Basic scope questions...

[jBam]

Hi all - sorry for a "the basics" thread... I've been looking into scopes, have read many posts here and a lot online, particularly in the last 24 hours (most of which I've been awake thanks to my new daughter lol), and my head is spinning a little. I'm not sure if I'm missing terminology to better search forums and specs, but think it's time to ask some basics.

I'm keen to buy a scope... 50/50 education/practical use. To narrow down options let's just assume for now I'll go digital (I've read a heap, I get the benefits either way).

Uses:
*General scoping of Points in circuits for learning, including comparing to spice sims.
*line level scoping of i/o signals to assess circuit performance - e.g. I've designed a variable harmonic saturator - hoping to see i/o simultaneously on one screen.
*smaller signal traces like control voltages for a compressor, simultaneous with e.g. an input, output or both. How small?... e.g. a diode compressor (or any other small signal that an audio design might need)... so thinking: scope noise vs tiny signal might be a problem. Would like ability to scope DC only too, but I assume that's common?

Affordable 12 bit seems to be new. Something like the Rigol 802 or 804 has caught my eye... upper end of financials for though. Then there's the siglent sds1104x-e or keysight edux1052a... and there's usb options etc etc. I'd also like to future proof a bit... let's assume it's the only scope I buy for the foreseeable future, unless advice is "buy a cheap shitty one until you can afford an xyz".

Cost... has to be under AUD $1000, preferably under AUD $750 (roughly 500usd)... ideal would be much much cheaper... under AUD$500 (300usd) is my target and any decent scope in that range might be an insta-buy. But I degrees slightly... I'm particularly after understanding a few things first:

1) In lots of places it seems advice eventually shifts to "if you can afford 4 scopes, do that instead of two"... any thoughts? I'm imagining i/o and a control signal (3 traces) makes sense... maybe even two control signals at once (4 traces)... but do you guys use typically 2 or less on any range of project?? Is 4 a preferred solution for audio design, or is that more for other applications... I think I want 4 for the examples above... but not sure if I'm being over the top!? EDIT: Actually... I think I'll "need" 4... some of my hairbrained ideas have 3 control voltages ... still keen to hear general thoughts though as it might help me use a scope more effectively
2) so I might want to look at a sine wave through a circuit... sure... but what is the max display time called so i can look up specs in e.g the rigol 802/804.... could I zoom out and see a whole kick drum? Is that something a computer usb connection allows (e.g. long recordings of data)? Or is this just not how these digital scopes work?
3) does the rigol 802/804 generate sine waves etc... this called a "function generator" yeah?... and do i need one if it doesn't? E.g. could i use 24bit 44.1k audio... or a synthesiser oscillator? Or is a precision generator really more useful in audio design / calibration etc? If it is, I'd rather it was built in if possible...
4) for what I'm hoping to do (refer initial discussion above) am I even aiming for the right thing? Or is there a much much cheaper starting point that might get me 70-80% there while I see if 12 bit digital proces drop quickly throughout 2024?? E.g. I've just not looked into usb via computer... many post saying "no good for this and that" turned me off... but some of these comments might be out of date in late 2023.

Sorry for the long post. As always, any help would be greatly appreciated

Edits: grammar etc due to being stupidly tired lol

 

[Philbo King]

Even after using a Keysight 13 GHz 4 chan scope (which IS useful for microwave and fiber optic pulsed laser design work) I still preferred a simpler Tektronics scope for analog/audio electronics, mostly due to a simpler and more effective user interface. YMMV. Note that, for audio, a 4 or 8 channel audio interface on a PC is far cheaper and higher resolution than any scope out there.
Other scopes I've used and didn't like very much are LeCroy and Rohde-Schwarz. Both are capable but have inefficient and complex operator interfaces.

 

[Brian Roth]

I have "dipped my toes" into using audio interfaces and a computer for typical/mundane testing/measurements/etc. Guess I'm too old, but it was a PITA to get correctly operating reliably.

