Jefimenko, Coulomb and Lorentz
Bill Miller
among physicists is as rare as unanimity among cats. But a general consensus
was that, as it relates to the equations, Jefimenko got them "right." It
also appears doubtful that Jefimenko "copied" the concept from P&P.
There was less unanimity as to what these equations mean -- especially as it
relates to induction and causality.
That's why I have launched a separate thread. Please bear with me while I
state some obvious concepts.
There are two basic laws upon which most, if not all electromagnetics is
founded. These are Coulombs law and Lorentz's force equation.
Coulomb's law reads: F = (q1*q2)/r
Lorentz's equation reads: F = q(vXB)
These two equations define how charges behave in the presence of Electric
and Magnetic fields.
In words, Coulombs law states that the force exerted between two particles
is proportional to the product of the magnitude of the charges and inversely
proportional to the distance between them.
(Yes, I know there is a problem with Coulomb as r ---> zero and the
force ---> infinity!)
In words, Lorentz's equation says that the force exerted on a charge is
defined by the product of the magnitude of the charge and the vector product
of the relative velocity of the charge with respect to an applied magnetic
field.
The force exerted by the Coulomb effect is operative whether the two charges
are at rest, or in motion. The closer they are, the bigger the force.
In contrast, the Lorentz force *only* occurs when there is relative motion
between the charge and the applied field.
Our moderator, Fred, pointed out that whereas texts carefully scrutinize
Lorentz in the context of a moving charge and a fixed B, they seem to pay
scant attention to a charge in the presence of a varying, but not
necessarily moving, B.
Let me suggest that Jefimenko's expression for B provides the necessary
information.
Jefimenko's equation for B consists of *two* terms. The first is what we
might call "magnetostatic" B field. This is the B generated in response to a
time-invariant current.The second term is the B generated in response to a
time-variant current. I suggest the term "magnetokinetic" B field, or as
Fred has suggested, B_k.
This term is similar to Jefimenko's E_k, but is not identical to it, since
B_k contains a cross product with the unit radius vector.
So... if we "plug" these two B terms from Jefimenko into the Lorentz
equation, we *should* be able to gain some insights into the effect of a
time-variable magnetic field on a charge. And maybe that will help us decide
if a changing magnetic field can *somehow* cause an E field.
Or not.
(I just "gotta" believe that someone has already done this! But the texts I
own seem bereft of information on this subject.)
All the best,
Bill Miller
FrediFizzx
news:8eaj6c...@mid.individual.net...
Not a moderator of this group. Well, it is easy to see that the force
will be varying with time also.
> Let me suggest that Jefimenko's expression for B provides the
> necessary information.
>
> Jefimenko's equation for B consists of *two* terms. The first is what
> we might call "magnetostatic" B field. This is the B generated in
> response to a time-invariant current.The second term is the B
> generated in response to a time-variant current. I suggest the term
> "magnetokinetic" B field, or as Fred has suggested, B_k.
No, I was suggesting originally that E_k should be labeled B_k because
the cause looks to be something that would cause a magnetic type field.
Even though it is appearing in J's E equation also. In fact, as I have
said before, J's causal equations are not decoupled. I think that is a
serious problem that needs more investigation. I don't really know if
it is a flaw in Jefimenko's reasoning about E_k but suspect it may be.
> This term is similar to Jefimenko's E_k, but is not identical to it,
> since B_k contains a cross product with the unit radius vector.
>
> So... if we "plug" these two B terms from Jefimenko into the Lorentz
> equation, we *should* be able to gain some insights into the effect of
> a time-variable magnetic field on a charge. And maybe that will help
> us decide if a changing magnetic field can *somehow* cause an E field.
>
> Or not.
Hmm... I thought that was already settled. A changing magnetic field
certainly does cause an E field in my example of a magnet through a wire
loop, but it is not the Faraday E field it is causing. It is an E field
from the induced current. But yes, we need to know if it was the
changing B field or the associated Faraday E field that really induced
the current. Or both? Well, thinking about it a bit more and as I have
mentioned before, we can plug in J's B equation into curl E = -dB/dt
then we do have a cause for curl E. And can lump it all together
macroscopically to say the changing B is also causing the Faraday E
field. It is that sort of thing that makes me remain a bit skeptical
about what he is saying.
> (I just "gotta" believe that someone has already done this! But the
> texts I own seem bereft of information on this subject.)
I've got an example from my Schaum's Electromagnetic book of a moving
(rotating) wire loop and changing B field that I will post tomorrow.
Not sure it is really going to help much.
Best,
Fred Diether
John Polasek
wrote:
>A previous thread discussed the validity of Jefimenko's Equations. Unanimity
>among physicists is as rare as unanimity among cats. But a general consensus
>was that, as it relates to the equations, Jefimenko got them "right." It
>also appears doubtful that Jefimenko "copied" the concept from P&P.
>
>There was less unanimity as to what these equations mean -- especially as it
>relates to induction and causality.
>
>That's why I have launched a separate thread. Please bear with me while I
>state some obvious concepts.
>
>There are two basic laws upon which most, if not all electromagnetics is
>founded. These are Coulombs law and Lorentz's force equation.
>
>Coulomb's law reads: F = (q1*q2)/r
>Lorentz's equation reads: F = q(vXB)
Are you sure?
>These two equations define how charges behave in the presence of Electric
>and Magnetic fields.
>
>In words, Coulombs law states that the force exerted between two particles
>is proportional to the product of the magnitude of the charges and inversely
>proportional to the distance between them.
I wasn't sure you meant to write Coulombs law as you did, until you
substantiated it in words as being inversely proportional to the
separation.
This could be a breakthrough and looks like a new way to get rid of
Coulombs constant.
It is in direct competition with F = q1q2/4pieps0*r^2, my personal
favorite.
You might want to entertain the possibility that you are beyond your
depth.
>(I just "gotta" believe that someone has already done this! But the texts I
>own seem bereft of information on this subject.)
>
>All the best,
>
>Bill Miller
John Polasek
Benj
> (I just "gotta" believe that someone has already done this! But the texts I
> own seem bereft of information on this subject.)
> Bill Miller
I think the problem has been that for hundreds of years nobody has
bothered to do this sort of thing because everything "seemed" to work
so well experimentally.
For example. I know you agree with me and J that a changing electric
field "D" (proportional to E in free space) has never been found
experimentally to create a B field. This may seem like some minor
thing having to do with magnetic fields around capacitors, but it's
far more serious than that! The so-called "displacement current" is
the very "clever" basis that meant that Maxwell's equations could
predict radio! If there is no experimental basis for displacement
current, then clearly there is something very very wrong at a very
fundamental level with Maxwell!
A couple of more comments about Fred's unshakable belief that E and B
fields can cause each other. He says that in the case of a tossed
magnet, the E field comes from the induced current. But what if there
is no current? I believe one can show that in the case of magnetic
induction OR relative motion there is in essence an E field throughout
space CAPABLE of driving a current and/or moving charges but it's
existence doesn't depend upon it's doing so anymore than the existence
of an electrostatic E field depends upon the charges that it is
producing force upon! Or maybe it does!!! Now that would be
interesting! Whoa! You measure an E_static by inserting a "test
charge", but the minute you do that the E field suddenly springs into
existence! How do you get around that one to prove it?
Proof that E can exist without current comes from an other way of
looking at things. Current creates a B field about itself. If we look
at J's equations we can note the similarity to the definition of the
magnetic vector potential A. Making that connection we obtain a
relation E_k = - dA/dt. A is clearly related to current and caused by
it. It is retarded as it goes away from it as well. Hence we find that
E_k = - dA/dt is very much like Curl E = -dB/dt or it's integral which
is Faraday's law. So clearly A is ALSO another field caused by
current! So the real question now would be are all these "retarded"
fields real or are they simply aspects of some more unified quantity
and J is simply chasing his tail in confusion? I suggest that we
ignore the old term "EMF" which comes from a previous age. It's
confusing. What we really are talking about is voltage which is
electric potential. That is the integral of E along some path. Note
the difference in E_Static (where integral around a loop always gives
zero) and E_k where the same integral gives a value...which is the
basis of Hooper and J declaring these fields to NOT be the "same". The
calculation of potential shows that the E_k does not depend on induced
currents flowing. Similarly we know that B = curl A. All of this
shows that given current definitions there is a lot of interaction
going on between these quantities, but much is not clear! J has
started the process of trying to sort it all out, but as Bill found,
much is still left in confusion.
But I think Bill is exactly correct in trying to sort through this
stuff!
The late Professor A. H. Benade once told me that he could take over
the freshman physics lab. Give everyone a pendulum and a stop watch
and have them all doing cutting edge physics for the whole year! (He
also noted they'd never let him do it, either!) The point is that the
leading edge of physics IS right there in your freshman textbook IF
you know where to look and IF you know what to question!
FrediFizzx
news:2e083ed5-310d-4739...@w15g2000pro.googlegroups.com...
> On Sep 2, 5:29 pm, "Bill Miller" <kt...@yahoo.com> wrote:
>
>> (I just "gotta" believe that someone has already done this! But the
>> texts I
>> own seem bereft of information on this subject.)
>> Bill Miller
>
> I think the problem has been that for hundreds of years nobody has
> bothered to do this sort of thing because everything "seemed" to work
> so well experimentally.
:-) Hundreds? Bothered? Not "seemed" as there has so far been no
experimental evidence that the current formulation of Classical
Electrodynamics is wrong in the least bit in its domain of
applicability.
> For example. I know you agree with me and J that a changing electric
> field "D" (proportional to E in free space) has never been found
> experimentally to create a B field. This may seem like some minor
> thing having to do with magnetic fields around capacitors, but it's
> far more serious than that! The so-called "displacement current" is
> the very "clever" basis that meant that Maxwell's equations could
> predict radio! If there is no experimental basis for displacement
> current, then clearly there is something very very wrong at a very
> fundamental level with Maxwell!
There is experimental evidence for displacement current. There is just
no good evidence that it creates a B field. There would be something
very fundamentally wrong if displacement current didn't exist; div J
would not be equal to -drho/dt and charge would not be conserved. :-)
But let's see here; if we plug J's causal equation for E into curl B =
dE/(c^2 dt), why doesn't that give us a causal relationship for curl B?
No one has really answered that problem yet. Yeah, you took a stab at
it but I wasn't at all satisfied with your answer. But if we rearrange
the equation,
(curl B -dE/(c^2dt))/mu0 = J
Then we can see that a curled B field and -dE/(c^2dt) should be produced
by J simultaneously. If J is constant, seems weird that it would
produce a changing E field though.
> A couple of more comments about Fred's unshakable belief that E and B
> fields can cause each other. He says that in the case of a tossed
> magnet, the E field comes from the induced current.
The *caused* E field.
> But what if there
> is no current?
No induced current, then no caused E field.
> I believe one can show that in the case of magnetic
> induction OR relative motion there is in essence an E field throughout
> space CAPABLE of driving a current and/or moving charges but it's
> existence doesn't depend upon it's doing so anymore than the existence
> of an electrostatic E field depends upon the charges that it is
> producing force upon! Or maybe it does!!! Now that would be
> interesting! Whoa! You measure an E_static by inserting a "test
> charge", but the minute you do that the E field suddenly springs into
> existence! How do you get around that one to prove it?
I was thinking of the same thing. Does the caused E field depend on
charged matter existing? For that matter, does the Faraday field also
depend on charged matter existing? I suspect it is impossible to tell
since it takes charged matter to detect an E field. Something else to
toss out there; *all* elementary charged matter has a magnetic moment.
> Proof that E can exist without current comes from an other way of
> looking at things. Current creates a B field about itself. If we look
> at J's equations we can note the similarity to the definition of the
> magnetic vector potential A. Making that connection we obtain a
> relation E_k = - dA/dt. A is clearly related to current and caused by
> it. It is retarded as it goes away from it as well. Hence we find that
> E_k = - dA/dt is very much like Curl E = -dB/dt or it's integral which
> is Faraday's law.
Easy one to see. It doesn't even depend on E_k because if the scalar
potential is zero then you have,
E = -dA/dt
Take the curl of both sides.
curl E = - d(curl A)/dt
curl A = B
curl E = -dB/dt
> So clearly A is ALSO another field caused by
> current!
Yes. What is A in terms of B?
> So the real question now would be are all these "retarded"
> fields real or are they simply aspects of some more unified quantity
> and J is simply chasing his tail in confusion?
