New, Inexpensive DSO now shipping

Seeed Studio’s line of hacker-friendly tools has expanded by one, they’ve announced that beta units of their DSO Quad oscilloscope are now available for shipping. The DSO Quad is about the size of a thick iPod yet packs impressive features such as two 72MSPS analog channels and a signal generator. By far the coolest ‘feature’ of the DSO Quad is that it’s completely open source.

DSO Quad is a pocket size four-channel digital oscilloscope designed “for common electronic engineering tasks.” It’s based on an ARM Cortex M3 that provides 72MSPS analog bandwidth on two channels with an integrated FPGA and a high speed ADC. In addition to the four data acquisition channels (two analog, two digital), the DSO Quad has a signal generator.  This lets you put out square, triangle, saw, and sine waves from 10Hz to 1MHz. An internal 2MB USB stick can be used to store sampled data, upgrade the firmware, or run custom programs. Since the device is still in beta, some of the software specifications aren’t firm, but if you’re looking for a relatively inexpensive ‘scope, this could be the one for you. Just bear in mind that for this pre-production run you’re not getting any documentation, so be prepared to be off the reservation and on your own.

24 thoughts on “New, Inexpensive DSO now shipping

  1. Yeah if you haven’t been following their twitter feed lately, you’re too late. I’d wait until some of the little kinks are worked out if you are seriously considering adding this as a lab tool though. I have a Rigol DS1052E and couldn’t be happier with the performance for home and office use, although it’s double the price of this pocket gem. The guys at Seeedstudio make some really cool stuff!

  2. They claim 72 MHz analog bandwidth with a 72 MSPS sampling rate (on one channel). This seems surprising. I tried to download the schematic from the wiki but couldn’t. Can anyone speak to the veracity of the analog bandwidth claim?

  3. Analog bandwidth can be higher or lower than Nyquist – it’s just a measure of how much is lost before the ADC. With only 72 MSa/s, you’re mainly limited by the digitization, but if you band-limit the signal from 36 MHz to 72 MHz you could just work in the second Nyquist zone if the analog bandwidth is that high.

    That said I’d bet they just overdesigned the front end.

  4. @Patrick & DanJ

    They are probably achieving 72 MHz analog bandwidth using a technique called “equivalent time sampling”, in which the samples that make up a waveform on the screen actually come from different triggering events. In my experience, ET sampling is really, really annoying to use compared to real-time sampling–you really have to be much more careful about setting up the scope parameters instead of just trusting that the trace on the screen is an accurate representation of the signal you are probing. The problem boils down to the fact that the signal is usually not exactly the same at each trigger event, so (especially with a complex waveform) the image you see on the screen tends to be a lie.

    Basically, this scope will be useful to about 10 MHz for general-purpose stuff, and a skilled operator might be able to drive it to 72 MHz under the right conditions.

  5. @Philippe,

    You’ll notice that I used the word ‘about.’ Most ‘real’ scopes aim for 10 points per Hz of real-time bandwidth. You can actually get away with only 2.5 samples per cycle with sinc(x) interpolation, but you’d better have a really good anti-aliasing filter on the analog front end.

  6. beginners don’t get fooled, open source in no way make up for such toy-like performance. Chinease now producing something you can really measure with in similar price range

  7. @therian might be referring to the much discussed DS1052 from Rigol which is a little over 300. If someone can spend the money, i recommend the rigol. It’s a real scope, not quite a toy like this.
    I would buy this one here only if the price was half…

  8. It does seem like an expensive toy, however it could be quite adequate in some circumstances. What I like about it is that it resembles a PDA. I do contract work in facilities that have strict policies (read: paperwork) for bringing in gear, and more for taking it out again. Doesn’t apply though to cell phones and PDAs, and most security personnel couldn’t tell the difference with this kit.

  9. Does anyone know where to get scope probes with a 1/8″ TRS plug? I think that is what this uses. As much as I want to chop up one of my own, I don’t want any high frequency issues. And I don’t really know the proper way to connect the coaxial cable from a probe to a TRS plug. I’d imagine at higher frequencies the typical way of doing it would cause some issues.

  10. I’ve been following the development of this and I, like others, have some difficulty in getting behind it. I have a softDSP scope that works nice (for the work that I do). A DS1052E would be a nice upgrade in my situation.

    The one thing going for this is it’s size and portability. Some automotive applications could justify getting one, but I don’t do enough of that work to make it worthwhile, unfortunately.

  11. The DSO nano is not a toy its an impressive little scope. It samples in real time. It does have limitations, as does any scope. I own a Picoscope, which is one of the best you can buy. But the nano has a big buffer for its size. Buffer means just as much as the sample rate in a DSO

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.