A Wireless Oscilloscope Isn’t As Dumb As It Sounds

The latest CrowdSupply campaign is a wireless, Bluetooth oscilloscope that doesn’t make a whole lot of sense until you really think about it. Once you get it, the Aeroscope wireless oscilloscope is actually a pretty neat idea.

If the idea of battery-powered, Bluetooth-enabled test and measurement gear sounds familiar, you’re not dreaming. The Mooshimeter, also a project on CrowdSupply, is a multichannel multimeter with no buttons, no dial, and no display. You use the Mooshimeter through an app on your phone. This sounds like a dumb idea initially, but if you want to measure the current consumption of a drone, or under the hood of your car while you’re driving, it’s a really, really great idea.

The specs of the Aeroscope aren’t bad for the price. It is, of course, a one-channel scope with 20 MHz bandwidth and 100Msps. Connection to the device under test is through pokey bits or grabby bits that screw into an SMA connector, and connection to a display is over Bluetooth 4.0. You’re not getting a scope that costs as much as a car here, but you wouldn’t want to put that scope in the engine bay of your car, either.

The Aeroscope is currently on CrowdSupply for $200. Compared to the alternatives, that’s a bit more than the no-name, USB scopes. Then again, those are USB scopes, not a wireless, Bluetooth-enabled tool, and we can’t wait to see what kind of work this thing enables.

39 thoughts on “A Wireless Oscilloscope Isn’t As Dumb As It Sounds

  1. Why would it be a dumb idea? This is especially useful in cases like; you don’t want to fry your expensive test equipment just to test a stupid (might be high voltage) power circuitry, and you don’t want to mess with isolation transformers, preventing ground loops, etc.

  2. Totally agree, wireless tools makes a big difference. Nowadays, both energy storage technologies and data transmission technologies are cheap and powerful enough to allow wireless hand tools and instruments.

    Once you touched a battery powered dremel/soldering iron/mooshimeter/tablet/whatever, you’ll hate to go back to their wired versions.

      1. My Mooshimeter has *ridiculous* battery life. It doesn’t even bother with a power switch, there’s no point. A couple of AA’s every couple of years isn’t a burden.

        And at one point I’d even forgotten where it was (hadn’t lost it, just hadn’t needed it in ages), until someone walking past the member storage at my hackerspace said “hey, what’s this Mooshimeter thing I’m seeing on my phone?”. I’m never going to lose that thing.

    1. “Once you touched a battery powered dremel/soldering iron/mooshimeter/tablet/whatever, you’ll hate to go back to their wired versions.”

      I own two battery powered drills.
      Neither of them work, as their batteries have died.
      So, I need to buy new batteries, and maybe pay to have tabs welded on them, before they are useful again.
      In the meantime, I’ll use a corded drill or a bit and brace.

      1. Old/cheap battery powered drills are awful, yes. Often because the manufacturer deliberately used a crappy battery charging circuit so batteries would die faster.

        So it should more accurately be “Once you’ve touched a *good* battery powered tool”. The recent lithium-ion powered non-budget tools are good. At least, they can’t get away with shitty charging circuits anymore due to the whole “going up in flames” thing.

    2. It’s often preferable to have a dedicated display and UI though.

      I suppose you could get a cheap tablet only for use as a display for measurement equipment, but I think the UI would still not be as good as dedicated buttons and knobs.

          1. Only when the scope and DUT are wired and sharing a ground, so you need two channels for the two differential lines. But a wireless scope floats, which *automatically* gives you common mode rejection. So just connect the ground and probe to the two differential lines and you’re golden.

  3. I wonder how many waveforms/sec it captures & renders?

    For anyone not familiar with this, the capture rate gives you a good idea of how likely a scope will show you an infrequent anomaly. If you use your scope for troubleshooting modern microcontroller-based systems, where things usually work but occasionally go awry, perhaps due to a rare race condition in the code, this spec (which many scopes don’t even publish because they are so slow) matters as much or perhaps more than bandwidth.

    For example, imagine you have the horizontal scale set to 500 ns/division, so the entire image width is 5 microseconds. You’d see 5 cycles of a 1 MHz waveform, or a couple bits of 400kbps I2C communication. This is certainly well within the range of a scope with 20 MHz bandwidth. Also imagine whatever you’re triggering on happens pretty much continuously. A theoretically perfect oscilloscope would be able to trigger again immediately after the 5 us. If the infrequent error happen in the next 5 us time period, you’d be sure to see it a faint blip different from the bright line representing all the other cases where the signal was correct.

    No scope is that good. The very best scopes have capture rates in the 500,000 to 1,000,000 waveforms/sec range. With such a scope, you’re likely “blind” for 1 to 2 microseconds, before the scope can trigger again and begin collecting the next 5 microseconds of data. Some pretty decent budget scopes exist. Rigol DS1054Z ($399) can to 30,000 waveforms/sec. Keysight EDUZ1002 ($448) can do 50,000 waveforms/sec. With those, you’d expect the scope to be “blind” for approx 25 us, before it can capture the next 5 us. If your anomaly happens at random times (because you don’t understand it yet and don’t have a strategy to trigger on it), the budget scopes give you about a 1 in 6 chance of seeing such a glitch. A top of the line scope would probably give you a 5 in 6 chance, since it’s (probably) blind for only 1 us before it can capture the next 5 us.

    Of course, if a scope has a capture rate in the few hundred Hz range, or even as slow as your screen’s 60 Hz refresh rate, odds of seeing an infrequent anomaly are virtually zero. Such scopes are still quite useful for viewing regularly occurring waveforms and single-shot capture.