Jefimenko was definitely not confused; he presented an extremely good
argument for E_k with many examples. But he kind of dropped the ball on
"Induction by Moving Magnets" Sect. 2-6 since he went back the
magnetostatic Lorentz force. But he did a neat derivation of the
magnetostatic Lorentz force from E_k starting in Section 2-5 ending in
2-6.
Best,
Fred Diether
Szczepan Bialek
"Benj" <bja...@iwaynet.net> wrote
news:2e083ed5-310d-4739...@w15g2000pro.googlegroups.com...
>
> The so-called "displacement current" is
the very "clever" basis that meant that Maxwell's equations could
predict radio! If there is no experimental basis for displacement
current, then clearly there is something very very wrong at a very
fundamental level with Maxwell!
The displacement current is in dielectric. But only in form of oscillations
of electrons. The electric waves go through dielectrics and it is the
"experimental basis".
The oscillations need the mass and inertia. Electrons have them.
In the Maxwell model electric particles (masless) cooperate with the massive
magnetic particles.
It had sense. Heaviside's EM do not.
You all should read how Heaviside wrote the "Maxwell's equations":
http://en.wikisource.org/wiki/Electromagnetic_effects_of_a_moving_charge
S*
Bill Miller
"FrediFizzx" <fredi...@hotmail.com> wrote in message
news:8ebl5l...@mid.individual.net...
>>
>> Our moderator, Fred, pointed out that whereas texts carefully scrutinize
>> Lorentz in the context of a moving charge and a fixed B, they seem to pay
>> scant attention to a charge in the presence of a varying, but not
>> necessarily moving, B.
>
> Not a moderator of this group. Well, it is easy to see that the force
> will be varying with time also.
Sorry for the unwarranted promotion. (Or would it be a demotion?) :-)
>
>> Let me suggest that Jefimenko's expression for B provides the necessary
>> information.
>>
>> Jefimenko's equation for B consists of *two* terms. The first is what we
>> might call "magnetostatic" B field. This is the B generated in response
>> to a time-invariant current.The second term is the B generated in
>> response to a time-variant current. I suggest the term "magnetokinetic" B
>> field, or as Fred has suggested, B_k.
>
> No, I was suggesting originally that E_k should be labeled B_k because the
> cause looks to be something that would cause a magnetic type field. Even
> though it is appearing in J's E equation also. In fact, as I have said
> before, J's causal equations are not decoupled. I think that is a serious
> problem that needs more investigation. I don't really know if it is a
> flaw in Jefimenko's reasoning about E_k but suspect it may be.
OK. What is the difference between an Electric Field and a Magnetic Field?
Since we cannot look, smell or taste a field, let me suggest that our tool
must be what effect a field has on a charge. An Electric field is one that
exerts a force on a charge whether or not it is moving. A Magnetic field is
one that exerts a force on a charge IFF there is a relative motion between
the charge and the field.
Or is this too simple?
If it is too simple, then where is it in error as it relates to J (the
durrentnot the physicist abbreviated)??
In particular, how can E_k *be* a magnetic field if E_k does not contain a
cross product relationship between the moving current J and H?
>> This term is similar to Jefimenko's E_k, but is not identical to it,
>> since B_k contains a cross product with the unit radius vector.
>>
>> So... if we "plug" these two B terms from Jefimenko into the Lorentz
>> equation, we *should* be able to gain some insights into the effect of a
>> time-variable magnetic field on a charge. And maybe that will help us
>> decide if a changing magnetic field can *somehow* cause an E field.
.
>
> Hmm... I thought that was already settled. A changing magnetic field
> certainly does cause an E field in my example of a magnet through a wire
> loop, but it is not the Faraday E field it is causing.
Not sure how you concluded that! Jefimenko. has a step-by-step analysis of
how the E_k term is responsible for induction. I invited you to provide a
detailed explanation of where he went wrong. Perhaps you missed my
invitation?
I think that we agree that Jefimnko's expressions accurately describe the
relationship between fields, charges and currents. And we agree these are
accurate renditions of Maxwell. So... what set of equations would you use
to explain magnetic causality of electric current?
>It is an E field from the induced current. But yes, we need to know if it
>was the changing B field or the associated Faraday E field that really
>induced the current. Or both? Well, thinking about it a bit more and as I
>have mentioned before, we can plug in J's B equation into curl E = -dB/dt
>then we do have a cause for curl E.
Upon further thought, I don't believe I agree with the idea that one can
freely "move around" a causal equation in the manner that you suggest. It
seems to me that allowing this technique to be accepted might lead to some
pretty silly "causal" expressions. I think Benj has looked at this in
greater detail.
>And can lump it all together macroscopically to say the changing B is also
>causing the Faraday E field. It is that sort of thing that makes me remain
>a bit skeptical about what he is saying.
>
>> (I just "gotta" believe that someone has already done this! But the texts
>> I own seem bereft of information on this subject.)
>
> I've got an example from my Schaum's Electromagnetic book of a moving
> (rotating) wire loop and changing B field that I will post tomorrow. Not
> sure it is really going to help much.
I look forward to it. I'm also still looking forward to the paper you told
us about that proved that E causes H. : -)
Bill Miller
"Benj" <bja...@iwaynet.net> wrote in message
news:2e083ed5-310d-4739...@w15g2000pro.googlegroups.com...
> (I just "gotta" believe that someone has already done this! But the texts
> I
> own seem bereft of information on this subject.)
> Bill Miller
I think the problem has been that for hundreds of years nobody has
bothered to do this sort of thing because everything "seemed" to work
so well experimentally.
>For example. I know you agree with me and J that a changing electric
>field "D" (proportional to E in free space) has never been found
experimentally to create a B field.
Actually, Fred allowed as how he had seen a paper that reported on an
experiment that *did* prove that D caused H. But he has not yet been able to
locate it. (!) Perhaps, Fred's position on this may have changed?
This may seem like some minor
thing having to do with magnetic fields around capacitors, but it's
far more serious than that! The so-called "displacement current" is
the very "clever" basis that meant that Maxwell's equations could
predict radio! If there is no experimental basis for displacement
current, then clearly there is something very very wrong at a very
>fundamental level with Maxwell!
I'm not quite ready to toss out Maxwell. But yes, there is a fundamental
issue involved here.
>A couple of more comments about Fred's unshakable belief that E and B
fields can cause each other. He says that in the case of a tossed
magnet, the E field comes from the induced current. But what if there
is no current? I believe one can show that in the case of magnetic
induction OR relative motion there is in essence an E field throughout
space CAPABLE of driving a current and/or moving charges but it's
existence doesn't depend upon it's doing so anymore than the existence
of an electrostatic E field depends upon the charges that it is
producing force upon! Or maybe it does!!! Now that would be
interesting! Whoa! You measure an E_static by inserting a "test
charge", but the minute you do that the E field suddenly springs into
>existence! How do you get around that one to prove it?
Herein lies the challenge. E and H require charges and currents to exist.
But can the charges and currents exist without E and H?
>Proof that E can exist without current comes from an other way of
looking at things. Current creates a B field about itself. If we look
at J's equations we can note the similarity to the definition of the
magnetic vector potential A. Making that connection we obtain a
relation E_k = - dA/dt. A is clearly related to current and caused by
it. It is retarded as it goes away from it as well. Hence we find that
E_k = - dA/dt is very much like Curl E = -dB/dt or it's integral which
is Faraday's law. So clearly A is ALSO another field caused by
>current!
A as a field? I'm not so sure!
>So the real question now would be are all these "retarded"
fields real or are they simply aspects of some more unified quantity
and J is simply chasing his tail in confusion? I suggest that we
ignore the old term "EMF" which comes from a previous age. It's
confusing. What we really are talking about is voltage which is
electric potential. That is the integral of E along some path. Note
the difference in E_Static (where integral around a loop always gives
zero) and E_k where the same integral gives a value...which is the
basis of Hooper and J declaring these fields to NOT be the "same". The
calculation of potential shows that the E_k does not depend on induced
currents flowing. Similarly we know that B = curl A. All of this
shows that given current definitions there is a lot of interaction
going on between these quantities, but much is not clear! J has
started the process of trying to sort it all out, but as Bill found,
much is still left in confusion.
But I think Bill is exactly correct in trying to sort through this
>stuff!
I suspect you are in the minority in that opinion!
>The late Professor A. H. Benade once told me that he could take over
the freshman physics lab. Give everyone a pendulum and a stop watch
and have them all doing cutting edge physics for the whole year! (He
also noted they'd never let him do it, either!) The point is that the
leading edge of physics IS right there in your freshman textbook IF
you know where to look and IF you know what to >question!
An example is looking at the essentials of E and H fields and defining what
they can, and cannot do.
John Polasek
>
> some obvious concepts:
>
>These are Coulombs law and Lorentz's force equation.
>
>Coulomb's law reads: F = (q1*q2)/r
unchallenged. Are the units of force coulombs squared per meter?
>Lorentz's equation reads: F = q(vXB)
>
>These two equations define how charges behave in the presence of Electric
>and Magnetic fields.
>
>In words, Coulombs law states that the force exerted between two particles
>is proportional to the product of the magnitude of the charges and inversely
>proportional to the distance between them.
>
>(Yes, I know there is a problem with Coulomb as r ---> zero and the
>force ---> infinity!)
snip
>
>Let me suggest that Jefimenko's expression for B provides the necessary
>information.
>
>and maybe that will help us decide
curl E = -mu*dH/dt
curl H = eps0*dE/dt displacement current density
>(I just "gotta" believe that someone has already done this!)
Your faith has made you whole-they have done this. Quo vide!
>
>All the best,
>
>Bill Miller
Bill Miller
> "Benj" <bja...@iwaynet.net> wrote in message
> news:2e083ed5-310d-4739...@w15g2000pro.googlegroups.com...
>> On Sep 2, 5:29 pm, "Bill Miller" <kt...@yahoo.com> wrote:
>>
>>> (I just "gotta" believe that someone has already done this! But the
>>> texts I
>>> own seem bereft of information on this subject.)
>>> Bill Miller
>>
>> I think the problem has been that for hundreds of years nobody has
>> bothered to do this sort of thing because everything "seemed" to work
>> so well experimentally.
>
> :-) Hundreds? Bothered? Not "seemed" as there has so far been no
> experimental evidence that the current formulation of Classical
> Electrodynamics is wrong in the least bit in its domain of applicability.
Correct. But does the domain of applicability include the area between the
plates of a capacitor? Experimental evidence says that there"ain't" a
magnetic field that can be unequivocally tied to Displacement Current.
>
>> For example. I know you agree with me and J that a changing electric
>> field "D" (proportional to E in free space) has never been found
>> experimentally to create a B field. This may seem like some minor
>> thing having to do with magnetic fields around capacitors, but it's
>> far more serious than that! The so-called "displacement current" is
>> the very "clever" basis that meant that Maxwell's equations could
>> predict radio! If there is no experimental basis for displacement
>> current, then clearly there is something very very wrong at a very
>> fundamental level with Maxwell!
>
> There is experimental evidence for displacement current.
The problem is the IMHO poor choice of the name.
Since it is called Displacement *Current* it is easy to undrstand why many
folks attribute *all* of the characteristics of current (except charge flow)
to it.
Displacement Current is simply the rate of change of the E field WRT time.
>There is just no good evidence that it creates a B field. There would be
>something very fundamentally wrong if displacement current didn't exist;
>div J would not be equal to -drho/dt and charge would not be conserved. :-)
>But let's see here; if we plug J's causal equation for E into curl B =
>dE/(c^2 dt), why doesn't that give us a causal relationship for curl B? No
>one has really answered that problem yet. Yeah, you took a stab at it but
>I wasn't at all satisfied with your answer. But if we rearrange the
>equation,
>
> (curl B -dE/(c^2dt))/mu0 = J
>
> Then we can see that a curled B field and -dE/(c^2dt) should be produced
> by J simultaneously. If J is constant, seems weird that it would produce
> a changing E field though.
In another post, I mentioned that indiscriminate moving around of causal
elements could lead to "silly" results. I suspect you may have just proven
that!
It took me several readings of Jefimenko's derivation to appreciate the
skill with which he separated correct (but non-causal) relationships from
those that were causal. In that regard, Jefimenko is probably the single
most exasperating expository writer that I have ever encountered.
Most such authors, upon describing a new or diffferent concept or result
will -- with decorum, of course -- point out the new and different with the
appropriate amount of bugles and rockets. It took we three readings before I
said, "Oh S--T" as I realized he had just "dumped" magnetic induction in the
trash!
