so say you have 80 channels, so including settling time that kicks you down to 1 reading/second over 80 channels and those channel switches have a pretty finite lifespan! Also it only has 512 memory slots for readings. When you start adding options, it makes some of the other competitors from Keysight look cheap.

On top of that you get to program it via a shitty touch screen UI made by the lowest bidder or some weird ass proprietary language and development tool chain. And I know something from experience; test gear companies can't write software.

We had similar stuff with two 34401A's and two channel multiplexers driven from a rack mount PC and visual basic that could do this sort of stuff back in the late 1990s.  :-//

Composition > monolithic devices.
 
The following users thanked this post: vk6zgo, 001

Offline Mr. Scram

  • Super Contributor
  • ***
  • Posts: 8741
  • Country: 00
  • Display aficionado
Re: Do You still use analog oscilloscopes in 2019?
« Reply #178 on: August 13, 2019, 10:31:23 am »
Not sure I get the use cases for it. I built similar test platforms years ago using GPIB and composition. Composing instruments is always more powerful than having a black box. You can reuse the instruments for other cases later once your production run is over.

Also as always with these things, they tout 6.5 digits, 1MS sample rate but look at the datasheet. To sample all those channels you need a channel switch with a 3ms rate change so say you have 80 channels, so including settling time that kicks you down to 1 reading/second over 80 channels and those channel switches have a pretty finite lifespan! Also it only has 512 memory slots for readings. When you start adding options, it makes some of the other competitors from Keysight look cheap.

On top of that you get to program it via a shitty touch screen UI made by the lowest bidder or some weird ass proprietary language and development tool chain. And I know something from experience; test gear companies can't write software.

We had similar stuff with two 34401A's and two channel multiplexers driven from a rack mount PC and visual basic that could do this sort of stuff back in the late 1990s.  :-//

Composition > monolithic devices.
Again, it's all about COTS versus having to roll your own. Needing four devices, a rack mount PC and a coder who knows what he's doing versus one DMM and a monkey poking at the screen able to make all kinds of involved measurements without external postprocessing. The software to do it on a PC is supplied too if that tickles your fancy. But none of this was the point. It was all about making changes intuitively visible and a graph is essentially doing what a needle does except that it also shows a historical trend.

I'm not sure where you get the 512 readings from but the datasheet says the following. "Maximum Reading Memory (volatile) Up to 7 million readings with standard buffer (includes channel and formatting information). Internal (non-volatile) Memory for Saved Scripts and Scan Configurations 6 MB, enables hundreds of scan configurations or TSP scripts to be saved in non-volatile memory" You can also use your own storage. The number of times a relay is operated is stored and the relays themselves are standard fare so can be replaced when their millions of operations are used up.
 

Offline Mr. Scram

  • Super Contributor
  • ***
  • Posts: 8741
  • Country: 00
  • Display aficionado
Re: Do You still use analog oscilloscopes in 2019?
« Reply #179 on: August 13, 2019, 10:32:51 am »
You are referring to this, I assume?

http://thesignalpath.com/blogs/2019/08/04/keithley-daq6510-6-5-digit-data-acquisition-multimeter-review-teardown-experiments/
That's a pretty good video. The SMU and DMM7510 reviews too as a lot is shared across all models.
 

Offline Fungus

  • Super Contributor
  • ***
  • Posts: 10488
  • Country: 00
Re: Do You still use analog oscilloscopes in 2019?
« Reply #180 on: August 13, 2019, 10:48:40 am »
On dso's, I guess I don't get much of what they are about.  Oscilloscopes are for looking at  waveforms, right?

The key is in the 'S' part of "DSO". You can capture waves and examine them at leasure (zoom/pan). You can also set up complex conditions for when to trigger so you only see the stuff you want to see.

This alone is worth the price of entry (eg. a Rigol DS1054Z).

If you have special needs that aren't covered by a basic model then make a list of them and spend more money on a better DSO.

Or ... if it's something you do a lot then buy a dedicated device for that function and use the two devices together. You'll get better results (as you noted).

But in what cases is it needed to store waveforms rather than just looking at them in realtime?

Whenever a waveform isn't a simple repeating shape that fits on screen.

Meters are for accurately checking voltage and current.  Spectrum analyzers are  for looking at the frequency domain.   Logic analyzers are for looking at digital signals.  So then, why is so much piled into these low cost dso's?