    For troubleshooting intermittent problems which happen on the time scale of several microseconds (well within 10-20 MHz bandwidth and the capability of even 8 bit microcontrollers), fast waveform capture rates are important. Many folks compare oscilloscopes only on bandwidth and number of channels, which is a real shame, because the scope’s capture+render rate is often the spec which matters most.

  4. >>This sounds like a dumb idea initially, but if you want to measure […] under the hood of your car while you’re driving,

    Actually *this* sounded like a dumb idea initially, but I was picturing somebody hanging on for life while holding the probe to the desired test point, while a driver was driving around with the hood up. But I suppose you *could* just use the sprung hook (spring loaded hook) and leave it under the hood with the hood closed.

  5. Not a dumb idea. I think it’s great that wireless is _finally_ showing up in these tools. I’m liking the newish Fluke meters at work that have wireless pickups. They’re especially useful with regard to arc flash requirements where you need to suit up in protective coats, gloves, and hoods to do anything.

  6. I love SMA connectors for all sorts of things where BNC doesn’t cut it, but BNC is totally fine for 20MHz, and it’s the standard on oscilloscopes, so the choice of SMA is perplexing here. It’s marginally smaller but it doesn’t look like BNC would be awkwardly big, and it holds on a bit tighter but BNCs aren’t exactly wobbly. Maybe they are when used like this?

    I dunno, I’m trusting their judgment and I preordered one. I’ll just carry a BNC-SMA adapter for when I want to use traditional probes. I hope the firmware running in the scope itself is upgradeable, but for $200 I’m not super picky.

    1. I just finished asking the development crew a question on that very topic. How do you then calibrate your traditional probe once attached? If they start to include a square wave calibration output, I’m sold.

      1. By using the 1kHz squarewave output of my benchtop ‘scope, or a function generator, or lashing up a 74HC14 to make my own portable one. I wonder if a headphone output would have the necessary slew rate, because I might be carrying a function generator in my pocket already…

  7. I’m with most other commenters here — “wireless ‘scope” doesn’t look like a dumb idea at all.

    The one crappy gotcha with the Aeroscope is the single-channel aspect; it’s not enough to buy two of these to emulate a two-channel scope, unless there’s some way to synchronize them. It’s not as bad as it sounds, since about half the time I’m using two channels it’s just to get a differential measurement where neither side is grounded; a battery-powered wireless probe has no true ground, so in these cases you can just clip the “ground” to one side and the probe to the other. But the other half, I actually need to look at two different signals side-by-side, or to trigger off one signal while watching the other, and this just won’t cut it.

    I could also whine about the +/- 40V maximum input rating (so no poking at line voltage, unless you make a 10:1 probe) and 100mv/div minimum resolution, but really, I rarely do anything where those would be an issue.

    It’s tempting, but I’ll wait and hope version 2.0 has either 2 channels (halving the sampling rate; TANSTAAFL) or better yet a sync connection allowing you to chain multiple Aeroscopes together, and have them all trigger and sample in lockstep, at full rate.

  8. I’ve got an idea, plain old scope probes that are wireless. Works with any oscilloscope. It would have to be a high bandwidth radio to encode a fast digital signal. ADC->DAC type thing. Something that operates on RC channels maybe. Each probe set would have rotary switches to select the channel. Then you get into the problem of power. But not really if the distance is limited. Hmmm…..

    1. Let me plug my dormant fiber-optic isolated oscilloscope probe project. https://hackaday.io/project/12231-fiber-optic-isolated-voltage-probe (not wireless, but VERY isolated with differential inputs.)

      I bet a direct analog coupling could be done on the 2.4GHz or 5GHz ISM bands. keeping a constant probe gain and offset would be a challenge. Analog video has ~6MHz bandwidth. Might be able to make something useful out of a FPV video transmitter?

  9. This scope looks entirely useful. Hasn’t an almost identical product been launched before? (or is this a 2.0 launch?)

    Fairly dissapointed at the 1Mohm/17pf input loading. While this will allow the use of standard 10x probes with a BNC adapter, they should be able to do FAR better because the input amplifier is just 10-20mm from the probe tip. Something like 1Gohm/4pf input loading should be feasible with a switch for 1Mohm/17pf when using a standard 10x probe.

  10. There’s the DSO203, which can sample 1 channel at 144m samples/sec, or two at 72m. It’s small, portable, battery operated, has it’s own screen, a built in function generator and is considerably cheaper (Seeed are currently listing it at $139).

    1. DSO203 and it’s cousins are good devices, they have terrible analog bandwidth though. (DSO203 looks like 8MHz, which isn’t hopelessly terrible, but the DSO nano is only 200kHz!)

  11. Same problem here as with the dso nano. Only one channel and it has the spec’s of a toy.
    That wouldn’t be too bad if it also had the price of a toy, especially because a toy like this can be quite usable.
    Main market for a toy like this is for hobbyists on a low budget or who don’t know which spec’s they’d need from a more mature product.
    USD200 is however far to much for a toy.

    Something like the Xprotolab or the labrador are also toys, but because of their toy prices they deliver quite decent performance for their price.

    Wireless (option) is a bonus.
    Fortunately the manufactories of USB scopes are slowly beginning to understand this and they are beginning to add WiFi to their instruments. So if they also support my Linux box I might be tempted to buy one someday but untill that time I’l just stick to my Rigol scope.

  12. I and a few other members of my group had this idea while working at Tektronix… in 2006 or 2007. We presented it at an innovation conference thing that they had going on.

    I’ll never forget the laughs we got. It never went anywhere. I got laid off in 2009.

  13. More like wireless data acquisition.

    Otherwise it’s like those Bluetooth-enabled tape measures that have no display, so they *need* an Android phone for a display (but you *can* share you measuring info on Facebook!).

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s