PopularFX
Home Help Search Login Register
Welcome,Guest. Please login or register.
2019-11-12, 23:55:12
News: Registration with the OUR forum is now by invitation only.

Pages: 1 [2] 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 ... 25
Author Topic: Professor Walter Lewin's Non-conservative Fields Experiment  (Read 252423 times)
Hero Member
*****

Posts: 1146
KVL does not hold for circuits where there's magnetic induction into the circuit loops itself, becuase this creates an EMF that is not accounted for on the circuit diagram as an explicit voltage source, and so when you sum up voltage drops around the circuit it does not add up to zero.    

Assume the current flow around the loop is 1mA, and the loop has two resistors, 100 ohms and 900 ohms.   The voltage drop due to the current is .1mV and .9mV,   and so as you sum up the voltage drops ( - 0.1 - 0.9 = - 1V) you do not obtain zero.

This is not some mysterious circuit, just an example to get the student to realize there are ASSUMPTIONS to every equation they encounter in their books.

EM

P.S.  The professor says it's a non-conservative because this example violates the definition of conservative fields, meaning that as you add up voltage drops along any path that returns to the starting point,  or it's closed,  you should sum up to zero, and for this example it's obviously not the case.    But I would not say the magnetic field is non-conservative, just the voltage around the loop in this specific example.

This example is a bit more involved then the professor illustrates, and I have seen the full explanations.  for example it depends on the loop the probe wires form and what angle they are respective to the plane of the paper, because we have flux changes here so we need to think 3D.   for example if you place the probes plane verticaly in the center, the flux from the solenoid will not flow through your probe plane, so what voltage will it measure?   0.1V  or 0.9 V?    This always baffles my mind, but there is an answer, can you figure it out?   :) 
« Last Edit: 2011-02-24, 15:52:30 by EMdevices »
   
Hero Member
*****

Posts: 1083
@Poynt99
Quote
The voltage law says that the sum of voltages around every closed loop in the circuit must equal zero. A closed loop has the obvious definition: Starting at a node, trace a path through the circuit that returns you to the origin node. KVL expresses the fact that electric fields are conservative: The total work performed in moving a test charge around a closed path is zero. The KVL equation for our circuit is

I would have to disagree on several points, first the circuit you have depicted does not resemble the circuit in question in any way as you have incorrectly placed a single fixed voltage source in the circuit. Why would you think you can completely change a circuit to justify the action of another that is nothing like it?, that's crazy.

Quote
In writing KVL equations, we follow the convention that an element's voltage enters with a plus sign if traversing the closed path, we go from the positive to the negative of the voltage's definition.
I would disagree again, I have used electron flow notation exclusively for the last 15 years ever since I learned that conventional flow notation (+ to -) is in fact incorrect. To this day I still find it bizarre that so many talk of proof, truth and doing things correctly yet still use an antiquated and incorrect form of flow notation.

@Emdevices
I would agree this is too easy --- maybe that is the problem. ;D
Regards
AC


---------------------------
"Great minds discuss ideas; average minds discuss events; small minds discuss people." - Eleanor Roosevelt.

Be careful when you blindly follow the Masses... sometimes the "M" is silent.
   
Group: Guest
I would say that KVL applies.  There is no hocus-pocus going on with some kind of exception.  I haven't watched the Walter Lewin clip but I will just state a simple example based on what EMdevices said.

Quote
Assume the current flow around the loop is 1mA, and the loop has two resistors, 100 ohms and 900 ohms.   The voltage drop due to the current is .1mV and .9mV,   and so as you sum up the voltage drops ( - 0.1 - 0.9 = - 1V) you do not obtain zero.

The net voltage drop around the loop is zero because the interconnect wires are not just dumb equipotential nodes anymore, they become voltage sources.  If there is a total voltage drop of 1 volt due to the two resistors then there has to be a total voltage increase of 1 volt between the two interconnect wires which are acting as EMF sources.