It's cheap to add those functions to a DSO because it's just software, it gives useful results even if it's not perfect, so... why not?
« Last Edit: August 13, 2019, 10:54:18 am by Fungus »
 
The following users thanked this post: BillyD

Offline 2N3055

  • Super Contributor
  • ***
  • Posts: 2239
  • Country: hr
Re: Do You still use analog oscilloscopes in 2019?
« Reply #181 on: August 13, 2019, 11:05:36 am »
There are no people saying "we should be using old analog ampermeters, those are better" That would be just silly.

Ooh, a challenge.

Try measuring charge with a modern meter.

At school we measured charge using a ballistic galvanometer. Principle: dissipate the charge through the meter in a time that is much shorter than the meter's response time. The needle is kicked and the maximum deflection is proportional to the current*time, i.e. charge.

Well I'm sure you will have trouble measuring liters or kilos with digital multimeter too..  Ballistic galvanometer is specialized instrument. If it was something commonly used, multimeters would have a measurement for that. 
You could use a meter with high Z and know capacitor, and measure  peak voltage.
 

Offline rsjsouza

  • Super Contributor
  • ***
  • Posts: 3690
  • Country: us
  • Eternally curious
    • Vbe - vídeo blog eletrônico
Re: Do You still use analog oscilloscopes in 2019?
« Reply #182 on: August 13, 2019, 12:29:33 pm »
Analog multimeters mostly suck at resistance measurements - highly non-linear. For the rest, IMO they are alright (not that I use one regularly anymore) .

One thing that I like about analog meters is that it exercises your brain to interpret what you see - after all, our brains need as much exercise (even in mundane things) as the muscles in our body.

The fact they also operate without batteries is an interesting factor.

Regarding DSOs, to me the "S" is the first advantage one faces when moving from the analog to the digital domain (others follow suit the more one works with it). Never to forget: the digital domain is a simplification of the analog domain and carries intrinsic limitations with it. Good understanding of these is critical to get the most of the two techniques.
Vbe - vídeo blog eletrônico http://videos.vbeletronico.com

Oh, the "whys" of the datasheets... The information is there not to be an axiomatic truth, but instead each speck of data must be slowly inhaled while carefully performing a deep search inside oneself to find the true metaphysical sense...
 

Offline tggzzz

  • Super Contributor
  • ***
  • Posts: 10960
  • Country: gb
    • Having fun doing more, with less
Re: Do You still use analog oscilloscopes in 2019?
« Reply #183 on: August 13, 2019, 04:09:15 pm »
There are no people saying "we should be using old analog ampermeters, those are better" That would be just silly.

Ooh, a challenge.

Try measuring charge with a modern meter.

At school we measured charge using a ballistic galvanometer. Principle: dissipate the charge through the meter in a time that is much shorter than the meter's response time. The needle is kicked and the maximum deflection is proportional to the current*time, i.e. charge.

Well I'm sure you will have trouble measuring liters or kilos with digital multimeter too..  Ballistic galvanometer is specialized instrument. If it was something commonly used, multimeters would have a measurement for that. 
You could use a meter with high Z and know capacitor, and measure  peak voltage.

Nonetheless, analogue ammeters are better for measuring charge, and using them for that purpose isn't silly.

That meter+capacitor technique won't work in many interesting cases.
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 

Offline Old Printer

  • Frequent Contributor
  • **
  • Posts: 555
  • Country: us
Re: Do You still use analog oscilloscopes in 2019?
« Reply #184 on: August 13, 2019, 04:34:18 pm »
Never to forget: the digital domain is a simplification of the analog domain and carries intrinsic limitations with it.

This would make a great tee-shirt. Sure would burst a lot of bubbles :)
 
The following users thanked this post: rsjsouza

Offline Mr. Scram

  • Super Contributor
  • ***
  • Posts: 8741
  • Country: 00
  • Display aficionado
Re: Do You still use analog oscilloscopes in 2019?
« Reply #185 on: August 13, 2019, 04:59:31 pm »
This would make a great tee-shirt. Sure would burst a lot of bubbles :)
What bubbles would it burst?
 

Offline 2N3055

  • Super Contributor
  • ***
  • Posts: 2239
  • Country: hr
Re: Do You still use analog oscilloscopes in 2019?
« Reply #186 on: August 13, 2019, 05:02:34 pm »
There are no people saying "we should be using old analog ampermeters, those are better" That would be just silly.

Ooh, a challenge.

Try measuring charge with a modern meter.

