-
Most measurements I do even on my 4ch scope are with 1 or 2 probes.
But not 100% of them, right?
If 1% of measurements use 3 probes and only 0.01% of them use 4 probes then you still need a 4 channel scope.
The time you should consider a 2 channel 'scope is when you can do 99% of your work with just 1 channel.
... and then you find you need a 5 channel scope
Better to apply understanding and imagination to the problem, and find a way to use the tools at your disposal. Being able to understand and differentiate between necessary and sufficient and convenient is a valuable engineering skill. -
You might need to work with a 4 channel scope, just for Analog. You debug an amplifier, looking at the input, the output, and any other node. You want to record that your power supply sequencing is good. You want to debug any differential transceiver, input 2 outputs...
I dont think 4 channels is an overkill. I actually think, 4 CH MSO is something which is needed for electronics. -
You can get almost always get away with 2 channels. Sure, it's less convenient, you will have to switch probes around more in the middle of a test, but you can almost always make the measurement.
But if you need that extra bandwidth, you can't get away with a lower bandwidth.
So I guess the question is easy, do you need the extra bandwidth? -
You can get almost always get away with 2 channels. Sure, it's less convenient, you will have to switch probes around more in the middle of a test, but you can almost always make the measurement.
Sure, but what if you're still at the stage where you can still choose?But if you need that extra bandwidth, you can't get away with a lower bandwidth.
So I guess the question is easy, do you need the extra bandwidth?
And the more difficult followup question is, "how much bandwidth do you need"?
-
You can get almost always get away with 2 channels. Sure, it's less convenient, you will have to switch probes around more in the middle of a test, but you can almost always make the measurement.
Sure, but what if you're still at the stage where you can still choose?
I don't think there's any doubt in my own mind that personally speaking, 4ch has been almost a prerequisite since I first started using a Tek 2465 CRO twenty years ago, but before then I'd been perfectly fine with 2ch for a couple of decades before that. this was a case of what you don't know, you don't miss!
While the extra two channels on the 2465 had restrictions (pretty much limiting them to digital levels), I found I used those extra channels enough of the time for developing stuff that I don't like going back to 2ch, for example the rare occasions I use a 2ch handheld scope I find it tedious to use, but that's as much to do with the UI as it is to do with it only having two channels.
While you can in many situations supplement your scope channels with an LA, I much prefer using the scope than messing about with setting up a PC based LA, but that's my personal preference.
Putting my neck out, I'd say that two channels is a much more relevant for equipment repair, compared R&D where you're going to find four channels more useful.QuoteBut if you need that extra bandwidth, you can't get away with a lower bandwidth.
So I guess the question is easy, do you need the extra bandwidth?
And the more difficult followup question is, "how much bandwidth do you need"?
Never enough ;-)
Bandwidth can quickly end up being an arms race, particularly when you start having to throw money at probes. Even at 200MHz, passive probes will introduce significant artefacts. However, you will still be able to see stuff on a 200MHz scope that you wouldn't with 100MHz BW. Whether you need to see that stuff or not though is open to question: W2AEW did a great video (#25) where he had a transistor amplifier circuit go into oscillation, and why you might want more bandwidth.
FWIW, from what I've seen, I agree that for what the OP is doing, and at the level at which they're doing it, in their shoes unless they have bad eyesight the DS1054Z would be the one to go for: the screen on the 1054Z is pretty small. On the other hand, its diminutive size is a plus point if you have limited bench space. -
I still regret getting a 2 channel scope.
i still regret buying the 1054z.
i'd rather have a 2 channel scope that works as it should (2+ext trig, 2 chan on its own is very limiting)
i can't comment on the channel requirements as i don't do power electronics. My 2 cents on the 1054 itself
What aspects did you find bad about the 1054z that made you regret your purchase? -
If you spend that wasted time working in a supermarket you'll be surprised how little time you have to work to earn enough money to make up the price difference between a 2 channel or 4 channel scope.If 1% of measurements use 3 probes and only 0.01% of them use 4 probes then you still need a 4 channel scope.
