Do you find Oscilloscope useful in FRC?

Hello all,
Our team was wondering if we should upgrade and get a new oscilloscope. What is your experience with using them in FRC? And how often do you use them?
Also, if you have a specific recommendation, would like to hear.

I bring in my Rigol DS1054Z sometimes, but I find that we don’t really use it that often, if ever. The most useful thing it’s used for within our team is debugging PWM, but we don’t use PWM on our robot anymore anyways, so it’s kind of defunct there anyways. At home though, I use it a lot. Everything from reverse engineering serial communications protocols cough cough ws2812b cough cough to just general screwing around with microphones and such.

2 Likes

My previous team had an electrician who’s day job consisted of troubleshooting industrial machines (plastic injection, die cast, etc) as a mentor. He brought a lot of nice tools to the table, but the tool I found most useful was the clip on amp meter. Super nice to clip on each motor in a gearbox to see if motor controller following was working as it should, or quickly see how hot things were running. He also had another higher end Fluke multimeter that we’d use to verify frame electrical isolation, but I don’t think it gave us anything a low end multimeter wouldn’t.

But to touch on oscilloscopes: we never needed one. We moved away from PWM a few years back, and never needed it before that switch.

4 Likes

The only valuable thing I can think of that we use an o’scope for is to measure how long the software is taking to do something, i.e. toggle a pin, capture an image and do math on said image then toggle the pin, measure time between toggles.

If you want to understand the time characteristics of a signal, you use a scope.
Multi-meters will give you a long duration averaged value only.
Suppose that you are looking at a discrete signal - the multi-meter shows 3 Volts and you expected 5. The scope can show you the actual waveform on the discrete line, perhaps a mix of 5 Volts and 0 Volts.
Perhaps you are trying to debug an analog measurement - the scope can show you if there is noise or perhaps some sort of localized drop-out in the indicated signal.
You need to understand how you are connecting the scope to the system you are testing - hopefully you have a mentor that is comfortable working with a scope.

Would I use a scope to measure Voltage on a battery after charging? no, probably a beak or a multimeter.
Would I use a scope to measure Voltage on a battery operating a robot? Perhaps, this could provide visibility into the Voltage droops as a particular actuator kicks in (even then only with the robot “up on blocks”)

For use in FRC, my favorite scope is my scopemeter. I have a Fluke 199C. I would prefer to have the 4 channel version, and I would really prefer to have a dual time base. The scopemeter is battery powered and isolated from the circuit being tested.

From my experience in FRC, I would consider a scope to be a specialty tool. But if you have a mentor that is interested and the funds, perhaps purchase a replacement

Only time I can think of that we used one was 2011, used to tune a potentiometer we had on our arm. Safe to say, unless you’ve run out of other necessities to spend your budget on, an advanced oscope for the purposes of FRC is pretty low on the totem pole.

If you’re in the market to spend money (as schools tend to be around the summer time), I recommend using the Spectrum First $10k as a checklist of sorts.

1 Like

Unless the people using the oscilloscope understand what it can and cannot do and how to use it fluently, an oscilloscope will most likely be a distraction from a productive troubleshooting procedure. Mistakes in how the probes are grounded can lead to many wasted hours chasing phantom noise (like some of my coworkers).

If one is going to purchase an oscilloscope, get at least a 2-channel Rigol. Alternatively, a good, used Tektronix could be a good option. From the last time I looked, none of the inexpensive USB scopes were worth considering.

The meter linked is a professional-grade one capable of measuring DC as well as AC currents and may be too expensive for many FRC teams. Most less expensive clamp meters can only measure AC currents. For FRC applications, it is worth finding one that is capable of measuring DC currents.

We use an o’scope once or twice per year and I bring in my own 100 MHz Hantek digital storage scope on those occasions. If the team had one stored in our high school it would be exposed to unspeakable horrors during school days.

A few times in the past, when we were trying to get home brew encoders working…but now that encoders are so much easier, haven’t had the scope to the school for quite a few years.

Of course, building a scope just to play with vacuum tubes was fun when I was a kid, and it’s still fun, although sourcing parts works a bit differently than it used to in the early 70s.