>> A couple of more comments about Fred's unshakable belief that E and B
>> fields can cause each other. He says that in the case of a tossed
>> magnet, the E field comes from the induced current.
>
> The *caused* E field.
>
>> But what if there
>> is no current?
>
> No induced current, then no caused E field.
>
>> I believe one can show that in the case of magnetic
>> induction OR relative motion there is in essence an E field throughout
>> space CAPABLE of driving a current and/or moving charges but it's
>> existence doesn't depend upon it's doing so anymore than the existence
>> of an electrostatic E field depends upon the charges that it is
>> producing force upon! Or maybe it does!!! Now that would be
>> interesting! Whoa! You measure an E_static by inserting a "test
>> charge", but the minute you do that the E field suddenly springs into
>> existence! How do you get around that one to prove it?
>
> I was thinking of the same thing. Does the caused E field depend on
> charged matter existing? For that matter, does the Faraday field also
> depend on charged matter existing? I suspect it is impossible to tell
> since it takes charged matter to detect an E field. Something else to
> toss out there; *all* elementary charged matter has a magnetic moment.
Once again showing that E and H are inextricably linked. But is a magnetic
moment sufficient to account for the supposed generation of E from a
changing H? How?
I'm sorry. I didn't see the ball dropped. Could you expand on that a bit?
>But he did a neat derivation of the magnetostatic Lorentz force from E_k
>starting in Section 2-5 ending in 2-6.
Yep. Another example of an interesting result that he tossed out there with
no fanfare!
Bill Miller
"Szczepan Bialek" <sz.b...@wp.pl> wrote in message
news:4c81f43f$0$21008$6578...@news.neostrada.pl...
>
> "Benj" <bja...@iwaynet.net> wrote
> news:2e083ed5-310d-4739...@w15g2000pro.googlegroups.com...
>>
>> The so-called "displacement current" is
> the very "clever" basis that meant that Maxwell's equations could
> predict radio! If there is no experimental basis for displacement
> current, then clearly there is something very very wrong at a very
> fundamental level with Maxwell!
>
> The displacement current is in dielectric. But only in form of
> oscillations of electrons. The electric waves go through dielectrics and
> it is the "experimental basis".
This is true for dielectrics other than vacuum. In many physical
dielectrics, you obtain some interesting phenomena involving polarization
and also the electric equivalent of hysteresis.
But these phenomena all disappear when the physical medium is replaced by
vacuum.
FrediFizzx
news:s1q4861qvnqr8e7ke...@4ax.com...
> On Thu, 2 Sep 2010 17:29:06 -0400, "Bill Miller" <kt...@yahoo.com>
> wrote:
> snip
>>
>> some obvious concepts:
>>
>>These are Coulombs law and Lorentz's force equation.
>>
>>Coulomb's law reads: F = (q1*q2)/r
> Could you take time to go over this one please? We can't let it go
> unchallenged. Are the units of force coulombs squared per meter?
John, it is just typos on Bill's part. We all know what Coulomb's law
is.
>>Lorentz's equation reads: F = q(vXB)
>>
>>These two equations define how charges behave in the presence of
>>Electric
>>and Magnetic fields.
>>
>>In words, Coulombs law states that the force exerted between two
>>particles
>>is proportional to the product of the magnitude of the charges and
>>inversely
>>proportional to the distance between them.
>>
>>(Yes, I know there is a problem with Coulomb as r ---> zero and the
>>force ---> infinity!)
> That's not the only problem
> snip
>>
>>Let me suggest that Jefimenko's expression for B provides the
>>necessary
>>information.
> snip
>>
>>and maybe that will help us decide
>>if a changing magnetic field can *somehow* cause an E field.
> It can, and both can cause each other, and without Jefimenko
> curl E = -mu*dH/dt
> curl H = eps0*dE/dt displacement current density
Well, that is what this discussion is all about. Jefimenko says that
they aren't causal expressions and it means that both sides of those
expressions are created at the same time by a third "cause". So what is
the exact third cause for each of those expressions? IOW, the
mathematical expressions. The only thing I can think of is plugging in
his causal expressions for H and E in the right hand side then we have
causes for curl E and curl H. Bill and Benj say no; can't do that. If
they are right, then what is the cause for them?
Best,
Fred Diether
Szczepan Białek
"Bill Miller" <kt...@yahoo.com> wrote
news:8efcu9...@mid.individual.net...
>
> "Szczepan Bialek" <sz.b...@wp.pl> wrote in message
> news:4c81f43f$0$21008$6578...@news.neostrada.pl...
>>
>> "Benj" <bja...@iwaynet.net> wrote
>> news:2e083ed5-310d-4739...@w15g2000pro.googlegroups.com...
>>>
>>> The so-called "displacement current" is
>> the very "clever" basis that meant that Maxwell's equations could
>> predict radio! If there is no experimental basis for displacement
>> current, then clearly there is something very very wrong at a very
>> fundamental level with Maxwell!
>>
>> The displacement current is in dielectric. But only in form of
>> oscillations of electrons. The electric waves go through dielectrics and
>> it is the "experimental basis".
>
> This is true for dielectrics other than vacuum. In many physical
> dielectrics, you obtain some interesting phenomena involving polarization
> and also the electric equivalent of hysteresis.
In the space is the rare plazma. The plazma is not a dielectric.
No permanent polarization. There is the embedding of electrons and the
electrostriction.
Electrects are made by embedding the electrons. In such approach the
hysteresis is obvious.
> But these phenomena all disappear when the physical medium is replaced by
> vacuum.
Vacum (rare plasma) is a conductor.
FrediFizzx
>
> "FrediFizzx" <fredi...@hotmail.com> wrote in message
> news:8edqqa...@mid.individual.net...
>> "Benj" <bja...@iwaynet.net> wrote in message
>> news:2e083ed5-310d-4739...@w15g2000pro.googlegroups.com...
>>> On Sep 2, 5:29 pm, "Bill Miller" <kt...@yahoo.com> wrote:
>>>
>>>> (I just "gotta" believe that someone has already done this! But the
>>>> texts I
>>>> own seem bereft of information on this subject.)
>>>> Bill Miller
>>>
>>> I think the problem has been that for hundreds of years nobody has
>>> bothered to do this sort of thing because everything "seemed" to
>>> work
>>> so well experimentally.
>>
>> :-) Hundreds? Bothered? Not "seemed" as there has so far been no
>> experimental evidence that the current formulation of Classical
>> Electrodynamics is wrong in the least bit in its domain of
>> applicability.
>
> Correct. But does the domain of applicability include the area between
> the plates of a capacitor? Experimental evidence says that
> there"ain't" a magnetic field that can be unequivocally tied to
> Displacement Current.
Sure, the volume between plates of a capacitor is in the domain. Ok,
from now on we will assume -dE/(c^2dt) is not a cause of curl B just to
take it off the plate for this discussion. As I said before, since
there is a c^2 in the denominator, it is a very very small effect
anywise unless the time period is very high frequency or E is enormous.
>>> For example. I know you agree with me and J that a changing electric
>>> field "D" (proportional to E in free space) has never been found
>>> experimentally to create a B field. This may seem like some minor
>>> thing having to do with magnetic fields around capacitors, but it's
>>> far more serious than that! The so-called "displacement current" is
>>> the very "clever" basis that meant that Maxwell's equations could
>>> predict radio! If there is no experimental basis for displacement
>>> current, then clearly there is something very very wrong at a very
>>> fundamental level with Maxwell!
>>
>> There is experimental evidence for displacement current.
>
> The problem is the IMHO poor choice of the name.
>
> Since it is called Displacement *Current* it is easy to undrstand why
> many folks attribute *all* of the characteristics of current (except
> charge flow) to it.
>
> Displacement Current is simply the rate of change of the E field WRT
> time.
Well, Maxwell originally had it as "... a displacement of actual
electric charges residing in dielectric media." Quote from J's
Causality book. I suspect this is still true even for the "vacuum".
Quantum Vacuum Charge = +,- sqrt(4pi eps0 hbar c). But what J is
objecting to is calling its supposed action "Maxwell Induction". Quite
frankly, I had never heard of that term before J mentioned it. What
would you call it instead of "displacement current"?
>>There is just no good evidence that it creates a B field. There would
>>be something very fundamentally wrong if displacement current didn't
>>exist; div J would not be equal to -drho/dt and charge would not be
>>conserved. :-) But let's see here; if we plug J's causal equation for
>>E into curl B = dE/(c^2 dt), why doesn't that give us a causal
>>relationship for curl B? No one has really answered that problem yet.
>>Yeah, you took a stab at it but I wasn't at all satisfied with your
>>answer. But if we rearrange the equation,
>>
>> (curl B -dE/(c^2dt))/mu0 = J
>>
>> Then we can see that a curled B field and -dE/(c^2dt) should be
>> produced by J simultaneously. If J is constant, seems weird that it
>> would produce a changing E field though.
>
> In another post, I mentioned that indiscriminate moving around of
> causal elements could lead to "silly" results. I suspect you may have
> just proven that!
Are you now saying that dE/dt *is* a "causal" element? :-) Griffiths
says, "Maxwell's equations tell you how *charges* produce *fields*;
reciprocally, the force law [Lorentz] tells you how *fields* affect
*charges*." If that is true then J must be producing curl B and -
dE/(c^2dt) simultaneously. So for the flip side of this, has anyone
experimentally detected -dE/(c^2dt) with a constant J? Or what make it
go to zero in the case of constant current density? But we can also see
that two of Maxwell's equations are causal, div E = rho/eps0 and J =
curl B with dE/(c^2dt) taken to zero, if Griffiths is right. Well
actually there is no doubt that they are causal since there is
experimental evidence for it.
> It took me several readings of Jefimenko's derivation to appreciate
> the skill with which he separated correct (but non-causal)
> relationships from those that were causal. In that regard, Jefimenko
> is probably the single most exasperating expository writer that I have
> ever encountered.
>
> Most such authors, upon describing a new or diffferent concept or
> result will -- with decorum, of course -- point out the new and
> different with the appropriate amount of bugles and rockets. It took
> we three readings before I said, "Oh S--T" as I realized he had just
> "dumped" magnetic induction in the trash!
He didn't; see below.
No. I just put that out to reinforce the notion that E and H *are*
linked. Something that is perhaps not well known is that an electron
also has a curled E field in addition to its "static" E field. The
energy of its B field and curl E field cancel each other out observing
from a distance. In fact, the B field of a magnet is from the magnetic
moments of charged elementary matter all aligned in the same direction
and adding up. You can model that magnets have microscopic currents
that all add up but the B field of a magnet is actually more direct.
[snip uncommented text]
>>> So the real question now would be are all these "retarded"
>>> fields real or are they simply aspects of some more unified quantity
>>> and J is simply chasing his tail in confusion?
>>
>> Jefimenko was definitely not confused; he presented an extremely good
>> argument for E_k with many examples. But he kind of dropped the ball
>> on "Induction by Moving Magnets" Sect. 2-6 since he went back the
>> magnetostatic Lorentz force.
>
> I'm sorry. I didn't see the ball dropped. Could you expand on that a
> bit?
Right after eq. (2-6.10) F = - v x B he says, "If q were within a
conductor, then the force given by Eq. (2-6.10) would create a
conduction current in this conductor." So *macroscopically* he is
admitting that a B field can cause induction current in the conductor.
Of course the ultimate explanation for the B field of the magnet is the
magnetic moments of elementary charged matter in the magnet. No one
doubts that.
>>But he did a neat derivation of the magnetostatic Lorentz force from
>>E_k starting in Section 2-5 ending in 2-6.
>
> Yep. Another example of an interesting result that he tossed out there
> with no fanfare!
But it is another example of why I think E_k is more magnetic than
electric!
Best,
Fred Diether
FrediFizzx
[snip uncommented text]
[ fixing a typo in eq. (2-6.10)]
Right after eq. (2-6.10) F = - q(v x B) he says, "If q were within a
FrediFizzx
>
> "FrediFizzx" <fredi...@hotmail.com> wrote in message
> news:8ebl5l...@mid.individual.net...
>
>>>
>>> Our moderator, Fred, pointed out that whereas texts carefully
>>> scrutinize Lorentz in the context of a moving charge and a fixed B,
>>> they seem to pay scant attention to a charge in the presence of a
>>> varying, but not necessarily moving, B.
