I would not shy away from used, if you find a reputable calibrator to calibrate/verify it. There are alot of them around. Heck they may even have some they dont/ cant use anymore!!
That’s a pretty awesome memory to have, for sure!
Yeah, that’s definitely a universal experience! I have a couple of cheap multimeters that are my ‘I don’t care if I drop this down a hole’ ones, but even then I found out the hard way that they don’t warn me when the battery runs low, they just get super inaccurate!
That’s good, I think that’s definitely the place to start.
The scope I have at the moment is used but refurbished by Keysight. I’d definitely have no issue with that. There are also places that deal specifically in used test equipment which I would trust as they’re likely to have good checks.
Other than that, I’m 50/50 on the used route. Oscilloscopes are relatively sensitive pieces of test equipment and it’s very possible to abuse them. On the other hand, they’re also pretty easy to verify functionality of. Pretty much all scopes have a built in calibration waveform that you can connect the probes to and verify that you’re getting the correct voltage/frequency square wave out and as long as it has clean edges, you can be reasonably confident there’s nothing terrible wrong with it.
I guess another thing to consider is that scope technology has moved relatively quickly alongside other digital tech over the past decade to the point where there’s not likely to be ‘that’ much supply of gear that I would recommend. The prices have also come down significantly so it’s not likely to be a $250 used Rigol vs a $500 new Rigol, it might be a $500 used Tektronix vs a $500 new Rigol, likely with the same performance but just worse user experience on the older gear. Have a look around and feel free to post up anything that you’d like an opinion on.
It may be cheaper over there but here the calibration likely costs more than the scopes we’re discussing above, sadly. Thankfully I think as long as the waveforms on the internal calibration hookup look good then it’s probably pretty safe, outside of weird stuff like power supply issues or crashing.
1 Like
Thanks so much for sharing your expertise. You’ve given me a few more intelligent questions to consider as I explore this.
I have one of these (not exactly this, but very similar):
It has helped me out in a pinch. It is not very fun to use. It really makes me think the rigol is worth the money.
But it does technically work and it is cheap.
I actually have a USB one too. I planned on getting it to work with the pi I keep on my electronics desk and I got it sort of working. But it prefers windows. I haven’t needed it since.
1 Like
I’ll say! I, too, was a fire control (radar) tech in the Navy (E5/AQB2). Went in Feb '67 and spent nearly a year and a half in various electronics/avionics schools. Lots of avionics equipment was still vacuum-tube based and transistors were the new kid on the block… so was trained in both. The ASB1/7 radar in the KA/EKA-3B (Skywarrior) aircraft I eventually had to maintain was housed in several (50-100 lb each) modules spread from nose to tail. The modulator in the “hell hole” near the tail was especially prone to failure… large rectifier tubes were mounted horizontally and often snapped at the neck with a rough arrested landing. Fun times, struggling to remove/replace it when the plane was parked at deck-edge with tail over-hanging the safety nets and nothing else… 
– David
2 Likes
Not high-performance but handy in many cases… my $300 Rigol scope side-by-side with $20 (now $$28) Banggood scope, monitoring a mini-Rambo’s FAN1 signal, remapped to Z-MIN, while running a Garfield raster engraving file…
Same setup, closer look…
From a long-lost thread here on the forum…
– David
2 Likes
How’s that “Opening Source” working on it? 
1 Like
I use this scope at work, I prefer usb since it doesnt take up much space in my service truck and does what I need it to do. If anything picotech has some good information on their site to look through about using and choosing a scope.
As usual, other people have more time, motivation and attention to detail than me. At the time (and this was a while ago) there was something that worked, but it was missing some features. I saw some low hanging fruit, but I was focused on other things. I haven’t even looked since then to see if there is a better open source gui or proprietary interface.
Open sourcing something is like a hole in a dam though. Once some flow starts, it can pop the whole thing with just a little momentum. That first hole is usually the hardest.
Wierd, I was also a Fire Controlman in the Navy. Surface side for me with Weapons Direction systems and downlink form missiles and then fire control radars (55B Mod10). Then when those ships went away I did Rolling Airframe Missile and Nato Seasparrow. Lat 4 years I changed rates to Information Systems. Been retired for 19 years now. Thanks for your service shippy!