Then I will use my brain to simplify the measurement channel requirement, or design another measurement plan. -
Mostly one, the slowness of the user interface.
the scope is slow on its own.. but let's say bearable. The moment you enable one of the features it's not usable anymore, to me.
So at work i finally had to interface with an SPI EEPROM because need for speed, this meant that during development i needed to actually use four channels while in 99.5% or more of the cases i need one or two to check the spi lines. Since the scopes on the work bench are a picoscope 2000 series and a tektronix tps i decided to bring my 1054z. Setting up serial decoding is frustrating to say the least and waiting 5 to 10 seconds to decode the new packets is even worse, after 10 minutes i switched off the damn thing, decoded by hand a couple of packets at the tektronix to see if timings and signals were correct then only used the picoscope, one channel at the time. it was that faster to proceed in this way.
you see, even tough on paper a picoscope (for example) has shortcomings over a 1054z, like
- lack of physical controls
- 48kS memory (this is an older pico, new ones start at 32MS i think for fast acquisition, in slow/continous acquisition the limit is your pc memory)
- very low wfm/s at timebases higher than 1 ms because USB2.0
- thousands of acquisition and decoded data already there, at a click.
in the real world it is a much more powerful beast.. and i can work much better with it.. beside having serial decoding that works, math that works, measurements that don't crap out, fft with some usable resolution, ETS, more decoding standards that you'll ever dream of in this price range
infact, now i use the tek only if i can't get around using two channels or if i need to do isolated measurements
I haven't sold my 1054z yet because i haven't used the hobby bench in quite a while but when i'll start playing at home again and when i'll go all deep with canbus and other automotive stuff at home too, 1054 will be unloaded to someone that will still want to buy it.
what will i get? i'd like to try a pico 5000, or a GDS-2104. if budget will be lower one of the siglent 2 chans until i will be able to afford a lecroy/keysight (but keysight only if they will update the 3k series with more memory)
no more cheap crap, that's sure. -
I agree the UI is slow and unresponsive but it's not unusable. It's not dissimilar to a Tek MDO3000 UI in this respect, although at least the rotary encoders keep up reasonably well on the Tek. The encoders on the Rigol are pretty bad but you get used to its shortcomings.
I had a Pico a few years ago, but it sat in a drawer most of its life. I bought it thinking that a USB scope it would be convenient, I found I used it maybe three times over the five or six years I had it. I still have a 200MHz 4ch TDS2024B which I think is basically the same as a TPSxxxx. I keep it for sentimental reasons, and I can't remember the last time I actually used it for real work, whereas the Rigol I use pretty regularly even though it's only 100MHz.
One thing I would say is that the Rigol will decode high speed SPI that the Keysight 3000 won't. The biggest frustration I have with the Rigol decoding is that it only decodes what's on the screen, and you can't decode more than a few bytes as it only subsamples what's in memory. I'd prefer it that there was some optional manual control over the decoding, but I found it's still better than doing it manually (although sometimes it gets it wrong!).
Understood regarding the memory limitation of Keysight, it does seem a bit limiting nowadays. Although there is always segmented memory, even though it's straightforward, it's a bit of a pain to have to set up. Practically speaking though it's pretty rare I find the memory on the Keysights to be a real limitation. The reason my daily driver scope is a Keysight is quite simply the highly responsive and straightforward UI. But we are talking about an order of magnitude difference in price here!
It'll take an awful lot more convincing to make me buy another Pico USB scope over a Rigol bench scope though. -
It'll take an awful lot more convincing to make me buy another Pico USB scope over a Rigol bench scope though.
Pico is sort of "scientific calculator" of scopes. But currently most people are ok with +-/* and want knobs to play with. I see only way Pico going mainstream - packing all the same stuff into touchscreen box like Micsig, spice that with sharp looking UI and ext video/keyb/mouse option - it'll be all the rage. Then it would just come down to attaching ext keyb/mouse for power users and all the rest would be happy just stroking that touchscreen for basic functionality.
Edit: Actually they are already moving there:
https://www.picotech.com/library/picoscope/picoscope-beta-release-6.12.3
Latest betas indeed come with "tappable" stuff. So "early adopter" could just take random Windows (or even Linux) tablet, tape Pico to it and it'll be all the rage already today
-
You can get almost always get away with 2 channels. Sure, it's less convenient, you will have to switch probes around more in the middle of a test, but you can almost always make the measurement.