At school we measured charge using a ballistic galvanometer. Principle: dissipate the charge through the meter in a time that is much shorter than the meter's response time. The needle is kicked and the maximum deflection is proportional to the current*time, i.e. charge.

Well I'm sure you will have trouble measuring liters or kilos with digital multimeter too..  Ballistic galvanometer is specialized instrument. If it was something commonly used, multimeters would have a measurement for that. 
You could use a meter with high Z and know capacitor, and measure  peak voltage.

Nonetheless, analogue ammeters are better for measuring charge, and using them for that purpose isn't silly.

That meter+capacitor technique won't work in many interesting cases.

My english fails me. Sorry. What I wanted to relay is that fact that there are separate instruments to measure current, voltage or resistance doesn't make multimeters stupid. If anything, most of us only use multimeters for those kind of measurements, and we have no separate voltmeters and ammeters and voltmeters, for pretty much  anything that multimeters do.

What you are mentioning is a measurement that doesn't exist on multimeter. So it is not only not silly but is only logical thing to use specialized instrument made for that particular measurement. 

Of course you can measure quantity of charge with capacitor. It is actually most common method today in today's electrometers, by using known capacitor and connecting it either in integrator mode or in a shunt mode (the one I mentioned). Ballistic galvanometer uses mass to integrate charge pulse into mechanical displacement, capacitor uses stored charge to integrate to voltage directly. Actually, ballistic galvanometers are calibrated by discharging calibrated capacitors into it, making it mathematically equivalent, proven by physical experiment.
Yes, I'm old enough, we did play with those in school too. They are fascinating, very very cool. I love old tech.

Scopes can measure frequency, but my TF930 does it better. But most of the time, I can quickly, while scoping around, verify that crystal on my board runs at 10MHz and that the waveform is OK. And 8 digits and 2-3 ppm frequency accuracy on a scope is better than what that cheap 20-30 ppm crystal will do.
If I need it more precise, I fire up TF930.. But 80-90% of time it's not necessary.
 

Offline Old Printer

  • Frequent Contributor
  • **
  • Posts: 555
  • Country: us
Re: Do You still use analog oscilloscopes in 2019?
« Reply #187 on: August 13, 2019, 05:14:13 pm »
This would make a great tee-shirt. Sure would burst a lot of bubbles :)
What bubbles would it burst?

For starters, anyone who looks at a digital display and assumes it more accurate than an analog one. I could get into recorded music, but I think you get my drift.  ;)
 

Online CatalinaWOW

  • Super Contributor
  • ***
  • Posts: 3509
  • Country: us
Re: Do You still use analog oscilloscopes in 2019?
« Reply #188 on: August 13, 2019, 07:26:05 pm »
CatalinaWOW, I assume that you are looking for glitches in digital signals.  I don't know if that will ever be a practical use case for myself.  And if I do have a need for that in the future, I think I would have a better idea of what I need at that point (rather than now).  And as tech marches on, there will likely be better tools for that task in the future.

That happens, but my most frequent use case is in systems with multiple servo loops, all operating at the same time, but with different loads and frequencies.  I haven't personally done this but I would assume that CNC setups of various types run into this sort of thing.  Any place signals have been mixed would seem likely to find uses for this.  All those sum and difference frequencies make searching for the culprit fun.
 

Offline rsjsouza

  • Super Contributor
  • ***
  • Posts: 3690
  • Country: us
  • Eternally curious
    • Vbe - vídeo blog eletrônico
Re: Do You still use analog oscilloscopes in 2019?
« Reply #189 on: August 13, 2019, 10:08:15 pm »
Never to forget: the digital domain is a simplification of the analog domain and carries intrinsic limitations with it.
This would make a great tee-shirt. Sure would burst a lot of bubbles :)
What bubbles would it burst?
For starters, anyone who looks at a digital display and assumes it more accurate than an analog one. I could get into recorded music, but I think you get my drift.  ;)
It is interesting to see how many younglings think that analog is useless or "out" while digital is "in". I've burst many bubbles with my sentence above and with this other one: "we are analog by nature, thus any digital gizmo is useless if it does not have an analog way to interact with us". But that shouldn't be a surprise for anyone here.

There's a reason as to why working with digital is so prevalent these days: it is easier to process, obtain information, transform, reconstruct, restore, etc. All that at the expense of acceptable levels of uncertainty and many orders of magnitude of increased circuitry complexity.