>>
>> Not a moderator of this group. Well, it is easy to see that the
>> force will be varying with time also.
>
> Sorry for the unwarranted promotion. (Or would it be a demotion?) :-)
LOL! Neither; this group doesn't really need a moderator as it is one
of the last sci.physics Usenet un-moderated groups that is not totally
overrun by extreme nut cases. Quiet as that is kept. :-)
>>> Let me suggest that Jefimenko's expression for B provides the
>>> necessary information.
>>>
>>> Jefimenko's equation for B consists of *two* terms. The first is
>>> what we might call "magnetostatic" B field. This is the B generated
>>> in response to a time-invariant current.The second term is the B
>>> generated in response to a time-variant current. I suggest the term
>>> "magnetokinetic" B field, or as Fred has suggested, B_k.
>>
>> No, I was suggesting originally that E_k should be labeled B_k
>> because the cause looks to be something that would cause a magnetic
>> type field. Even though it is appearing in J's E equation also. In
>> fact, as I have said before, J's causal equations are not decoupled.
>> I think that is a serious problem that needs more investigation. I
>> don't really know if it is a flaw in Jefimenko's reasoning about E_k
>> but suspect it may be.
>
> OK. What is the difference between an Electric Field and a Magnetic
> Field?
In what context? Initial and boundary conditions are important.
> Since we cannot look, smell or taste a field, let me suggest that our
> tool must be what effect a field has on a charge. An Electric field is
> one that exerts a force on a charge whether or not it is moving. A
> Magnetic field is one that exerts a force on a charge IFF there is a
> relative motion between the charge and the field.
Ok, but probably too simplistic for all cases.
> Or is this too simple?
>
> If it is too simple, then where is it in error as it relates to J (the
> durrentnot the physicist abbreviated)??
Just because it is too simple, does not mean it is in error. Magnets
also exert forces on each other a concept that can actually be modeled
by "magnetic charge". So what are the two B fields of the two magnets
doing to each other? And what effect does a Faraday E field have on
charges? But we have to go back to saying what condition the charges
are in such as in a conductor or are they just free in space. See what
I mean?
> In particular, how can E_k *be* a magnetic field if E_k does not
> contain a cross product relationship between the moving current J
> and H?
I am not saying that E_k is a magnetic field. I have been saying that
looks to be more magnetic-like. How does J produce an electric field?
What is the equation for that other than the E_k term? Actually it
seems to only produce an E field when it is changing in a conductor. So
I think we have a conundrum here. We have an expression that looks to
be magnetic-like producing an E field. And J even admits that it
simultaneously produces -dA/dt, which is a magnetic type component.
Now, is there actually something more to -dA/dt that we are missing
here?
>>> This term is similar to Jefimenko's E_k, but is not identical to it,
>>> since B_k contains a cross product with the unit radius vector.
>>>
>>> So... if we "plug" these two B terms from Jefimenko into the Lorentz
>>> equation, we *should* be able to gain some insights into the effect
>>> of a time-variable magnetic field on a charge. And maybe that will
>>> help us decide if a changing magnetic field can *somehow* cause an E
>>> field.
> .
>>
>> Hmm... I thought that was already settled. A changing magnetic
>> field certainly does cause an E field in my example of a magnet
>> through a wire loop, but it is not the Faraday E field it is causing.
>
> Not sure how you concluded that! Jefimenko. has a step-by-step
> analysis of how the E_k term is responsible for induction. I invited
> you to provide a detailed explanation of where he went wrong. Perhaps
> you missed my invitation?
No, I think you missed what I said about it.
> I think that we agree that Jefimnko's expressions accurately describe
> the relationship between fields, charges and currents. And we agree
> these are accurate renditions of Maxwell. So... what set of equations
> would you use to explain magnetic causality of electric current?
The same way Jefimenko ended up in section 2-6 with the magnetostatic
Lorentz force causing the induced current.
>>It is an E field from the induced current. But yes, we need to know
>>if it was the changing B field or the associated Faraday E field that
>>really induced the current. Or both? Well, thinking about it a bit
>>more and as I have mentioned before, we can plug in J's B equation
>>into curl E = -dB/dt then we do have a cause for curl E.
>
> Upon further thought, I don't believe I agree with the idea that one
> can freely "move around" a causal equation in the manner that you
> suggest. It seems to me that allowing this technique to be accepted
> might lead to some pretty silly "causal" expressions. I think Benj has
> looked at this in greater detail.
Then you need to show exactly how the causality disappears when doing
that. You or Benj have yet to explain that mathematically.
>>And can lump it all together macroscopically to say the changing B is
>>also causing the Faraday E field. It is that sort of thing that makes
>>me remain a bit skeptical about what he is saying.
>>
>>> (I just "gotta" believe that someone has already done this! But the
>>> texts I own seem bereft of information on this subject.)
>>
>> I've got an example from my Schaum's Electromagnetic book of a moving
>> (rotating) wire loop and changing B field that I will post tomorrow.
>> Not sure it is really going to help much.
>
> I look forward to it. I'm also still looking forward to the paper you
> told us about that proved that E causes H. : -)
This is for a closed conducting loop in motion with a B field changing
as a function of time.
V = - d/dt $B.dS = -$dB/dt.dS + $(U x B).dl
V is induced voltage, $ is integral sign, the dS integral is surface;
the dl integral is line. U is velocity; dB/dt is partial derivatives.
The first term on the right is the voltage due to the change in B, with
the loop held fixed; the second term on the right is the voltage arising
from the motion of the loop, with B held fixed. Like I said above, not
sure this is going to help much.
Best,
Fred Diether
John Polasek
<fredi...@hotmail.com> wrote:
>"John Polasek" <jpol...@cfl.rr.com> wrote in message
>news:s1q4861qvnqr8e7ke...@4ax.com...
>> On Thu, 2 Sep 2010 17:29:06 -0400, "Bill Miller" <kt...@yahoo.com>
>> wrote:
>> snip
>>>
>>> some obvious concepts:
>>>
>>>These are Coulombs law and Lorentz's force equation.
>>>
>>>Coulomb's law reads: F = (q1*q2)/r
>
>> Could you take time to go over this one please? We can't let it go
>> unchallenged. Are the units of force coulombs squared per meter?
>
>John, it is just typos on Bill's part. We all know what Coulomb's law
>is.
>
>>>Lorentz's equation reads: F = q(vXB)
>>>
>>>These two equations define how charges behave in the presence of
>>>Electric
>>>and Magnetic fields.
>>>
>>>In words, Coulombs law states that the force exerted between two
>>>particles
>>>is proportional to the product of the magnitude of the charges and
>>>inversely
>>>proportional to the distance between them.
Bill went out of his way to clarify 1/r. His next statement confirms
it.
>>>(Yes, I know there is a problem with Coulomb as r ---> zero and the
>>>force ---> infinity!)
>> That's not the only problem
>> snip
>>>
>>>Let me suggest that Jefimenko's expression for B provides the
>>>necessary
>>>information.
>> snip
>>>
>>>and maybe that will help us decide
>>>if a changing magnetic field can *somehow* cause an E field.
>
>> It can, and both can cause each other, and without Jefimenko
>> curl E = -mu*dH/dt
>> curl H = eps0*dE/dt displacement current density
>
>Well, that is what this discussion is all about. Jefimenko says that
>they aren't causal expressions and it
> means that both sides of those expressions are created at the same time by a third "cause".
Saying so is not enough.
>So what is
>the exact third cause for each of those expressions? IOW, the
>mathematical expressions.
H is the cause in the first equation. I can make H with coil and
current and I can vary these to get Hdot. mu goes to work to make a
curly E-field which is the only possible way. The latter can be
measured with a coil that is the analogue of the solenoidal field.
E is the cause of the second equation. The simplest is parallel
plates with E = V/g. I can vary these to get Edot. Eps0 goes to work
on it to produce a curly H field which in this case would be a torus
between the plates.
Faraway in free space you can't point to an overt source of charge or
current. The energy has already been carried away between E and H,
supported by eps0 and mu.
I think if you look back through these messages you will find that
only Jefimenko and I ever use eps0 and mu.
All of this is about physics. When I write D = E I can't assign units
nor make any sensible comment about it. But when I write D=eps E, I
have C/m^2 on one side and V/m on the other side. Then I can talk
cause and effect, eps0 being the mediator.
>The only thing I can think of is plugging in
>his causal expressions for H and E in the right hand side then we have
>causes for curl E and curl H.
Why not use my causal expressions; H amp turns/meter in my coil which
carries AC current. And for E, the capacitor with applied AC voltage.
Connect them in series or parallel. In the far field these effects are
distributed, with the energy retained but under attrition by geometric
spreading.
John Polasek
Bill Miller
> "John Polasek" <jpol...@cfl.rr.com> wrote in message
> news:s1q4861qvnqr8e7ke...@4ax.com...
>> On Thu, 2 Sep 2010 17:29:06 -0400, "Bill Miller" <kt...@yahoo.com>
>> wrote:
>> snip
>>>
>>> some obvious concepts:
>>>
>>>These are Coulombs law and Lorentz's force equation.
>>>
>>>Coulomb's law reads: F = (q1*q2)/r
>
>> Could you take time to go over this one please? We can't let it go
>> unchallenged. Are the units of force coulombs squared per meter?
>
> John, it is just typos on Bill's part. We all know what Coulomb's law is.
Yes and sorry! I typed it in, forgot the squared sign and went on from
there. #BLUSH#
Bill
Bill Miller
The "fly" in this "ointment" is that -- despite many efforts -- no one has
been able to measure the "curly H field" between the plates. Fred rightly
points out that, because of the c sqrd in the denominator, the "curly H
field" will be *very* small. But experimental phsicists know this, have
constructed apparatus (apparati?) with adequate shielding and sensitivity to
eliminate outside contamination *and* measure the expected "curly H fields"
and have come away with null results.
> Faraway in free space you can't point to an overt source of charge or
> current.
You have just put your finger on a major issue with Maxwell. Maxwell's
equations (withoud retardation) are "action at a distance" equations. Once
you are far enough away from a charge that has caused a field, the charge
could disappear (E) or stop flowing (H) and your measurements would not
reflect that change for a finite amount of time.
>The energy has already been carried away between E and H,
> supported by eps0 and mu.
> I think if you look back through these messages you will find that
> only Jefimenko and I ever use eps0 and mu.
I can't speak for the others, but when i use E H D and B in an equation in a
chat session like this, I am being lazy and non rigorous since I believe
that others reading my posting will recognize this and look at the sense of
my message rather than the rogourous nature of my exposition. (I went too
far with this with my Coulomb's law posting and then compinded the error by
treating my own mis-typed equation as if it were correct. I "gotta" watch
that.)
In most of the discussions, I and the others are discussing properties in
free space. In that environment, Mu and Epsilon are constant and linear WRT
frequency and energy level. I understand that your Pair Space theory assumes
a different situation. But the conclusions that you reach are not consistent
with baseline assumptions regarding the nature of the EM universe that we
others are using.
It seems unlikely that this will be resolved, at least in the short term.
> All of this is about physics. When I write D = E I can't assign units
> nor make any sensible comment about it. But when I write D=eps E, I
> have C/m^2 on one side and V/m on the other side. Then I can talk
> cause and effect, eps0 being the mediator.
>>The only thing I can think of is plugging in
>>his causal expressions for H and E in the right hand side then we have
>>causes for curl E and curl H.
> Why not use my causal expressions; H amp turns/meter in my coil which
> carries AC current. And for E, the capacitor with applied AC voltage.
> Connect them in series or parallel. In the far field these effects are
> distributed, with the energy retained but under attrition by geometric
> spreading.
> John Polasek
>
>>Bill and Benj say no; can't do that. If
>>they are right, then what is the cause for them?
>>
>>Best,
>>
>>Fred Diether
Bill Miller
>> Correct. But does the domain of applicability include the area between
>> the plates of a capacitor? Experimental evidence says that there"ain't" a
>> magnetic field that can be unequivocally tied to Displacement Current.
>
> Sure, the volume between plates of a capacitor is in the domain. Ok, from
> now on we will assume -dE/(c^2dt) is not a cause of curl B just to take it
> off the plate for this discussion. As I said before, since there is a c^2
> in the denominator, it is a very very small effect anywise unless the time
> period is very high frequency or E is enormous.
As I noted in a prior post, experimental physicists know all that, built
stuff to deal with it and found nothing.
