Switching from bench to USB scope - bad idea?

[Hagrid]

Hello folks, a bit ago I started a thread where I have asked for a scope with good analytical features.
Some Lecroy devices looked really awesome first, but I now tend to get a Picoscope 5444D.
I have a hacked Rigol DS1054Z at the moment (and a crusty analog 20MHz Hameg) and plan to sell the Rigol
in order to get the Picoscope.
Is this a good Idea? The Rigol device is awesome at this price point and great for general purpose use, to probe around a prototype or when repairing a device. I think this is true in general for most bench scopes since you have great accessibility by a dedicated front panel. On the other hand it is almost limited to the internal features. Exporting data is really slow painful and ugly. This means that I am locked down to the features it offers. For the most part that's more than enough, but when doing more analytical things and statistics its basically useless.

It seems that the 5444D with its huge segment able memory is a nice solution here. Even when the memory is full I can just get the data on my computer in a reasonable time via USB 3 and acquire more. For further processing I plan to use Matlab or GNU Octave. The switchable resolution is also a killer feature. The only thing I can't tell is how good it will replace my Rigol when doing general purpose tasks.

What do you think, should I sell my Rigol and get the Picoscope? Should I get the cheapest (about 100USD) Picoscope first to check if I like the way it works? Or are these devices completely different meaning I should keep both? I am not really a fan of wasting 100 bucks, but after all its a fraction of the price of the 5444D.


Greetings, Hagrid

[Fungus]

a) It depends on your work flow. If your Rigol is always used where there's a powered-on PC then it might work out.

b) You can download the sample data from your Rigol to your PC. After that the analysis is up to the software on your PC. Trying a few programs before you make a decision won't cost anything.

[Hagrid]

b) You can download the sample data from your Rigol to your PC. After that the analysis is up to the software on your PC. Trying a few programs before you make a decision won't cost anything.

I have tried out the python scripts but this is just really slow in general (for example: set single trigger, measure peak after a trigger, transmit data and repeat). Is there a way to download the whole memory when segmenting it? I couldn't find anything.

[2N3055]

If you can afford to keep ds1000Z AND get Picoscope would be best option...
Little Z machine is excellent for service/ repair scope (like you said), and Picos are great as analytic scopes and protocol decodes.

If  would have to chose which one to have as a single one, I would choose Pico.
I don't care for buttons, GUI is on enough, I always have a PC ready at the desk, and USB3 series from Pico are fast enough to be interactive. Not great but kinda OK. And much better in everything else.

But again, best option is to have a standalone interactive scope and Pico. That combination replaces scopes that cost waaay much more than those two combined...

[MrW0lf]

What do you think, should I sell my Rigol and get the Picoscope? Should I get the cheapest (about 100USD) Picoscope first to check if I like the way it works?

I have 4 USB scopes incl 3 Picos. For brief period had Rigol also. Each scope is different and they form complementary system.

Oldest PS 2205 25MHz 2ch (which is similar to 2204A) is still often used. It runs very smooth with PS6 software because it is generation of scopes for which software was originally designed. Also it does all stuff on full memory buffer. I like to use it as V*I integrator.

PS 2408B is new gen with large memory. There they are forced to do more onboard processing and in some cases transfer only some tens of kSa worth of snapshots until trigger/zoom whatever. When using automated measurements, roll mode etc you must know thresholds for "snapshot mode" transition.

Common issues are lack of good DPO and serial trigger. DPO is separate mode which sort of works but measurements are disabled and memory buffer is tiny.

ETS might turn to be useful because it will run full blast on all channels at the same time eg 4x20GSa/s.

Full capability of scopes can be unlocked only with SDK because PS6 software is bit overwhelmed by new gen scopes at times. Especially at data transfer rates and wfm/s.

Rigol has more-less normal DPO and serial trigger. You can use trigger out help Pico.

So all in all if you keep Rigol you wont care for Pico issues because it covers them. Also if you get 2204A it probably wont be left gathering dust in future because you can run 2 PS6 instances to get sort of independent timebase mode with 2 scopes etc. It is sized as probe adapter so will not run of desk space also no matter how much of USB gear you pile up.

And when collect little funds can swap Rigol for Siglent X-E to cover DPO type of tasks.

[ebastler]

Nobody has mentioned "look and feel" yet, I believe.

Personally, I like knobs and buttons on a scope. While measuring something, I often hold a probe in my right hand. I feel comfortable adjusting knobs on the scope with my left, but am really clumsy at moving a mouse pointer left-handed (be it via a mouse or touchpad). Your preferences may be different, of course; but I think it is important to be aware of this aspect.

[tkamiya]

Unless I have a certain application in mind which PC based product is suitable, I tend to stay away from them.  Reason is longevity.

For example, TEK 465 is perhaps 35+ years old and still going.  I don't recall any product that require a computer to work still used from that period.  (yes, they existed)  I have one of those "automated" equipment from late 70s.  Software is long gone and so as a platform that can run it.  It's an automated VNA.

