What do a Rogowski coil, a magnetic core, and a hall effect sensor have in common? They are all ways you can make oscilloscope probes that measure current. If you think of a scope as a voltage measurement device, you ought to watch the recent video from Keysight Technology (see below). It is true that Keysight would love to sell you a probe, but the video is not a sales pitch, just general technical information about making current measurements with an oscilloscope.
Of course, you can always measure the voltage across a shunt resistor — either one that is naturally in the circuit or one you’ve put inline just for measuring purposes. But if you add a resistor it will change the circuit subtly and it may have to handle a lot of power.
The Keysight video points out that there are different probes for different current measurement regimes. High current, medium current, and low current all use different probes with different technologies. The video is only about 6 minutes long and if you’ve never thought about measuring current with a scope, it is worth watching.
The video shares some high-level details of how the current probes work — that’s where the Rogowski coil comes in, for example. Of course, you can’t expect a vendor to tell you how to build your own current probes. That’s OK, though, because we will. Current probes are often expensive, but you can sometimes pick up a deal on a used one.
Good job Mrochta! ;)
Why does she have this weir affect?
It’s so funny. I checked out their youtube channel. Aly gets about two times more views than the other redhead they have on their stream and four times more views than the videos with guys, despite the latter being actually the more informative. lmao
Doesn’t surprise me in the least. Most people interested in this are male (whether that’s good is another question, but it can’t hurt to have some female role models).
Informative is quantifiable in more than one way. If no one watches, how informative was a video? It might be great, but it didn’t inform anyone who didn’t watch it…
This one was good enough and actually prompted me to drag out my current probes and measure noise and bandwidth on them, FWIW. And the ones from the company she’s representing are BTW, better than the ones I own (which weren’t cheap).
So you are sying clickbait is also informative because a lot of people look at it? But regardless- I’m not making any judgements, I was just posting this funny observation.
Of course I’m not saying that. What I *AM* saying is that if no one sees it at all, it cannot be effectively informative – at all, in a sense of the results in the real world. Clickbait is named that in part because it leads to worthless stuff, else it’s called a useful link. You called that clickbait. I found it informative. I didn’t call it clickbait – you just did.
Don’t be a dick.
Well said.
This.
Relevant: Keithley publishes a wonderful little book describing how to do precision low-noise measurements, including current (of course). Available free from their site as a 4 MB pdf: http://download.tek.com/document/1KW-30010-0%20KeithleyLowLevelHandbook_7Ed.pdf Highly recommended if you’re interested in this sort of thing.
Wow. On a quick glance, that looks awesome.
The Keithley Low Level Measurements Handbook is still in its 7th Edition and as of 26-September-2023 (when I wrote this comment) it can be freely downloaded here:
https://www.tek.com/en/documents/product-article/keithley-low-level-measurements-handbook—7th-edition
In a circuitous way, Keithley Instruments [1] over the years became a subsidiary of Agilent (ex-hp) then a subsidiary of Tektronix [2] (now owned by Fortive [3]).
1. Keithley Instruments
https://en.wikipedia.org/wiki/Keithley_Instruments
2. Tektronix
https://en.wikipedia.org/wiki/Tektronix
3. Fortive
https://en.wikipedia.org/wiki/Fortive
I want one of those new flux gate probes, those are really cool.
I tried this shunt resistor idea. Result: a blown fuse in the oscilloscope. I assume a short between voltage on the probes’ GND and the grid GND, where most affordable scopes are connected to.
Lesson learned: either use a battery powered scope or some indirect measuring device.
Why would there ever be a voltage on the probe’s ground?
The probe’s ground goes to ground, it should never be connected to a voltage.
Probe ground is connected to grid ground, that’s true. And also the problem. Because because both sides of the shunt resistor can have a voltage on them.
Doesn’t apply if one side of the shunt resistor is connected to ground anyways, of course, but in that case (stepper motor driver -> motor) it wasn’t.
Yes, this. It is safest to never connect line-powered scopes’ ground leads to anything other than a ground (you can get away with it in special cases, but…). Generally to measure voltage differentials, use two probes and take the difference between them (most digital scopes have a math mode that will add or subtract the two channels automatically).
And most (all?) digital scopes implement this A-B difference mode extremely poorly. Because they do the difference in math, on the already-digitized data, the common mode voltage must be within the range of the digitizer. Any difference of two signals, in any useful real-world application, will be a small fraction of that, just a few bits of the digitizer. Couple that awful dynamic range with the positively abusive user interface most new scopes enjoy inflicting on the user, and the “difference” feature is a frustrating, pointless exercise. You really, really need an analog difference amp if you plan to do any of this.
Good Lord, give me a decent UI and front end on modern ‘scopes please. (not that I’m complaining about noise and bandwidth vs. cost though).
So, according to the video, the main reasons for not using a shunt resistor is because you have to do math in your head? Not all the other crappy things with it, like having to break up your circuit, introduce a resistance in your circuit (which, given high current, might generate a lot of heat) or the inability measure small currents. Or the fact that if the low side of your shunt resistor isn’t GND, you must use two channels to measure each side (which comes with its own problems).
This also touched on another question: Why doesn’t oscilloscopes have a math mode (maybe they do, apparently the one she uses doesn’t) where you take your signal and divide/multiply it by a known factor, and let the oscilloscope calculate ohms law for the case of using a shunt resistor. This cannot not be hard to implement.