I understand that you would like to take this "off the plate." That's kinda
like saying, "Other than that, Mrs. Kennedy, did you enjoy your stay in
Dallas?"
If we adopt your assumption above, then the question needs to be, "Since a
changing E does not cause a changing H, then isn't it highly likely -- in
spite of the (mostly) conclusive evidence -- that something other than a
changing H is causing a changing E?" (I said "mostly" in the preceding
question because Maxwell's equations -- as caisal descriptions -- do not
explain Lentz's Law. E_k does.
I *still* have not seen a convincing derivation using your suggested M and
N. Are you still working on that? :-)
>> Displacement Current is simply the rate of change of the E field WRT
>> time.
>
> Well, Maxwell originally had it as "... a displacement of actual electric
> charges residing in dielectric media." Quote from J's Causality book.
J was corect. "The Dynamical Theory Of The Electromagnetic Field" provides a
fascinating view of Maxwell's process, from gears to ether. So, maxwell
originally had it not as a dispalcement of chrges, but of gears!
The charge displacement concept is understandable, given the technology of
the 1800s. Building a parallel plate capacitor with air as the dielectric
poses some real mechanical problems. With thin, closely-spaced plates, it
takes only a slight amount of force to cause a short circuit -- or even
worse -- an ittermittent short. But if you insert a small piece of
non-conducting material between the plates, the "shorting" problem is
resolved. *But* those materials *do* exhibit electron displacement,
polarization, etc. It might not have been difficult to confuse what was
going on in a material dielectric from what was going on with it removed.
Plus, I suspect that very few experiments were performed in vacuum. Air has
molecules and electrons, so it would have been very easy to assume that
since air had electrons, it would act just like a solid dielectric.
Suppositions on my part, and unsupported by actual data. But my Dad built
radios back in the 20's and showed me how to make a capacitor out of
foil-backed-paper salvaged from cigarette packages when i was about 7 or 8.
>I suspect this is still true even for the "vacuum".
> Quantum Vacuum Charge = +,- sqrt(4pi eps0 hbar c).
Maybe, but is there experimental evidence that demonstrates that the QVC
actually exhibits displacement? It is my understanding that Mu and Epsilon
sub zero have been tested extensively at "all" frequencies and power levels
with no measurablechange?
But what J is
> objecting to is calling its supposed action "Maxwell Induction". Quite
> frankly, I had never heard of that term before J mentioned it. What would
> you call it instead of "displacement current"?
As I said, displacement current is simply epsilon times the rate of change
of E field. Nothing special about that.
>>>There is just no good evidence that it creates a B field. There would be
>>>something very fundamentally wrong if displacement current didn't exist;
>>>div J would not be equal to -drho/dt and charge would not be conserved.
>>>:-) But let's see here; if we plug J's causal equation for E into curl B
>>>= dE/(c^2 dt), why doesn't that give us a causal relationship for curl B?
>>>No one has really answered that problem yet. Yeah, you took a stab at it
>>>but I wasn't at all satisfied with your answer. But if we rearrange the
>>>equation,
>>>
>>> (curl B -dE/(c^2dt))/mu0 = J
>>>
>>> Then we can see that a curled B field and -dE/(c^2dt) should be produced
>>> by J simultaneously. If J is constant, seems weird that it would
>>> produce a changing E field though.
>>
>> In another post, I mentioned that indiscriminate moving around of causal
>> elements could lead to "silly" results. I suspect you may have just
>> proven that!
>
> Are you now saying that dE/dt *is* a "causal" element? :-) Griffiths
> says, "Maxwell's equations tell you how *charges* produce *fields*;
> reciprocally, the force law [Lorentz] tells you how *fields* affect
> *charges*." If that is true then J must be producing curl B and -
> dE/(c^2dt) simultaneously. So for the flip side of this, has anyone
> experimentally detected -dE/(c^2dt) with a constant J?
Not that I am aware of. But it would be very small (obviously) and would be
masked by both environmental noise and even by ripple in the most well
regulated power supply I can think of. On the same note, I don't hold my
breath waiting for someone to detect "curly H" associated with a constant
time-dependant J in a conductor. Many introductory texts contain a sample
problem asking the student to calculate the value of displacement current
associated with a straight conductor. They do that, I suspect, to
demonstrate how teeny weeny it is! Only between the plates of a capacitor,
wherein the E is concentrated because of the close spacing, does D become
significant. And, of course, that's why all the (failed) experiments to
measure "curly H" have concentrated on the volume between the plates.
Or what make it
> go to zero in the case of constant current density? But we can also see
> that two of Maxwell's equations are causal, div E = rho/eps0 and J = curl
> B with dE/(c^2dt) taken to zero, if Griffiths is right. Well actually
> there is no doubt that they are causal since there is experimental
> evidence for it.
>
>> It took me several readings of Jefimenko's derivation to appreciate the
>> skill with which he separated correct (but non-causal) relationships from
>> those that were causal. In that regard, Jefimenko is probably the single
>> most exasperating expository writer that I have ever encountered.
>>
>> Most such authors, upon describing a new or diffferent concept or result
>> will -- with decorum, of course -- point out the new and different with
>> the appropriate amount of bugles and rockets. It took we three readings
>> before I said, "Oh S--T" as I realized he had just "dumped" magnetic
>> induction in the trash!
>
> He didn't; see below.
>
>>
>> Once again showing that E and H are inextricably linked. But is a
>> magnetic moment sufficient to account for the supposed generation of E
>> from a changing H? How?
>
> No. I just put that out to reinforce the notion that E and H *are*
> linked. Something that is perhaps not well known is that an electron also
> has a curled E field in addition to its "static" E field.
Whoa! Where did this statement come from?
> The energy of its B field and curl E field cancel each other out observing
> from a distance. In fact, the B field of a magnet is from the magnetic
> moments of charged elementary matter all aligned in the same direction and
> adding up. You can model that magnets have microscopic currents that all
> add up but the B field of a magnet is actually more direct.
IOW Jefimenko's analysis of how a magnet works is totally incorrect?
>
> [snip uncommented text]
>
>>>> So the real question now would be are all these "retarded"
>>>> fields real or are they simply aspects of some more unified quantity
>>>> and J is simply chasing his tail in confusion?
>>>
>>> Jefimenko was definitely not confused; he presented an extremely good
>>> argument for E_k with many examples. But he kind of dropped the ball on
>>> "Induction by Moving Magnets" Sect. 2-6 since he went back the
>>> magnetostatic Lorentz force.
>>
>> I'm sorry. I didn't see the ball dropped. Could you expand on that a bit?
>
> Right after eq. (2-6.10) F = - v x B he says, "If q were within a
> conductor, then the force given by Eq. (2-6.10) would create a conduction
> current in this conductor." So *macroscopically* he is admitting that a B
> field can cause induction current in the conductor.
Huh? You have a cross product term between the (relative) velocity (of the
charge) WRT B. If the charge is in a conductor *and* it is moving, then yes
you get a change (in direction) conduction current. If there is no relative
motion, then there cannot be an effect on the charge. I'm at home so i don't
have access to J's book, so i don't know the context. But if that is what he
is saying then it is a clear mis-statement.
> Of course the ultimate explanation for the B field of the magnet is the
> magnetic moments of elementary charged matter in the magnet. No one
> doubts that.
>
>>>But he did a neat derivation of the magnetostatic Lorentz force from E_k
>>>starting in Section 2-5 ending in 2-6.
>>
>> Yep. Another example of an interesting result that he tossed out there
>> with no fanfare!
>
> But it is another example of why I think E_k is more magnetic than
> electric!
>
Again, how can E_k be magnetic if its *only* effect on a charge is to
accelerate it linearily? It seems that the only evidence that you have is
the existence of a Mu term. Is that sufficient to redefine it? Magnetic
fields operate on charges to change the direction of an existing velocity.
Nothing in the definition of E_k even implies that it does this.
John Polasek
wrote:
>
>"John Polasek" <jpol...@cfl.rr.com> wrote in message
>news:27p586dbbmh3fepp0...@4ax.com...
>> On Sat, 4 Sep 2010 10:34:31 -0700, "FrediFizzx"
>> <fredi...@hotmail.com> wrote:
>>
>>>"John Polasek" <jpol...@cfl.rr.com> wrote in message
>>>news:s1q4861qvnqr8e7ke...@4ax.com...
>>>> On Thu, 2 Sep 2010 17:29:06 -0400, "Bill Miller" <kt...@yahoo.com>
>>>> wrote:
>>>> snip
I looked up one comment where they said they had been able to measure
it:
D. F. Bartlett and T. R. Corle
Department of Physics, University of Colorado, Boulder, Colorado 80309
Received 25 February 1985; published in the issue dated 1 July 1985
We have measured the magnetic field directly inside a thin, circular,
parallel-plate capacitor as it is being charged. We find that this
field varies linearly with distance from the axis, as is to be
expected if a uniform displacement current flows between the plates.
The measured slope of B vs r agrees with predictions to within 5%.
But the paper costs $30.
One way of testing it would be to slip a mu metal washer with a number
of turns of #44 wire between plates and apply AC voltage to the
capacitor and reading the coil voltage.
That would probably need at least a gap of 4 mm and with 4 inch plates
.01 m², then C = eps0*area/gap = eps0*.01/.001=88 micro-micro-farads
which would give a puny current in any case.
Fred rightly
>points out that, because of the c sqrd in the denominator, the "curly H
>field" will be *very* small.
I think Fred is talking the Gaussian curlH = 1/c dD/dt but not c^2.
>But experimental phsicists know this, have
>constructed apparatus (apparati?) with adequate shielding and sensitivity to
>eliminate outside contamination *and* measure the expected "curly H fields"
>and have come away with null results.
>
>> Faraway in free space you can't point to an overt source of charge or
>> current.
>
>You have just put your finger on a major issue with Maxwell. Maxwell's
>equations (withoud retardation) are "action at a distance" equations. Once
>you are far enough away from a charge that has caused a field, the charge
>could disappear (E) or stop flowing (H) and your measurements would not
>reflect that change for a finite amount of time.
The fact that there is a time delay X/c is a snoozer.
>
>>The energy has already been carried away between E and H,
>> supported by eps0 and mu.
>> I think if you look back through these messages you will find that
>> only Jefimenko and I ever use eps0 and mu.
>
>I can't speak for the others, but when i use E H D and B in an equation in a
>chat session like this,
I don't see any H. or D. being used.
In fact let me give you some quotations from Griffiths that come from
this paper: (I am preparing a reply to the paper)
"The quantum vacuum at the foundations of classical electrodynamics"
G Leuchs, A S Villar, L L Sánchez-Soto" Max Planck Institute
Applied Physics B, Volume: 100 , Issue: 1 (2010
http://arxiv.org/abs/1005.0131. They pointed to some of the
limitations of electrodynamics quoting** that "the constant e0 has
the unimpressive rule of matching units, just happening to have the
value e0 = 8.8542 x 10-12 As/(Vm)". Similarly, they lamented that "The
electric displacement D enjoys no better status among us physicists
than that of e0. It is mostly seen as a mathematical tool of limited
use, because it would combine two physically different quantities".
And, further, "Sometimes the electric displacement is even considered
completely dispensable…".
**D. J. Griffiths: Introduction To Electrodynamics, 3d edition
(Prentice-Hall, New Jersey 1999) p. 180
Freddy puts a lot of stock in Griffiths-is it true what they say
about eps0? Then Griffiths is a worthless book.
The paper is trying to dope out quantum vacuum using crippled theories
of electrodynamics. I am able to help them out. Once they get my
correct values for and omega0 they will be all set.
>In most of the discussions, I and the others are discussing properties in
>free space. In that environment, Mu and Epsilon are constant and linear WRT
>frequency and energy level. I understand that your Pair Space theory assumes
>a different situation. But the conclusions that you reach are not consistent
>with baseline assumptions regarding the nature of the EM universe that we
>others are using.
>
>It seems unlikely that this will be resolved, at least in the short term.
>
>
>> All of this is about physics. When I write D = E I can't assign units
>> nor make any sensible comment about it. But when I write D=eps E, I
>> have C/m^2 on one side and V/m on the other side. Then I can talk
>> cause and effect, eps0 being the mediator.
>>>The only thing I can think of is plugging in
>>>his causal expressions for H and E in the right hand side then we have
>>>causes for curl E and curl H.
>> Why not use my causal expressions; H amp turns/meter in my coil which
>> carries AC current. And for E, the capacitor with applied AC voltage.