Vbe - vídeo blog eletrônico http://videos.vbeletronico.com

Oh, the "whys" of the datasheets... The information is there not to be an axiomatic truth, but instead each speck of data must be slowly inhaled while carefully performing a deep search inside oneself to find the true metaphysical sense...
 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 11122
  • Country: us
Re: Do You still use analog oscilloscopes in 2019?
« Reply #190 on: August 13, 2019, 10:30:29 pm »
This would make a great tee-shirt. Sure would burst a lot of bubbles :)
What bubbles would it burst?

For starters, anyone who looks at a digital display and assumes it more accurate than an analog one. I could get into recorded music, but I think you get my drift.  ;)

I would argue that in most cases a digital display is more precise, now whether it's also more accurate depends on a lot of things.
 

Offline Old Printer

  • Frequent Contributor
  • **
  • Posts: 555
  • Country: us
Re: Do You still use analog oscilloscopes in 2019?
« Reply #191 on: August 14, 2019, 12:36:56 pm »
This would make a great tee-shirt. Sure would burst a lot of bubbles :)
What bubbles would it burst?

For starters, anyone who looks at a digital display and assumes it more accurate than an analog one. I could get into recorded music, but I think you get my drift.  ;)

I would argue that in most cases a digital display is more precise, now whether it's also more accurate depends on a lot of things.
Yes, a lot of things, things non EE types probably don't even know exist like ADC's, hence the "bubbles". So to be precise, I was not referring to a lot of people here.
 

Offline rsjsouza

  • Super Contributor
  • ***
  • Posts: 3690
  • Country: us
  • Eternally curious
    • Vbe - vídeo blog eletrônico
Vbe - vídeo blog eletrônico http://videos.vbeletronico.com

Oh, the "whys" of the datasheets... The information is there not to be an axiomatic truth, but instead each speck of data must be slowly inhaled while carefully performing a deep search inside oneself to find the true metaphysical sense...
 
The following users thanked this post: 001, Tom45

Offline james_s

  • Super Contributor
  • ***
  • Posts: 11122
  • Country: us
Re: Do You still use analog oscilloscopes in 2019?
« Reply #193 on: August 23, 2019, 04:14:32 pm »
Now if only automotive manufactures would get the memo.

I loathe touchscreens. I put up with one on my phone because it's the only practical way to make a pocked sized device like a smartphone work, but beyond that, no thanks. I like physical controls that have a tactile feel, I should be able to operate everything without looking at it.
 
The following users thanked this post: 001

Offline rsjsouza

  • Super Contributor
  • ***
  • Posts: 3690
  • Country: us
  • Eternally curious
    • Vbe - vídeo blog eletrônico
Re: Do You still use analog oscilloscopes in 2019?
« Reply #194 on: August 23, 2019, 06:07:03 pm »
james_s, I think touchscreens are a very good fit for specific purposes that do not require excessive precision - tracing/highlighting regions (for mask, for example) or quickly accessing menus or functions that float around the screen. Feedback mechanisms such as button highlighting or haptic also help a lot with the operation of touchscreens - usually haptic is quite responsive, but visual aids are heavily dependent on the CPU load of the HMI and are rarely well implemented.

Bring touchscreens to a precision fight is a very bad idea IMHO... Setting values on a dial or slider interface, or even trying to highlight or move a cursor on a line of text (so common when posting on EEVBlog from the cellphone, for example) is an exercise in patience with the occasional frustration. For those tasks, bring me a mechanical/physical interface any day.

Oh well... I wonder if our line of thought will become extinct as our generation ages...
Vbe - vídeo blog eletrônico http://videos.vbeletronico.com

Oh, the "whys" of the datasheets... The information is there not to be an axiomatic truth, but instead each speck of data must be slowly inhaled while carefully performing a deep search inside oneself to find the true metaphysical sense...
 

Offline guenthert

  • Frequent Contributor
  • **
  • Posts: 342
  • Country: us
Re: Do You still use analog oscilloscopes in 2019?
« Reply #195 on: August 23, 2019, 06:26:39 pm »
[..]
Disadvantages

1) DCOs are untruthful by their nature. Unless you know beforehand what you are looking for, what is on the screen may have nothing to do with the reality. Which creates a particular problem when you are troubleshooting... . You have to check and cross-check what you see to be sure.
[..]
  And I'd argue, you need to know the same when using analog scopes.  What will the 12pF capacity of your probe doing to the signal under test?  The scope (analog or digital) will also be a low pass filter -- what are the exact characteristics?  If you blindly trust what either kind of scope is showing you, you're in for a ride.