Ugh....I'm just being an Olde Phart.... LOL....

Bri

 

[ruffrecords]

For 99% of audio work a 20MHz two channel scope will do.

Cheers

Ian

 

[rock soderstrom]

Cost... has to be under AUD $1000, preferably under AUD $750 (roughly 500usd)... ideal would be much much cheaper... under AUD$500 (300usd) is my target and any decent scope in that range might be an insta-buy.

https://groupdiy.com/threads/i-need-a-new-oscilloscope-300€.77401/

 

[abbey road d enfer]

Point 1:As to "if you can afford 4 scopes, do that instead of two", it seems odd to me.
It is true that digital and analog scopes have distinct advantages.
Some digital scopes make you forget they are digital, but tend to be expensive.
Now is it what it means?
If it actually means 2 or 4 inputs(traces), for most audio work, a dual-trace scope is perfectly adequate. A 4-trace or more) scope is useful for digital applications.
Typically, digital oscilloscopes can be inexpensive and available in many different versions.
Analog ones tend to be rarer; there's not much choice for cheap analog ones, and the others are quite expensive.
Analog scopes have an instant response, when digital ones need to fill a buffer before displaying anything, that is a problem when tracing elusive phenomenons, such as clicks and noise.
I have two scopes, one digital that is small, doesn't require time to heat, and offers a coplement of visualiing a number of parameters.
I also have a "vintage " analog one, that I use only for tracing intermittent signals.

Point 2: You may want to check this:
https://www.electronicdesign.com/home/article/21199767/understanding-long-memory-in-a-dsoCheck the Rigol 1202 user manual; read pages 62-64
You will see that there is a trade-off between duration and time resolution.
E.G. with 20 000 points of 1ms, you can visualize a 20s sequence, but 1ms resolution does not allow seeing very sharp transients.

Point 3: If you already have signal generator, a scope with built-in DDS is worthy only if the scope includes a true Bode plot option, I mean one that displays amplitude vs. frequency. Many scopes just display amplitude and phase as numerical values, no graph.

Point 4: any USB oscilloscope is close to a PITA, if only becausee it has no physical knobs and controls to do the basic operations like setting level, position and time-base.
I had a Pico; its only advantage was size. It was too limited in performance and ergonomics.

 

[Cqwet Dbdfte]

Some digital scopes have a low pass button. The Rigol scopes are OK. Vintage scopes have maintenance issues, but those old tube scopes make nice auxiliary room heaters.
Much have happened in digital scopes due to improved semiconductor tech.
USB scopes never fell for that. Not for me.
Very low signal measurements may require a differential probe kit. The ground terminal on the probe is always connected to AC mains and dangling it around a circuit is not recommended.
Using two probes for differential measurements, while possible, has limitations, probe matching etc.

 

[Ivan K.]

I have a 4 Ch., 1 GHz LeCroy and an Owon 4-ch, 200MHz 8-14bit scopes but would recommend that you look at Picoscopes Oscilloscope | Pico Technology. All you need is a laptop and you have a complete, portable scope. They have great, free software, which you can download and use in a demo mode. For audio, I highly recommend a higher than an 8-bit vertical resolution-at least 12 bits or higher for best dynamic range and lowest noise, like their 20 MHz, 4000 series, PicoScope 4224A, 4424A and 4824A Oscilloscopes, or even their 5MHz, 4262, a 16-bit scope: 16 bit high resolution oscilloscope | Overview. Both of these scopes do spectrum analysis via FFTs and have built-in signal generators.

 

[abbey road d enfer]

I have a 4 Ch., 1 GHz LeCroy and an Owon 4-ch, 200MHz 8-14bit scopes but would recommend that you look at Picoscopes

Funny how one's favorite can be another's most disliked!
The lack of a proper front end is what made me reject the Picoscope I had (still have but available for a symbolic $ + shipping).

