Digital and analog oscilloscope

Hello,
I'm planning buying a cheap oscilloscope for home projects.
I'm considering the following options:

Analog 60MHz bandwidth dual channel.
Digital 100Mhz bandwith, 40Msamples/s, dual channel.

A 100Mhz digital oscilloscope with just 40Msamples/s can work with aperiodic signals of up to 10 or 20MHz, but would be able to get 100MHz periodic signal using sub-sampling methods.

But what does 60Mhz really mean for an analog oscilloscope?? Can It work with signals of up to 60Mhz??

Thanks!

There was a recent thread on this topic, see "PC Ocsilloscopes" in this forum.

The advantage of an analogue scope is that it does not need a computer.

But, unless you buy one with storage, you can only view repetative waveforms. Whereas, a digital one can store "once off" waveforms and print them via a computer.

The bandwidth figure for both the analogue and digital scopes indicates where the -3dB point is. So at 60 MHz, the signal seen on the analogue scope will be 3 dB lower than it really is. So you cannot rely on the voltage measurement of signals approaching the bandwidth limit.

40MS/s is a crappy sampling rate. Tektronix' low-end 60MHz DSO have a 2GS/s sampling rate. Don't let the ad copy fool you: a low sampling rate required repetitive operation to display and store a decent waveform.

A DSO only requires a computer if it's a PC board installed in the computer. Decent DSOs are stand-alone instruments that can be optionally interfaced to a computer.

There are arguments for and against analog vs. digital scopes. I prefer analog. When you really get down to it, storage is not an oft-used option as a necessity -- folks only use it all the time and determine that it's a necessary function only because they have it. In 40 years of hobby and professional electronics, I've never found the lack of a storage function to be a detriment to what I'm doing. High-end, lab-grade analog scopes are available these days for less than a tenth of their original value and are still reliable, high-quality instruments that can only be replaced by new models costing thousands.

Dean

I find the storage function of my scope to be a definite advantage. I wouldn't say it was a necessity because I could easily use a scope without that function, but to be honest now that I've owned one that does have storage capability, I probably wouldn't look to buy one without it. The reason I find it so useful is because of the help it can be during diagnosis. If ever I'm working on a product which is new to me, I can take some measurements on a good unit, store them to disk (or print them out) and I instantly have a record of what a signal should look like, the amplitude it should normally be, it's frequency, and any noise or distortion which can be considered normal at that point in the circuit. Then, when I'm diagnosing a faulty unit, I can use those records as a look-up. If I go to a point in the circuit and I feel that the signal looks a bit noisy, or perhaps it's just not the sort of signal I was expecting to see, I can refer to my good set of examples to find out if it's correct or not. So yes, I think it's fair to say that the storage function can become an invaluable tool.

Brian

ljcox said:

There was a recent thread on this topic, see "PC Ocsilloscopes" in this forum.

The advantage of an analogue scope is that it does not need a computer.

Click to expand...

There is a difference between a PC and digital oscilloscope. An analog oscilloscope is just that, a digital oscilloscope is a stand-alone digital version of the analog oscilloscope, and a PC oscilloscope is a digital oscilloscope that must run through the PC.

I personally like mixed-signal, but it was $5000

Well, I know 40MSamples/s is crappy. But, for periodic signals, the efective sampling is 1Gb/s. And the analog bandwidth is 100Mhz. And it is dual channel, with integrated waveform generator.
I can get one of this for 300-500 dollars.
this is my budget. I can't afford more.

the alternative is a second hand big heavy analog oscilloscope, 60-100 Mhz bandwith,...

In 1983 I needed an oscilliscope, and I convinced a bank to loan me $1800.00 for 12 months to purchase a Tektronix 2236. I got the scope I wanted, paid off the loan, and established a banking relationship that endures to the present.

Don't think small, just make sure you can carry the freight.

But I'm a student, not working now, and I just cant afford $1800

Ebay can be quite good for scopes sometimes, although you do have to be very careful about what you're buying and what the returns policy for each seller is.

Brian

patroclus, the specs you are listing for the digital scope sound suspiciously like those of the bitscope (https://www.bitscope.com)

Is that the one you are talking about? because if it is, I HIGHLY suggest you look elsewhere, I own one and I do not at all think it is worth the money. First off, one thing they don't make clear is that it cannot capture on both channels simultaneously for one-shot signals. It captures first one channel, then the other, making it OK for repetitive signals but worthless for one-shots. it only samples with 8 bit resolution (which makes it very coarse, waveforms look pretty fuzzy and you can't measure voltages with any decent resolution). The software is amateur at best, it is not very user-friendly, and it has been pretty unstable for me; if I leave the software alone for a few minutes with the scope running, often it is unable to capture anything from the scope anymore, and I have to quit and restart the software to continue using it. Also, every time you start the software, both channels display an offset from the 0v level on the grid, which you have to manually adjust to zero, and the setting does not save so you have to do it every time you start the program, which is a lot considering how often it crashes. And, it has poor support for scanning/zooming an already captured signal; generally to do this you have to just adjust the view and re-capture, which is a pain for one-shot signals that may be difficult to reproduce. I really think the software is an extreme weak point for the bitscope, and most of the issues should not be very hard to fix if they actually tried (like saving the offset values), but I've had mine for a couple of years now and only a few very minor software updates have come out, and none of them seem to have fixed a single thing; every time I email them about it I don't get very helpful responses.

