You can get one for a decent price these days, Ive bene using the digilent analog discovery 2 lately and its pretty amazing for its proce. Also it allows for much more freedom because of its nature as USB oscilloscope. i hooked it up to an RPi and with a battery and screen its now a proper mobile oscilloscope.
Just out of curiosity - is there a viable USB oscilloscope for $35 out there? I'd love to have a scope for a few projects but I'm hesitant to jump right into one of the $300+ ones.
Yes. I got one from AliExpress for 30$ including the 12V psu. It works fine if you aren't working with super high frequencies. I use it mostly as a voltmeter.
I've never found a USB scope that I particularly liked working with.
Gauging from the job postings on their website, I have a hunch/dream that the folks at Saleae are cooking a decent USB scope up as their next product offering. Would love it if they did. Saleae's Logic Pro 16 is without a doubt the best designed EE tool I've ever used.
For $35 most of the scope you see are going to be repurposed sound card chips acting as a digitizer. They work ok-ish but you get what you pay for.
My big problem with a USB oscilloscope is that they are only ground referenced through the USB ground to your computer and if you measure something with a different ground reference you can blow up the USB port on your computer.
I've been looking for an oscilloscope for almost a decade now. Every year the cheapo ones get better but they're still very limited IMO and feel very expensive for what little they do.
As far as I can tell the Analog Discovery you mention is quoted by Digilent at $279 and for this price you get two 100MS/s analog inputs, so that means that you can't sample anything above 50MHz max (and in practice you probably won't be able to do a lot of useful work above ~40Mhz). With modern circuits routinely having signals in the hundreds of MHz that's fairly limiting for me.
And if I just want to probe "slow" digital signals like SPI or I2C I can get 90% of the features of this module by using a cheap-o ~$10 Saleae clone off aliexpress. It's only a digital analyzer but it goes up to 24MHz and with 8 channels you can already do quite a lot of work for a tiny fraction of the price of an entry-level scope.
Of course if you do a lot of analog work then a digital analyzer won't cut it and the ~$300 price tag for a basic scope is almost a required purchase at this point.
As far as I can tell the Analog Discovery you mention is quoted by Digilent at $279 and for this price you get two 100MS/s analog inputs, so that means that you can't sample anything above 50MHz max (and in practice you probably won't be able to do a lot of useful work above ~40Mhz). With modern circuits routinely having signals in the hundreds of MHz that's fairly limiting for me.
This comment dismisses a lot of measurement use cases out of hand, at least as far as a hobbyist or rudimentary manufacturing test case is concerned. An input bandwidth of 50MHz is still more than sufficient to evaluate all kinds of useful circuits - a few I can think of offhand:
* Buck regulator loop compensation and transient response
* Large signal audio debug (not suitable for small signal just due to the limits in the analog frontend and the number of bits in the scope ADC)
* Basic digital protocol debug
If you're tuning a PLL, or need to evaluate DDR signal integrity, or trying to design an audio system with -120dB of THD+N, then yeah, an Analog Discovery is probably not the right tool. Then again, if you need to do any of those work items, then you probably already know that.
Yeah I've been looking for equipment for a few projects with rather wide bandwidth and frequency range requirements (not neccesarily oscilloscopes, but digitizer cards, waveform generators, etc.) and what I've found is that even if I could afford the equipment they dont sell to consumers. My best bet is going to buy some very old used things on Ebay, but in worried about software support. In some cases the software is so old I'm not sure if i could find it, or it's still exceedingly expensive.
The scope he's screencapping from is a Rigol digital scope. Very full featured little lunchbox scope, for only $350 USD. And you can hack it for extra bandwidth and protocol decoders!
Or buy Siglent SDS1202X-E. 200MHz and protocol decoders out of box with no hacks, intensity grading and the UI isn't laggy as hell (have used both extensively)
Simple: The germanium diode was the first suitable diode I found in my parts collection. :-)
A regular silicon diode will not work. An appropriate Schottky silicon diode will do the job; see the Infineon app-note in the "PPS" of the website. Also, the point-contact (regardless of the semiconductor material) reduces the diode's junction capacitance, making it faster, i.e., more suited for RF applications.
"If one wants to make it more sensitive to certain frequencies than to others, one needs to add a resonant circuit, usually a combination of an inductor and a capacitor"
Coincidentally for OP, the ideal antenna element length for a 2.4 ghz dipole antenna is around an inch and a quarter, which handwavy looks like about how long his diode leads are.