For example, use drops for the resistors and increases for the EMF sources:

-0.1 + 0.1 - 0.9 + 0.9 = 0

or

-0.1 + 0.5 - 0.9 + 0.5 = 0

or

-0.1 + 0.6 - 0.9 + 0.4 = 0

As long as EMF1 + EMF2 = 1 volt then 1 mA of current will flow through the loop.  The two EMFs only exist if there is changing flux going through the loop.

MileHigh
   
Hero Member
*****

Posts: 1146
MH, I would suggest watching the videos first.   C.C
   

Group: Elite
Hero Member
******

Posts: 1859
Folks,

First I'll say, 'please forgive me, in advance'.

Throughout history we've lived with incorrect rules. When we learn more or use better judgment the rules are corrected or we discover new ones.

What happens when we misapply the rules or just stop teaching the details? Even worse, what happens when we teach these details by providing false examples?

I know KVL has exceptions. This was part of my education. It was in the course material and worked on the bench. (On what might be a funny note: I was also taught that magnetic fields are neither conservative(non-dissipative) nor non-conservative(dissipative).  ??? )

I'm just not so sure Lewin's example was truly an example of this case. If it is then the KVL shouldn't be applied and the example is of how laws can be misapplied.

From the little bit of detail we have.... .99 could be completely correct by considering the loop inductance to cover a more common solution.

... more thoughts on other similar issues but maybe another day and thread  :-X


---------------------------
"As far as the laws of mathematics refer to reality, they are not certain; as far as they are certain, they do not refer to reality." - Einstein

"What we observe is not nature itself, but nature exposed to our method of questioning." - Werner Heisenberg
   

Group: Administrator
Hero Member
*****

Posts: 2905
It's not as complicated as it may seem...
EM, AC, KVL does indeed hold.


First, lets look at a few definitions of what KVL is:


1) Electrical Engineering Dictionary - CRC Press, 2000

Kirchoff’s voltage law (KVL) a fundamental
law of electricity that states that the
sum of the voltage drops and rises in a closed
loop must equal 0.


2) Fundamentals of Electric Circuits - 3ed. - p.37

Kirchhoff’s voltage law (KVL) states that the algebraic sum of all voltages
around a closed path(or loop) is zero.

or alternatively;

Sum of voltage drops = Sum of voltage rises


3) Modern Dictionary of Electronics_7ed - p.408

Kirchhoff’s laws- No.2:

The algebraic sum of the voltage drops in any closed path in a circuit is equal to the algebraic sum
of the electromotive forces in that path.


4) The Resource Handbook of Electronics - CRC Press, 2001 - sect. 6.2

Kirchoff’s voltage law (KVL). The algebraic sum of instantaneous voltages
around a closed loop is zero. **

** In my opinion, this is the best one of the four offered here.



The first observation one should make by examining what professor Lewin has presented along with his objection, is that his explanation of KVL and his use of the law, is incorrect. Applying KVL involves using voltage drops or losses across the circuit elements, but always using the same direction or polarity. Lewin has arbitrarily flipped the voltage meter across one resistor and this accounts for his opposite polarity problem. In fact, the voltage, if assumed to be correctly measured across the resistors, will sum to either +1V or -1V, not -0.1V and +0.9V. That is his error number 1. How could a prestigious college professor mess up something as basic as KVL?

MH, you are on the ball my friend.  :)  O0

The second observation one should make, and this is how Lewin skewed the experiment, is that there are two (or four, depending on the position of points A and D) voltage drops that are not accounted for. The interconnecting wire between the resistors must in fact be included in this KVL computation, as per the above definitions, but they were not, and no one except MH has recognized this.

The sum of the instantaneous voltages around this closed loop do indeed equal zero volts!

But you won't get this result if you do not measure all the voltage drops!

See the attached scope shots of all four voltages in the first diagram of my simulation. The sum is 0V.

KVL holds regardless if the emf is induced by induction, or if impressed by a source or emf such as a battery.

@EM, this experiment has nothing to do with induction in the measurement probe cabling or wires. That is a misnomer I'm afraid. @AC, the conventional vs. electron current debate is irrelevant in regards to the results of this experiment. The results are equal either way.

.99


« Last Edit: 2011-02-25, 04:08:34 by poynt99 »
   
Group: Guest
I'll bank some Brownie Points in case I make seven years worth of mistakes!  lol
   
Hero Member
*****

Posts: 1083
@poytn99
I would disagree, first you are not using a real world test only a simulator, second you have incorrectly used a coil acting as a point source--why not justify your point by just simulating a battery as the source and be done with it?. Changing all the parameters to get the results you like is not what I would call good science and when you do a real experiment with the correct parameters then we can talk about facts.

Quote
@AC, the conventional vs. electron current debate is irrelevant in regards to the results of this experiment. The results are equal either way.
I didn't imply that it applied here specifically I just wanted to clarify the fact that most everyone is still doing something incorrectly by repeating a mistake that was corrected decades ago, electron flow notation is correct. So why is most everyone still repeating this mistake?, because people are creatures of habit and they fear change almost as much as death. To be honest I find it kind of comical that so many are ready to defend their normalacy to the bitter end, to defend the past and present no matter how silly or pointless it is because we consider it normal. Then we speak of change but do nothing to change ourselves, we demand change but shun anyone who acts differently, who speaks differently or promotes technology which is different. I think anyone who is brave enough to take that one big step backwards and take a look at the bigger picture may find as I have that there is nothing normal about what we call normal.
There is a good book to justify this, Gustav LeBon--The Crowd: A Study of the Popular Mind, I think old Gustav had it right that if you convince people (or sheep) that something no matter how @#$%up or bizarre is "normal" they will not only do it but also defend it to the bitter end. It is said that Adolf Hitler kept a copy of this book on his nightstand, you see the problem with normal is that you have to step outside the box in order to recognize it, normal people don't have a god damn clue what "normal" is.
Regards
AC


---------------------------
"Great minds discuss ideas; average minds discuss events; small minds discuss people." - Eleanor Roosevelt.

Be careful when you blindly follow the Masses... sometimes the "M" is silent.
   
Group: Guest
I'm not sure if he knows what he is claiming. Have a look at his lecture supplement (attachment). His first diagram and assumption that all the currents are clockwise can not be correct.



.99

Hi Poynt99,

The solenoid has an expanding field during the time frame of the test.  That expanding field and its associated flux moves outward from the center of each winding radially in such a way so as to induce a specific voltage in the circuit everywhere simultaneously. It is the change in magnetic flux that does this.  The field is round and radial, the circuit is squared, so there are points where the flux does not intersect the conductors at a true 90° and there are places where it does. But because the coil is centered in the middle of the circuit, it induces a current flow in the same direction everywhere, and in this case the orientation of the flux as it passes through the conductors the current flow (conventional opposite electron direction) is clockwise everywhere.

The voltage around a given looped conductor is the the closed loop integral of E·dl. I'm sure the conductor has some resistance, I think the professor used something like a 4AWG wire at about 4" long between the resistors IIRC (please check this to be sure), so you could factor that in and using this calculator you could even get some measure of inductance to plug in there too as tiny as that would be. But the end result is simply that the coil puts energy into the circuit everywhere at the same time, and it all flows in the same direction.

A changing magnetic field is not conservative. Thank goodness that is the case because all of our motors and generators depend on that fact. If they were conservative we would never be able to get any transformations from them. A uniform static magnetic field is conservative though. That's why static magnetic fields cannot be used in transformers. It has to change in order to get energy to transform.

Assuming no voltage is being induced in the resistors themselves (even though in reality we know that some is in this experiment) we can treat the two conductors as voltage sources. So there are two voltage sources that have equal voltage induced in them because they are identical, and the fields react with them identically. It just so happens that each represents 1/2 Volt end to end. So there is 1/2V across each resistor. Relative to point A, one end of the wire is -0.25V and the other end is +0.25V, same with D -0.25V and +0.25V. And that is what the professor was trying to impress on the students, is that near zero impedance can still exhibit a voltage drop across it when it is acting as a power source - it is a path dependent exercise. And it is very non intuitive because it causes you to think there is an inversion in your measurements, like "somebody must have made a mistake". But there is no mistake, it just is what it is - two instantaneous power sources each feeding two resistors of different values in a loop.

So the total series voltage in the circuit then is 1V. Each resistor has +0.25V and -0.25V across it, or 0.5V. So using this measurement method we can now say that the voltage drop from A to D is 0V. Therefore, I can use either A or D to be my reference to any of the four nodes at the resistors and I will get the same reading for each.

How is it possible then, that we measure -0.1V and +0.9V ? If KVL holds, then KCL breaks and if KCL holds the KVL breaks. Or if we somehow fix those two to hold, then we must throw out our laws of induction because in this case the only way we can get the circuit to agree, is if we do the math with path dependence, aka Faraday's law which allows that to happen. Or are we to fool ourselves into thinking that somehow the wires make an adjustment to be nonlinear?

You have only probably watched the two videos that whet your appetite for the explanation. But the whole lesson is nearly an hour long (51 minutes) and fully explains it mathematically by building up to how the current and voltage are induced. http://videolectures.net/mit802s02_lewin_lec16/ Step by Step derivation PDF http://videolectures.net/site/normal_dl/tag=28248/non-conservative_fields-do_not_trust_your_intuition.pdf

I love the part where he says "Their brains couldn't handle it"

Essentially we have an EMF of 1V in the loop and the total loop has 1000 Ohms so the current in the loop is 1/1000A. 1mA * 900 Ohms is 0.9V, and -1mA * 100 Ohms is -0.1V. What is the true voltage across each voltage source (wires) and why is it path dependent?

 :)
   
Group: Guest
Assuming no voltage is being induced in the resistors themselves (even though in reality we know that some is in this experiment) we can treat the two conductors as voltage sources. So there are two voltage sources that have equal voltage induced in them because they are identical, and the fields react with them identically. It just so happens that each represents 1/2 Volt end to end. So there is 1/2V across each resistor. Relative to point A, one end of the wire is -0.25V and the other end is +0.25V, same with D -0.25V and +0.25V. And that is what the professor was trying to impress on the students, is that near zero impedance can still exhibit a voltage drop across it when it is acting as a power source - it is a path dependent exercise. And it is very non intuitive because it causes you to think there is an inversion in your measurements, like "somebody must have made a mistake". But there is no mistake, it just is what it is - two instantaneous power sources each feeding two resistors of different values in a loop.

You should rethink this one out Harvey because you are over-complicating things.  You are implying that the centers of the wires are at a kind of ground reference for the loop when that's not the case.  You are implying that there is a constraint where the voltage across each resistor is 0.5 volts and that's not the case.  The entire loop is floating relative to the outside world and it is a reference unto itself.

Just go back to basics and do the walk around closed loop integral of E·dl.

Look at my one of my examples:  
Quote
-0.1 + 0.5 - 0.9 + 0.5 = 0

You start your walk and you drop 0.1 volts.  At this point along the walk, it's a free-for-all, the EMF can go anywhere.  So you continue and find yourself up 0.5 volts and now you are sitting 0.4 volts above from where you started.  You continue the walk around the loop and you drop 0.9 volts.  Finally, you go up 0.5 volts and you connect the dots.  It's really very simple.

MileHigh
   

Group: Administrator
Hero Member
*****

Posts: 2905
It's not as complicated as it may seem...
@poytn99
I would disagree, first you are not using a real world test only a simulator, second you have incorrectly used a coil acting as a point source--why not justify your point by just simulating a battery as the source and be done with it?. Changing all the parameters to get the results you like is not what I would call good science and when you do a real experiment with the correct parameters then we can talk about facts.

I either don't get your points, or they are not valid.

First, a simulator can be successfully applied to many situations, this one included.

Second, why are you wanting to change the experiment by using a voltage source? I am simulating the experiment in such a way that it models what is actually happening, minus the negligible inductance that might occur in the measurement wires. But this can be included as well. However, professor Lewin said himself that the currents in the measurement wires is negligible compared to that through the inner loop where the resistors and interconnecting wires are.

The experiment involves the production of an emf within a loop, and this emf is produced via induction from a solenoid coil located inside that loop. It would seem that proposing to use a battery instead of what the true scenario involves, is unscientific.

The KVL holds.

.99
   

Group: Administrator
Hero Member
*****

Posts: 2905
It's not as complicated as it may seem...
Harvey,

I'm not really certain what your final point might be, but my posts stand.

I've shown that:

1) professor Lewin is using KVL incorrectly to sum his voltages; the total using ONLY the resistors should be 1V total, not 0.8V.

2) the professor has omitted two of the voltage measurements that clearly if accounted for, show that KVL holds for this case.

3) the two "missing" voltages cancel out the resistor voltages for a net loop voltage of 0V, as per KVL.

It really isn't any more complicated than this.

.99
   

Group: Administrator
Hero Member
*****

Posts: 2905
It's not as complicated as it may seem...
Harvey, AC, EM,

Your challenge, should you choose to accept it, is to successfully argue against the 3 points in the above post #36.

.99
   
Group: Guest
Hi Darren,

You need to place your source coil inside the loop, not outside - you have inverted one of the sources there. This is why you keep making the mistake with voltage sums thinking that the Professor made a mistake when it is you that have made the mistake. The current all flows in a single direction in the center loop, it is induced by an expanding magnetic field and has a specific direction in the wire. Using a single reference point (D for example) the polarities are inverse for each resistor as current flows into D from one, and out from D through the other.

The Professor shows the math for both Kirchhoff and Faraday. Kirchhoff's method results in 1V for loop (not 0.8 as you mistakenly claim). Faraday's method clearly shows 0V as it should be.

I am confident that once you get your polarities sorted out correctly in your mind and the induced values properly modeled in your simulator that you will gain a much greater appreciation for this demonstration and the professor's methods of teaching.

:)

It would also be good consider this point found on Wikipedia for KVL: http://en.wikipedia.org/wiki/Kirchhoff%27s_circuit_laws#Limitations

« Last Edit: 2011-02-27, 03:07:26 by Harvey »
   
Group: Guest
Harvey:

Let's assume that the current flow is clockwise and you travel clockwise around the loop.  Every time you encounter a resistance you observe a voltage drop.  So in this case you observe two voltage drops and they add together to give you a larger voltage drop.

The polarities are not inverse for each resistor as long as you continue in the same direction along the path (or loop).  I will try to watch the clip again.  I watched it several months ago and seem to recall being uncomfortable with what he was saying for this specific example.  Of course in general his series is excellent.

MileHigh
   
Group: Guest
Harvey:

Let's assume that the current flow is clockwise and you travel clockwise around the loop.  Every time you encounter a resistance you observe a voltage drop.  So in this case you observe two voltage drops and they add together to give you a larger voltage drop.

The polarities are not inverse for each resistor as long as you continue in the same direction along the path (or loop).  I will try to watch the clip again.  I watched it several months ago and seem to recall being uncomfortable with what he was saying for this specific example.  Of course in general his series is excellent.

MileHigh

You are correct. Thus the closed loop integral according to KVL results in 1 rather than 0. This is why it fails. In order to apply KVL you must introduce the non-existent power source(s). As the Professor states, that 1V is exactly the induced EMF.