  I (just a hobbyist) got an old (slow) analog one for fun (essentially as X-Y vector monitor) and for nostalgia.  And a buddy just tasked me with repairing his Tek 475.  For troubleshooting or measurements I wouldn't think twice to use one of my digital ones (I'd love to have one of the old analog Tek 7 series ones with their quite impressive specifications, but I just don't have the space for such and I'm afraid by now, they require too much TLC).
 

Offline tggzzz

  • Super Contributor
  • ***
  • Posts: 10960
  • Country: gb
    • Having fun doing more, with less
Re: Do You still use analog oscilloscopes in 2019?
« Reply #196 on: August 23, 2019, 06:30:48 pm »
james_s, I think touchscreens are a very good fit for specific purposes that do not require excessive precision - tracing/highlighting regions (for mask, for example) or quickly accessing menus or functions that float around the screen. Feedback mechanisms such as button highlighting or haptic also help a lot with the operation of touchscreens - usually haptic is quite responsive, but visual aids are heavily dependent on the CPU load of the HMI and are rarely well implemented.

Bring touchscreens to a precision fight is a very bad idea IMHO... Setting values on a dial or slider interface, or even trying to highlight or move a cursor on a line of text (so common when posting on EEVBlog from the cellphone, for example) is an exercise in patience with the occasional frustration. For those tasks, bring me a mechanical/physical interface any day.

Oh well... I wonder if our line of thought will become extinct as our generation ages...

Touchscreens require that you are looking at the screen to see which control you are changing.

That's potentially lethal in a car, when you should be looking at the road in front of you. Hence it is illegal here to use a non-handsfree cellphone while driving.

"Heads down" syndrome is implicated in many aircraft crashes, as can be determined from the two flight recorders. Unfortunately cars don't have those, so it will be difficult to prove.
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 
The following users thanked this post: rsjsouza

Offline Gyro

  • Super Contributor
  • ***
  • Posts: 5288
  • Country: gb
Re: Do You still use analog oscilloscopes in 2019?
« Reply #197 on: August 23, 2019, 06:42:52 pm »
That's why many modern cars stick up an annoying disclaimer on the centre screen at startup, pointing out the risk and asking you to confirm.
Chris

"Victor Meldrew, the Crimson Avenger!"
 

Offline tggzzz

  • Super Contributor
  • ***
  • Posts: 10960
  • Country: gb
    • Having fun doing more, with less
Re: Do You still use analog oscilloscopes in 2019?
« Reply #198 on: August 23, 2019, 08:13:36 pm »
That's why many modern cars stick up an annoying disclaimer on the centre screen at startup, pointing out the risk and asking you to confirm.

So they actually say "This equipment is dangerous if correctly used in the way designed"?

I wonder if that is sufficient. Mind you, I did once write a contract to develop an X and included a clause stating that it was not fit to be used as an X, and I pointed out the clause in the covering letter. The client signed the contract :)
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 

Offline rsjsouza

  • Super Contributor
  • ***
  • Posts: 3690
  • Country: us
  • Eternally curious
    • Vbe - vídeo blog eletrônico
Re: Do You still use analog oscilloscopes in 2019?
« Reply #199 on: August 23, 2019, 08:35:37 pm »
"Heads down" syndrome is implicated in many aircraft crashes, as can be determined from the two flight recorders. Unfortunately cars don't have those, so it will be difficult to prove.
Certainly. It is ridiculous the amount of people that I see zipping at 120km/h while their heads and eye are plugged to the stupid phone. There are accidents and there is recklessness.

I recall years ago many studies and prototypes of safety systems that monitored the movement of a driver's eyes and/or head to detect tiredness - they should instead detect lack of attentiveness. I am pretty sure there are much more accidents caused by phones than by sleep.

That's why many modern cars stick up an annoying disclaimer on the centre screen at startup, pointing out the risk and asking you to confirm.
I wonder if that is sufficient.
My friend's car has an in-dash GPS that prevents it from being operated at all when the car is moving - even by a passenger! (which would be difficult to tell form the GPS's perspective)

Unless the car is in full stop (at a traffic light or jam), the GPS accepts no inputs.
« Last Edit: August 23, 2019, 08:38:13 pm by rsjsouza »
Vbe - vídeo blog eletrônico http://videos.vbeletronico.com

Oh, the "whys" of the datasheets... The information is there not to be an axiomatic truth, but instead each speck of data must be slowly inhaled while carefully performing a deep search inside oneself to find the true metaphysical sense...
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf