Why getting an analog oscilloscope?

[little_carlos]

Hello guys
Some people say (even dave says) that its a good idea to have an analog oscilloscope along with a digital one, but why? what does analog oscilloscopes have that are so special?

[tautech]

Hello guys
Some people say (even dave says) that its a good idea to have an analog oscilloscope along with a digital one, but why? what does analog oscilloscopes have that are so special?

The only real benefit of a CRO for newbies is the manual UI. It helps users understand basic principles and use of a scope. Even entry level DSO's now have sufficient capability and specs that there is little need for a CRO as a second scope.
But this does presume a user knows how to get the most out of their equipment, most newbies don't and a CRO can be more forgiving for the uninitiated.

There has been much discussion on this and there are some that insist a CRO is a must, however much of this POV is dated and getting more so as DSO specs improve.

Their will be others that have the entirely opposite POV and you must consider all arguements.

[MistaMun]

I feel like they have easier interface and they look cooler! I'm a beginner too and I had an opportunity to use one of the digital ones and they were just too complicated for what I wanted to do. I remember that it took good 2 minute to initialize everything for the digital scope while my Tektronix 455 turns on in couple seconds.

[Len]

I feel like they have easier interface and they look cooler! I'm a beginner too and I had an opportunity to use one of the digital ones and they were just too complicated for what I wanted to do. I remember that it took good 2 minute to initialize everything for the digital scope while my Tektronix 455 turns on in couple seconds.

I made the other choice. I was also a beginner when I got my Rigol DSO, and I'm glad I went for digital instead of analog. I had no trouble learning how to use it - finding the signal, setting the trigger, using trigger holdoff, etc. (Although I usually use the Auto button.) I like the extra capabilities of the digital scope - measurements, etc.

IMO, the only advantage of an analog scope is that with luck you can find a good, cheap used one.

[John Coloccia]

It depends what you're doing and what kind of DSO you're actually considering. IMHO, until recently, analog scopes were simply more useful for looking at things like audio signals. It was just easier to tell the difference between some random noise, insignificant transients, significant noise/transients, and the actual signal.

That said, the latest generations of intensity graded scopes are so good that my analog scope is just collecting dust now. Even the lowly Rigol 1054Z is so good that it's completely replaced my analog scope, even for looking at audio signals (or any other analog like or modulated signal). To get the full benefits though, it really has to have the intensity graded display. IMHO, that was the missing like that finally made old analog scopes truly obsolete for most practical purposes.

That, and nostalgia, are the only two reasons I can really think of to get an old analog scope. They are a bit simpler to use in some respects, but in some respects they're a little more finicky too. You should be able to pick up an old analog scope, in very good condition with probes, for about $100US, and it wouldn't hurt to play with one just to say that you did.

[SkyMaster]

... what does analog oscilloscopes have that are so special?

The total absence of software bug 

[little_carlos]

... what does analog oscilloscopes have that are so special?

The total absence of software bug 

LOL you totally said what i was looking for hahaha

[fivefish]

I have both... I have to say with an analog scope, if it breaks, you can fix it... and learn more about it's internal operations in the process. Of course, we hope it's not the custom hybrid chips that failed, and that you can buy good/used replacement parts on eBay if needed. 

[DimitriP]

If you have to ask , you should get one
Use the extra money to buy paaaaaats.
And if you decide not to keep it, sell it. Someone will take it.

[ez24]

If an analog has a handle then it would make a good boat anchor.

[DimitriP]

... what does analog oscilloscopes have that are so special?

The total absence of software bug 

Not to mention the endless  ,    and videos that follow when someone discovers something is 

[fubar.gr]

One thing I like with analog oscilloscopes (and other instruments in general) is that all current settings are neatly laid out in the front panel.

For example: I am getting weird voltage readings on my digital oscilloscope. Is the probe attenuation set correctly? To check it I will have to probably dig into at least a couple of menu layers.

With an analog scope all I have to do is to check the relevant switch position!

[veedub565]

From a professional point of view I'd personally prefer the digital scope, far more capability than an analogue one. That said though I used to have both a digital and analogue spectrum analyser on my bench, 99% of the time I used the digital one, but the analogue one was capable of displaying certain things that the digital one couldn't show very well. Also it's good for a beginner to learn on an analogue scope, it teaches you to read the waveform properly instead of relying on the digital brains to work it all out for you.

[Circlotron]

When CD players were new, some audiophools flocked to them, demonstrating to the rest of the world just how wise and discerning they were. Later on, every man and his dog had a CD player so the audiophools now switched back to analog vinyl discs because according to them they were so unbelievably superior to CDs, yet again demonstrating to everyone else their splendidly superior intellect and discernment.

I suspect there might be a hint of this with regard to analog vs digital scopes with some people.
I have one of each BTW.

[GNU_Ninja]

I have an ancient Philips PM3240 dual channel 50Mhz analog scope sat on my bench that I picked up for about £15 three years ago. It's a nice scope, easy to use, and has served me well ... But! ... it does take up a HUGE amount of bench space, so, when funds are available, I intend to replace it with a much smaller digital one (Siglent or Rigol). Certain signals were easier to interpret on an analog scope, but as has been pointed out, the latest digital scopes now have intensity graded displays too.