 

[zamproject]

still have but available for a symbolic $ + shipping

If it run on Mac I take it

As Ian say no GHz needed for audio
Last year I had to replace my old Metrix, tempted for a new digital scope, I finally catch an Hameg 30Mhz (303-6)
This is the last gen analog entry level, but great quality, you can get them in pristine shape for less than 200 bucks...
The only possible issue is size if you work on-site (not my case), but I don't care moving it from workbench room to studio room.

 

[Ivan K.]

Hi abbey,

Please day more about the lack of a "proper input"
I gave my 2408A Picoscope to a scopeless friend in need and have occasionally regretted it. It always worked for me...

 

[Gold]

Hi abbey,

Please day more about the lack of a "proper input"
I gave my 2408A Picoscope to a scopeless friend in need and have occasionally regretted it. It always worked for me...

I’m not Abbey but I have one and have the same opinion. Strike one was that it came with grabbers and not a scope probe. There is no easy way to use a scope probe with it. No thanks.

Strike two is that it’s fiddly to change parameters. Changing them on the fly is particularly annoying.

I’m not sure there is a strike three but the first two were enough to make me never use it.

 

[wst3]

full disclosure, I am old, and stubborn, and cranky!

I've tried a number of DSOs over the years, and they are getting better, but about the most challenging task for me is aligning tape decks.

I own two Tek and two Philips o'scopes, at that time one of the Teks stayed in my car, the other on my bench, One of the Philips also sat on the bench, but the number of times I needed two o'scopes was... very small. The fourth was a backup, back then I really did depend on a scope for my work.

So, let me echo the analog recommendation. A good, solid state, 2 channel, 40 MHz oscilloscope can be purchased used for very little money (I hate to quote prices because invariably they are wrong anyway.) If you stick with the old guard - Tektronix and Philips are really reliable. I also had a good experience with a Kikusi, don't remember the model, but the trace was bright and beautiful.

For audio work there is little need for dual timebase, and many will tell you the built in multi-meter is a waste, but I use it a lot more than I would have guessed.

Oh well, I had to look! There are several Tek 2205 models most with probes, all under $200. That's a very useful oscilloscope. There are about an equal number of Philips 32xx oscilloscopes in the same price range.

If you are patient I will probably list a couple of mine one of these days - but don't depend on that<G>.

 

[FunkyTek]

So, let me echo the analog recommendation. A good, solid state, 2 channel, 40 MHz oscilloscope can be purchased used for very little money (I hate to quote prices because invariably they are wrong anyway.) If you stick with the old guard - Tektronix and Philips are really reliable. I also had a good experience with a Kikusi, don't remember the model, but the trace was bright and beautiful.

For audio work there is little need for dual timebase, and many will tell you the built in multi-meter is a waste, but I use it a lot more than I would have guessed.

Oh well, I had to look! There are several Tek 2205 models most with probes, all under $200. That's a very useful oscilloscope. There are about an equal number of Philips 32xx oscilloscopes in the same price range.

I second the recommendation to at least consider a Tek 22xx analog scope or equivalent. They lack the digital features, but if you are looking for noise, they can be way better, due to resolution. They have a deep chassis though, so they take more bench space. They can be had very cheap, or at least the last time I looked.
There is also the option of a Tek465 or 475 2ch 100MHz scope. These are smaller than the 22xx series and built like a tank. But definitely vintage.

I have, at last count 15 scopes, mainly Tektronix, of all types, from battery powered to huge "mainframe" lab scopes , but I would not go that route.

Mostly I end up using either a "smaller" analog scope like my Tek 2215, 2246A, 2445, or 2465. The 22xx series can be had cheap and are great scopes.
The best for a small bench are the flat panel digital scopes, of which there are hundreds to choose from. I end up using my TDS2214 (4ch color, 100MHz) a lot because of it's size and versatility. I have standalone spectrum analyzers but the convenience of an FFT in the digital scope is nice.
In my opinion, the best analog scope (Biased cuz I worked in the IC group for it) is the Tektronix 2465. But if you kill a preamp, that channel is a gonner because you will not likely find a replacement. They stopped making them in the 80's I think.

