Is iPad scope worth professional pricetag?
As a scope aficionado, I’m always excited by new product introductions, whether from established brands, or startups. Here’s a new scope module that recently caught my eye. The packaging is slick, and it uses an iPad for its UI. But as I delved into the published specs, I became somewhat incredulous at the design and cost. One might imagine the company’s principals newly arrived from another planet, and hence unfamiliar with what Earthlings look for in their oscilloscopes.
Anyway, here’s a hands-off review, based purely on specs (and the iPad app, which can run in demo mode), that you should find interesting. And after reading my surprised impressions, do leave your own comments. Someone’s gotta give these newcomers to our world decent market research data.
The Moku:Lab from Liquid Instruments retails for USD 4,990, and is positioned as a professional instrument – obviously. They currently include an iPad in that price, but that may be an introductory offer only.
The key specs are:
- 2 channels
- 200 MHz 12-bit input at 500 MSa/s
- DDS: 2 channels, 200 MHz sine (100 MHz others) 16-bit at 1 GSa/s
- External (TTL) trigger (5 MHz bandwidth)
Moku:Lab wireless scope module with iPad (source: Liquid Instruments)
No, that’s not a typo: 5 MHz external trigger bandwidth, and fixed threshold. The current version of the app doesn’t seem to have a setting for external trigger anyway, so, not a problem! Two-channel scopes are bearable when they have a good external trigger…especially if you can view it. This has neither. At least the app is pretty slick.
There is a single “Math” channel which can compute the four basic math functions, integrate, differentiate, and FFT. I can’t see any way to control the FFT parameters though!
The DDS is admittedly superior to what’s normally found in scope “add-ons”, though only the basic waveshapes are available. There are AM/FM/PM modes, but no sweep. A Bode Analyzer feature does sweep up to 120 MHz, so I guess that could be used in a pinch, though presumably sans sync pulse.
That brings us to the “feature set”. Besides Bode, there are phasemeter, PID controller, data logger, and lock-in amplifier “instruments” included. Does this extra-scope functionality count for much? In some cases, perhaps it does.
Another instrument is a spectrum analyzer. At first, it appeared to be implemented using traditional SA architecture, but no, it IS an FFT. This standalone FFT does have a very limited set of controls (like window selection), but like I mentioned above, when FFT is used in scope mode, there are no controls, nor do the settings from SA mode carry over.
Before I discuss further scope details, let’s do some price comparisons. I’ll use the modules of Pico Technology for reference, as they are arguably the key – if not only – player in the professional scope module arena.
A comparable module is the Pico 5244B – for $1,975 – 40% of the Moku’s cost. Key specs are very similar, though Pico’s DDS runs at 20% of the frequency. But it IS a true ARB.
Coming from the other direction, a Pico 6403D costs 15% more than the Moku, but provides four channels, has 350 MHz BW, and samples at 5 GSa/s. Hmm.
What’s NOT there
You may have noticed I haven’t touched on the more subtle scope specs yet. Here’s where things get particularly troubling given the professional positioning of this module.
- No logic inputs
- No trigger view
- No trigger holdoff
- No AC trigger coupling
- No advanced trigger modes; just basic Edge
- No pan/zoom view
- No data decodes
- 16k sample memory
- 8k waveforms/s
Once again, those last two are not typos. For comparison, the Pico scopes above have 512 MSa and 1 GSa buffers, and capture 130k & 170k waveforms/s. Here we see the difference between a professional instrument and a toy, albeit a very snazzy and expensive toy. Even scopes marketed to hobbyists and makers do far better in all these areas.
Screen capture in scope (demo) mode
But wait…there’s more. Or, should I say, less. The inputs have only two sensitivity settings! You can select 1 VP-P or 10 VP-P. So, even with a wire as a probe, at 1mV/div, we’re left with 5 bits of on-screen resolution (though TONS of dynamic overload range)!
Using a ×10 probe as is the fashion these days, you’ll be seeing the scope’s full 12-bit resolution at 1 V/div. At a lower volts/div setting, resolution will be diminished. At, for example, 0.1V/div, resolution will roughly match a standard 8-bit scope. Interesting.
At the other end of the input range, the maximum signal with a ×10 probe is 100 VP-P. That’s not nearly enough to look at AC line voltage. In Japan. During a brownout. Oops.
Conclusions & puzzlements
With software and FPGA updates, and a drastic price reduction, the Moku:Lab scope module could be a reasonable piece of gear. But I must surmise that the hardware upon which it’s based is fundamentally incapable of anything approaching professional needs, or even hobby needs. Nor would I ever suggest buying hardware or software based on promises of future improvements – unless those promises are for icing on an already capable cake. But I’m rambling…there’s no evidence here of forthcoming improvements to the cake.
One must speculate on how such a product comes into being. Presumably, the company’s principals aren't really from another planet. Some good ideas were clearly on the minds of the designers, yet a real understanding of scopes, and the market, seems completely missing. I hope they can manage to regroup, go into hiding for a year, and come back with a real product. Then we’ll all win.
- The scope...from Hell!
- How Many Bits Would You Like With That?
- A practical take on FFTs
- Designer's Notebook: How NOT to do it
- Product Tryout: NI VirtualBench
- Tryout: LabNation Smartscope
- Tryout: Teledyne LeCroy WaveSurfer 3000 oscilloscope
- Review: The Read Oak Canyon "RF Pro Touch" RF Generator
- Review: Picotest J2150A Harmonic Comb Generator
—Michael Dunn is Editor in Chief at EDN with several decades of electronic design experience in various areas.