Walter Lewin 8.02 Lect 16 Super Demo (Correction)

[TheOCDengineer]

Dear All

I recently watched Walter Lewins lecture on Faraday's law. In this lecture at the end he does a super demo in which he measures the same point with 2 different voltmeters and gets 2 results that are different.

Video Lecture here


He has a pretty large youtube channel and I posted a comment under this video

https://youtu.be/wz_GqO-Urk4

Here is the dialog
 
TheEngineer
9 hours ago
The math and theory hold up but in the physical set up how did you determine which oscilloscope measured which path? If they are measuring the same exact spot would it be defined? If I added a third oscilloscope what would it measure? Does the orientation of the probes have any effect? Do you have a picture of the set up, with the placement of the probes? Thanks for your time.?
REPLY



Lectures by Walter Lewin. They will make you ? Physics.
Lectures by Walter Lewin. They will make you ? Physics.
8 hours ago
watch this video. The current through each voltmeter determines the reading of the voltmeter and those currents vary in my case by a factor of about 10. watch this video   watch this video?
REPLY



TheEngineer
TheEngineer
8 hours ago
The video does not answer which meter is V1 and V2. If I were to repeat the experiment is it possible that they would switch places? If I add another meter could it measure something completely different, or effect the measurements of the others??
REPLY



Lectures by Walter Lewin. They will make you ? Physics.
Lectures by Walter Lewin. They will make you ? Physics.
7 hours ago
>>>The video does not answer which meter is V1 and V2>>>
you are mistaken. Watch the video again. V1  and V2 are clearly indicated. I1 goes through V1 and I2 goes through V2 and my calculations show that the ration I2/I1 is about 9 and that the currents are in opposite direction; I1 goes through V1 in the -- to ++ and I2 goes in ++ to -- direction of V2. Thus the ratio of the magnitude of V2/V1 is 9 and the polarities are REVERSED This is my last response. Please learn to watch!?



I found him to be quite rude "Please Learn to watch" and he also did not answer my question.


So here is my question to everyone.

1. Has anyone made a replica of his set up exactly? Was anyone able to replicate his results?

2. What would you measure with only one voltmeter?

3. Assuming the voltmeters are exactly the same how would you determine which voltmeter would be V1 and which would be V2.

4. What would happen if I add a third meter? What would it measure?

Please assume the theory is correct and that he is violating KVL. Also assume he is measuring the same exact point.


Thanks

[IanB]

When you see something you do not understand, it means you did not understand what you saw.

It is for you, as the one seeking knowledge, to go away and work on the problem, and think about it, and watch the video over and over, and think about it some more, until you get it.

I think the question you asked means you really didn't understand the experiment, and therefore you asked the wrong question. If you ask the wrong question, how can it be answered to your satisfaction?

[IanB]

Let me give you a hint. If you set up two voltmeters to measure the same thing, they both will show the same reading.

In the experiment demonstrated, the two voltmeters show different readings, therefore they are measuring different quantities. If you configured both meters to measure the V1 quantity they both would show the (same) V1 measurement. Equally, if you set them both up to measure the V2 quantity they both would show the (same) V2 measurement.

So now, what your hypothetical third meter would measure would depend on how you configured it. If you configured it to measure V1 it would show V1. If you configured it to measure V2 it would show V2. If you configured it some other way it would show some other (unspecified) voltage.

[rfeecs]

Looks like a double post?  So I'll move my comments here.

He is trying to amaze and impress his students by confusing them.

At the end he claims that he connects two oscilloscope probes to exactly the same point and measures different voltages.  "Lo Lehman" in the comments calls him out on this in a very long comment and points out that voltage is induced in the connecting wires, and Dr. Lewin responds:

>>>>For the record, it is impossible for two identical voltmeters with the same settings and hooked up to the same 2 points of measurement to yield different results at the same point in time.>>>
                                               YOU ARE DEAD WRONG!

