Scope for mostly digital stuff, Pico, Siglent, Rigol or what?

[mat85]

Hello everyone,

This is my first post here so welcome everybody, I'm glad to be here as this board is very often suggested by uncle Google.

Don't kill me please for this question as I know it's very often here but AGAIN - just another newbie is looking for scope advice.
I've read some topics here but I decide that maybe my case I more unique

What I will be doing mostly:

- LA
- Raspberry Pi projects (and using gpio, pwm, etc) and debugging them
- programming and debugging microcontrollers (stm32, arduino, esp32), I do not rule out making some small projects for profit (I'm a professional programmer)
- some electronic reverse engineering (eg. decoding some protocols like remote control for garage doors, etc - just for fun)
- EDIT: I have plans to work on FPGA in near future

Only on special occasions or rarely:
- repairing some electronics stuff (old laptops, radios, pcbs, etc)
- maybe one day I will try to understand car electronics and diagnose this (just for fun, nothing special)

I'm a passionate and like electronics. I have some mid-school theory and practical background (soldering, measuring, etc). Generally I like measuring staff, plots. numbers - sometimes just for nothing, sometimes for specific purpose. I know, kinda odd, but that's me. I just want to have better fun from this passion and sometimes very useful devices.

My budget can stretch to:
- Siglent SDS2104X (full option for same price as below)
- Siglent-SDS2104X Plus + some staff for LA (with hack possibility)
- Rigol MSO 5000 (5074 + hack possibly)
- ? (I don't know other good options like maybe keysight? I'm green here)

I completely don't know what about pure USB scopes like PicoScope, there are not too many opinions about them, monitor as scope screen seems attractive if provided software is good (I don't know if it's - that's assumption), but they are expensive considering parameters.

Important note which may affect some recommendations: I already have LA DSLogic Plus 16ch

I've watched many movies and articles about "Siglent vs Rigol", so my question is (assuming one of them will be recommended): which is better for my profile (mostly LA stuff). I know Rigol has noisy frontend but allegedly is "more digital oriented" (whatever it means). Noisy frontend does not seem to be a big problem as I won't be working with small mV scales (correct me if I'm wrong). On the other hand Siglent frontend looks much more polished and fluent, firmware updates are more frequent (my impression) and it's generally better for analog stuff but more expensive of course. This is what I know from my amateur research.

Thanks in advance, appreciate any comments and suggestion from more experienced folks!

[tautech]

Welcome to the forum.

SDS2000X is older design with individual channel controls and the later SDS2000X Plus models use a shared vertical control and offer more and better features.
I had SDS2304X for some years and the older SDS2304 before that.
SDS2104X Plus is a lot of scope for what you plan to use it for.
To save a few $ there is a DIY SPL2016 equivalent thread here on EEVblog.

[egonotto]

Hello,

if you buy Siglent SDS2104X Plus with digital probes it costs 1826€. And you need hack to use the digital probes.
If you buy Rigol MSO5074 with digital probes it costs 1426€
I think USB scopes are not so useful for you.

As the noisy frontend of Rigol is not important for you, I believe that Rigol has a better price/performance ratio for you.

But for me a low noisy frontend is very important, so I would by the Siglent SDS2104X Plus

Best regards
egonotto

[nctnico]

There could be limits to the use of digital channels on these models (operating speed, versatility, etc) so make sure to investigate that thouroughly before buying.

[mat85]

There could be limits to the use of digital channels on these models (operating speed, versatility, etc) so make sure to investigate that thouroughly before buying.

Hi nctnico! Thanks for your reply. I know about some limitations.

From my research I know that on Rigol, for instance, search function can be only done in analog channels, which is cumbersome process when I use a digital probe for testing. This can be described as "limitation" but I don't know if you have same problems in mind. On the other hand Siglent, allegedly, does not allow to search in LA at all (correct me someone if this is a wrong statement)

Thanks in advance,

[tautech]

From my research ............

Did you find these:
https://www.eevblog.com/forum/testgear/low-cost-logic-analyzer-probe-for-rigol-mso5000-easyeda-project/
https://www.eevblog.com/forum/testgear/diy-logic-analyzer-probe-and-pods-for-siglent-scopes/

Do note there are zero active components in SPL2016 whereas for the Rigol one there are and at this time apparently quite hard to obtain.

[nctnico]

The hardware for the digital channels won't do you much good if the oscilloscope itself doesn't allow to use these to the fullest extend. Hence my suggestion to research that thouroughly first.

[thm_w]

The DSlogic plus is already better than anything you'll find in this class of oscilloscope.
Any sort of data decoding is far easier to do on a PC. And if you are programming/debugging, its going to be on the PC anyway.