BTW: do research on ENOB (effective number of bits). Many digital scope manufacturers play games with the resolution. The realizable resolution can be less than the banner spec.

I have used 2 time bases in audio, for looking at signal and say power supply noise, at the same time. Looking to find correlation with odd behavior. Signal & control is another usage, although typically I will just use two different scopes...

I use my Fluke 867BE battery powered scope (basically a graphing DMM) for the convenience of no AC plug, so field work is a plus, also when I need a true floating scope. Something like that might be an option, but I think you are looking for more channels and features.
I just recently picked up a tiny Chinese DSO/component analyzer. I am jonesing to try it out. For $20 on Temu, it was too temping to pass up. My basic "is it alive" test went well. I can just keep it in a pocket of my go-bag, it is so small. It's tiny, low res, and only 1 channel. But a fully contained DSO in a shirt pocket?

I worked in scope design groups for a number of years. So yeah, I have a stupid collection of scopes. Eventually I will cull the heard, but it will be hard choices.
In all fairness, some are paired with test setups, like my HP vector scope and it's matching signal generator and distortion analyzer, My Tek503 is in a rack shelf with it's signal generator and distortion analyzer. These are not used as general purpose scopes and never powered up separately.

 

[Jeffw5555]

I second the vote above for the 22XX Tektronix series, however I don't know their availability Down Under. The 465 and later the 2215 series were the workhorses of industry so there are tons of them out there at stupidly low prices, you can buy a working 2215 here in the US for $50 used.

At one point I had maybe a dozen different Tek scopes here in my lab, at this point the one that gets used most often is a 2215A, followed by a modern Siglent SDS1104X. My favorite scope is the 485, followed by the 7904A mainframe, but I haven't used either one of them in maybe 15 years. Don't know if they even work now, been meaning to clean out and sell all my old Tek test equipment as it just sits around.

The reason the 485 is my favorite is from my first job 45 years ago, I used a 485 that the owner of the company bought new when it came out in the early 70's and he would brag how much he paid for it new. (The 2023 real dollar value calculates out to $29,000 in today's money. You can buy a working 485 on eBay for under a $100 now, sheesh)

About 20-25 years ago, the US military offloaded all the 22XX scopes for newer stuff, and I bought pallets of them for stupidly low prices, fixed up and calibrated them as best I could and resold them. The most common failure in a 2215 is the power supply, and those are easy fixes. I also bought 485's a lot and those are diabolically hard to repair, but amazing how they fit what they did in those chassis! (Almost all 485 failures are due to crappy shorted tantalum caps, consequently over the last 20 years I have not designed in a tantalum cap in any of my designs)

 

[FunkyTek]

here is a reasonable choice for a new cheap 100MHz 2ch scope. At $240US, it fits your grab it range. Not sure of it's vertical resolution though. I just glanced at the specs and on paper, it should do what you want.
Hanmatek DOS1102 Temu has it for US$236.99
They claim it has 10K storage, which on paper is better than my TekTDS2014
It has a USB interface that supports both thumb drive and SCPL communication with a PC. Run LabView on your PC and you should be able to both dump captures to the PC but also control the scope.
No personal experience with their scopes, but it looks temping

 

[FunkyTek]

I second the vote above for the 22XX Tektronix series, however I don't know their availability Down Under. The 465 and later the 2215 series were the workhorses of industry so there are tons of them out there at stupidly low prices, you can buy a working 2215 here in the US for $50 used.

At one point I had maybe a dozen different Tek scopes here in my lab, at this point the one that gets used most often is a 2215A, followed by a modern Siglent SDS1104X. My favorite scope is the 485, followed by the 7904A mainframe, but I haven't used either one of them in maybe 15 years. Don't know if they even work now, been meaning to clean out and sell all my old Tek test equipment as it just sits around.