The demo I show has become a classic - it is shown all over the world. Two identical  voltmeters connected to the same 2 points in a circuit show very different values and the polarities are reversed. That's the power of Faraday’s Law which runs our entire economy. It's amazing how many people have a religious INCORRECT believe that Kirchhoff's loop rule always works. If that were true my two voltmeters should read IDENTICAL values with the same polarity but that is not the case. K’s loop rule is only valid if d(phi)/dt=0 but that is not the case here. The induced emf (Faraday!) is in the closed loop D-R2-A-R1-D. The direction of the emf at time t depends on the sign of dphi/dt at time t. That determines the direction of the E-field in the closed loop. If the the E-field is clockwise then the E-field in R2 is in the direction from D to A and the E-field in R1 is in the direction from A to D. Thus the closed loop integral of E dot dl in the loop D-R2-A-R1-D is NOT zero. The integral of E dot dl in going from D to A depends on the path; one path is through R2 and the other path is through R1. Thus the 2 voltmeters should read different values and reversed polarities. If you don’t get it, you are not alone. Occasionally I get comments from people including science teachers who want to argue with me about this demo  - they clearly never FULLY understood Faraday's Law. I never argue with anyone about this as this is too basic.
                                      THIS IS THEREFORE MY ONLY RESPONSE. 

By the way, in class I made the length of the wire at D zero and also of the wire at A.  Thus there was only one point D and one point A and both voltmeters are attached to D and A. However, it is not at all a necessary condition that the wires at A and at D have zero length as the resistance of copper wire is much much lower than the resistances of R1 and R2. Thus my diagram on the blackboard is fine.

Here's his supplement to the lecture:
http://web.mit.edu/8.02/www/Spring02/lectures/lecsup3-15.pdf

Clearly in the experiment there must be wires between the two resistors to form the loop.  So the meters are not actually connected to the same physical points as he seems to insist in his comments.

[IanB]

Clearly in the experiment there must be wires between the two resistors to form the loop.  So the meters are not actually connected to the same physical points as he seems to insist in his comments.

I think he is just saying that points A and D are wired as star connections, and therefore four wires meet together at a single point.

[rfeecs]

Clearly in the experiment there must be wires between the two resistors to form the loop.  So the meters are not actually connected to the same physical points as he seems to insist in his comments.

I think he is just saying that points A and D are wired as star connections, and therefore four wires meet together at a single point.

Yes, then he is saying that two meters connected to the same points A and D measure different voltages.  Absolute bullshit.

[TheOCDengineer]

So I have been thinking about this problem the last couple of days and I think I have an idea to the confusion. Normally as engineers we think of the schematic as the components and their connections. Thus it holds no relationship of physical orientation. However in Mr. Lewin diagram it is also showing the physical relationship.
I found this interesting video in which at the end he moved the pig tail loops from one side or another.

https://youtu.be/-AjdUuq8JNY

His explanation is that there is a emf generated in the loop. But if we assume that there is no EMF generated in the loop ( most likely not the case) then this may be related to path selection. The problem that I have with Mr. Lewins demo is that meters and leads orientation aside there would be nothing to influence which meter reads what. I need to keep thinking about this.

[IanB]

What Walter Lewin's demo shows is that the measuring instrument itself can be part of the circuit, in particular the wires that go from the test point to the instrument.

This is actually not at all surprising, as it is commonly known that attaching a measuring instrument can affect the response of a circuit, and that the test leads can pick up signals that you are not intending to measure. In particular with oscilloscopes you have to be very careful to avoid picking up stray signals in the test probe, especially at higher frequencies. That is why the loop area of the probe tip and ground connection at the test point needs to be kept as small as possible.

All Walter Lewin has done is to take this widely known aspect of measurement and dress it up to look like a party trick.

[TheOCDengineer]

Maybe, but if you look closely to the lead the area is not large when in either the left most or right most. That is why it is very confusing.

[IanB]

The problem that I have with Mr. Lewins demo is that meters and leads orientation aside there would be nothing to influence which meter reads what.

The lead orientation has absolutely everything to do with the outcome though. Consider that if a changing magnetic flux intercepts the test lead it will induce a current in that lead, and therefore the position of the test lead in space affects how and in what direction it is exposed to a changing magnetic field.

[IanB]

Maybe, but if you look closely to the lead the area is not large when in either the left most or right most. That is why it is very confusing.

But it must be enough. The test leads have to go all the way to the oscilloscopes, and the oscilloscopes are far away from the experiment. So you have a large loop area.

[TheOCDengineer]

Not really the area is actually really small. The pigtail is on top of the resistor and the rest of the probe is a coax and thus no outside magnetic field will have much influence. That's why it is so interesting, and makes me think that it is path selection rather than induced emf.