3 Likes
That’s impressively cheap. That’s exactly the type of hardware that I’d normally say to avoid completely because the specs are SO abysmal, but it’s cheap enough that it’s probably worth having around even just as a way to avoid needing to unplug and move the benchtop unit a couple of times a year. 200kHz bandwidth, 1MSa/S, only 1kpoint memory depth, no stated waveform capture rate, fixed trigger position. That’s legitimately the worst spec list I’ve EVER seen.
My attitude is that an oscilloscope is a generalist tool for insight. They can also mislead you and trick you. That small handheld unit is perfect for something like what you’re doing there. Watching a known-good signal and checking PWM frequency/duty cycle. You could also do that with a decently advanced multimeter or tons of other standalone gear. A cheap USB logic analyzer etc. It is already relying on the signal being relatively correct.
Where an oscilloscope actually shines is looking for things that are a bit more odd or complicated like PWM pulses on stepper drivers infrequently cutting out in the middle due to overheating. That’s the type of thing that benefits from more complex triggering, or longer memory depth.
I was mostly just poking fun of the ‘opening’ source instead of ‘open’ source.
But thanks for the extensive review!
1 Like
I knew I liked you! My dad said one day he had to go up and repair a 150’ antennae in rough seas. The mast was saying and he still had to go. (not for me.)
I joined the army to stay on land, but then my motorpool almost had to go to Korea for maneuvers, guess who would have went with the vehicles, That got cancelled thankfully!
1 Like
If I don’t even know what I’m doing, the first thing the oscilloscope is doing is just helping my mental model understand what all the jargon means. Being able to see the PWM or PPM or Serial lines. I don’t think you need a tool like that, because you already have a picture in your mind of what it should be doing.
To maybe use a different analogy, the banggood one is like a cheap hand saw. The rigol is the dewalt table saw, and the professional ones are like sawstops. If you only have to cut one board per year, a cheap hand saw will do a terrible job, but it will be over soon. If you are a professional, the dewalt will screw something up every couple dozen cuts, and you would pay for a sawstop in no time.
My personal opinion is that people should get the banggood one, use it once or twice, realize that the rigol is worth the money, and then buy that one. It would be less waste and less energy to just get the rigol. But the banggood one is enough to whet your appetite and motivate you to get something nicer.
If you plan on doing a lot of EE hobby work, then skip the banggood.
That is mostly based on my cheapo DIY version, not the banggood directly.
For sure, the Banggood one is cheap enough and different enough to a benchtop unit that I think there’s space for both. My real concern would be anything in the middle of those two points. Something that’s a bit more expensive and looks like a Benchtop unit but may be large/expensive enough that it will discourage upgrading.
Perhaps a good analogy is a mitre box and hand saw vs a cheap mitre saw vs a decent mitre saw. To start with, get the mitre box and spend $10. Then if you’re using it lots, get a reasonable mitre saw. If you buy the cheapest mitre saw you can get, you may be less inclined to upgrade it while also forever wondering why you’re cutting shitty mitres. A skilled cabinetmaker can likely make good stuff with a crap mitre saw if they’re careful. A beginner needs all the help they can get. Edit: And, most critically, doesn’t necessarily have the skills to know that their saw is the problem. A crappy scope will show you that everything looks fine and mislead you into missing the runt pulses or double-edges in a waveform that make everything not work right.
Helping that mental model to start with is great, but as I said, you’re looking at things that are likely working. The real power and usefulness of a scope is when you’re figuring out why things might not be working. That’s where I think the lesser specced units will fall down significantly.
2 Likes
So all this talk. I honestly have no idea how to use an oscilloscope. Are there any rudementary trainings recommended? But you know i have never NEEDED one yet, maybe i dont need the answer to my question, lol. I am sure there has to be youtubes out there a plenty.
Fundamentally, an oscilloscope is just a voltmeter that displays voltage vs time in the form of a plot. Almost all of the adjustment you’re doing when working with the scope is zooming in and out, both in time and voltage, setting a trigger point which is what causes the scope to start displaying what it sees on the screen and the position of where that start point is, relative to the screen.
So if you were looking at an unknown digital signal, you’d start by adjusting the volts per division knob. This changes the ‘zoom’ in the Y or voltage direction, so making the scope more sensitive means that the same waveform will be ‘taller’ on the screen. If the waveform is too small then you’d make it more sensitive to take up more space. If you can’t see the waveform or it’s disappearing off the screen, you’d lower the sensitivity. You can also adjust the voltage offset which essentially moves the viewport of the screen up and down. In the case of our random digital signal, you might adjust it so that the volts per division is at 2.5V per division and the waveform is taking up 2 divisions on the screen, meaning it’s a 0-5V waveform.