You could theoretically use a flat piece of copper with two rectangular slots in it, and solder the resistors across those slots and get the same effect. As long as the change in flux induces an eddy current circular through the path of those two resistors of 1mA there will be those voltage drops. Of course the actual demonstration was reduced by a factor of 10, so that would permissible in any real world experiment. This matter can be very counter intuitive because in that situation the resistor is shorted out by copper all around it. But the magnetic field if properly centered would induce those currents circularly, and independent of the material on either side of the path.

At the end of the PDF by Lewin, he has a section called "Test Yourself" that was added to impress on the students what is happening. If you try to use Kirchhoff's math on that problem and compare those results to Faraday's method, you will see a large difference. The only way to make Kirchhoff's method work is to add inductors as Darren has, and include them in the equation and at a very fundamental level this is why the simulators perform as well as they do, because they do break things down to those levels.

But it is not necessary to do that if you use Faraday's law to calculate the EMF :)
   

Group: Administrator
Hero Member
*****

Posts: 2905
It's not as complicated as it may seem...
Harvey,

In SPICE, locations of the inductors is immaterial. All the inductors are coupled (like a transformer) with a chosen coupling factor K. I used 0.5 for K, but this is not critical.

It may appear that one inductor is inverted, but this is not the case. The two inductors representing the wire are connected in series-adding. See my new diagram (attached). The loop is a series loop, and therefore not only must the inductors be placed in series adding, but when you go around with your voltage meter, you must pay careful attention to the polarity of the leads. Professor Lewin has inverted the leads measuring R1, and that is why his voltage measurement is -0.1V rather than the correct +0.1V. Mine is in fact correct. Do you agree now?



It is not I that has implied the loop voltage for the resistors ONLY is 0.8V, it is in fact professor Lewin that has done so. Carefully read my post again. I tried to state that the correct loop voltage is 1V, not 0.8V (+0.9V - 0.1V = +0.8V) as Professor Lewin has implied with his stated measurements.

It can not be denied that professor Lewin failed to mention and measure the two other voltages present in the circuit. By definition all voltage gains and losses must be measured in the loop and summed to formulate the result. This is what is shown in the simulation, and this is what would be measured in real life as well. Not executing the law properly by omitting measurements, does not constitute proper grounds to invalidate that law.

It can also not be denied that professor Lewin has made an error with his KVL loop calculation by inverting one voltage meter.

KVL holds.  O0

.99
   
Group: Guest
Darren,

Where exactly in your diagram above do you place the common ground reference for your scope probes?

Professor Lewin was very precise in showing that the reference remained unchanged. It is you who are moving your references.

In addition, you are adding imaginary power sources to the circuit, these do not exist in his diagram or his demonstration and that is why they are excluded from the KVL calculations. As I stated in my last post, it is necessary to add them in to pretend that KVL has not failed when the math clearly shows that it has. Without the power sources (reality) you get 1V around the loop, not 0V. Faraday does not need those imaginary inductors that you have included falsely. Did you even think to measure the real inductance of those wires and model that?  :-\

Pretending that the good professor has made some mistake because you don't fully comprehend the magnitude of the demonstration is bad form. It is well known that KVL cannot be used where a changing magnetic field is concerned. Look again at the beginning of the first video when the battery is in the circuit. The professor clearly shows how KVL holds in this case because the battery is included in the calculations that net to zero. By removing the battery, KVL fails both in the math and in real life.

I have tried to help you learn this as tactfully as I can, but your I guess your "brain just cant handle it" and that tells us something about you ;)

EDIT to add:
You may wish to fully avail yourself of the discussion already recorded here which takes this to the most accurate levels:
http://www.physicsforums.com/showthread.php?t=405700
   
Group: Guest
Darren:

You have the voltage polarities for the inductors in your diagram backwards.

Harvey:

Quote
The only way to make Kirchhoff's method work is to add inductors as Darren has, and include them in the equation and at a very fundamental level this is why the simulators perform as well as they do, because they do break things down to those levels.