[rfeecs]

So I have been thinking about this problem the last couple of days and I think I have an idea to the confusion. Normally as engineers we think of the schematic as the components and their connections. Thus it holds no relationship of physical orientation. However in Mr. Lewin diagram it is also showing the physical relationship.
I found this interesting video in which at the end he moved the pig tail loops from one side or another.

https://youtu.be/-AjdUuq8JNY

His explanation is that there is a emf generated in the loop. But if we assume that there is no EMF generated in the loop ( most likely not the case) then this may be related to path selection. The problem that I have with Mr. Lewins demo is that meters and leads orientation aside there would be nothing to influence which meter reads what. I need to keep thinking about this.

That video gives a very clear and complete explanation.  It shows very clearly the voltage induced across the wires connecting the resistors.

Dr. Lewin makes a very clearly incorrect statement in his lecture.  At around 40:00, he claims that going from D to A by two different paths gives two different voltages.  This is clearly wrong.

It is a shame that his students must have walked out of that lecture scratching their head wondering what was going on, instead of getting a clear explanation for this simple concept.

[chris_leyson]

What he is implies in his lecture is shown in the first schematic where the inductors represent the wire or the single turn secondary and V1 and V2 are the voltmeters. If this is the case then both voltmeters would read the same.I couldn't clearly see his experimental setup in the video but what he must be doing is shown in the second schematic.

That second video was a very good explanation of what is really going on. It seems Walter Lewin was implying one thing but measuring another, it's no wonder people got confused. Also a very good demonstration of what happens if you've got stray magnetic fields and are using scope probe ground leads.

Lectures by Walter Lewin. They will make you ? Physics. More like ? Walter Lewin

[rfeecs]

He's still at it:


First he says Kirchoff's law is violated.  So what?  It is well known that Kirchoff's law has limitations.

Then he goes on with the same old argument.

This is one of those "paradox" problems like the "Capacitor paradox", where you take an ideal lumped circuit and get a strange result.  You get strange results because the ideal circuit requires things like zero resistance and infinite current.

In this video he says you have superconducting wires with no resistance, and you can't have E-fields in them or you get infinite current.  Same problem.

In any real circuit, you have electrical length in the wires, and the resistors too for that matter.

There is no physical experiment that can reproduce what he is proposing:  zero length wires and resistors; zero ohm zero inductance wires; zero inductance resistors; yet there is a finite loop with a magnetic field.  It is a totally theoretical toy problem.

You would think he would want to explain why his statement makes no physical sense.  How do the meters know where they are placed?  It's nuts.  I'm wondering if he is going a little nuts as well.

[IanB]

The argument that there can't be a large voltage between the ends of a low resistance wire because of Ohm's law doesn't hold water if you consider that an ideal voltage source in theoretical models has zero impedance, yet still represents an arbitrary potential difference between its terminals.

What's more, this theoretical model would allow an infinite current to flow if you shorted its terminals together.

In the real world you can replace zero impedance with low impedance, and replace infinite current with large current, and everything still holds together.

[rfeecs]

Where he makes his mistake in the "Is Kirchhoff's Loop Rule for the Birds?" video is he says that the wires have no resistance, so they can have no voltage across them.  This is wrong.

If he uses his favorite Faraday's law, and integrates from one end of the wire to the other, he will get an instantaneous voltage across it.  For example, from the previous video, +0.5V across the top wire from end to end, then -.9V across one resistor, 0.5V across the bottom wire, and -.1V across the other resistor.  He is wrong when he says that there is zero volts across the wire.

Any electrical engineer knows that even an ideal transformer that has zero resistance in the secondary turn can generate a voltage across it.  Same thing.

[KE5FX]

Didn't this guy get fired for hitting on freshman chicks, or something? 

[rfeecs]

Didn't this guy get fired for hitting on freshman chicks, or something? 

Yes.  He was already pretty much retired.  They pulled down his online courses:
http://news.mit.edu/2014/lewin-courses-removed-1208

[orin]

FWIW, Maxim agree with Walter Lewin:

https://www.maximintegrated.com/en/app-notes/index.mvp/id/2238

Also, from Kirk T. McDonald at Princeton:

http://www.physics.princeton.edu/~mcdonald/examples/lewin.pdf

"...which is particularly surprising in the case that points a and c are the same, and points b and d are the same" (i.e. a star wiring configuration for the resistors and voltmeters).  Though he does use Kirchhoff with L*dI/dt + M*dIp/dt for the analysis, he effectively goes back to the the Faraday's law form equation (2).  Kirchhoff and Faraday do give the same equations - if you account for _all_ inductances in the Kirchhoff analysis, but Dr Lewin considers it a cheat.