Also, back on the hardware side, I have my doubts about some of their claims. When I have used mine for looking at digital signals such as serial data streams (using the analog inputs) I have noticed that they often look very rounded (like they've been low-pass filtered), even when they are only in the range of 10khz or so. The SAME signals, when I look at them on a real digital scope in the lab at school, look perfectly square and nice. When it's not giving an accurate view of a simple low-frequency digital signal, I really have my doubts about how well it would perform with high-frequency signals.

Honestly, I prefer to simply wait and bring my projects to school/work the next day and use the real digital scopes we have in the labs, rather than spend a couple of hours at home getting increasingly angry at the bitscope every time I need to test something. I have not yet finished putting together a digital buffer board to use the digital inputs (which sound much more reliable than the analog inputs), but if that works out well I may very well use the bitscope as nothing but a logic analyzer.

Yes, I'm talking about bitscope..
I heard good reviews from other people. Just keep in mind that a good and expensive oscilloscope will beat it up, by hundrends of miles. Just remember that they have a budget oscilloscope for 350 dollars... you cannot just put it along a 2000 dollar piece of equipment...
I don't have a very high expectation. And I don't need a great resolution or performance, as I'm very specialized in digital works. I want it to simple debugging and repairs,...
By the way, which model do you have??
other PC based scopes seem even worse.

**broken link removed**

500$ dollars.

- Dual channel and 80 Ms/s, double the rate of bitscope.
- 32k samples per channel... fair, a bit better than the budget bitscope (and the $500 dollar bitscope has 128k samples).
- 16 channel Logic Analyzer, synchronously sampled with the DSO (8 channels in bitscope).
- 100 Ms/s Arbitray Waveform Generator ($500 bitscope has similar)

It seems better than bitscope, BUT, now the main disadvantage; bandwidth is 60MHz. It seems good tool, but I wonder if lowering bandwidth from 100MHz to 60Mhz would "hurt" a lot...

any advice??

patroclus said:

But I'm a student, not working now, and I just cant afford $1800

Click to expand...

Of course you can't afford $1800.00 all at once. That is why you borrow the money from a bank and pay them a smaller amount every month. You gotta choose what you are going to spend your limited resources on.

While I do like having a digital storage scope, I'm also kind of cheap. So at home I use an old analog scope and find that I can do most of the things I want with it. In fact, back when I was just out of school, I bought a brand new Leader LBO524 30 MHz scope for $1000 and thought I was doing well with that. Nowadays, while i still use that Leader scope, I also like my HP1725A and I find it possible to find a TEK 465 or HP1725A or HP1740A for $50 if I am patient and persistent. These are all good performers and much better value than any digital scope out there.

I've used a lot of digital scopes and really dislike most of the older Tektronix models because they have really clunky controls. But I must admit I do like their TDS2012 and similar TDS20xx models, the ones with the colour displays. These are very popular in commercial electronics firms in the US.

A 20Mhz or 30MHz analog double-beam scope will do 99+% of all jobs you need a scope for, and usually do it better than a digital scope, and FAR better than a PC scope.

I would suggest getting a proper analogue scope first, then perhaps later getting a PC scope for the 0.0x% of occasions where it might be more useful?.

I like the mixed-signal scope I used in one the labs at the university. I knew it wasn't analog because it could do a bunch of digital functions, auto-measurements, and analysis math and stuff like that, but I could tell it wasn't digital either because it was the waveform *too* smooth and clean cut on the displays to be digital.

Just my luck, to fall in love with the $5000 scope.

Can analog scopes do FFT, and other math as well as auto-measurements? THe ones I saw couldn't.

Nigel Goodwin said:

A 20Mhz or 30MHz analog double-beam scope will do 99+% of all jobs you need a scope for, and usually do it better than a digital scope, and FAR better than a PC scope.

I would suggest getting a proper analogue scope first, then perhaps later getting a PC scope for the 0.0x% of occasions where it might be more useful?.

Click to expand...

Why do you say that an analog scope will do a better job than a digital scope?

They seem to respond faster and have finer resolution. THey are smoother too. Try one out.

Nigel Goodwin said:

A 20Mhz or 30MHz analog double-beam scope will do 99+% of all jobs you need a scope for, and usually do it better than a digital scope, and FAR better than a PC scope.

I would suggest getting a proper analogue scope first, then perhaps later getting a PC scope for the 0.0x% of occasions where it might be more useful?.

Click to expand...

Clearly Nigel, you use scopes for different things than I do, and I'm sure many others are no different. If you're doing a lot of RF and filter/amplifier types of circuits, then yes, an analog scope is perfectly useful... However, personally I do a lot of digital stuff, where I do a lot of one-shot captures of non-repetitive digital waveforms, and then use cursors to examine the signal timing, etc. And I have also made quite a bit of use of the FFT. And, since I like to try and document my projects (especially those for school, where it's necessary) I do a LOT of screen captures, saved to image files on either a connected computer or a floppy disk inserted in the scope (depending on the scope I'm using). Of course, I do also quite often have the scope in continuous trigger, looking at a signal that's periodic, which an analog scope can do, but I would estimate that more than 50% of the things I use a scope for are things that can't be done on an analog scope.

Not to badmouth analog scopes, and certainly not to say they aren't useful, they definitely have their uses, but saying an analog scope is suitable 99+% of the time is a tremendous exaggeration for the kind of things many hobbyists wish to do.

My university lab tech says we have more digital than analog scopes because it costs more to get a decent analog scope than a digital one...so don't think you will be saving money by getting analog.

If you only need digital waveform monitoring with storage, very decent logic analyzers are common place in the used market. They are obvoiusly not state of the art, but for what most people will get up to during hobby use they are great, maybe even overkill. Some are on the large/heavy side, but prices are low. Pair one with a used but good quality analogue scope and you're all set.