You made reference to the fact that the wires themselves have the property of inductance.  That's why Poynt added them to his circuit, it's just to model the physical reality on paper and make it a tangible entity that a layperson can recognize and PSpice can crunch on.  The wires are a power source because of the changing magnetic flux going through the loop that they form.

Anyway, it looks pretty straightforward to me.   Let's look at a few cases:

1.  Ordinary wire loop with the two series resistors:  Here you have current flow and the voltage drops across the resistors in balance with the EMFs from the wires for net zero voltage around the loop.  If you stick ideal voltage probes in two spots along a given wire you will measure a voltage.

2.  Ordinary wire loop with no resistors:  Here you have current flow where the linear EMF in the wire and the distributed resistance in the wire cancel each other out.  I'm pretty sure if you stick ideal voltage probes at any two points along the loop of wire you will read zero volts.

3.  A superconducting loop of wire with no resistors:  Here you are in pure current loop territory.  The net flux through the loop is always zero, and your ideal voltage probes will measure zero volts anywhere along the wire.

Anyway, I hope to really look at the clip again soon.

MileHigh
   

Group: Administrator
Hero Member
*****

Posts: 2905
It's not as complicated as it may seem...
Darren:

You have the voltage polarities for the inductors in your diagram backwards.

Actually they are correct. On the voltage meters , I have indicated the polarity of the meter, NOT the polarity of the voltage that will be measured. ;)

Harvey:

Either one is attempting to sum the voltage drops and gains in a loop per KVL, or they are not.  Is professor Lewin attempting to compute the KVL loop or not? If he IS, and it appears to me that he is, then he is clearly going about it incorrectly. If he is not, then he is attempting to perform a different experiment, and he is mixing apples and oranges. What I have shown is the correct way to compute and use KVL.

We may disagree about how KVL applies, and about what the professor was or was not trying to demonstrate, but there is absolutely nothing mysterious or unexplainable in the demonstration professor Lewin gave, and that really is the bottom line. And the electric fields are conserved. ;)

.99
   
Newbie
*

Posts: 43
<sigh> I really, really don't have time for this (apologies ahead of time for not being available to respond)... but...

If I were Walter giving that lecture and demonstration the thing I would be hoping people would get is loop integrals in changing magnetic fields. Don't focus on just the changing magnetic field in the central circuit -- consider also the changing field in the loops of the measurement circuits.

If you wanted to accurately measure the potential across R1 and R2 in practice you would have to take extreme measures to reduce the loop area for the measuring circuit -- to as close to zero as possible. If there is any area in the measuring loop then you have to include the induced EMF for that loop in any calculation of measured potentials. Even if you wanted to measure the potential between two points A and D accurately you would have to ensure the the measuring loops were as close to zero area as possible no matter where you located the measuring device.

Also -- don't forget the potential induced as you traverse the line integral that the circuit follows. There is an induced EMF per distance (flux cut actually) that is independent of any circuit resistances -- that's what induction is... the sections of wire joining R1 and R2 are EMF sources. To see what's happening don't model this in Spice - use FEM package like Maxwell (or any other competent FEM package) that will capture the induced EMF's in the actual topology.
   

Group: Administrator
Hero Member
*****

Posts: 2905
It's not as complicated as it may seem...
Thanks for your input Mark, and thank you for affirming the emf sources in the wiring as I have shown and been arguing for.

First, SPICE does an excellent job showing us how this circuit behaves, but yes it would be great to confirm with some magnetic modeling software indeed.

Second, I have run simulations taking into account induction in the measurement wiring, and the contribution to the net current through each resistor is inconsequential. There would have to be hundreds of turns with the measurement wire before its induced emf had a significant impact on the opposing current in each resistor.

Again, professor Lewin is mixing apples and oranges here by introducing this dynamic to the experiment, but even so, it can be accounted for and shown that it has little effect on the measured voltages.