>> Connect them in series or parallel. In the far field these effects are
>> distributed, with the energy retained but under attrition by geometric
>> spreading.
>> John Polasek
John Polasek
>>>Best,
>>>
>>>Fred Diether
John Polasek
FrediFizzx
Bartlett later recanted. Seems you can't do the experiment with a
capacitor because the B field from the conduction current on the plates
is too strong. Or something like that.
http://puhep1.princeton.edu/~mcdonald/examples/EM/bartlett_ajp_58_1168_90.pdf
http://puhep1.princeton.edu/~mcdonald/examples/EM/bartlett_prl_55_59_85.pdf
http://puhep1.princeton.edu/~mcdonald/examples/EM/bartlett_pra_39_938_89.pdf
> One way of testing it would be to slip a mu metal washer with a number
> of turns of #44 wire between plates and apply AC voltage to the
> capacitor and reading the coil voltage.
> That would probably need at least a gap of 4 mm and with 4 inch plates
> .01 m², then C = eps0*area/gap = eps0*.01/.001=88 micro-micro-farads
> which would give a puny current in any case.
>
> Fred rightly
>>points out that, because of the c sqrd in the denominator, the "curly
>>H
>>field" will be *very* small.
> I think Fred is talking the Gaussian curlH = 1/c dD/dt but not c^2.
I was talking about curl B = dE/(c^2dt) in SI units. Bill is talking
about curl H = eps0dE/dt where there is no 1/c^2 so that mistake is on
him. :-)
That is not what Griffiths said on page 180. The only part they are
quoting from Griffiths is "...just happening to have the value e0 =
8.8542 x 10^-12 As/(Vm)". They were even paraphrasing. Here is the
whole statement from Griffiths, "That's why eps0 is call the
*permittivity of free space*. I dislike the term, for it suggests that
the vacuum is just a special kind of linear dielectric, in which the
permittivity happens to have the value 8.85 x 10^-12 C^2/(Nm^2)". But
you have to remember that this is a textbook on Classical
Electrodynamics where the vacuum is taken to be zero in every regards.
We know better than that.
Similarly, they lamented that "The
> electric displacement D enjoys no better status among us physicists
> than that of e0. It is mostly seen as a mathematical tool of limited
> use, because it would combine two physically different quantities".
> And, further, "Sometimes the electric displacement is even considered
> completely dispensable.".
> **D. J. Griffiths: Introduction To Electrodynamics, 3d edition
> (Prentice-Hall, New Jersey 1999) p. 180
> Freddy puts a lot of stock in Griffiths-is it true what they say
> about eps0? Then Griffiths is a worthless book.
Sheesh John, it is a textbook on Classical Electrodynamics where the
vacuum is taken to be zero in every regards.
> The paper is trying to dope out quantum vacuum using crippled theories
> of electrodynamics. I am able to help them out. Once they get my
> correct values for and omega0 they will be all set.
What "crippled theories"??? They are using your favorite SI system! I
think you misread their intent in the first few paragraphs.
Best,
Fred Diether
Benj
> Bartlett later recanted. Seems you can't do the experiment with a
> capacitor because the B field from the conduction current on the plates
> is too strong. Or something like that.
>
> http://puhep1.princeton.edu/~mcdonald/examples/EM/bartlett_ajp_58_116...
> http://puhep1.princeton.edu/~mcdonald/examples/EM/bartlett_prl_55_59_...
> http://puhep1.princeton.edu/~mcdonald/examples/EM/bartlett_pra_39_938...
Wow Fred, you beat me to it! I'm surprised since you've been the one
claiming that displacement current and the magnetic field it allegedly
generates are real...
John Polasek
<fredi...@hotmail.com> wrote:
I'd like to see their apparatus. I think my mu metal washer would
work.
>
>http://puhep1.princeton.edu/~mcdonald/examples/EM/bartlett_ajp_58_1168_90.pdf
>http://puhep1.princeton.edu/~mcdonald/examples/EM/bartlett_prl_55_59_85.pdf
>http://puhep1.princeton.edu/~mcdonald/examples/EM/bartlett_pra_39_938_89.pdf
These Links Don't Seem to open
>> One way of testing it would be to slip a mu metal washer with a number
>> of turns of #44 wire between plates and apply AC voltage to the
>> capacitor and reading the coil voltage.
>> That would probably need at least a gap of 4 mm and with 4 inch plates
>> .01 m², then C = eps0*area/gap = eps0*.01/.001=88 micro-micro-farads
>> which would give a puny current in any case.
>>
>> Fred rightly
>>>points out that, because of the c sqrd in the denominator, the "curly
>>>H
>>>field" will be *very* small.
>
>> I think Fred is talking the Gaussian curlH = 1/c dD/dt but not c^2.
>
>I was talking about curl B = dE/(c^2dt) in SI units.
You are correct but those are not SI units, See below
> Bill is talking
>about curl H = eps0dE/dt where there is no 1/c^2 so that mistake is on
>him. :-)
Bills got it right, curl H has no c^2 but curl B does.
curl H = eps0dE/dt. Multiply by mu to get curl B
curl B = mu*eps0 dE/dt = 1/c^2 dE/dt
But there's no advantage to using 1/c^2, you can't get a feel for it.
c is all they had to work with until somebody established mu and eps0
and it became the preferred mode in the '60's. But most authors never
'got it'.
Okay suppose I take that back. But since they are logically impelled
to incorporate permittivity into the quantum vacuum somehow, they made
a poor choice with Griffiths. And they are handicapping themselves by
converting everything into their paltry quantum set of variables: e,
c, hbar, m, and alpha but handicapped by two unknowns omega0 and r. I
have the correct value for these in my book, after which all their
equations work.
FrediFizzx
What is the mathematical formulation of Lentz's Law? :-)
>>> Displacement Current is simply the rate of change of the E field WRT
>>> time.
>>
>> Well, Maxwell originally had it as "... a displacement of actual
>> electric charges residing in dielectric media." Quote from J's
>> Causality book.
>
> J was corect. "The Dynamical Theory Of The Electromagnetic Field"
> provides a fascinating view of Maxwell's process, from gears to ether.
> So, maxwell originally had it not as a dispalcement of chrges, but of
> gears!
>
> The charge displacement concept is understandable, given the
> technology of the 1800s. Building a parallel plate capacitor with air
> as the dielectric poses some real mechanical problems. With thin,
> closely-spaced plates, it takes only a slight amount of force to cause
> a short circuit -- or even worse -- an ittermittent short. But if you
> insert a small piece of non-conducting material between the plates,
> the "shorting" problem is resolved. *But* those materials *do* exhibit
> electron displacement, polarization, etc. It might not have been
> difficult to confuse what was going on in a material dielectric from
> what was going on with it removed. Plus, I suspect that very few
> experiments were performed in vacuum. Air has molecules and electrons,
> so it would have been very easy to assume that since air had
> electrons, it would act just like a solid dielectric.
The quantum "vacuum" also acts a dielectric IMHO.
> Suppositions on my part, and unsupported by actual data. But my Dad
> built radios back in the 20's and showed me how to make a capacitor
> out of foil-backed-paper salvaged from cigarette packages when i was
> about 7 or 8.
>
> >I suspect this is still true even for the "vacuum".
>> Quantum Vacuum Charge = +,- sqrt(4pi eps0 hbar c).
>
> Maybe, but is there experimental evidence that demonstrates that the
> QVC actually exhibits displacement? It is my understanding that Mu and
> Epsilon sub zero have been tested extensively at "all" frequencies and
> power levels with no measurablechange?
It is totally up to interpretation as to whether or not the quantum
"vacuum" is a relativistic medium that has bound charge. The
experimental evidence is there but interpreted differently. The quantum
"vacuum" is so perfect, you can disregard it if you wish to. It's the
modern version of Maxwell's aether. And why would you think mu0 and
eps0 would change with displacement?
Actually, Jefimenko shows eq. (1-3.7) how a *changing* J
causes -dE/(c^2dt) but it is kind of strange because it goes to second
order for dJ/dt. Also ends up with a drho/dt term for a cause
of -dE/(c^2dt) but I think he might have made a mistake on eq. (1-3.8);
seems like the div J term would be cancelled out by part of the second
term so there is no drho/dt. But another strange thing is that what he
ends up with for dE/dt is not the same as the cause for curl H (B).
When he was showing the cause for curl E and dB/dt, they ended up being
identical. Well, there is no source in Maxwell for curl E + dB/dt = 0
so it is different.
>> Or what make it
>> go to zero in the case of constant current density? But we can also
>> see that two of Maxwell's equations are causal, div E = rho/eps0 and
>> J = curl B with dE/(c^2dt) taken to zero, if Griffiths is right.
>> Well actually there is no doubt that they are causal since there is
>> experimental evidence for it.
>>
>>> It took me several readings of Jefimenko's derivation to appreciate
>>> the skill with which he separated correct (but non-causal)
>>> relationships from those that were causal. In that regard, Jefimenko
>>> is probably the single most exasperating expository writer that I
>>> have ever encountered.
>>>
>>> Most such authors, upon describing a new or diffferent concept or
>>> result will -- with decorum, of course -- point out the new and
>>> different with the appropriate amount of bugles and rockets. It took
>>> we three readings before I said, "Oh S--T" as I realized he had just
>>> "dumped" magnetic induction in the trash!
>>
>> He didn't; see below.
>>
>
>>>
>>> Once again showing that E and H are inextricably linked. But is a
>>> magnetic moment sufficient to account for the supposed generation of
>>> E from a changing H? How?
>>
>> No. I just put that out to reinforce the notion that E and H *are*
>> linked. Something that is perhaps not well known is that an electron
>> also has a curled E field in addition to its "static" E field.
>
> Whoa! Where did this statement come from?
Weisskopf's famous paper, "On the Self-Energy and the Electromagnetic
Field of the Electron".
http://cos.cumt.edu.cn/jpkc/dxwl/zl/zl1/Physical%20Review%20Classics/particle/004.pdf
>> The energy of its B field and curl E field cancel each other out
>> observing from a distance. In fact, the B field of a magnet is from
>> the magnetic moments of charged elementary matter all aligned in the
>> same direction and adding up. You can model that magnets have
>> microscopic currents that all add up but the B field of a magnet is
>> actually more direct.
>
> IOW Jefimenko's analysis of how a magnet works is totally incorrect?
His mathematical model is not incorrect because you can model a magnet
as having bound current macroscopically that is all added up from
microscopic currents. But the true microscopic explanation is what I
said. So... what is the cause of the magnetic moments of charged
elementary matter? :-)
>> [snip uncommented text]
>>
>>>>> So the real question now would be are all these "retarded"
>>>>> fields real or are they simply aspects of some more unified
>>>>> quantity
>>>>> and J is simply chasing his tail in confusion?
>>>>
>>>> Jefimenko was definitely not confused; he presented an extremely
>>>> good argument for E_k with many examples. But he kind of dropped
>>>> the ball on "Induction by Moving Magnets" Sect. 2-6 since he went
>>>> back the magnetostatic Lorentz force.
>>>
>>> I'm sorry. I didn't see the ball dropped. Could you expand on that a
>>> bit?
>>
>> Right after eq. (2-6.10) F = - v x B he says, "If q were within a
>> conductor, then the force given by Eq. (2-6.10) would create a
>> conduction current in this conductor." So *macroscopically* he is
>> admitting that a B field can cause induction current in the
>> conductor.
>
> Huh? You have a cross product term between the (relative) velocity (of
> the charge) WRT B. If the charge is in a conductor *and* it is moving,
> then yes you get a change (in direction) conduction current. If there
> is no relative motion, then there cannot be an effect on the charge.
> I'm at home so i don't have access to J's book, so i don't know the
> context. But if that is what he is saying then it is a clear
> mis-statement.
I was just repeating what J said. But there is relative motion; the
magnet B field is moving wrt the charges in the conductor.
>> Of course the ultimate explanation for the B field of the magnet is
>> the magnetic moments of elementary charged matter in the magnet. No
>> one doubts that.
>>
>>>>But he did a neat derivation of the magnetostatic Lorentz force from
>>>>E_k starting in Section 2-5 ending in 2-6.
>>>
>>> Yep. Another example of an interesting result that he tossed out
>>> there with no fanfare!
>>
>> But it is another example of why I think E_k is more magnetic than
>> electric!
>>
>
> Again, how can E_k be magnetic if its *only* effect on a charge is to
> accelerate it linearily? It seems that the only evidence that you have
> is the existence of a Mu term. Is that sufficient to redefine it?