PC and particularly OS change so often.  Maintaining compatibility with previous releases tend to be hit and miss, and there is no body even remotely concerned about doing anything to maintain Windows 3.1 compatibility.  This trend will likely accelerate. 

Since test equipment for me is a "forever" purchase, I don't want my test gear to depend on a PC.

I guess it's different if you are building an assembly line where product cycle is rather short, or money is a prime concern. 

[2N3055]

Unless I have a certain application in mind which PC based product is suitable, I tend to stay away from them.  Reason is longevity.

For example, TEK 465 is perhaps 35+ years old and still going.  .........

None of today's standalone scopes will last that long. And no spare parts will be available. And higher end ones are just fancier version of Picoscope (digitizer board) + PC + screen put in  a box.
I have a Picoscope 212/100 from 2003..Still works, I have old laptop dedicated for it.

[MrW0lf]

Unless I have a certain application in mind which PC based product is suitable, I tend to stay away from them.  Reason is longevity.

My 2205 is 8 years old. All new stuff still supported, bugfixes continue etc. How long till typical low end DSO firmware dev drops. completely 3-4 years max? So it's relative. Also I could be possible to run some older opsys as virtual machine depending how well drivers will like it.

[tkamiya]

Unless I have a certain application in mind which PC based product is suitable, I tend to stay away from them.  Reason is longevity.

For example, TEK 465 is perhaps 35+ years old and still going.  .........

None of today's standalone scopes will last that long. And no spare parts will be available.

That's why I stick with older ones.  I can't imagine buying something in excess of ten thousand dollars and not be able to get it repaired.   

[_Wim_]

If you can afford to keep ds1000Z AND get Picoscope would be best option...
Little Z machine is excellent for service/ repair scope (like you said), and Picos are great as analytic scopes and protocol decodes.

I second that. I also have a pico 5000 series scope, and a rigol ds1000z. Both get used quite often. Apart from the physical buttons on the Rigol, I also like the fact that is has a much higher max input voltage, so less likely do damage the front end when probing unknown circuits.

[VEGETA]

I am not an expert but I always find the bench scope a better choice overall, you should find a workaround to get what you want from it if that is your case.

However, if you want to sell the analog old scope (provided it is working), then I am interested. Especially if price + shipping is reasonable enough.

Good luck!

[Hagrid]

Thanks all for the answers. It looks like its mostly personal preference. I think the lower voltage range is a somewhat interesting point, but usually I don't work with line voltage or similar. Maybe I should keep the Rigol first and decide then to sell it or not. I don't want it to sit in a corner and collect dust.

However, if you want to sell the analog old scope (provided it is working), then I am interested. Especially if price + shipping is reasonable enough.

Sorry, it was my fist scope and I learned so much with it. Selling it would break my heart. 

If there are any important points about the Picoscope please fell free to share  .


Greetings, Hagrid

[james_s]

I don't like PC based instruments, they save very little in cost, the display and controls is no longer a major part of the cost of an instrument. I find that a laptop takes up quite a lot of bench space and it's awkward to find somewhere to balance it. Cables and cords strung around get in the way, usability of scope type controls with a touchpad is tedious, you can have problems with drivers, and if you do need to use a PC say to look at a datasheet or load code into a microcontroller or something like that you either need a large monitor or you're awkwardly flipping between applications. A standalone instrument is just so much nicer to use, I only keep my USB scope around when I think I might want a scope but don't want to drag around a real one.

[_Wim_]

Typically I would agree about the cost of usb instrument vs bench instrument, but:

try to find an affordable scope with:
- very high resolution (>12 bit)
- lots of decoding options without additional cost
- large memory depth (512MS)
- built in AWG
- possibility to run bode plots
- a large display (>10")

I think the Picoscope 5000 series are here in a kind of niche, they are not cheap, but nothing else in this price segment (<2000€) has this combination available.  I would typically not recommend their offerings below 500€, because in this segment Rigol and Siglent rule.

As to the laptop being akward, I would recommend a USB docking station with two monitors attached. I use one monitor for browsing and datasheet, and the other for virtual instruments or devellopment software. This does indeed mean you are kind of fixed to a single location (mostly that does not bother me), but for that reason it can indeed be very handy to keep the Rigol around also.

[eugenenine]

Thoughts on the digilent scopes? https://store.digilentinc.com/usb-scopes-instruments-circuits/

I don't have a bench so when I want to do anything I have to pull equipment from the shelf in my closet.  I have a Tek 2211 but it takes a lot of space.

[james_s]

If those are things you need then a USB scope may be a reasonable option however none of that is something I need. Even older DSOs typically have more than enough memory for typical use, and it's rare that I'd need a built in AWG. Having real knobs and buttons is *so* much nicer from a usability standpoint than clicking virtual controls on a screen. I also very much prefer an all in one unit with a handle on top that I can carry around and position as needed, I hate having to deal with dongles and external devices plugged together.

