23
votes
\$\begingroup\$

I'm learning electronics (and basic electricity) and I wondered if a 1MHz oscilloscope like the DSO Nano is enough for (very) basic electrical and/or electronic circuits. I'm almost broke and I need a cheap portable oscilloscope to try various circuits (wherever I am, hence the required portable property) with resistors, capacitors, transistors, and microcontrollers (if I study hard enough). I don't thing I need a more powerful oscilloscope because I'm learning and I'm just following the exercises in various books right now. This would be my first oscilloscope and I would eventually buy a 20MHz portable oscilloscope later.

Thank you!

\$\endgroup\$
  • \$\begingroup\$ Also look at the DSO Nano v2 and the DSO Quad (it's currently in beta) \$\endgroup\$ – drxzcl Jan 27 '11 at 8:46
  • 1
    \$\begingroup\$ @Ranieri - v2 is now in stock at Sparkfun as well, admittedly for $10 more than seeedstudio. BTW, you can make better links with [text](http:\\link) (edited your comment for you) \$\endgroup\$ – Kevin Vermeer Jan 27 '11 at 17:38
  • \$\begingroup\$ @reemrevnivek - Thanks, I had no idea that worked in comments too. \$\endgroup\$ – drxzcl Jan 27 '11 at 20:52

10 Answers 10

12
votes
\$\begingroup\$

IMHO for its low price, the DSO Nano is actually a good first-time scope. As others have mentioned, you will have problems with showing anything faster than 100-250KHz coherently, but this is pretty high for quite a lot of uses. Yes, you won't be able to see system clocks or PWMs with it, but in my experience a scope is more useful for analog signals, and as long as they don't rise and fall too quickly, you'll be fine with the DSO Nano.

Once you get more money and become more serious about the projects, you should invest in a 2nd hand scope, something real with at least 100MHz bandwidth.

\$\endgroup\$
25
votes
\$\begingroup\$

The DSO Nano is a poor choice for a hobbyist's oscilloscope.

Capabilities

It's really only good for audio bandwidth signals - anything above 100kHz is going to be visible, but unmeasurable. By most standards you want the sampling rate to be 10x the bandwidth of the signal in order to get close to measuring significant parameters of the signal (peaks, rise time, frequency, duty cycle, etc). While you can see signals with as poor as a 5x sampling rate, you can't reliably measure such signals and expect to get good results.

Competition, New Oscilloscopes

Keep in mind that a very low end entry oscilloscope NEW is only about $400 - so if you can spend more then a 2 channel 50MHz Rigol would be a fine choice as a beginner:

http://www.dealextreme.com/details.dx/sku.30573

Note that the 50MHz rating is the fastest signal it can measure. It measures at 1 billion samples per second for one channel, or 500 million samples per second for both channels, which means you'll be able to see signals much faster than the rated 50MHz.

Compared to the single channel, 1 million samples per second Nano this is worlds better.

Competition, Used Oscilloscopes

But even if that's too much, you can get a $50-$100 used oscilloscope on ebay that is significantly better than the nano. For the same price you can easily find a 2 channel 20MHz scope (which can be used to measure the clock of a 40MHz micro, just not precisely) and there are many scopes in the 50-60MHz range for under $100.

What is it good for?

The nano is NOT an oscilloscope replacement. The major reasons to get one are

  • You want one (it's open source, and could be fun to hack)
  • You need a portable device that can function as a simple oscilloscope for audio bandwidth applications
  • You are only interested in high-level overview of the signals you measure, and won't really be doing significant work with it (ie, beginning hobbyist) so don't want something bulky that you might only use a few times a year.

That being said, despite its severe limitations, I can see it as an additional tool for the workbench after you've invested in a good oscilloscope and other gear. Keep in mind that many tasks, such as SPI, can be slowed down to within the range of the oscilloscope, and even though it's only good up to 100kHz analog signals, one could reasonably measure up to 500kHz digital signals (if they wrote software for the Nano that synchronized the ADC clock to the input signal).