Get whatever scope you prefer outside of LA use, IMO. Unless you expect to be using more than 3 analog channels at a time (1 for digital trigger).

[mat85]

@tautech: yes, I know about these threads. From my POV additional spend for the genuine probes is not a problem (generally), but sometimes software option+probes is too expensive in comparison with a scope price.

The DSlogic plus is already better than anything you'll find in this class of oscilloscope.
Any sort of data decoding is far easier to do on a PC. And if you are programming/debugging, its going to be on the PC anyway.

Get whatever scope you prefer outside of LA use, IMO. Unless you expect to be using more than 3 analog channels at a time (1 for digital trigger).

Thanks for your interesting opinion. Separate LA like DSLogic is really better option than dedicated MSO?

[tggzzz]

Welcome to the forum, and thanks for a thoughtful question.

My rule of thumb is based on using digital tools for digital signals, analogue tools for analogue signals, RF tools for RF signals, and applying the aphorism in my .sig

In practice that means:

• first use a scope to check that the analogue waveforms are correct
• if the analogue waveforms are TTL/CMOS/etc conveying digital signals, then use a scope to ensure the analogue waveforms will be correctly interpreted as digital signals by the receiver - especially voltages and setup and hold times
• once you have ensured that "signal integrity" and you have valid digital signal, change to debugging them with digital tools
• for digital logic signals, use a logic analyser. They are cheap, capture many signals, and decent ones have triggers and filters that allow you to ignore uninteresting "noise" and concentrate on what matters
• where you are interested in the information in a digital signal, use a protocol analyser. These are specialised for specific protocols with useful triggering and filtering, but never overlook the power of a simple printf() statement (or logging equivalent)

[nctnico]

@tautech: yes, I know about these threads. From my POV additional spend for the genuine probes is not a problem (generally), but sometimes software option+probes is too expensive in comparison with a scope price.

The DSlogic plus is already better than anything you'll find in this class of oscilloscope.
Any sort of data decoding is far easier to do on a PC. And if you are programming/debugging, its going to be on the PC anyway.

Get whatever scope you prefer outside of LA use, IMO. Unless you expect to be using more than 3 analog channels at a time (1 for digital trigger).

Thanks for your interesting opinion. Separate LA like DSLogic is really better option than dedicated MSO?

The advantages of an MSO are:
- time relation with analog signals (think looking at ADC input level and digital output code)
- realtime screen updates (if you are lucky, decoding updates quickly as well but the models you choose are not the best at decoding anyway)
- one tool that does it all so no need to swap probes around and so on
- reasonably high bandwidth & samplerate compared to typical USB logic analyser

In general an MSO can do all that a USB logic analyser can do. If you need more features like deep memory, advanced triggering, etc then you'll need to look at the high-end logic analysers from Tektronix or Agilent.

[tggzzz]

@tautech: yes, I know about these threads. From my POV additional spend for the genuine probes is not a problem (generally), but sometimes software option+probes is too expensive in comparison with a scope price.

The DSlogic plus is already better than anything you'll find in this class of oscilloscope.
Any sort of data decoding is far easier to do on a PC. And if you are programming/debugging, its going to be on the PC anyway.

Get whatever scope you prefer outside of LA use, IMO. Unless you expect to be using more than 3 analog channels at a time (1 for digital trigger).

Thanks for your interesting opinion. Separate LA like DSLogic is really better option than dedicated MSO?

The advantages of an MSO are:
- time relation with analog signals (think looking at ADC input level and digital output code)
- realtime screen updates (if you are lucky, decoding updates quickly as well but the models you choose are not the best at decoding anyway)
- one tool that does it all so no need to swap probes around and so on
- reasonably high bandwidth & samplerate compared to typical USB logic analyser

In general an MSO can do all that a USB logic analyser can do. If you need more features like deep memory, advanced triggering, etc then you'll need to look at the high-end logic analysers from Tektronix or Agilent.

Yes, it does come down to use cases, and the OP probably doesn't have sufficient experience to know where the boundaries are.
 
There is another class of equipment that you didn't mention, and which might be relevant given the OP's first post. There is, of course, a lot of overlap between the different classes.

Protocol analysers, by my definition, are digital signal instruments that (unlike LAs) presume/ignore detailed timing relationships, in favour of concentrating on the messages being transmitted. If you want to understand the information flowing between an MCU and a peripheral (e.g. over SPI/USB/I2C/RS232/GPIB/CANbus/etc), then a protocol analyser could be ideal.

While you can get dedicated analysers for those "protocols", I suspect the OP would find it preferable to dip their toe in the water with something cheap and cheerful like a Bus Pirate.

There's a lot to be said for trying something quickly and cheaply - and then spending significant money on something you know does what you need.

