I’ve been looking for a good oscilloscope for a long time.
Unfortunately, they are expensive, and most of them don’t really work the way I want to anyways. In fact, most oscilloscopes seems to be modeled on analog oscilloscopes, which is stupid.
Analog oscilloscopes were originally made with CRT displays, and instead having a TV signal drive the electron cannon, the inputs from the oscilloscope would drive the Y axis of the electron cannon directly. The only memory these oscilloscopes had was in the phosphor on the inside of the CRT, which would persist the drawn line for a fraction of a second.
On oscilloscope like this is fantastically useful, as long as whatever you’re observing is repeatable. In fact, ideally you want it repeat many many times per second, and you want to set up the oscilloscope so that it draws each repetition on the screen. Now when you adjust something, the line on the screen moves, and you can see what happens immediately.
Nobody puts CRT displays in oscilloscopes anymore. Oscilloscopes nowadays are basically computers. They read data from an input sampler and draws lines on the screen accordingly. A new oscilloscope can do all the same things that old analog ones did, but they can do so much more, IF they have enough memory. (Which most oscilloscopes do not.)
IF you have enough memory, then you can skip all the tedious mucking around with triggers and repeated signals, instead you can just start sampling, do whatever you need to do to trigger the thing you want to measure, then stop sampling. After that you have the data in memory, and you can zoom in/out and study it for as long as you want. True, doing it this way means that you miss out on the instant feedback that you get from having the image update immediately when you change something, but not having to have a repeated signal is a HUGE advantage, because making things repeat can sometimes be very difficult.
The other things that most modern oscilloscopes do wrong is that they have lots of buttons, and they come with a screen. Not sure why anybody would want that, because it’s so much better to just hook it up to a computer, use a large monitor and change the controls with a mouse. I guess it makes sense for field equipment, but who uses an oscilloscope in the field??
The final thing that makes buying an oscilloscope hard is that most companies that make oscilloscope are evil. Or… very greedy at least. Not sure if there is a difference… Most high-end oscilloscopes cost thousand of dollars, but once you get it, half the features only works for a month, and if you want to keep using them you have to buy license keys… to use hardware you already bought… I’d feel like Lando if I got an oscilloscope like that.
Anyways, I didn’t post this rant in “Equipment” for no reason, I do have a recommendation; the DreamSourceLabs U3P100:
- open-source software
- 100Mhz bandwidth
- 1 Gsample / s
- 2Gbit memory
- 5Gbit/s usb3 interface
The memory limitations means that I still need to fiddle with triggers and stuff sometimes. However, it’s fast enough to read WS2811 data, and give me an idea not just of the bits that are being sent, but the quality of the signal as well. The open-source software (DSView) is fast and works well on linux.
It’s fairly modest as oscilloscopes go, but the price is reasonable, it’s well made and well supported. I really wish it had more memory though. If it had 8Gb of ram, I could sample at full speed for 10 seconds, which would be enough to perform just about any operation I’m interested in.
Alternatively, if it has a faster interface (10Gbit or higher) then it would be possible to read the data to the computer at full speed, and the memory size would just not matter.
It’s definitely a compromise from what I want, but it’s the best I can find for now. If DreamSourceLabs makes a better one in the future, I will probably upgrade.
DreamSourceLabs also makes very good logic analyzers. I have a DSLogic Plus which has served me very well for a few years. I’m tempted to upgrade to their latest offering, even though I don’t need it…