> Magnetic fields operate on charges to change the direction of an
> existing velocity. Nothing in the definition of E_k even implies that
> it does this.
You keep forgetting that the magnetic field is *changing*. This is not
a static situation! E_k (B_k?) is changing also. Plus we are talking
about charges in wire-like conductors; where else can a charge go in a
wire conductor? Ya gotta remember the boundary conditions. Also... we
need to investigate what effect a changing A field has on charges in
conductors since there is no doubt that a changing A field is created by
the changing current density. And... it is not only the mu0 term; it is
also the J term itself that makes this look magnetic-like. Plus the
fact that J's E causal equation is *not* decoupled. You have to have
second order terms for decoupling.
Best,
Fred Diether
FrediFizzx
They were using SQUIDs to detect B.
>> http://puhep1.princeton.edu/~mcdonald/examples/EM/bartlett_ajp_58_1168_90.pdf
>> http://puhep1.princeton.edu/~mcdonald/examples/EM/bartlett_prl_55_59_85.pdf
>> http://puhep1.princeton.edu/~mcdonald/examples/EM/bartlett_pra_39_938_89.pdf
> These Links Don't Seem to open
They work just fine for me. Try this and scroll down to find the
papers.
http://puhep1.princeton.edu/~mcdonald/examples/EM/
>>> One way of testing it would be to slip a mu metal washer with a
>>> number
>>> of turns of #44 wire between plates and apply AC voltage to the
>>> capacitor and reading the coil voltage.
>>> That would probably need at least a gap of 4 mm and with 4 inch
>>> plates
>>> .01 m², then C = eps0*area/gap = eps0*.01/.001=88 micro-micro-farads
>>> which would give a puny current in any case.
>>>
>>> Fred rightly
>>>>points out that, because of the c sqrd in the denominator, the
>>>>"curly
>>>>H
>>>>field" will be *very* small.
>>
>>> I think Fred is talking the Gaussian curlH = 1/c dD/dt but not c^2.
>>
>>I was talking about curl B = dE/(c^2dt) in SI units.
> You are correct but those are not SI units, See below
Sure they are; see below.
>> Bill is talking
>>about curl H = eps0dE/dt where there is no 1/c^2 so that mistake is on
>>him. :-)
> Bills got it right, curl H has no c^2 but curl B does.
> curl H = eps0dE/dt. Multiply by mu to get curl B
> curl B = mu*eps0 dE/dt = 1/c^2 dE/dt
1/c^2 dE/dt is the same as dE/(c^2dt) so it is SI units for sure.
> But there's no advantage to using 1/c^2, you can't get a feel for it.
> c is all they had to work with until somebody established mu and eps0
> and it became the preferred mode in the '60's. But most authors never
> 'got it'.
Well, 1/c^2 is easy to see that it is a very small quantity for a
factor. How would you easily notice that with eps0*mu0?
[snip uncommented text]
>>> The paper is trying to dope out quantum vacuum using crippled
>>> theories
>>> of electrodynamics. I am able to help them out. Once they get my
>>> correct values for and omega0 they will be all set.
>>
>>What "crippled theories"??? They are using your favorite SI system!
>>I
>>think you misread their intent in the first few paragraphs.
> Okay suppose I take that back. But since they are logically impelled
> to incorporate permittivity into the quantum vacuum somehow, they made
> a poor choice with Griffiths.
LOL! All they did was to get a numerical value of eps0 from Griffiths.
> And they are handicapping themselves by
> converting everything into their paltry quantum set of variables: e,
> c, hbar, m, and alpha but handicapped by two unknowns omega0 and r. I
> have the correct value for these in my book, after which all their
> equations work.
Yeah, the paper is not all that great. If you want to put "The quantum
vacuum at the foundations of classical electrodynamics", then Quantum
Vacuum Charge = +,-sqrt(4pi eps0 hbar c) does a much better job. :-)
Ya don't even really need to talk about eps0 and mu0 that way with QVC.
I find that there is only a 4pi difference between building E and B
fields from QVC and what Maxwell has to offer. I think it is due to
photon spin since I did not take that into consideration.
Best,
Fred Diether
John Polasek
<fredi...@hotmail.com> wrote:
Have you ever wondered why your expression for vacuum charge is 11.71
times that of an electron? There's a good reason: if you would include
alpha, the fine structure constant, under the radical, they would
become equal. Do you have a good reason for not including alpha?
I forget the line of reasoning you used to come up with vacuum charge.
It seems to me that every new equation needs to come up with a solid
story.
FrediFizzx
[snip]
>>Yeah, the paper is not all that great. If you want to put "The
>>quantum
>>vacuum at the foundations of classical electrodynamics", then Quantum
>>Vacuum Charge = +,-sqrt(4pi eps0 hbar c) does a much better job. :-)
> Have you ever wondered why your expression for vacuum charge is 11.71
> times that of an electron?
See below.
> There's a good reason: if you would include
> alpha, the fine structure constant, under the radical, they would
> become equal. Do you have a good reason for not including alpha?
Of course there is a good reason for not including alpha. It would
simply be the charge of an electron or positron then and not QVC.
> I forget the line of reasoning you used to come up with vacuum charge.
> It seems to me that every new equation needs to come up with a solid
> story.
The simple story is that the quantum "vacuum" can be modeled as a
relativistic medium composed of "less than virtual" fermionic pairs just
like they were describing in their paper. QVC is the bound microscopic
charge of that medium per "cell". It can't simply be electron or
positron charge because there are more than just "less than virtual"
electron-positron pairs making up the "cells" of that medium. The
geometric topology of the cell structure determines QVC and gives us an
explanation for alpha. And also an explanation for the other coupling
"constants".
Best,
Fred Diether
John Polasek
<fredi...@hotmail.com> wrote:
>"John Polasek" <jpol...@cfl.rr.com> wrote in message
>news:3iv986h8e4eecudi1...@4ax.com...
>> On Sun, 5 Sep 2010 23:57:57 -0700, "FrediFizzx"
>> <fredi...@hotmail.com> wrote:
>
>[snip]
>
>>>Yeah, the paper is not all that great. If you want to put "The
>>>quantum
>>>vacuum at the foundations of classical electrodynamics", then Quantum
>>>Vacuum Charge = +,-sqrt(4pi eps0 hbar c) does a much better job. :-)
Could your QVC cast any light on their equations? Their formulations
are fairly logical, but naïve.
>> Have you ever wondered why your expression for vacuum charge is 11.71
>> times that of an electron?
>
>See below.
>
>> There's a good reason: if you would include
>> alpha, the fine structure constant, under the radical, they would
>> become equal. Do you have a good reason for not including alpha?
>
>Of course there is a good reason for not including alpha. It would
>simply be the charge of an electron or positron then and not QVC.
You want to create a new particle very very cautiously.
>
>> I forget the line of reasoning you used to come up with vacuum charge.
>> It seems to me that every new equation needs to come up with a solid
>> story.
>
>The simple story is that the quantum "vacuum" can be modeled as a
>relativistic medium composed of "less than virtual" fermionic pairs just
>like they were describing in their paper. QVC is the bound microscopic
>charge of that medium per "cell". It can't simply be electron or
>positron charge because there are more than just "less than virtual"
>electron-positron pairs making up the "cells" of that medium.
Your QVC charge is e/sqrt(alpha) = 11.71*e. This seems to be an
additional questionable object in the fermion zoo. What is its mass?
>The geometric topology of the cell structure determines QVC and gives us an
>explanation for alpha.
Topology is never affected by scale factor of which alpha is 1/137.
>And also an explanation for the other coupling
>"constants".
That's good. What is the size of your cell?
(I think we both see how hard it is to sell an idea!)
FrediFizzx
> On Mon, 6 Sep 2010 16:01:23 -0700, "FrediFizzx"
> <fredi...@hotmail.com> wrote:
>
>>"John Polasek" <jpol...@cfl.rr.com> wrote in message
>>news:3iv986h8e4eecudi1...@4ax.com...
>>> On Sun, 5 Sep 2010 23:57:57 -0700, "FrediFizzx"
>>> <fredi...@hotmail.com> wrote:
>>
>>[snip]
>>
>>>>Yeah, the paper is not all that great. If you want to put "The
>>>>quantum
>>>>vacuum at the foundations of classical electrodynamics", then
>>>>Quantum
>>>>Vacuum Charge = +,-sqrt(4pi eps0 hbar c) does a much better job.
>>>>:-)
> Could your QVC cast any light on their equations? Their formulations
> are fairly logical, but naïve.
Not really for casting any light. For polarization of the quantum
"vacuum", all you need is a relativistic medium of bound charge. Their
paper is mainly about justifying eps0 and mu0 to have physical meaning.
Which is OK. But QVC is not unit system dependent.
>>> Have you ever wondered why your expression for vacuum charge is
>>> 11.71
>>> times that of an electron?
>>
>>See below.
>>
>>> There's a good reason: if you would include
>>> alpha, the fine structure constant, under the radical, they would
>>> become equal. Do you have a good reason for not including alpha?
>>
>>Of course there is a good reason for not including alpha. It would
>>simply be the charge of an electron or positron then and not QVC.
> You want to create a new particle very very cautiously.
Not a new particle.
>>> I forget the line of reasoning you used to come up with vacuum
>>> charge.
>>> It seems to me that every new equation needs to come up with a solid
>>> story.
>>
>>The simple story is that the quantum "vacuum" can be modeled as a
>>relativistic medium composed of "less than virtual" fermionic pairs
>>just
>>like they were describing in their paper. QVC is the bound
>>microscopic
>>charge of that medium per "cell". It can't simply be electron or
>>positron charge because there are more than just "less than virtual"
>>electron-positron pairs making up the "cells" of that medium.
> Your QVC charge is e/sqrt(alpha) = 11.71*e. This seems to be an
> additional questionable object in the fermion zoo. What is its mass?
It is not a particular object but the combination of objects of the
"zoo".
>>The geometric topology of the cell structure determines QVC and gives
>>us an
>>explanation for alpha.
> Topology is never affected by scale factor of which alpha is 1/137.
??? I don't know what you mean.
>>And also an explanation for the other coupling
>>"constants".
> That's good. What is the size of your cell?
> (I think we both see how hard it is to sell an idea!)
Why would exact "size" matter? But we would need a geometrical math
wizard on a super computer to try to figure that out. On the order of
the electron compton wavelength divided by 2pi cubed will probably be
close. It might not even be possible to figure out a size if space is
becoming emergent at this level. But we could definitely say that a
cell is smaller than a hydrogen atom. :-)
Best,
Fred Diether
Benj
> It took me several readings of Jefimenko's derivation to appreciate the
> skill with which he separated correct (but non-causal) relationships from
> those that were causal. In that regard, Jefimenko is probably the single
> most exasperating expository writer that I have ever encountered.
I did want to comment on my total agreement with the above assessment
of Jefimenko's writing. It is HIGHLY condensed, and he doesn't give an
inch! He'll casually refer to some major derivation tucked away in an
appendix as if it's obvious and pull some of the most amazing math out
of his butt to solve in a few steps what seemed (or were to me!)
problems of nightmare complexity! He is Amazingly meticulous! I've
never found a mistake! And that also is totally unlike me. I've got
to work things over and over until I get two answers that agree before
I start to have confidence in what I did. And even then...
The bottom line is Jefimenko was not a guy like me. But he commands a
LOT of respect in my book. The folks saying he's a kook who doesn't
know what he's doing are simply nuts. But he's NOT an easy read. Which
is why I guess it's simpler to just not bother to see what he's done
and criticize it without ever having read it. Personally I'm totally
tickled that someone has started back at the 19th century and said,
"let's take another look at this stuff!" And what he came up with is
very interesting, indeed!
Vince Morgan
On Sep 4, 1:08 pm, "Bill Miller" <kt...@yahoo.com> wrote:
> It took me several readings of Jefimenko's derivation to appreciate the
> skill with which he separated correct (but non-causal) relationships from
> those that were causal. In that regard, Jefimenko is probably the single
> most exasperating expository writer that I have ever encountered.
[quote]
I did want to comment on my total agreement with the above assessment
of Jefimenko's writing. It is HIGHLY condensed, and he doesn't give an
inch! He'll casually refer to some major derivation tucked away in an
appendix as if it's obvious and pull some of the most amazing math out
of his butt to solve in a few steps what seemed (or were to me!)
problems of nightmare complexity! He is Amazingly meticulous! I've
never found a mistake! And that also is totally unlike me. I've got
to work things over and over until I get two answers that agree before
I start to have confidence in what I did. And even then...