[nctnico]

DSOs are going more and more into the direction of protocol analysers as well. R&S for example has the ability to annotate decoded data with labels. Also, protocol analysers aren't particulary cheap as well so you'll need to use one often to make the expense worthwhile.

[tggzzz]

DSOs are going more and more into the direction of protocol analysers as well. R&S for example has the ability to annotate decoded data with labels. Also, protocol analysers aren't particulary cheap as well so you'll need to use one often to make the expense worthwhile.

Agreed. I'm not up to date on all recent scopes' capabilities, but would expect the blurring of the boundaries to continue.

Bus Pirates are cheap, and might be suitable for some of the OP's use cases. "But you can buy [wing]span"

[Adrian_Arg.]

for this
- programming and debugging microcontrollers (stm32, arduino, esp32), I do not rule out making some small projects for profit (I'm a professional programmer)
- some electronic reverse engineering (eg. decoding some protocols like remote control for garage doors, etc - just for fun)
  I have a rigol ds1054 + an 8-dollar 8-channel logic analyzer, two multimeters, a unit 30-05 source and a yihua 8786D, and a 20Mhz FY 6900, and I assure you that I am very satisfied with my little laboratory.

in spanish
para esto
- programación y depuración de microcontroladores (stm32, arduino, esp32), no descarto hacer algunos pequeños proyectos con fines lucrativos (soy programador profesional)
- algo de ingeniería inversa electrónica (por ejemplo, decodificar algunos protocolos como el control remoto para puertas de garaje, etc., solo por diversión)
  yo tengo un rigol ds1054 + un Analizador logico de 8 canales de 8 dolares, dos multimetros, una fuente unit 30-05 y una yihua 8786D, y un FY 6900 de 20Mhz, y te aseguro que estoy muy conforme con mi pequeño laboratorio.

[thm_w]

Thanks for your interesting opinion. Separate LA like DSLogic is really better option than dedicated MSO?

Yes, far better for what I've done. Good luck finding all of this on an oscilloscope: https://sigrok.org/wiki/Protocol_decoders

I'm sure you know the process:
- hit start capture in dslogic software
- hit Run in debugger
- view capture in dslogic software, search, pan, save setup, save capture if needed

Very simple process, all on the PC.
Getting options, naming, triggers setup on an oscilloscope is more hassle than that.

Sure if you want to time correlate analog and digital signals, there is no easy way around that. But often its a single SPI/I2C bus which easily fits within your 4 analog channels. and of course all of these oscilloscopes have a LA option, so, the option is there should you need it down the road.

[Fungus]

Thanks for your interesting opinion. Separate LA like DSLogic is really better option than dedicated MSO?

They have their pros and cons.

The showstopper for me with those logic analyzers is no real-time display on the screen. I want to be able to see signals changing in real time but they work by pressing record, waiting a while, then pressing stop and seeing some data.

This is good for snooping on things and decoding of long data transfers but bad for just poking around circuits and for debugging your programs.

You don't need a DSLogic for hacking garage door openers. I'd say go with the MSO.

(and maybe get one of those $6 logic analyzers from eBay if you want to run Sigrok)

[Fungus]

I'm sure you know the process:
- hit start capture in dslogic software
- hit Run in debugger
- view capture in dslogic software, search, pan, save setup, save capture if needed

It's a real pain to have to do all those steps every single time you edit your program and want to see the results.

An MSO can be set up for real time display so you don't have to touch anything. It's sooooo much faster to work with when you're coding.

[nctnico]

I'm sure you know the process:
- hit start capture in dslogic software
- hit Run in debugger
- view capture in dslogic software, search, pan, save setup, save capture if needed

It's a real pain to have to do all those steps every single time you edit your program and want to see the results.

An MSO can be set up for real time display so you don't have to touch anything. It's sooooo much faster to work with when you're coding.

I 100% agree with this.

[thm_w]

You don't need a DSLogic for hacking garage door openers. I'd say go with the MSO.

(and maybe get one of those $6 logic analyzers from eBay if you want to run Sigrok)

OP already has the Dslogic, if you read the thread.

It's a real pain to have to do all those steps every single time you edit your program and want to see the results.

An MSO can be set up for real time display so you don't have to touch anything. It's sooooo much faster to work with when you're coding.

Its the one step of hitting "start capture" in dslogic, hardly much effort compared to having to deal with the MSO to pan and zoom, which is a pain.
That way you stay on the PC, don't have to move your hands back and forth between the PC and the scope.
Not to mention easier saving and viewing of captures, easier to type in the name, etc.

DSView has repetitive capture as well anyway.

I get it, hardware guys like hardware, you want knobs and buttons, but OP is a software guy.