To answer the question about what was special about my analog scope: It only cost me £15 

[tggzzz]

... but the analogue one was capable of displaying certain things that the digital one couldn't show very well. ...

Precisely. And vice-versa of course - they both have different advantages and disadvantages.

A good workman knows their tools, when to use a fork and when to use a spade (or nail/screw, MCU/FPGA etc etc).

A workman with limited experience makes absolute statements.

[Gyro]

IMHO, until recently, analog scopes were simply more useful for looking at things like audio signals.

And RF signals too, the sort of thing that pushes the bandwidth  sample rate of a DSO. I still much prefer to look at an HF signal on my 250MHz Analogue. Agreed, DSOs have much improved their graded intensity displays etc, they are still a s/w 'simulation' of what you see on an analogue CRT (and it's a lot easier to tweak the brightness control on an analogue).

I do have a bias towards devices where the whole design effort has historically been focused on getting the analogue performance as good as possible.  It's important to remember that all scopes are analogue, it's just about whether they're analogue as far a the screen or analogue as far as the ADC, analogue performance is totally critical for both! Some manufacturers and scopes can call on a greater 'heritage' of analogue experience, rather than coming into it from a digital and software background (how many S/W controlled PSUs have analogue glitch / overshoot issues!).

The advent of DSOs was much like the arrival of PVR Set-Top-Boxes.... The availability of magic 'Pause' function. It's the killer USP.... the painless capture of non repetitive signals. Let's face it, analogue storage always was a total pain in the a**, you could spend all day getting the persistence etc set just right to capture a difficult event, and photographing it before the screen bloomed. With repetitive signals, the gap between Analogue and Digital is a lot closer, both bring their conveniences, be they intuitive manual controls or autoset and automatic measurements different screen qualities, price (and let's face it, one of the reasons for manufacturers switching from analogue to digital is that they're cheaper to make), etc. Both will have their advocates and followers (fanatical or otherwise).

At the end of the day, there isn't going to be a totally right answer - Are DSOs better than they were 10 years ago? Definitely. Do they still have a way to go yet before they are the perfect choice for every situation? Yes probably.

For High frequency repetitive I still like my 250MHz analogue, for audio measurements I like my 16bit low sample rate, low noise, Picoscope, for the stuff in between (non repetitive and digital) a combination of a modest USB DSO (not a Hantek!) and a logic analyser. It's a combination that works well for me. YMMV 

Edit: And 100% with tggzzz 

[TimFox]

For troubleshooting, I have found that the true analog oscilloscope is better when you don't know what the problem is, since it is less likely to confuse you.  If you need an accurate measurement (such as timing) of a signal that you already understand, the digital oscilloscope is better.  As stated by Gyro, the capture of a non-repetitive waveform is far better with a DSO:  I have no nostalgia for analog storage oscilloscope technology.

[tggzzz]

For troubleshooting, I have found that the true analog oscilloscope is better when you don't know what the problem is, since it is less likely to confuse you.  If you need an accurate measurement (such as timing) of a signal that you already understand, the digital oscilloscope is better.  As stated by Gyro, the capture of a non-repetitive waveform is far better with a DSO:  I have no nostalgia for analog storage oscilloscope technology.

That is a good summary of analogue scope's advantages.

A classic example of a non-repetitive waveform is PSU sequencing and transients, for which a digitising scope is ideal. For non-repetitive digital signals, I prefer to debug in the digital domain, For eye diagrams, DSOs are marginally preferable to analoge scopes since they don't forget the 1 in 10,000 worst-case pattern.

As for analogue storage scopes... they were more usable than "analogue" storage computer terminals such as the Tek 4014

[AG6QR]

An analog scope has no aliasing artifacts.  In X-Y mode, the trace is continuously updated with no downtime.  There are no hidden menu settings.

The DSO has a whole lot of advantages, too, and on balance, it's a better tool most of the time.  But every once in a while, an analog scope shines.

[vk6zgo]

For troubleshooting, I have found that the true analog oscilloscope is better when you don't know what the problem is, since it is less likely to confuse you.  If you need an accurate measurement (such as timing) of a signal that you already understand, the digital oscilloscope is better.  As stated by Gyro, the capture of a non-repetitive waveform is far better with a DSO:  I have no nostalgia for analog storage oscilloscope technology.

In troubleshooting,it is a "given" that you don't know what the problem is.
If you did,you would fix it,without resort to an Oscilloscope!

It also doesn't follow that "you don't understand" the signal in a fault situation.

For example,I have a pretty good handle on analog video signals,but there are a whole lot of things which can happen to such a signal,other than simple loss,all of which can be diagnosed fairly easily with the analog 'scope.

Early DSOs were fairly useless for this,due to their abysmal memory depth leading them to reduce the sample rate radically at long time/div settings.

Modern DSOs could do this sort of work,but it's taken a long time to get to this level,hence the perennial.advice to "get an analog 'scope".

You need a real DSO,though!