Sure, but what if you're still at the stage where you can still choose?But if you need that extra bandwidth, you can't get away with a lower bandwidth.
So I guess the question is easy, do you need the extra bandwidth?
And the more difficult followup question is, "how much bandwidth do you need"?
That is a reasonable question. And the answer to that is I don't know.
All of my work so far has been done on either an ancient Tek single channel 20MHz tube scope or the 2ch 40MHz version of that same scope. I kind of took for granted that a modern 100MHz (even if it drops to 250MSa/s on all four channels (Nyquist says it'd be acceptable up to 50MHz in that scenario IIRC?) would even provide a certain amount of headroom for the coming years.Mostly one, the slowness of the user interface.
the scope is slow on its own.. but let's say bearable. The moment you enable one of the features it's not usable anymore, to me.
So at work i finally had to interface with an SPI EEPROM because need for speed, this meant that during development i needed to actually use four channels while in 99.5% or more of the cases i need one or two to check the spi lines. Since the scopes on the work bench are a picoscope 2000 series and a tektronix tps i decided to bring my 1054z. Setting up serial decoding is frustrating to say the least and waiting 5 to 10 seconds to decode the new packets is even worse, after 10 minutes i switched off the damn thing, decoded by hand a couple of packets at the tektronix to see if timings and signals were correct then only used the picoscope, one channel at the time. it was that faster to proceed in this way.
you see, even tough on paper a picoscope (for example) has shortcomings over a 1054z, like
- lack of physical controls
- 48kS memory (this is an older pico, new ones start at 32MS i think for fast acquisition, in slow/continous acquisition the limit is your pc memory)
- very low wfm/s at timebases higher than 1 ms because USB2.0
- thousands of acquisition and decoded data already there, at a click.
in the real world it is a much more powerful beast.. and i can work much better with it.. beside having serial decoding that works, math that works, measurements that don't crap out, fft with some usable resolution, ETS, more decoding standards that you'll ever dream of in this price range
infact, now i use the tek only if i can't get around using two channels or if i need to do isolated measurements
I haven't sold my 1054z yet because i haven't used the hobby bench in quite a while but when i'll start playing at home again and when i'll go all deep with canbus and other automotive stuff at home too, 1054 will be unloaded to someone that will still want to buy it.
what will i get? i'd like to try a pico 5000, or a GDS-2104. if budget will be lower one of the siglent 2 chans until i will be able to afford a lecroy/keysight (but keysight only if they will update the 3k series with more memory)
no more cheap crap, that's sure.
I'll be completely honest, I've flat out discarded PC-based scopes without even looking at them. I like standalone equipment and don't quite fancy the idea of having a computer kicking around the bench that I rely on for measurements.
Touch screens are also something I'd like to avoid for basic operation stuff, for the simple reason that I tend to get grubby fingers when sat at my bench and scratches would quickly make the display unreadable (or if I use protectors, having to replace them constantly).
Responsiveness does seem to be a common criticising point for the 1054Z, and to be honest I appreciate snappy UIs a lot. I'll do a bit more research on this before ordering, just to make sure I won't kick into frustration mode.
Again, really appreciate the feedback, it's been invaluable so far
-
All of my work so far has been done on either an ancient Tek single channel 20MHz tube scope or the 2ch 40MHz version of that same scope.
In that case a DS1054Z will seem amazing by comparison.Responsiveness does seem to be a common criticising point for the 1054Z, and to be honest I appreciate snappy UIs a lot. I'll do a bit more research on this before ordering, just to make sure I won't kick into frustration mode.
'Fixing' it requires spending another $600 or so. Your call...
A lot of the complaints seem to center around the vertical movement. Everybody seems to see something different and a lot of that seems to be in your knob twisting technique. I made a separate thread on this:
https://www.eevblog.com/forum/testgear/the-ds1054z-vertical-response-discussion-thread/
It's never going to feel like an analog 'scope, they're just a potentiometer connected to the vertical amplifier so response is at the speed of light.