Once you have an idea of the voltage of the signal, you can adjust the trigger. This is what the scope looks for and tells it where the ‘start’ is for the waveform you want to look at. The simplest form of this would be a fixed voltage and then a transition direction. For our 0-5V waveform above, we can set the trigger level to be halfway between the high and low and then set it to only trigger on a positive going edge. That way the scope will wait until it sees the waveform go from say 2.4V to 2.6V and will then trigger, showing what it has recorded to you with that trigger point in the middle of the screen.
The next thing would be adjusting the timebase. This is zoom in the X or ‘time’ direction. If you’re seeing nothing but 2 lines with an occasional brief blip between them, you’re probably too zoomed in. If you’re seeing a ‘fuzz’ of loads of transitions between 0V and whatever a digital 1 is, you’re probably too zoomed out. Once you’ve got it so that you can comfortably see a bunch of pulses on the screen then you can look at the timebase setting and figure out what the frequency/duty cycle is. If we’re seeing a few pulses on the screen with 2 divisions between each positive going edge, we can look at the timebase to see what that means. If the timebase is set to 1ms then we know that the period of the signal is 2ms and the frequency is 500Hz. We can also see what the duty cycle is just by looking at it.
The next things we can do are things like adjusting the trigger offset, which essentially moves the trigger point left and right on the screen. A modern digital scope is always recording so you can have the trigger set either ‘before’ what you can see on the screen or after. Maybe you have a pulse that only occurs sometimes and you want to see the 100ms of time that leads up to that pulse, or what happens 1s after the pulse. You can do that by moving the trigger.
You can also then start to set up other stuff like math functions that will measure what’s on the screen for you. These can give you an easily readable number that shows you the maximum voltage, minimum voltage, peak-to-peak voltage, the frequency, the duty cycle etc. It can also give you statistics on those numbers so you know what the average, minimum, maximum and standard deviations of those measurements are.
You can have the scope in ‘free running’ mode, where it will always show you the latest measurements, regardless of whether it’s triggered or not. You can have it in trigger only mode where it only shows you stuff when it’s seen a trigger event. You can have the pause the scope to keep what’s on the screen from changing. You can set it to single shot where it will only run once when it triggers and will then pause.
Once the scope is paused, you can zoom in and out on parts of the signal. The screen might only show you 1000 points, but the scope may have captured millions of points in that same time period, so you can find a spot you want to look at and zoom in up to thousands of times ‘closer’.
You can also adjust the trigger type to be a little more complex. You might adjust it to look for pulses that are below a minimum width, which means the scope will need to see a positive going edge and then a negative going edge within say 1us. If you have a communications bus like SPI or the switch in a switch-mode power supply, this would be a situation that should never occur. Then you can set the scope to one-shot mode and wait to see if it triggers. If it doesn’t, you don’t have an issue, but if it does you can now clearly see that issue as it occurs.
You can then hook up other channels as well, which are always synchronized to one another. So you might be looking at 2-4 different voltages all at the same time. You choose to trigger off one, perhaps the clock on a data bus, but you can see what the other signals are doing at the same time. You could put one probe on a chip select line and use that as a trigger, then put the other probes on the clock and data lines. When the chip select line gets asserted, the scope can trigger and capture you an entire communications event. You can use the runt pulse trigger described above and look at the feedback lines of a switch-mode power supply to see what might have caused it. You could look at the supply voltage for a chip with one channel and the output of the chip with another.
All sorts of stuff like that.
This is kinda why it’s hard to describe what a scope does. It’s ultimately just showing voltage vs time, but what you can do with that is the fundamental building block of almost all electrical engineering.
From a quick google, Sparkfun seem to have a decent bit of information:
https://learn.sparkfun.com/tutorials/how-to-use-an-oscilloscope/all
Edit:
There are also a bunch of virtual oscilloscopes in web, app and downloadable executable form for training.
This one seems reasonable:
https://automatisierung.fh-aachen.de/virtuallab/scope/dist/
You can control the knobs by mousing over them and moving the scroll wheel, can adjust the trigger slope, all the basics. You can also adjust the type of waveform you’re feeding into it.
3 Likes
Whoa!!! Impressive reaponse!
2 Likes
It got out of hand remarkably quickly…
2 Likes
Yeah i just had someone ask me how network communications work. I got away on them and i am not really sure how much they understood, i get it, lol.
1 Like