Does Kirchhoff’s Law Hold? Disagreeing with a Master

Incredible. Within a few seconds of hearing how he did the measurements I identified the measuring mistake.

Rav

Remember the cold fusion discovery a few decades ago. It turned out the measurements were done incorrectly. The difference between a physicist and an engineer. A designer vs a tradesman. Theory vs. practice.

Rav

Okay. I'll definitely try the shorter shunt route first. Thank you for the advice!

For steady state currents that's fine. But for fast spikes, it can round the edges or create ringing (oscillation) and make your reading uglier than it is. If you don't mind those, go ahead.

ElectroBOOM [Mehdi Sadaghdar]

Thank you so much for the response! Okay, I will look into that. I was concerned about folding the wire, as I wasn't sure I could get wire thick enough to carry the current without getting too hot or excessively dropping the voltage, but flexible enough to fold. If I am unable to find appropriate wire gauge, would it work just as well to keep the battery cable/sense lines straight during measurement? Of course, I will be unable to filter out mains hum or other background interference, but I i'm really just looking for large amplitude transients and steady state current measurements.

You could do that, or for example twist the sense wires around the battery wires, but the battery cable is too long. I don't like such long sense resistor, because you don't know were it could pick noise from that easily. Instead use a lower gauge much shorter wire as your shunt and fold it like I showed in my SHUNT video.

ElectroBOOM [Mehdi Sadaghdar]

Hi Mehdi! I humbly request your advice if you have a moment to read the following: Firstly, I would like to thank you for the great content! Using this video and your previous video about shunt resistors, I have been able to apply your teachings at work in an effort to measure current production of an automotive alternator. You are making me look great :D! My question is as follows. I have a very thick battery cable about 3 meters in length that I would like to turn into a shunt. To do this, I will solder a signal wire to each terminal of the battery cable, and measure the voltage drop. As this is an alternator, I am expecting currents up to 400 amps, so I will need to be careful about the inductance of my signal wires, just as you described in your latest video. To prevent any induced voltages from they shunt current, I was thinking of laying the signal wires along the shunt wire, and carefully taping the signal wire such that it very closely follows the path of the shunt wire. In other words, I am trying to make the wires the exact same length, and keep the signal wires perfectly parallel to the shunt everywhere along its length. Is this the correct approach to minimize any induced voltages in my signal wiring? Thanks again !

Professors have been wrong before. I agree with you. The voltages induced in the sense lines have to be taken into account as well, and they explain everything. Otherwise we would have witnessed magic happening. Well done.

Jonas Otter

Excellent argument, great attention to detail -- I hope you get a response!

Travis Snoozy

We learn every day :)

While I have very little understanding of what you have said (I’m a political science major), I appreciate the way you said it. It is a great example of how to challenge a professor… with respect, data, and a clear challenge. I really hope he responds. Again, I only understood very little, but it is video’s like this that make me want to understand more. Thank you for that and keep them coming.

Mehdi, you just don't want to embarrase the poor doctor XD you're a humble guy, but you pretty much know you're right :) Now that I watch your video, it looks so obvious it's "just" bad probing, just look the positioning oficial those probes! I'm glad I watch this video, as an eletronic engineer myself, I wouldn't reach the same conclusion as you, not at first glance at least.

Steven da Silva

I became aware very early that selection of probe points and ground connections are very important. That simply selecting the wrong point to ground your 'scope probe can have you chasing ghost signals that aren't really there. I had the same thoughts as you when I saw that sequence, and the same possible solution. Place the probe wires so that they have as little induced voltage as possible.

Steven J Greenfield

These kind of videos are the pure reason why I continue as a Patron! Thank you for a great video and clear explanation of your opinion. I want to know what kind of answers you get.

I agree with you. In fact I noticed pretty much immediately that the probes were also creating loops, and then realized we came to the same conclusion. I don't understand how that professor could get fooled like that. Unless he was actually trying to teach a lesson about probing. If you wanted to do this properly the probes would need to be perpendicular to the loop in the 3rd dimension. And even then I'm not even sure.

Felix

I had seen that lecture before and taken it at face value. Thank you for encouraging us to question everything. The only question I would have relates to shielding the sense wires. Can this be done to a point that the fields can not affect the readings. I understand using twisted pair to try to cancel interference out but I don’t feel that is good enough when challenging something at this level.

Alex Taylor

I actually understood that! And I think you are right.

Science aside, What is the deal with his Licorice Allsorts necklace? LOL

Why do not you answer my publication?

The thing is that for the frequencies I'm working with (is you look at the scope pulses I get at mS speed) the inductance of the wire would be very small. Yes in mega Hz or higher frequency the inductance would be much more.