The very cheap USB type 'scopes you see on eBay,& the tiny pocket ones,occupy the same niche that  cheap low bandwidth analog 'scopes occupied for years,but the latter are still generally more  usable.

[oldway]

For troubleshooting, you don't need all the informations you have with a DSO.
The best is often the enemy of the good.
The only information you really need with precision is the frequency of the displayed signal.
In my analog scope, I have an Y output I use with a counter so I have this information.
DSO have poor XY , no Component Tester (very usefull for troubleshooting) and aliasing.
I don't like the "Parkinson like" displayed signal of the low end DSO : that's not the true signal pattern.

I think that for troubleshooting and analog electronics, analog oscilloscopes still are great values.

But DSO have a lot of advantages: storage, flexibility, compactness, ...they are now becaming "analog like" oscilloscopes with a lot of features.

NB: If you buy an analog oscilloscope, buy a good one with at least 50Mhz bandwith, dual time base and, if possible, component tester.
I don't like analog scopes with readout and cursors because you loose one of the greatest advantage of analog scopes: real time measurement.
The time you need to write the readout and cursors on crt, you can't write the waveform.
Analog scopes are no more supported ...for this reason, only buy a full analog scope, with switches and without sophisticated digital controls.
For example, Tek 465B, 2235, HP1740A, Hameg HM605...

[Fignuts70]

I think both still have their place depending on needs.  A couple other things that are often not considered in the digital versus analog discussion...  Depending on what your definition of "affordable" is, your experience may vary. 

Many older analog oscilloscopes have better Y-axis sensitivity ranges, some even down to the 10s of microvolts.  (Tek 502, 5030, 5031; 5A22N and 7A22 plugins, et. al.)  Good luck finding much better sensitivity than 5mV/div on affordable modern digital kit.  Granted, not something a lot of people need all that often, but if you do...  Which also brings up another thing: Differential input plugins for older analog oscilloscopes can generally be had for a fraction of the price of just a differential probe - often times you can buy the whole scope and plugins for less than a single differential probe. 

Many affordable modern digital oscilloscopes do not have the ability to slave 2 different time-bases to different channels and display them along side each other.  Again the keyword here is affordable!  (Tek 547, 7xx4, et. al. can do this if the correct plugins are used)

Most affordable modern digital oscilloscopes are 8-bit resolution on the Y-axis.  This means if you are measuring a signal and have the scope set for 1V/Div, the best resolution you're going to get is ~31.25mV, assuming an 8v x 10h graticule and my math is right (8V / 256 steps or 8 bits).  That means if you have a small amount of noise riding on your signal, it's going to be harder to see.  Again, whether or not this matters to what you're doing is another story, but it can make a difference sometimes.

As was mentioned before, X-Y mode on analog just seems to work better.

Aliasing.  It even happens sometimes when using the "Auto-set" feature, and that can be frustrating for anyone who hasn't trained themselves to double-check the time-base readout on a digital model - that or give the time-base wheel a spin back and forth just to verify.

That said, I love the measurement capabilities and a lot of the other features on the digital "dark side," especially if your digital kit has a built in function generator.  Having the ability to do bode plots is definitely nice to have.  Also the having FFT capability, generally nicer cursors, automatic math functionality, etc. all are pretty awesome.  Also "affordable" buys you a lot more in the digital realm nowadays than it used to. 

At the end of the day, it's all about what you need the oscilloscope for in the first place.  If you're troubleshooting digital circuits, using an analog model is bordering on masochism.  If you're troubleshooting power supply noise, that can be a very different matter.

Just my $.02 worth.

[JohnnyMalaria]

As was mentioned before, X-Y mode on analog just seems to work better.

Amen to that.

This shows the benefit of an analog CRO in XY mode.

The signal is the demodulated IQ pair from a phase analysis light scattering (PALS) apparatus. The amplitude and phase determined by the composite signal of light scattered by nanoparticles in a liquid. Random diffusion causes random fluctuation in both the amplitude and phase. Hence, the oscilloscope is very useful in visualizing this diffusion in real-time. Application of an electric field causes charged particles to move linearly. This manifests as a Doppler shift that is seen as rotation of the signal on the oscilloscope. The beeps correspond to when the field is turned on and off.

Trying to view this on a DSO such as a Rigol DS1104Z isn't even worth the effort...

[CaptDon]

One of the best reasons why I keep an analog scope with 80mhz bandwidth in my rack is that they don't alias!!! I need to look at the response of the control loop as it modulates a 100khz pwm signal to produce a locomotive diesel fuel injection drive signal. That signal going to the injector peaks at 26 amps of current and has a significant amount of 100khz riding the 'average' d.c. current. The waveform is horrendously distorted on every one of our digital scopes at the slow sweep speeds needed when the engine is at idle due to aliasing of a fast signal at a slow sweep speed. On the analog scopes I can see EXACTLY what the pwm is doing as well as the average current. Another reason to have an analog scope is if you work in the AM Broadcast Band and want to look at the modulated carrier with random music or voice modulation. When you slow the sweep to see complete voice envelopes the RF carrier aliases to look like a second audio frequency interference and makes the display unusable.