-
Responsiveness does seem to be a common criticising point for the 1054Z, and to be honest I appreciate snappy UIs a lot. I'll do a bit more research on this before ordering, just to make sure I won't kick into frustration mode.
Think about how you use the controls of your existing scope. Do you spend 1% of the time moving the display around? 5%? Whatever...
Now, suppose it took twice as long for the 1054Z UI. Does saving time on something you don't do very often make any difference in the world?
It's a DSO, it has a lot of memory and it's primary job function is filling the memory and putting something on the screen. You will never get a DSO to respond the same as an analog scope.
-
One step up from the Rigol 1054Z is the GW Instek GDS2000E series which is responsive in general unless you let it do some serious number crunching.
-
Responsiveness does seem to be a common criticising point for the 1054Z, and to be honest I appreciate snappy UIs a lot. I'll do a bit more research on this before ordering, just to make sure I won't kick into frustration mode.
Think about how you use the controls of your existing scope. Do you spend 1% of the time moving the display around? 5%? Whatever...
Now, suppose it took twice as long for the 1054Z UI. Does saving time on something you don't do very often make any difference in the world?
It's a DSO, it has a lot of memory and it's primary job function is filling the memory and putting something on the screen. You will never get a DSO to respond the same as an analog scope.
Oh, so it's controls responsiveness, not a very large representation delay.
That I can live with.All of my work so far has been done on either an ancient Tek single channel 20MHz tube scope or the 2ch 40MHz version of that same scope.
In that case a DS1054Z will seem amazing by comparison.Responsiveness does seem to be a common criticising point for the 1054Z, and to be honest I appreciate snappy UIs a lot. I'll do a bit more research on this before ordering, just to make sure I won't kick into frustration mode.
'Fixing' it requires spending another $600 or so. Your call...
A lot of the complaints seem to center around the vertical movement. Everybody seems to see something different and a lot of that seems to be in your knob twisting technique. I made a separate thread on this:
https://www.eevblog.com/forum/testgear/the-ds1054z-vertical-response-discussion-thread/
It's never going to feel like an analog 'scope, they're just a potentiometer connected to the vertical amplifier so response is at the speed of light.
From looking at your video it does seem to be quite annoying, but no, I'm not one to move things about all the time. I generally find a time base/amplitude what works for what I'm looking for, then go from there as required - not necessarily shifting the screen about all the time seeking ghosts.
I hope I'll be alright... If not, well, I think I'll figure it out during the period I'm allowed to return the scope
Thanks again! -
I have tried the Rigol DS1054z but was not impressed. I had it for 2 weeks and it hung up on me for 5 times in that short time. Als I found that the interface was not very responsive when having deep memory enabled and some times it took 2 minutes to recover. A hard reset was then faster. I exchanged the Rigol for the 2 channel Siglent SDS1102X and this scope is much more stable and faster. And the deep memory called History is realy nice when analyzing i2c and other protocols. And Siglent currently has an action that you get the serial decoding option free with your purchase.
-
Most of the time I use one channel for voltage (often with a HV diff probe) and a current probe for the second channel. More as two traces on a scope screen is not my thing. So my DSO is 2 channel (350 MHz Hameg) 4 traces is much to crowded. Besides that I can not remember the last time I needed more as 2 channels. (I have several scopes incl three 4 channel analog scopes so if I really need 4 channels I can use one of them.
The only reason I think 4 channel is handy, is so you do not have to set up the channel when you switch probes. But in my case that is only because I use all kind of probes. (and most times 2 scopes for this reason) On the other hand most of those probes cost more as 500 so that is no valid reason in your situation/budget)
If you only want one scope and think you will need 4 channels in the future then buy one. You get 4 probes and budget probes die rather fast if you use them a lot so having 4 is not bad.
-
... 4 traces is much to crowded...
That's a useful observation.QuoteIf you only want one scope and think you will need 4 channels in the future then buy one. You get 4 probes and budget probes die rather fast if you use them a lot so having 4 is not bad.
And when you need decent probes you'll find you can barely afford two! -
From looking at your video it does seem to be quite annoying, but no, I'm not one to move things about all the time. I generally find a time base/amplitude what works for what I'm looking for, then go from there as required - not necessarily shifting the screen about all the time seeking ghosts.