Conclusion

The price to performance of the device is just so low that I can't see it being worth the cost except to tinkerers that might want to hack it. Your money is better spent on a used oscilloscope.

\$\endgroup\$
  • 2
    \$\begingroup\$ Yeah, but all of those other scopes are 100x as big. \$\endgroup\$ – endolith Nov 4 '10 at 15:33
  • 1
    \$\begingroup\$ You are wrong where you say "see signals much faster than the rated 50MHz". If the analog chain isn't up to passing faster than that, it doesn't matter how fast you sample it. \$\endgroup\$ – Brian Carlton Jan 26 '11 at 23:54
  • 3
    \$\begingroup\$ @Brian - I didn't say that you'd be able to measure them, or see their exact representation, but the analog chain doesn't simply cut off everything above 50MHz. It's designed to be able to accurately render and measure signals at 50MHz, and the accuracy degrades beyond that as the analog filters attenuate higher frequency signals. So while it won't be a great representation, you would, in fact, be able to see if your 60MHz oscillator is at least oscillating, for instance. \$\endgroup\$ – Adam Davis Jan 27 '11 at 0:18
  • 3
    \$\begingroup\$ It's also the case that "50 MHz bandwidth" generally means "3 dB down at 50 MHz". The rolloff curve will be fairly smooth...not an ideal RC filter, but hardly brick-wall, either. So, if the signals are tall enough to survive some attenuation without getting pushed into the noise floor, the scope could well show things well beyond 50 MHz. \$\endgroup\$ – Warren Young Jan 21 '12 at 6:09
12
votes
\$\begingroup\$

Your typical hobbyist microcontroller can range anywhere from KHz up to 40MHz or more. So if you are wanting to have a microcontroller running at lets say 40MHz and be able to see what a PWM looks like, you may be out of luck.

However, if you are just wanting to see a general idea of what is happening on your analog circuits or simple digital circuits then the DSO Nano will work just fine for you.

Another option to consider is an OScope on a computer. You can use the sound card on your computer to interface with a program (Here is an example). The down side to this is that computer sound cards tend to be very noisy and many times only have a sample rate of about 96KHz or lower in some cases.

\$\endgroup\$
  • 6
    \$\begingroup\$ Even if the microcontroller itself (eg a PIC) runs at 40MHz, most likely the signals it is interfacing with the world will be nowhere near that speed. IIRC for PICs the fastest thing you could do is SPI, and that's at 1/4th the clock speed, so 10MHz in that case. Still much too fast for a 1MHz or even a 10MHz scope of course though, so your main point still stands. \$\endgroup\$ – davr Feb 18 '10 at 17:58
  • \$\begingroup\$ Thank you for the comment, I was considering buying some software like the SignalScope but I'll stick with the DSO Nano as my first oscilloscope. \$\endgroup\$ – interlol Feb 19 '10 at 8:51
  • 4
    \$\begingroup\$ You don't need to buy software to use your sound card as an oscilloscope. There are plenty of good free or open-source packages available. \$\endgroup\$ – Thomas O Nov 4 '10 at 15:34
7
votes
\$\begingroup\$

I have put together an oscilloscope comparison table where you can filter and sort devices to find the one best for your needs: Digital oscilloscopes for hobbyists

\$\endgroup\$
  • \$\begingroup\$ Your price estimate for the AVR DSO seems way off to me. The author quotes EUR 50-60, which is about USD 70-80. \$\endgroup\$ – drxzcl Jan 28 '11 at 10:00
  • \$\begingroup\$ @Gabriel - That is a fantastic table. If you're looking for more data, Cleverscope has a decent comparison here for the standalone Tek TDS2012B, and Agilent MS6012A, and the Pico 3206, Link Instruments DSO-8502, Bitscope 310 and TiePie HS3-100 PC scopes. \$\endgroup\$ – Kevin Vermeer Jun 20 '11 at 18:52
6
votes
\$\begingroup\$

A good rule of thumb is you need a minimum of 4x the speed of the signal you are watching, and higher (up to 10x) is better. So if you want to watch a signal that's running at 1MHz, you need a 4-10MHz scope.