ElectroBOOM [Mehdi Sadaghdar]

I thought so too. But considering Dr. Lewin made 4 videos insisting he is right, I thought otherwise. But again he seems too good to make a simple oversight like this, unless I'm a total idiot!

ElectroBOOM [Mehdi Sadaghdar]

thanks!

ElectroBOOM [Mehdi Sadaghdar]

on your first point, according to lens law the generated current contradicts the source, meaning it generates fields to contradict the original fields so its field is reverse of the source. So when you do the right hand rule, you have to consider that.

ElectroBOOM [Mehdi Sadaghdar]

BSEE/almost MSECE here with MS primary areas of power electronics, automatic control, and embedded systems. First, are you intending to draw the magnetic field outwards from the board? By curl direction and right-hand rule convention, shouldn't you currents traveling counter-clockwise? Secondly, what stands out to me most is that the direction of the measurement probes either completes or cancels the induction loops just like twisted pairs, similar to the probing conclusion you draw. This was notable in your description of the voltages as "reduced over the sense line" around 10:30. Third, if I were to draw the circuit, I'd definitely model it as sourced from a transformer as you do around 8:50, just with the dot convention to convey directionality of the current with respect to that of the magnetic field (because I think I missed points on a test at one point, maybe ;-) ). Thanks for a good video!

Joyous Uwu

The probes acting like a partial transformer winding makes sense to me. I reckon to probe the two points would require a pair of foil shielded coaxial probes. The inner core wires would be protected like a Faraday cage, as the outer foil will carry any induced voltage and the oscilloscope should read the same no matter which way the shielded probe wires run.

Seán Byrne

Hi, Mehdi. Good explanation. I'm a fan of Dr. Lewin's lectures as well, but I'm pretty sure you have the right explanation here. KVL does hold in this case. If two different voltages can appear at the same time across the same two points, we'd have to rethink all of EE.

M. Eric Carr

Do you realize that you just intuitively taught a concept of engineering that, while initially seems simple, requires considerable understanding to convey? Not only does your analysis "hold water" but it does so without boring the audience or without prior engineering level knowledge of electronics. You are a fantastic teacher!

I only have a bachelor's degree in computer engineering, not even electronics and not either a master, but I also thought about the sense wires affecting the reading from the first second I saw the experiment. Something doesn't match here for me, why a smart guy like Dr. Lewin would not tough about it? It have to be some kind of trick...

elias

I barely understood any of that but that's why I'm here, to maybe learn a bit? Oh....and to watch you shock the hell out of yourself. Who said education isnt entertaining? :)

I would assume that this is some professorial trick to get you to learn that the stuff you're measuring with also becomes part of the circuit, and that he would explain what you've explained in a later lecture (hopefully the students would have worked this out for themselves). Otherwise it looks like an example of how the smart can outsmart themselves.

You say at the beginning that "we agree that the inductance of the wire is negligible". Well, this device is basically an antenna (or the secondary of a transformer, which is more or less the same idea). The inductive effects are not actually negligible: as you show, you get a voltage on your wire when the current is changing!

It does not matter how clever you are or how many things you know, there is always a blind spot! It's clear to me I am among people far more educated and worthy than I, but I do for the most part find this field fascinating and being a retired electromechanical engineer it makes me miss the old days so I try and keep my hand in and finger on the pulse with the help of guys like you.

would it be possible for you to start a lecture series about circuit analysis? i believe you are the best teacher i know. theory combined with actual applications/experimentations is the best way to learn. i haven't been bored in any of you videos. you're so good! more voltage times current to you sir!

Sure you are right. A measurement setup should be build up in a way that it is possibly only affected by the ends of the connectors. If you wrap the lines around the circuit, you have to "measure" unwanted inductions and the measurement itself is useless. The professor should have known that...

I believe you are right, definitely a "transformer effect" resulting in bad probing. PLease update us if you get a reply.

Richard Boyce

Shield the sense wires, anr/or try probing with straight wires down the middle and find out? Makes sense they would become a transformer too. Twisting pairs to even out interference only work if they're twisted all the way through the fields.

Please keep us updated on the progress and hopefully he will contact you back.

Robs Repairs

I thought about the sense wires being influenced by the field the moment I saw them. You're absolutely right. But I have only a master's degree too, so it won't be much support that I agree with you. But I don't make Youtube videos. That may be on plus.

I think you are right, sense wire is a part of the "transformer"

Thomas Eriksen

Well all I can say is if my masters in bumblefuckery from the online institute of AvE is right, that sounds like a skookum choocher. 😁

Wow, I wonder what WL would say about it.