The bottom line is Jefimenko was not a guy like me. But he commands a
LOT of respect in my book. The folks saying he's a kook who doesn't
know what he's doing are simply nuts. But he's NOT an easy read. Which
is why I guess it's simpler to just not bother to see what he's done
and criticize it without ever having read it. Personally I'm totally
tickled that someone has started back at the 19th century and said,
"let's take another look at this stuff!" And what he came up with is
very interesting, indeed!
[/quote]
And, a nice book with pictures might come out of it. In fact I think it
should. Please!!!
Regards,
Vince
Bill Miller
<snip>
>> If we adopt your assumption above, then the question needs to be, "Since
>> a changing E does not cause a changing H, then isn't it highly likely --
>> in spite of the (mostly) conclusive evidence -- that something other than
>> a changing H is causing a changing E?" (I said "mostly" in the preceding
>> question because Maxwell's equations -- as caisal descriptions -- do not
>> explain Lentz's Law. E_k does.
>>
>> I *still* have not seen a convincing derivation using your suggested M
>> and N. Are you still working on that? :-)
>
> What is the mathematical formulation of Lentz's Law? :-)
This question took me a while, since I wanted to look at a few different
sources.
The first thing I learned is that thre is no "t" in the name. It's Lenz's
law.
The second is that Wiki simply iterates something like the verbal statement
I made earlier. Namely:
"An induced current is always in such a direction as to oppose the motion or
change causing it"
Panofsky ignores it. Others simply equate it with Faradays law, e = -
dphi/dt, pointing out that Lenz introduced the minus sign.
But this "law" was not derived from first principles. It is a law based on
observation and *interpretation.*
Of course, that technique has been used over the ages. For example,
Ptolemaic astronomers used it to establish the concept of epicycles. And QM
folks use it extensively to develop the descriptions of QED that, as we
know, are usable from the particle level all the way up to the... UM...
UH... well... the particle level.
<snip>
>
>> Air has molecules and electrons, so it would have been very easy to
>> assume that since air had electrons, it would act just like a solid
>> dielectric.
>
> The quantum "vacuum" also acts a dielectric IMHO.
Does the quantum "vacuum" exhibit polarization? Or displacement?
>
>>
>> Maybe, but is there experimental evidence that demonstrates that the QVC
>> actually exhibits displacement? It is my understanding that Mu and
>> Epsilon sub zero have been tested extensively at "all" frequencies and
>> power levels with no measurablechange?
>
> And why would you think mu0 and eps0 would change with displacement?
Why would they *not* change? After all, they change dramatically in *real*
media.
>
<snip>
>> IOW Jefimenko's analysis of how a magnet works is totally incorrect?
>
> His mathematical model is not incorrect because you can model a magnet as
> having bound current macroscopically that is all added up from microscopic
> currents. But the true microscopic explanation is what I said. So...
> what is the cause of the magnetic moments of charged elementary matter?
> :-)
I think *that* question is the subject of a thread that would be a whole lot
longer, and far more contentious than anything in this or similar threads.
However, in a (probably) futile attmpt to outdo Benj for "cookoo-dom," my
"vote" is for Mills's electron model. That one, for a bound electron, is an
infinitessimally thin "shell" of charge, (like the candy coating on an M&M)
whose characteristics are that its charge is dynamically moving in such a
way as to not have any Fourier transform components that are synchronous
with lightspeed (ie no radiation) and a charge flow that is consistent with
the measured magnetic aspects of the electron. Other charged particles are
similarly defined. The model -- unlike those of QED or anything else I
know -- provides startlingly high correlations with the characteristics of
all elements and a huge number of inorganic and organic molecules.
If anyone wants to talk about this further, I respectfully suggest a new
thread!
>>> [snip uncommented text]
>>>
>>>>> Jefimenko was definitely not confused; he presented an extremely good
>>>>> argument for E_k with many examples. But he kind of dropped the ball
>>>>> on "Induction by Moving Magnets" Sect. 2-6 since he went back the
>>>>> magnetostatic Lorentz force.
>>>>
<snip>
OK.. let's *assume* you are correct. Then we have a situation wherein J
*seems* to have "nailed it" WRT induction between coils -- moving or
static-- but is wrong about the tossed magnet. Then where is the
mathematical model that is *not* in conflict with J *but* accurately
describes the tossed magnet as entirely a magnetic phenomenon? (Please use
something other than dphi/dt!)
> Plus the fact that J's E causal equation is *not* decoupled. You have to
> have second order terms for decoupling.
You have mentioned this issue before. I do not see its relevence. But since
you clearly think it is relevent, could you please explain why? (No I am not
trying to "set you up." I respect your knowledge and clear thinking and
would like to see where this, in your judgment, is important. Maybe it's
something I should have learned in school, but wasn't paying attention!)
:-)
>
All The Best,
Bill Miller
Bill Miller
"Vince Morgan" <vin...@TAKEOUToptusnet.com.au> wrote in message
news:4c85f232$0$2080$afc3...@news.optusnet.com.au...
>
<snip>
> And, a nice book with pictures might come out of it. In fact I think it
> should. Please!!!
>
Nice to hear from you again.
And yes, a more "student friendly" version would definitely be a useful
addition. I've thought about tackling it -- maybe as a Power Point
presentation for some local friends. But time is never an asset!
When J gets to the examples, there are a number of illustrations -- "lifted"
from his very good 60s textbook -- that show equivalence between "old
fashioned" and "new" derivations.
But the initial statements of the problem and the solutions he proposes are
bereft of any visual aids.
Not for the faint-hearted, but I am glad I slogged my way through it.
FrediFizzx
>
> "FrediFizzx" <fredi...@hotmail.com> wrote in message
> news:8einrr...@mid.individual.net...
> <snip>
>>> If we adopt your assumption above, then the question needs to be,
>>> "Since a changing E does not cause a changing H, then isn't it highly
>>> likely -- in spite of the (mostly) conclusive evidence -- that
>>> something other than a changing H is causing a changing E?" (I said
>>> "mostly" in the preceding question because Maxwell's equations -- as
>>> caisal descriptions -- do not explain Lentz's Law. E_k does.
>>>
>>> I *still* have not seen a convincing derivation using your suggested M
>>> and N. Are you still working on that? :-)
>>
>> What is the mathematical formulation of Lentz's Law? :-)
>
> This question took me a while, since I wanted to look at a few different
> sources.
>
> The first thing I learned is that thre is no "t" in the name. It's Lenz's
> law.
>
> The second is that Wiki simply iterates something like the verbal
> statement I made earlier. Namely:
> "An induced current is always in such a direction as to oppose the motion
> or change causing it"
>
> Panofsky ignores it. Others simply equate it with Faradays law, e = -
> dphi/dt, pointing out that Lenz introduced the minus sign.
>
> But this "law" was not derived from first principles. It is a law based on
> observation and *interpretation.*
But -dA/dt also explains it as Jefimenko points out since that is equal to
his E_k. But since it is just a "rule" and doesn't really have a
mathematical description, then I can't really give you a derivation based on
Lorentz forces and Newton's laws.
> Of course, that technique has been used over the ages. For example,
> Ptolemaic astronomers used it to establish the concept of epicycles. And
> QM folks use it extensively to develop the descriptions of QED that, as we
> know, are usable from the particle level all the way up to the... UM...
> UH... well... the particle level.
>
> <snip>
>
>>
>>> Air has molecules and electrons, so it would have been very easy to
>>> assume that since air had electrons, it would act just like a solid
>>> dielectric.
>>
>> The quantum "vacuum" also acts a dielectric IMHO.
>
> Does the quantum "vacuum" exhibit polarization? Or displacement?
It totally depends on interpretations. I believe it exhibits both.
>>> Maybe, but is there experimental evidence that demonstrates that the QVC
>>> actually exhibits displacement? It is my understanding that Mu and
>>> Epsilon sub zero have been tested extensively at "all" frequencies and
>>> power levels with no measurablechange?
>>
>> And why would you think mu0 and eps0 would change with displacement?
>
> Why would they *not* change? After all, they change dramatically in *real*
> media.
Huh???? Not in linear media.
> <snip>
>>> IOW Jefimenko's analysis of how a magnet works is totally incorrect?
>>
>> His mathematical model is not incorrect because you can model a magnet as
>> having bound current macroscopically that is all added up from
>> microscopic currents. But the true microscopic explanation is what I
>> said. So... what is the cause of the magnetic moments of charged
>> elementary matter? :-)
>
> I think *that* question is the subject of a thread that would be a whole
> lot longer, and far more contentious than anything in this or similar
> threads.
>
> However, in a (probably) futile attmpt to outdo Benj for "cookoo-dom," my
> "vote" is for Mills's electron model. That one, for a bound electron, is
> an infinitessimally thin "shell" of charge, (like the candy coating on an
> M&M) whose characteristics are that its charge is dynamically moving in
> such a way as to not have any Fourier transform components that are
> synchronous with lightspeed (ie no radiation) and a charge flow that is
> consistent with the measured magnetic aspects of the electron. Other
> charged particles are similarly defined. The model -- unlike those of QED
> or anything else I know -- provides startlingly high correlations with the
> characteristics of all elements and a huge number of inorganic and organic
> molecules.
>
> If anyone wants to talk about this further, I respectfully suggest a new
> thread!
Well since you mentioned it, I will say something else about it here. :-)
A "brane" model such as above is not bad but there is no "circulating
charge" so to speak. I do believe there is circulation of something within
the brane and this is the emergence of charge. Coupled with the idea of the
quantum "vacuum" as a medium of bound charge, the magnetic moments of
elementary particles is due to the interaction of the particles with the
elements in the medium. IOW, we are talking about both the emergence of
charge and mag moments at this level.
>>>> [snip uncommented text]
>>>>
>>>>>> Jefimenko was definitely not confused; he presented an extremely good
>>>>>> argument for E_k with many examples. But he kind of dropped the ball
>>>>>> on "Induction by Moving Magnets" Sect. 2-6 since he went back the
>>>>>> magnetostatic Lorentz force.
>>>>>
> <snip>
>
> OK.. let's *assume* you are correct. Then we have a situation wherein J
> *seems* to have "nailed it" WRT induction between coils -- moving or
> static-- but is wrong about the tossed magnet. Then where is the
> mathematical model that is *not* in conflict with J *but* accurately
> describes the tossed magnet as entirely a magnetic phenomenon? (Please
> use something other than dphi/dt!)
??? It is simply the magnetostatic Lorentz force, F = -q(v x B) as J shows.
Remember, the B field *is* moving wrt the charges in the wire loop. I guess
he was just doing a derivation of it to show that it could be modeled as
originating from microscopic currents. But we all know better that the
primary model for a magnet is as I have said from mag moments of elementary
matter.
>> Plus the fact that J's E causal equation is *not* decoupled. You have to
>> have second order terms for decoupling.
>
> You have mentioned this issue before. I do not see its relevence. But
> since you clearly think it is relevent, could you please explain why? (No
> I am not trying to "set you up." I respect your knowledge and clear
> thinking and would like to see where this, in your judgment, is important.
> Maybe it's something I should have learned in school, but wasn't paying
> attention!) :-)
Well, a common derivation of J's causal equations are starting from the
potential formulation.
E = -grad phi - dA/dt
B = curl A
For the E equation above, we can see that it is not decoupled as the phi
part is electric and the A part is magnetic. The first term on the right
corresponds to J's rho terms and the -dA/dt term corresponds to his dJ/dt
term. So even though it is not readily apparent in his causal equation for
E, there should be electric term and magnetic term in it. For complete
decoupling, you would need equations of the form,
grad^2 E = d^2E/(c^2dt^2)
grad^2 B = d^2B/(c^2dt^2)
Of course, these are the decoupled wave equations for E and B with no
sources but you can put the sources on them. I just don't have that handy
right now. So that is why I think his E_k should perhaps be labeled B_k.
But maybe A_k is more appropriate? It is kind of a funny thing that a
magnetic term is in the E equation but of course as I have shown before it
leads right to the point version of Farday's law when grad phi is zero.
E = -dA/dt
curl E = -d(curl A)/dt
curl A = B
curl E = -dB/dt
So it is just part of like what J says that E and B fields are primarily
created simultaneously. That fact does not actually disappear in his causal
E equation.
Best,
Fred Diether