Section 2.2 may or may not help explain the confusion of Voltage vs EMF.

I do not consider the Mabilde experiment to be equivalent.  Lewin has the secondary loop approximately half way up and around a solenoid.  Mabilde puts his secondary loop on top of his solenoid and I'd tend to agree with "Icy Stranger"'s comments on the magnetic field lines.

What one should take from this is that all bets are off when there is a time varying magnetic field inside a loop of your circuit.  Indeed, the very first thing that was explained on MIT's 6.002 "Circuits and Electronics" course is the lumped circuit abstraction where one of the assumptions is that there are no time varying magnetic fields outside the circuit elements.  An old version is here:



BTW, the real fun starts in this lecture 35:30 in...

[orin]

Didn't this guy get fired for hitting on freshman chicks, or something? 

Yes.  He was already pretty much retired.  They pulled down his online courses:
http://news.mit.edu/2014/lewin-courses-removed-1208

It was to do with some interaction with an online student.  Having gone through "Anti sexual harassment awareness training"* on a previous job, I'd guess he'd been a bad boy before and had been warned.

To be honest, and in my opinion, as I recall, some of his behavior in some of the online lectures I've seen was over the line... but I can't verify that it was on an edX course or opencourseware where I saw it as they took them down.

None of the above is relevant to the Kirchhoff's vs Faraday's law issue where Faraday's Law always holds.  Kirchhoff's law works at DC where there are no time varying magnetic fields and can be made to work at AC if you account for all inductances in a loop as L*dI/dt (or M*dI/dt for mutual inductances).

Orin.

*That really was the title and my take from it was: "You mean I can get away with that!?".  But there again, I wasn't in any kind of supervisory position and you have to be much more careful in a supervisory position.

[orin]

Yes, I find this subject fascinating.

For the record, another paper on the matter:

http://www.phy.pmf.unizg.hr/~npoljak/files/clanci/guias.pdf

[rfeecs]

All three of these papers explicitly point out that the voltage measured on a voltmeter in the presence of a changing magnetic field depends on the path that the connecting wires to the voltmeter take.  Move the wires and you will get a different measurement.

Apparently this is what Dr. Lewin means when he says that tracing the loop around from A to D gives you two different voltages depending if you go around from the right or from the left.  In the later response videos, he doesn't even show the meters at all.  At best, his description is unclear.  At worst he is being deceptive to make his magic trick look more puzzling and amazing.

Sometimes he shows the direction of the magnetic field inside the main loop.  But he never mentions that there is a field in the opposite direction outside the solenoid, and this field passes through the measurement loops.  This field causes the difference in measurements between the right and left meter: 



To most of us, the EMF induced in the wires of the voltmeter represent a measurement error, and not an accurate measurement of the voltage from A to D.  (How many times has Dave said "that's not a real signal, the oscilloscope is just picking up noise from the equipment and lights in the lab?)  It's a matter of definition, as pointed out in some of the papers.

[rfeecs]

Yes, I find this subject fascinating.

For the record, another paper on the matter:

http://www.phy.pmf.unizg.hr/~npoljak/files/clanci/guias.pdf

Wow.  I read over this paper more carefully and I think I finally get it.  Mind blown.  You don't even need the opposite field outside the solenoid.  The outer loop containing the meters and the main inner loop both surround the same flux.  So the induced EMFs cancel out.

Thanks.  Very interesting.

[TheOCDengineer]

Yes, I find this subject fascinating.

For the record, another paper on the matter:

http://www.phy.pmf.unizg.hr/~npoljak/files/clanci/guias.pdf

Wow.  I read over this paper more carefully and I think I finally get it.  Mind blown.  You don't even need the opposite field outside the solenoid.  The outer loop containing the meters and the main inner loop both surround the same flux.  So the induced EMFs cancel out.

Thanks.  Very interesting.

I think that paper explains it very well. Still a little confused on how the path selection works physically. I wonder what the relationship is for lead position. Equation 3 suggests that for one loop path there should be a positive and a negative and experimentally we proved it.