2 Likes

Yeah, I don’t think we’ve had to use one in a couple of years. I have a 1052Z that pretty much permanently resides at our shop since I’m not doing much electronics anymore at home. But as you said, since we haven’t used PWM in a while, it’s not really useful. We still bring it with us wherever we go, because you just never know.

1 Like

Given the improvements to WPILib and the way our sponsors, suppliers, and the community have improved our tools to the point of “plug and play”, a scope is not in the first $10k I’d spend on FIRST.

If you’re using non-RIO microcontrollers or building analog circuits, it might be good to get one in. I’ve used my USB-based Digilent scope once in six years, to check what my microcontroller was doing, and was happy enough with the experience. It seems like a reasonable tool for circuit exploration.

Knowledge of digital & analog circuits to the depths that you need a scope to explore and observe them is not a requirement to compete in FRC, but does offer good learning opportunities.

1 Like

Yup, about twice a year. The two main functions are:

  1. Hunting down weird noise-related issues
  2. Sniffing I2C or SPI packets to debug sensor behavior.

You could totally get by without one, and, echoing others, the use you get out of it will directly correlate to how skilled of users you have around.

100% recommend Rigol scopes as best bang-for-the-buck in traditional (no-external-pc-required) scope format.

2 Likes

These are the ones we use at work, and they absolutely rock. That being said, the 100A rating is a bit low, you’d probably want to find a 500A one… I totally want to get one for the team. But budget…

I’ve used ones like that (bigger and smaller)as well as some Rogowski coil types at various jobs but they only work on scopes.

My Amazon shopping list has this clamp meter. The product description doesn’t give a lot of specs. The question “Can DC current be measured by the clamp” has an answer from the vendor that it is rated for “DC and AC current to 400A”.

I also just came across this clamp meter that is specced for 600 A AC and DC.

Both of these clamp meters are within my personal budget ($38.97 and $30.99) so they should be within the budgets most teams are working with. I am under no illusions that they are in the same league as the professional meters. I had a Craftsman clamp meter that was in the same price range and quality level that lasted a good 12 years before dying.

1 Like

Ooooh very nice. Since I’m having fun searching around… Slightly more expensive, but more scope friendly. Slightly outside the range of what I’m willing to buy on a whim.

That being said, if one showed up on my doorstep, I’d happily run it through its paces and post thoroughly about it here.

That’s cool. I have been looking for something like that for my scopes at home but don’t have the budget to get a Tektronix one.

I noticed this other one for $51.90 that has a higher frequency response (20 kHz vs 400 Hz).

Looking at the clamp meters after I got home, I came across the Klein Tools CL380 for $84.99 at Amazon and Home Depot for $89.97. It looks much better than the Tacklife and Proster units.

1 Like

I’ve used a scope to determine if a sensor was bad or software was buggy before, that being said it’s typically faster to just replace the sensor first to get the robot working and play with the sensor later / in the off season.

I’ve also used them to check out noise on a CAN bus (not for FRC robots), i don’t recommend doing this, turns out CAN is really robust and it’s not a very useful venture.

If you have one it’s a useful tool on occasion, spend your money on other stuff though - there are way more useful tools for FRC for the price.

1 Like

Yeah, I forgot about SPI/I2C debugging. I guess we’ve been lucky the past couple years, and haven’t needed it, but we have in the past. Since my Rigol 1052 isn’t mixed signal, it’s kind of annoying to decode, but it still makes it easy to detect the presence of data.

Also, speaking of mixed signal/decoding. Has anyone had much success using the saleae logic analyzers decoding the CAN bus for FRC? Does the security token mess with decoding?

Since the Saleae is a logic level analyzer you would need a CAN transceiver to convert from the CAN PHY to logic level. If you do that here is our FRC CAN decoding for Saleae, we use the Logic Pro 16.

Though if you just want to see the traffic on the bus, you can use the SPARK MAX over USB with any tool that works with the canable such as this one:

3 Likes

If your team really has the need for an oscilloscope, look into the DSO Nano. They’re under $100 and about the size of a credit card. I use one all the time with my modular synth setup.