The reason the 485 is my favorite is from my first job 45 years ago, I used a 485 that the owner of the company bought new when it came out in the early 70's and he would brag how much he paid for it new. (The 2023 real dollar value calculates out to $29,000 in today's money. You can buy a working 485 on eBay for under a $100 now, sheesh)

About 20-25 years ago, the US military offloaded all the 22XX scopes for newer stuff, and I bought pallets of them for stupidly low prices, fixed up and calibrated them as best I could and resold them. The most common failure in a 2215 is the power supply, and those are easy fixes. I also bought 485's a lot and those are diabolically hard to repair, but amazing how they fit what they did in those chassis! (Almost all 485 failures are due to crappy shorted tantalum caps, consequently over the last 20 years I have not designed in a tantalum cap in any of my designs)

Sounds like your shop and my trailer of equipment have a lot in common! Probably because we are the same age.
485's are great. I put a LOT of hours on a 465M (with the DMM on top), but didn't have a 485 around. Partially because I had a row of lab scopes (7834,7854,7903RM). I didn't need 1GHz BW, so I donated the 7903RM to the Tektronix museum, along with my complete collection of 547 plugin's. They didn't want the 547 though.

You are right though, the 2215 is the one on the test bench that gets turned on the most often. I spare the 2465 for only when I really need bandwidth or bizarre trigger condition.

You are also spot on about how hard it is to work on a 465/475/485. They are super compact, for a CRT-based analog scope. That made them really handy as a field scope. But the down side is they are a work of art, packing art. Also agree that tantalums can be an issue. They do not like stressful jobs.

 

[Jeffw5555]

Also agree that tantalums can be an issue. They do not like stressful jobs.

The tantalums on the 4XX series are all over the power supply rails, hundreds of them on a 485. They short to near zero ohms when they go. I ended up using a milliohm meter to zero in on the board to find the likely shorted one to minimize the work in pulling the various boards in the jam-packed cabinet. Wasn't unusual to find multiple shorted ones on the same scope, which means some probably went without power even applied.

 

[jBam]

Hi all - wow, thanks for all of the disussion and guidance. This thread sat undiscussed, so I'd assumed that my long, tired and rambling opening post had scared everyone off haha. I'd actually already bitten the bullet and purchased a Rigol DHO804. Just slightly wondering if I should have gone 904 (or from comments on "2 channels is enough" above, the 902) for the built in AWG - whilst I can buy a (likely better) function generator separately for a total cost less than a 904, the compact all in one seems like a good option in hindsight for my needs. For now, I'm thinking of using Daqarta (software) which seems pretty comprehensive, even in the free version.

I got a good deal on the 804, so I'm happy with that regardless. I went with... an Aliexpress option lol... I find Ali so sketchy, but after a lot of web searching for the store name I'd found, I noticed it came up on sites like EEV Blog as a legit / trusted option. Anyone interested could check out "Jiutian Instrument Store"... Arrived factory sealed doubled-boxed (both Rigol boxes), and was about AU$200 cheaper than purchasing new in Australia even after taxes etc... I think prices are similar to new US prices though, so may not have a huge saving for people in the US.

Still expecting to use 3-4 probes in some compressor designs I've been working on... Just hoping to see / record signal input/output and sidechain signals in one pass... but time will tell. If I only ever use 2 probes, perhaps I'll just sell it and buy a 902 instead

For now I haven't had any chance to REALLY test it out... I've calibrated it, and used it with some synth oscillators to learn basic triggering etc. About to build some mic pres and microphones though, so that's when it'll get its first proper run

 

[Walter66]

Using a Hantek 5072P since ten years. It's not what an excellent analog scope can offer, but for such low prices its good enough for an audio amateur. Just wanted to get a new one instead of a thirty year old one. Before that, I used a Hameg analog scope. It was really expensive, for the 1980's it was excellent. But today all those scopes look a bit outdated to me. They may have its strengths over digital ones but for most applications, I'm good with the ditigal one.