Picking a used oscilloscope

[vettedude]

I have been scouring our local used sites to find a sub $250 scope that I can use.  I have run into three types 1. Purely old analog (ie Tek 463) 2. Old digital storage (ie Tek 2232 and 3. USB scopes new (ie Hantek 6022BE).  This is a minefield of information.  I am a retired Electrical Engineer from the Telecomm industry and am rekindling my childhood interest in tube amplifiers.  I know that using a scope with a signal generator can help trace out the circuit.  I am also dabbling with arduinno and raspberry pi for fun.  So I am looking for something to tread in both worlds.  I got quite interested in the tek2232, but then when I read the blogs here on circuit complexity, age and repair issues I got a bit worried.  Seems a lot of these older scopes have proprietary chips in them and certain parts go they are a boat anchor.  Any thoughts suggestions to guide me would be appreciated, as at the moment I am sort of favouring the hantek, but the 2232 does look interesting.  I have looked at various scopes, multiple models of tek, Hp, Hitachi, gould, and others I had never heard of.  Thanks

[Siwastaja]

With there requirements, forget analog scopes. They are fine for many audio enthusiasts, though.

USB scopes are cheaper than equivalent stand-alone ones, but you are going to get a minefield of information, because this if anything is a matter of preference and personality. I prefer stand-alone scopes any many engineers and hobbyists share this opinion, but some are completely happy with USB scopes.

Is there an option to increase your budget to $350? That would bring you a new Rigol DS1054Z, which is a real thing you can do some serious work with later, if necessary; the specifications won't run out too soon. If this is too much money, I think you could be able to find a used digital storage oscillosscope for less than $250. I'm seeing a used Rigol DS1052E on Ebay for $99 in the U.S. of A right now, for example. This particular piece of equipment satisfies your requirements for a long time as well, I would say.

[Adrian_Arg.]

Hello friend, I have as a hobby, small electronics and arduino projects, I acquired a rigol ds1054z two years ago, today I think it was the best investment for my hobby, for price and benefits it is the ideal oscilloscope, you can enable 100Mhz and you It comes with all licenses enabled, but could not be ordered.

in spanish
hola amigo, yo tengo por hobby, pequeños proyectos de electronica y arduino, adquiri hace dos años un rigol ds1054z, hoy creo que fue la mejor inversion para mi hobby, por precio y prestaciones es el osciloscopio ideal, se le puede habilitar los 100Mhz y ya viene con todas las licencias habilitadas, mas no se pude pedir.

[RoGeorge]

For second hand oscilloscopes be aware that most of them come without probes, and the probes may cost much more than the second hand oscilloscope itself.

Maybe add a little extra money and buy a Rigol DS1054Z.  It comes with 4 probes, can be unlocked to decode serial signals, it is fully digital (i.e. can take a single trace, or can capture and download the signal with a PC, has SCPI, etc.), can do at most 4x100MHz (with 1x1GSa/s ADC).

Rigol DS1054Z is still the best value for the money, IMHO.

One pitfall of DS1054Z is when used for very low signals (in the 1mV or less range, still usable but not recommended), because the amplifiers are noisy when compared with old analog boat anchors like Tektronix.

Another thing to know is the XY mode of digital oscilloscopes is not as great as it use to be in analog oscilloscopes.  This is valid for any model.  Simple Lissajous figures are not a problem, but some other more tricky XY things, especially when intensity modulation of the spot is expected, are not so great.

[vettedude]

Thanks everyone, I will check out the Rigol scopes.  Quick look here in canada and the recommended one will be about $530 cdn all in, so I may have a budget issue, well that is my problem.  The one on ebay (smaller screen) looks interesting, thanks.

Couple more questions:
1. I get the use of 2 channels, but why would I need 4 channels
2. 50 mhz. or 100  In my readings so far I see 5x for digital  (so 10mhz for 50) and 2x for analog (so 25mhz for 50).  Lots of talk of 50 vs 100 and then it goes nuts with freq all the way up and the costs rising at the same rate.

[Siwastaja]

One pitfall of DS1054Z is when used for very low signals (in the 1mV or less range, still usable but not recommended), because the amplifiers are noisy when compared with old analog boat anchors like Tektronix.

You can turn on averaging on digital scopes, which is similar to analog scopes drawing the triggered signals over the top of each other, with the persistence of vision (and phosphor).

This averaging takes the input stage noise away very well. Of course won't work for non-repetitive signals, but they wouldn't be visible on the analog scope at all.

[rstofer]

Thanks everyone, I will check out the Rigol scopes.  Quick look here in canada and the recommended one will be about $530 cdn all in, so I may have a budget issue, well that is my problem.  The one on ebay (smaller screen) looks interesting, thanks.

Couple more questions:
1. I get the use of 2 channels, but why would I need 4 channels
2. 50 mhz. or 100  In my readings so far I see 5x for digital  (so 10mhz for 50) and 2x for analog (so 25mhz for 50).  Lots of talk of 50 vs 100 and then it goes nuts with freq all the way up and the costs rising at the same rate.

I can't see buying a relatively new DSO from eBay.  What did the owner do to it that causes them to sell something so new?  OK, I'm cynical...

4 Channels come in handy for uC projects where you want to decode the SPI protocol.  It is the reason I bought the DS1054Z.  Truth be known, I spend about 90% of my time with just one channel.  Sometimes 2 but rarely 4.  Maybe I could use a cheap Logic Analyzer and buy the Siglent SDS1202X-E scope with only 2 channels but 200 MHz bandwidth.  Same kind of money, newer model than the Rigol and purportedly a better user interface.

The real spec is rise time but that is tied to bandwidth by a constant so they aren't independent.  tr = 0.35 / bw.  So a 50 MHz scope will have a rise time of 7 ns.  You won't be able to measure rise time < 7 ns.  A 100 MHz scope will have a rise time of 3.5 ns and so on.

But, yes, the scope acts like a low pass filter and will only pass harmonics of square waves up to a certain point.  The question is:  How good does the square wave need to look with limited harmonic content.  Here's everything you need to know about the subject:

https://www.ednasia.com/find-a-signal-s-bandwidth-from-its-harmonics/

Even on my projects running 100 MHz inside an FPGA, I seldom have an external signal running much more than 10 MHz.  I certainly don't push the unlocked 100 MHz capability of the DS1054Z.  If I do, I have an old 350 MHz Tek 485 analog scope that mostly just stands on the floor under my bench.  It hasn't been used since I got the Rigol.

Since you are working on tube amps at higher voltages, I'm not going to get into the Digilent Analog Discovery but you might want to check it out for all your other projects.  Two channel scope, two channel Arbitrary Waveform Generator, +- 5V adjustable power supply, 16 channels of digital IO which can be used for a logic analyzer and capabilities that simply can't be matched by a DSO.

Start at Reply 52 here there are 4 projects which are nearly impossible to do with a simple scope:

https://www.eevblog.com/forum/testgear/starter-scope/50/

Using the BNC adaper and 100x probes, the AD2 could handle the higher voltages but there is too much risk of an 'oopsie' that destroys the unit.  It wasn't really made for watching B+.

[Siwastaja]

50MHz is likely enough for a "typical" advanced hobbyist, but this fully depends on what you are doing. OTOH, if 50MHz isn't enough, the chances are you need even more than 100MHz.

So if you can find a good 50MHz scope for a good price, go for it. If, after a few years, you advance to something requiring more BW, you may able to find a proper 300MHz scope or even a used 1GHz one, and then you may have larger budget to spend as well.

I wouldn't go much below 50MHz, though, even for a hobbyist! You may need to debug a 10MHz SPI, or look at a 16MHz clock signal, or look a parasitic LC ringing at 50MHz-ish...