.99

   

Group: Administrator
Hero Member
*****

Posts: 2905
It's not as complicated as it may seem...
Here are some numbers with a simulation run taking into account the measurement wiring on the left hand voltage meter across R1:

The wiring inductance was increased over that of the inner loop inductance by 1000 fold. So rather than 400nH in each leg, we have 400uH in each leg of the voltage meter. Also, the coupling factor is the same 0.5 from the solenoid to the voltage meter wiring. Meter impedance is 1M.

The voltage measured across R1 goes from the previous 100mV, down to about 98mV. Conversely, if the meter leads are swapped, the voltage measured is about 102mV. With the meter leads at a nominal inductance of 1uH or so, there is no detectable difference in the voltage across R1. Moving the scope probes across the meter's internal 1M resistor produces the same measurement as that directly across R1.

So it is obvious that the meter leads in this experiment have no appreciable effect on the voltage that the meter is measuring. This reinforces the fact that this experiment really is about the total induced inner-loop emf, and how it is distributed among the 4 inner loop components to produce a sum-zero net loop voltage, as prescribed by KVL.

.99
   

Group: Elite
Hero Member
******

Posts: 1859
I have tried to help you learn this as tactfully as I can....

.....You may wish to fully avail yourself of the discussion already recorded here which takes this to the most accurate levels:
http://www.physicsforums.com/showthread.php?t=405700

Harvey,

There have been a few other discussions about this. I was part of one. The problem is...  applying the modified/generalized KVL still yields the correct answers. This practice is also widely accepted. I don't understand how it can be. When this happens either the laws of Kirchhoff or Faraday are misapplied.

I don't see anyone denying the inductance of the wires. What does appear to me is the blurring of the definitions for a field and a force. It is undeniable that the summation will result in zero but only if you apply the correct law, or some generalized form. If the circuit elements include inductance Faraday must be used vs. Kirchhoff's KVL. I'm sure any circuit simulator uses a generalized form otherwise there would be no electric field difference across an inductor.

My memory says terms like 'EMF' & 'voltage drop' and the word 'zero' weren't part of the Kirchhoff's voltage law. I've been wasting a lot of time trying to find the original German text to prove that. I see there is no point. The only time someone will run into a problem with the misuse are discussions like this or the rare case where they should be using a network analyzer.

Maybe this argument is just another way of weeding out the young from the old?

Or the old from the young   ;D


---------------------------
"As far as the laws of mathematics refer to reality, they are not certain; as far as they are certain, they do not refer to reality." - Einstein

"What we observe is not nature itself, but nature exposed to our method of questioning." - Werner Heisenberg
   

Group: Administrator
Hero Member
*****

Posts: 2905
It's not as complicated as it may seem...
WW,

Sorry that you disagree with me also. I guess MH, ION and I stand against AC, EM, Harvey, and WW on this one.

It's not about young or old btw; age is irrelevant when it comes to actual observations.

I think what is important and what should be taken away from this discussion, is that observation is what counts, either on the bench, or in a simulator. Definitions and derivatives of one law to form another, is secondary. Also of importance is stating and understanding the goal of what is being demonstrated.

I gave 4 definitions of KVL, and not one mentioned anything about sources or static conditions etc. The experiment as presented by me in the simulation strictly obeys KVL as stated in those definitions, but there may be other definitions that DO state that induction does not apply. I would disagree, and the results I've shown prove this through.

If all this dissonance is simply the result of semantics, then this has perhaps all been a waste of time.

I stand by my results however, and the fact that professor Lewin has made one error with the voltage meter, and only revealed half of the actual story. What professor Lewin has proven imo, is that he is human, and that even professors make mistakes from time to time.

.99
« Last Edit: 2011-02-27, 16:20:56 by poynt99 »
   
Pages: 1 [2] 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 ... 25
« previous next »


 

Home Help Search Login Register
Theme © PopularFX | Based on PFX Ideas! | Scripts from iScript4u 2019-11-12, 23:55:12