\$\endgroup\$
  • 1
    \$\begingroup\$ This would depend wildly on why you are looking at the signal; is it some analog trace or are you trying to look at the edges of your digital signal for overshoot, jitter, etc. \$\endgroup\$ – Nick T Nov 4 '10 at 17:34
3
votes
\$\begingroup\$

Depends on what you're designing. If you only need 1 MHz bandwidth it should work fine. You can get a cheap analog scope on eBay though, and it will have a higher bandwidth. Just bulkier and harder to carry around. The shipping on my old 20 MHz analog scope was more than the scope itself. :)

I also have a Bitscope BS50U, which I carry around with me, but I almost never use it and don't recommend it. The software is the weakest link. It's hard to navigate, unintuitive, hard to get a good trigger, etc. A software-based oscilloscope should be more powerful than a hardware scope, not less.

\$\endgroup\$
3
votes
\$\begingroup\$

No, I wouldn't advise it.

Main reasons:

  • While it has a 1 MHz bandwidth, its sample rate is only 1 MS/s. Practically this means you can only measure up to 100 KHz square waves. It is only suitable for audio signals in my opinion.

  • You spend $90 on the DSO Nano but you could spend $200 on a second hand oscilloscope with at least 60 MHz. I spend £150 (around $200 at the time) on my HP 54501A which was a pretty good deal. It has everything a hobbyist should need: 100 MHz bandwidth (will measure up to 300 MHz clock signals), excellent trigger facilities, including video, pattern and state (very useful for what I'm doing), it's digitising and it has HP-IB. If you can find a similar deal, go for it.

  • Trigger: it only mentions auto and single. Out of luck for complex triggers like I mentioned. You don't think you need them, but you often will.

  • Input protection: or lack of it. Most good scopes can take at least 100V on the inputs. This one only goes to 80Vp-p (so 40V amplitude maximum.) Don't worry, I've done this one or two times to be thankful that my scope has 250V maximum inputs.

  • It's not earthed. Don't even touch equipment plugged into the mains or you won't be protected like a normal scope is.

\$\endgroup\$
3
votes
\$\begingroup\$

The best scope for beginners is hands down Rigol 50Mhz scope, I used it for professional and hobby projects. You can buy below 350$, it has math function, two channels, can be hacked to support 100Mhz, digital filter, a decent size memory, USB capture etc. etc. Cannot say enough about it, it is really good. It has an antialiasing problem at 1msec but you can work around it.

\$\endgroup\$
1
vote
\$\begingroup\$

The first great step in oscilloscopes functionality was triggered sweep. Ignoring ones which are not triggerable, A lot of old scopes SUITABLE FOR BEGINNERS can be set to trigger at a voltage level within the range displayed, selectable for rising or falling slope. More complicated triggering than line-frequency-synchronized and adjustable per the signal as described above is suitable for paying attention to once you figure out how to use these. I used to sell used test equipment, and the combination of adjustable trigger and delayed sweep in the old tektronix 530 and 540 series oscilloscopes is the reason Jim Williams kept on restoring them until he died, FYI he is a recently deceased master analog circuitry designer. I can't speak well enough about the simplicity of these scopes as a first scope for a beginner. And they keep the room warm while you use them.

\$\endgroup\$
0
votes
\$\begingroup\$

Just to throw this out there, if you're trying to learn from a book and tight on money, another option is to use a circuit simulator. I've had sucess with SPICE (free versions), but I'm sure there are other options.

I'm not trying to say that hands on practice with circuits is not needed. At some point you do have to try things with actual hardware. But if you're just trying to learn and money is a constraint, circuit simulators can be very helpful.

\$\endgroup\$
  • \$\begingroup\$ I wonder why this is down voted ! What Dr. Watson telling is the truth. Today's electronics is completely SPICE and mathematical modeling based. Specially when it goes to nano-scale electronics. \$\endgroup\$ – Standard Sandun Mar 23 '13 at 10:49

Not the answer you're looking for? Browse other questions tagged or ask your own question.