In case you have missed, note that Rigol DS1054Z "50MHz" is actually a 100MHz scope and the 100MHz BW can be unlocked by paying extra - or of course, like everybody does it, cracking it with a keygen you easily find on the 'net for free  .

I have a 4-channel scope, but 95% of the time, use 1 or 2 channels. 4% of the time, using more is just very convenient, but measurements could be rearranged with two channels if absolutely necessary: for example, with SPI, first look at SCK & nSS, then rerun the test and look at SCK & MISO, then SCK & MOSI. It's quite rare that you 100% absolutely must have 4 channels.

[David Hess]

I got quite interested in the tek2232, but then when I read the blogs here on circuit complexity, age and repair issues I got a bit worried.  Seems a lot of these older scopes have proprietary chips in them and certain parts go they are a boat anchor.

I really like my Tektronix 2232 for general purpose use.

Age is more of a problem than any custom integrated circuits or parts; they tend to be pretty reliable.  Most problems involve dried out capacitors in the power supply.

[vettedude]

Thanks everyone, a bit more reading on my part is needed clearly.  Appreciate all of the responses and your patience. 

My budget of 250 is CDN or about 185 US (not too big).  One more thing, based on the comment from David Hess, I have a bit of a line on the 2232 and may be able to get it from between 100 and 150 cdn (was 200).  Is this throwing good money after bad?  I think my budget is going to get in the way of a rigol for me!! Bit frustrating as out of reach at the moment.

[Siwastaja]

Really, don't get an analog scope. They shine in some areas and some people are very sentimental about them, but as you mentioned Arduino, that's a total turn-off. In modern "general-purpose" electronics hobby or profession, about 90% of what you do requires a digital storage oscillosscope.

DSOs have totally redefined what people do with oscillosscopes. Today, an oscillosscope is universally considered an instrument which records voltages as a function of time. Analog scopes just don't do that (there were analog storage oscillosscopes but let's ignore that because it's a very limited feature). What they do instead, is left as excercise for the reader .

So when people talk about oscillosscopes, they often actually talk about DSOs! So get a DSO to do things people are talking about.

You can (and maybe even should!) additionally get an analog scope because it may teach you in a different way, they are nice instruments, but if you buy it as your only instrument, you are going to prevent yourself from doing work with digital circuitry (such as microcontrollers). For this purpose (learning, seeing how an analog scope works) don't spend over $50 on it. You are going to use it very little later.

Yes, an old analog scope + a cheap USB logic analyzer able to decode most common communication buses, could do the trick, though. That's more about matter of taste.

In general, buying a digital storage oscillosscope, even the older 2-channel 50MHz Rigol, makes you able to achieve so much more. In comparison, analog scope is actually a toy (and I know many people don't like this being said. I like analog tech myself. But separate test&measurement from playing around. You want tools that help you, not tools that dictate how you can work.)

[Doctorandus_P]

If you are also interested in "arduino" and other similar digital stuff, then also buy a Logic analyzer.
I have both a scope (older Rigol DS1052E) and a handful of these LA's as I tend to give them away to friends and such.
For microcontroller related stuff I use my Logic Analyser more then I use my Oscilloscope.

These things cost < USD10. and they work quite nice with the Open Source Sigrok / Pulseview software.
For an example of such a thing look at:

https://sigrok.org/wiki/MCU123_Saleae_Logic_clone

But any board with a Cypress CY7C68013A will work nicely.

About an oscilloscope. Maybe I'll consider buying second hand if it's below EUR100 but if you're looking for an USD250 second hand. I'll reccomend to save a bit more and buy a new one.

6022BE is hardly a tool, but more of a toy. I thought it also had pins for logic anayser inputs, but I seem to miss a connector and some IC's on the PCB: https://sigrok.org/wiki/Hantek_6022BE It's hardware is pretty limited, but the biggest bummer with this thing is the low quality software. But still, it may be usable for the right application.

[rstofer]

If your budget won’t handle a DSO, so be it.  If the 2232 works and it has probes, go for it.  Any scope is better than no scope.  I have had lesser scopes and done nicely.

[Electro Detective]

If you don't want to mess with aging Tek scopes, and need instant plug and play with analogue dsiplay,
many vintage Made In Japan scopes are pretty good and reliable   

as are 20 to 25 mhz generics sold at electronics stores about 10 years ago, Taiwan made I think?

Simple, reliable, easy to CAL, and most internals look generic

DSO buyers dump them cheap all the time, convinced they won't need them again and or regain bench space
just make sure -everything- works.
Oscilloscopes with issues beyond the owners repair skill set, are a mega party pooper   

I keep a couple around for a fast no brainer setup, and cover them over to keep fine dust from entering when not used

Bonus apps > if boxed, they serve well as temporary bench risers,
and cat nap    door stops too

 

[tggzzz]

Tube amplifier => high voltages and questionable isolation and grounding.

Find the appropriate type of probe, then see how much money you have left for a scope.

Have a look at some of the safety references and probe-type references here https://entertaininghacks.wordpress.com/library-2/scope-probe-reference-material/

Think very carefully if you want *1/*10 probes, and strongly consider HV isolation probes.

As for the arduino/RPi stuff where you are interested in the digital signals, strongly consider using a logic analyser plus printf statements. A scope is necessary for signal integrity, i.e. ensuring that the analogue waveforms will be correctly interpreted by the digital receiver. Once that is ensured, flip to debugging in the digital domain.

Also consider the Analog Discovery 2; it combines several instruments and is small. But be very careful about what voltages you attach it to, since a mistake could also take out the attached computer.

[BravoV]

One word ... warranty.

Expanded explanation on above one simple word ...

A. In order to fix a broken scope, sometimes you need "other" working scope. Do you other working scope ?

B. Sometimes even an "experienced" and expert level owner can not fix when it comes to unobtainium chip that is broken inside. Do you feel confident enough to fix it ?

C. Will you able to sleep well, knowing once bought, used just for few days, and then suddenly its dead without warranty.


My 2 cents.

[NivagSwerdna]

Any thoughts suggestions to guide me would be appreciated

My 2c worth... Avoid boat anchors... you will spend your time frustrated and not have any desk space.  Any modern digital scope would do... I bought a Rigol 1052E when they appeared and it has transformed my dabbling.  Any digital scope will do.  I see 2nd hand 1052Es on ebay pretty cheap sometimes.  Moving up there are 4CH scopes like the now slightly long in the tooth Rigol DS1054Z; the advantage of many channels is it opens the world to serial decode which you might find useful as well as obviously being able to visualise more than 1 thing (relative to a clk) at a time.
Any digital scope of the last 5 years would be good for you... just keep checking the listings.
It's an expensive purchase but it is possibly the most enlightening one you will ever make.   

[wizard69]

If that scope is working I'd buy it no question, however I'm a tool nut so take that into consideration.   This especially if you can talk them down to $50 or $75 which is entirely possible for old scopes.   As others have stated any scope is better than no scope and for tube amplifier work the scope should be fine if it is working.   Yes it is big and a boat anchor might be lighter but at that price you get a lot of capability cheap.   

You will need more instrumentation to do serious Arduino / microprocessor work.   Here is the thing other instrumentation includes logic analyzer support along with digital storage scope functions.    Others have already hinted at this but a modern digital scope is far more than a "scope", even without dedicated logic analyzer inputs, they can be programmed to decode a whole bunch of serial protocols and display that data.   In a way the scopes screen has becomes a display device.   This display can support counter functions, DMM functions and all sorts of "stuff" including full blown logic analyst on more expensive scopes.   Basically the bigger the price tag the more info that screen can deliver to you!

This is where USB instrumentation ( or even PCI-Epress cards for that matter) come into play.   In many cases you can get a lot of functionality in a USB based device because some of the expense is left to the PC the USB device is connected too.   As has already been hinted there are strong opinions with regards to USB based instruments.   Some people really like the knobs and buttons of traditional instruments while others like that software has far fewer limitations with USB based hardware.   I'd give serious considerations to USB tools especially to get started with instrumentation.   An example of a logic analyzer for USB: https://store.digilentinc.com/digital-discovery-portable-usb-logic-analyzer-and-digital-pattern-generator/, which is just one example of many available these days.   Nothing is perfect of course, many would reject USB tools, but I see a lot of value in this approach.   Even Keysight has some nice USB based tools if you are willing to pay for them.   

With today's 27" and larger 4K displays you can have your IDE onscreen right next to your instrumentation and not suffer one bit with USB hardware.   The reality is modern computing has eliminated many of the shortcomings of working with USB based tools.   You literally can have a whole test rack on screen and still have room for a IDE.   It is a different world than even a few years ago, your computer will not slow to a crawl, screen real estate will not suffer and of course you get high resolution on those screens.   this might sound like I'm champion USB tools, this isn't the case I'm just trying to point out that we live in a different world these days.   In fact there are cases where more traditional instruments are probably a better path to follow.   USB is a personal choice and can be cost effective in a home shop.

Thanks everyone, a bit more reading on my part is needed clearly.  Appreciate all of the responses and your patience. 

My budget of 250 is CDN or about 185 US (not too big).  One more thing, based on the comment from David Hess, I have a bit of a line on the 2232 and may be able to get it from between 100 and 150 cdn (was 200).  Is this throwing good money after bad?  I think my budget is going to get in the way of a rigol for me!! Bit frustrating as out of reach at the moment.

[tggzzz]

If your budget won’t handle a DSO, so be it.  If the 2232 works and it has probes, go for it.  Any scope is better than no scope.  I have had lesser scopes and done nicely.

Precisely.

Use skill and imagination to get the most out of whatever tool is available to you.

Don't try to find one tool to do everything; select an appropriate tool for each job.

[David Hess]

One more thing, based on the comment from David Hess, I have a bit of a line on the 2232 and may be able to get it from between 100 and 150 cdn (was 200).  Is this throwing good money after bad?  I think my budget is going to get in the way of a rigol for me!! Bit frustrating as out of reach at the moment.

If the 2232 allows for a return if it is not working, then that is a reasonable price.  Of course this assumes that you are in a position to evaluate it.  At that price I would buy it immediately but I already have working oscilloscopes and know how to refurbish a 2232 if necessary.

Cheap but acceptable 100 MHz probes can be had for $15 (US) each.

But as others have pointed out, the risk is that you end up spending more time maintaining the oscilloscope than using it.

[Electro Fan]

One more thing, based on the comment from David Hess, I have a bit of a line on the 2232 and may be able to get it from between 100 and 150 cdn (was 200).  Is this throwing good money after bad?  I think my budget is going to get in the way of a rigol for me!! Bit frustrating as out of reach at the moment.

If the 2232 allows for a return if it is not working, then that is a reasonable price.  Of course this assumes that you are in a position to evaluate it.  At that price I would buy it immediately but I already have working oscilloscopes and know how to refurbish a 2232 if necessary.

Cheap but acceptable 100 MHz probes can be had for $15 (US) each.

But as others have pointed out, the risk is that you end up spending more time maintaining the oscilloscope than using it.

+1 If this is your max budget, for that price range, if you can evaluate it before purchasing it or return it for a full refund it will be a good scope that will get you going with your interest in tube amps.

If you can stretch to a Rigol 1054Z you will probably enjoy it more for your interest in Arduinos.  Not sure if tequipment.net ships to Canada or if there is any tariff fee but they will give you a discount for being an EEVer - might help bring the a 1054Z in range.

If the Rigol isn’t feasible (or even if it is), as suggeted above, you can add a $10-15 eBay logic analyzer plus free Sigrok Pulseview software to help with your digital interests.