I hope I'll be alright... If not, well, I think I'll figure it out during the period I'm allowed to return the scope
Thanks again!
notice i didn't bring up the vertical movement (well actually i said bearable, which it is)
did we ever talk about the horizontal there in that thread? another big can of worms.
notice how he doesn't address moving around the fft, the serial decode parameters, the alternative trigger modes
consider those "600" more (are they even 600 more? i don't think so anymore.) to me they will be more as the 1054 will be sold at a loss. definetly return it if you can and you don't like it. -
2 channels, 4 channels, 2 channels, 4 channels
Truth is, real truth.......unless you've quite some reasonable experience with a scope using 4 channels can be quite challenging and complex. Gnd loops is the obvious first trap for unwary, then triggering to make sense of all you have on screen.
Add to this a display that is not really appropriate for a 4 channel scope and by this time the scope novice is thoroughly confused.
There's a good # of experienced members that have commented in this thread that I for one respect their views that 2 channels is most often more than enough and when more might be needed it's time to review your measurement practices. Good advice IMO.
For years I had only 2 channel scopes (CRO's) and if I didn't have the chance to get 4 channels DSO's as part of my dealership I'd still only have 2 channel DSO's, even they can Decode serial protocols.
If there's argument for 4 channels for PSU design.........well decent power analysis options don't exist in this entry level class of scopes. Period. -
Just had a thought - 2ch+ext trigger at 500MSa/s would make the Siglent effectively twice as good as the Rigol in 2ch+ext trigger mode... wouldn't it?
Taking what tautech just said as an example - I also do believe from looking at Dave's videos wrt. probes, high frequency analysis and such that 250MSa/s at low memory depth doesn't provide enough sampling speed to get a realistic picture of what's actually happening in the higher order harmonics and at higher switching frequencies - and that is quite important when it comes to PSU and filter design (or so I've been taught up until this point and have tried to take it into account as much as possible).
So in 1ch/2ch with signal triggering both should be equal... But if an external trigger is used, the Siglent should actually be quite a fair bit better than the Rigol, kind of making the point about the extra channels a bit moot, and only really an advantage if I do use that fourth channel.
...or am I just overthinking stuff?
Bearing in mind this scope will have to do for at least the next four or five years. -
Just had a thought - 2ch+ext trigger at 500MSa/s would make the Siglent effectively twice as good as the Rigol in 2ch+ext trigger mode... wouldn't it?
Bugs and the decoding is just as limited as on the DS1054Z. For 100MHz bandwidth 250Ms/s (4 channels enabled on the DS1054Z) is enough so the extra samplerate doesn't give you any extra information.
...or am I just overthinking stuff?
I really don't get why people insist on trying to make do with 2 channels. I often find myself looking at an input signal, output signal and use a probe to measure various points in the circuit. In some cases I use a current probe as well but that needs configuring the channel for showing current. If you need to swap probes all the time it will become a nuisance and you'll likely try to skip a measurement which actually could give you a good insight on a problem in your circuit. -
That's pretty much what I wanted to hear, if I'm honest...
Enough brain-cracking is enough. I'll order the Rigol, if I don't like it I'll find something else.
But it'll be hard to not like coming from a 40 year old entry-level scope.
-
That's pretty much what I wanted to hear, if I'm honest...
Exactly. Congratulations in your decision. I have a similar walk of life as you (and as many others around) and couldn't be happier with the progression of my purchases.
Enough brain-cracking is enough. I'll order the Rigol, if I don't like it I'll find something else.
But it'll be hard to not like coming from a 40 year old entry-level scope.
One detail: despite I have a 4ch nowadays, I find it useful only on occasion. But when I had only two it was perfectly ok as well. -
Give us a review on this thread when you can. I think it important to buy local so you can return it should you choose, or you need warranty work. Rigol and Siglent have their share of quality control issues. Enjoy.
That's pretty much what I wanted to hear, if I'm honest...
Enough brain-cracking is enough. I'll order the Rigol, if I don't like it I'll find something else.
But it'll be hard to not like coming from a 40 year old entry-level scope.