[_Wim_]

The built in AWG (signal generator) is only usefull for the bode plotting, because it can be synchonised with the FFT of the input channels. I think I see this kind of scope more as a sort of network/dynamic/spectrum analyzer with the posibility to show scope traces simultaniously. For "pure" scope use, I also would recommend a bench scope.

[MrW0lf]

Thoughts on the digilent scopes? https://store.digilentinc.com/usb-scopes-instruments-circuits/

Here thread with some data:
https://www.eevblog.com/forum/testgear/digilent-analog-discovery-2-75742/
Strictly as a scope it is quite limited by 100MSa/s sampling rate and 2x16kSa memory. Frontend as such goes to about 50MHz for frequency domain related tasks. Software architecture is innovative because GUI is mostly accessible via scripting and can surgically add DIY functionality w/o need to write all the helper crap you would need with just SDK.
Directly from knobbed DSO, for non-programmer - quite steep change. Picos are somewhere in the middle.

[Hagrid]

What do you think, should I sell my Rigol and get the Picoscope? Should I get the cheapest (about 100USD) Picoscope first to check if I like the way it works?

Full capability of scopes can be unlocked only with SDK because PS6 software is bit overwhelmed by new gen scopes at times. Especially at data transfer rates and wfm/s.

I have not dived fully into the sdk capabilities, but do you man that the sdk can also make use of all features which are offered by the picoscope 6 software (like hardware acceleration and so on...)? Would it be able to build a complete custom gui?

[MrW0lf]

I have not dived fully into the sdk capabilities, but do you man that the sdk can also make use of all features which are offered by the picoscope 6 software (like hardware acceleration and so on...)? Would it be able to build a complete custom gui?

https://www.picotech.com/library/documentation
"Programmer's Guides"

Looking diagonally I can see at least all the main things covered, including downsampling in hardware etc. So dunno if 100% complete but pretty much in the ballpark.

Most questionable thing for me in PS6 are various limitations to ensure steady operation. But PCs are not created equal and some limitations seem to be over the top. For example why limit frame rate in main mode to "visually sensible". Ok for visual inspection perhaps no big difference if 25fps or 100fps but huge difference for blind time and measurement stats gathering time.
Also streaming (roll) mode speed limitations are quite severe.
These problems should be fixable in direct operation via SDK.

Edit: Found some numbers (tested on decent laptop):

Pico 2408B, all processing OFF, 200ns/div, 125MSa/s ~24wfm/s 0% misses with edge trigger ~32wfm/s no trigger
Pico 2408B, all processing OFF, 200ns/div, 1GSa/s ~18wfm/s 0% misses with edge trigger ~29wfm/s no trigger
AD2 (USB2) all processing OFF, 200ns/div, 100MSa/s ~50wfm/s 0% misses with edge trigger ~100wfm/s no trigger

As can see wfm rate has very little to do with data amount on Pico. Tests were done in main (analytical) mode, not DPO. On AD2 rate also seems throttled but at much more reasonable level. Worst of all Pico USB3 scopes are throttled on same level as USB2. Hope PS6 is EOL because about time for next gen GUI.

[Hagrid]

Interesting, thanks a lot. This are indeed some points I will have to have a closer look at.

[David Hess]

Many USB scopes have terrible front ends and rely on "digital" gain for adjusting attenuation factor which is not documented.  Also watch out for their overload specifications which may only be 50 volts.

I would be most concerned with the lack of galvanic isolation through a USB interface.  The right kind of ground loop through the probe ground may not only damage the USB oscilloscope but things connected to it through its USB interface like an attached computer.

[james_s]

I assumed most of them now would be isolated? Maybe not? My USB scope is an older Bitscope which is not isolated. It is very easy for spikes to cause the whole USB bus to lock up, a separate USB isolator solves this but it's clunky.

[Performa01]

Edit: Found some numbers (tested on decent laptop):

Pico 2408B, all processing OFF, 200ns/div, 125MSa/s ~24wfm/s 0% misses with edge trigger ~32wfm/s no trigger
Pico 2408B, all processing OFF, 200ns/div, 1GSa/s ~18wfm/s 0% misses with edge trigger ~29wfm/s no trigger
AD2 (USB2) all processing OFF, 200ns/div, 100MSa/s ~50wfm/s 0% misses with edge trigger ~100wfm/s no trigger

As can see wfm rate has very little to do with data amount on Pico. Tests were done in main (analytical) mode, not DPO. On AD2 rate also seems throttled but at much more reasonable level. Worst of all Pico USB3 scopes are throttled on same level as USB2. Hope PS6 is EOL because about time for next gen GUI.

I wonder if you maybe have accidentally throttled the waveform capture rate in the PS6 preferences?