Even a pitiful attempt at a tiny air wound coil choke would decouple his scope probe wires from interfering with the natural dipole antenna pattern.
I bet, that if OP decoupled his scope probe wires, he could plot the natural radiation pattern of a dipole by rotating the diode and its antenna around and noting the voltage on the scope at various angles relative to the transmitter. If I have time I'll try that in the lab today just for fun.
All it'll take is a couple inches of hookup wire... OR more sneakily could coil the diode leads themselves at the correct distance from the diode... Or I could tack solder some coaxial cable and BNC connector with a sloppy homemade choke balun ... hmm...
I bet for TWO EUR I could plot antenna radiation patterns pretty reliably. I wonder if OPs mouse has a horizontal or vertical polarized antenna? I bet I can test that for two EUR.
I've done something kind of similar (but much less low-level) using the nRF24L01+ attached via SPI on the GPIO pins of a Raspberry Pi Zero. I'm in the middle of writing it up, but it will probably be a couple of weeks.
Opinion: HN should have the equivalent of folders/tags for articles; this article and others like it would be perfect for a folder/tag marked "Cool Hack For Under $10" -- or some designation similar to that...
29 comments
[ 3.1 ms ] story [ 74.4 ms ] threadRPi: $35
Battery: $15
Screen: $35
hm.
Gauging from the job postings on their website, I have a hunch/dream that the folks at Saleae are cooking a decent USB scope up as their next product offering. Would love it if they did. Saleae's Logic Pro 16 is without a doubt the best designed EE tool I've ever used.
My big problem with a USB oscilloscope is that they are only ground referenced through the USB ground to your computer and if you measure something with a different ground reference you can blow up the USB port on your computer.
As far as I can tell the Analog Discovery you mention is quoted by Digilent at $279 and for this price you get two 100MS/s analog inputs, so that means that you can't sample anything above 50MHz max (and in practice you probably won't be able to do a lot of useful work above ~40Mhz). With modern circuits routinely having signals in the hundreds of MHz that's fairly limiting for me.
And if I just want to probe "slow" digital signals like SPI or I2C I can get 90% of the features of this module by using a cheap-o ~$10 Saleae clone off aliexpress. It's only a digital analyzer but it goes up to 24MHz and with 8 channels you can already do quite a lot of work for a tiny fraction of the price of an entry-level scope.
Of course if you do a lot of analog work then a digital analyzer won't cut it and the ~$300 price tag for a basic scope is almost a required purchase at this point.
This comment dismisses a lot of measurement use cases out of hand, at least as far as a hobbyist or rudimentary manufacturing test case is concerned. An input bandwidth of 50MHz is still more than sufficient to evaluate all kinds of useful circuits - a few I can think of offhand:
* Buck regulator loop compensation and transient response * Large signal audio debug (not suitable for small signal just due to the limits in the analog frontend and the number of bits in the scope ADC) * Basic digital protocol debug
If you're tuning a PLL, or need to evaluate DDR signal integrity, or trying to design an audio system with -120dB of THD+N, then yeah, an Analog Discovery is probably not the right tool. Then again, if you need to do any of those work items, then you probably already know that.
It will light up near wifi/bt sources.
Am I missing something? Does the diode have to be germanium for this trick to work?
A regular silicon diode will not work. An appropriate Schottky silicon diode will do the job; see the Infineon app-note in the "PPS" of the website. Also, the point-contact (regardless of the semiconductor material) reduces the diode's junction capacitance, making it faster, i.e., more suited for RF applications.
It's a neat hack! Long live the Rigol DS1054Z!
Coincidentally for OP, the ideal antenna element length for a 2.4 ghz dipole antenna is around an inch and a quarter, which handwavy looks like about how long his diode leads are.
Even a pitiful attempt at a tiny air wound coil choke would decouple his scope probe wires from interfering with the natural dipole antenna pattern.
I bet, that if OP decoupled his scope probe wires, he could plot the natural radiation pattern of a dipole by rotating the diode and its antenna around and noting the voltage on the scope at various angles relative to the transmitter. If I have time I'll try that in the lab today just for fun.
All it'll take is a couple inches of hookup wire... OR more sneakily could coil the diode leads themselves at the correct distance from the diode... Or I could tack solder some coaxial cable and BNC connector with a sloppy homemade choke balun ... hmm...
I bet for TWO EUR I could plot antenna radiation patterns pretty reliably. I wonder if OPs mouse has a horizontal or vertical polarized antenna? I bet I can test that for two EUR.