A Paper on the Propagation of Magnetic Waves in Soft Iron

[F6FLT]

@Smudge

I still don't see any sense in what you are saying. You are comparing a conventional line terminated by a load equal to its characteristic impedance, with a line of your design, which is not.
But a conventional line, terminated by a reactive load, also generates standing waves and also has an input impedance that can be equal to anything, depending on the impedance of the terminal load and the frequency, exactly like your line.

Then a negative resistor is not an LC circuit with negative L or C, which could generate an AC signal. A negative resistor produces only DC. This is inconsistent with the fact that the negative resistance seen at the input would depend on the frequency. On the frequency of what? Of a signal that we don't even have to generate?

Finally, multiple resonances are commonplace in classical lines when they are loaded by a reactive load. So it is not a specificity of your line. You seem to be unaware of how classical lines without a suitable load work and you attribute new effects to your line when, apart from negative R, it is commonplace in all lines.

[Smudge]

You seem to be unaware of how classical lines without a suitable load work and you attribute new effects to your line when, apart from negative R, it is commonplace in all lines.

I left school at age 16 and went straight into work as a Scientific Assistance at a research establishment.  My entire working life of 49 years was in electronics and electromagnetics during which I schooled myself on General Relativity in order to understand some of the weirder aspects of EM.  I rose to the giddy heights of Chief Systems Engineer in the division I worked for.  So I guess my knowledge of how classical lines work is at least as good as yours and possible better.  I was deeply involved in near-field radar as proximity detectors for shells, bombs, missiles, torpedoes and mines, so I understand the difference between far-field (where the E and H vectors are in space quadrature but in time phase) and near-field (where the E and H vectors ore not in phase).  The far-field has the so-called wave impedance of 377 Ohms and that is a real resistance, not a reactance.  The near-field has an impedance that varies with distance from the transmitting antenna; for the classical small electric dipole the impedance rises as you get nearer to the antenna and for the small magnetic antenna it falls (small meaning antenna dimensions smaller then a wavelength).  In both cases the resistance value changes while the reactance value rises so that close to the antenna the wave impedance is either capacitive reactance of inductive reactance.  The classical delay line that you and most people are familiar with has a characteristic impedance that is dominantly resistive.  I can find no reference to anyone with a knowledge and experience of a delay line that has a reactive impedance.  I am sure that you have no such experience.

Transmission line theory readily accepts a reactive line impedance and when you look into this it predicts a negative input resistance when the line is terminated with a capacitance.  Now you might argue that such a line with its mathematically imaginary impedance is actually imaginary, it doesn't exist in real life, it is only in one's mind.  You have not presented that argument, you have continually claimed that any line must be modelled by actual L, C and R components and these cannot result in a negative resistance.  I have argued that magnetic field propagation along a core or a pair of cores IS a transmission line having reactive line impedance, and that is something that has been overlooked in science.   It can NOT be modelled by the usual LC network.  With my self-taught methods of solving problems in the magnetic domain I have perhaps muddied the waters by using an LC network where the Ls and Cs are magnetic ones relating flux to mmf in the same manner that actual Ls and Cs relate current to voltage.  If that has created confusion I apologise.

Wave propagation along a core has been looked at and it is known as magnetic viscosity.  Its usual effect is to create losses, and this is easily modelled by introducing a time delay between input and output that converts the linear BH characteristic into a hysteresis loop that is traversed CCW.  A CCW loop represents a loss, and it is assumed that this loss must go as heat in the core.  I can find no evidence of calorific measurements that verify this input loss is actually converted into heat.  I can hear you challenging this saying where can the energy go except as radiation or heat?  We know that the electron dipoles are very active, they precess at the Larmor rate, free electrons are whizzing about, orbital electrons are hopping into and out of their stationary orbits, there is a lot of activity going on.  If we can somehow connect into that activity, is it not possible that we can both source and sink energy there.  If the activity really is a connection to an active aether why can't that supply or sink energy?

Within the core material magnetic wave propagation is very complex with electron dipoles obeying quantum rules involving their orbital or inherent spin, dipoles flipping, domain walls moving etc etc.  Barkhausen jumps are a known effect; before the jump, as an externally applied field increases value the total field obeys dB/dH = μ₀, there is no relative permeability until a dipole flips, then there is the field jump to an increased value.  What happens if the H field is pulsed yielding a dH over a time dt that takes B to the point of initiation of a Barkhausen jump, then the dH is suddenly reversed, do we get a jump in B while H returns to its original value?  Of course in normal transformers these many jumps are not seen, the material behaves as though the accepted permeability rules apply dB/dH = μ_R μ₀.  But within the material there could be small discrete volumes where B changes by μ₀dH in synch with H, then the jump by μ_Rμ₀dH occurs.  Who knows what happens when there are H waves traveling both ways within the core influencing those small regions, what is the overall effect?  Has this been studied?  Shouldn't we be studying these possibilities?  If transmission line theory predicts an anomalous effect for a line with reactive impedance, and our only known line of this type is a transformer core operating in an unusual manner, and that core's characteristics come from active elements within the core, isn't there a possibility, however slight, that we could make it behave in an active manner.  I certainly believe so.

Smudge

Edit, changed CW to CCW to correct an error brought about by a senior moment.

[F6FLT]

@Smudge

You go into digressions. I don't doubt your technical skills regarding engineering, but I don't doubt your way of embroidering around it either. The fact that you have experience in a field where I too have experience without ever having noticed the slightest anomaly, and where, I suppose, no OU of any kind has ever been noticed nor used by the companies you worked for, is therefore of little relevance when it comes to talking about new phenomena. More than ever, the argument of authority does not hold.

A line being made of real resistances, real inductances and real capacities, it can be modelled by these elements. The difference with a simple network is the propagation time which plays a role when it is not negligible compared to the period of the signals, which is the general case of the use of lines. For this reason, the input impedance of a line depends on the frequency. This is the same problem we have in designing broadband antennas to ensure a constant impedance throughout its frequency range.
The only case where the input impedance of a line does not depend on frequency is when the resistive terminal load is equal to its characteristic impedance, which is not the case here since the terminal load is assumed to be reactive.

So there is no sense in talking about negative input resistance if you don't specify at what frequency. I think I found the question of frequency in your text, you see I looked for it and I am of good will, it is according to me in the phase constant β and thus θ too: the length of line θ=β*x being expressed in radians, it is relative to the period of the signal.

Now according to the line impedance given in your equation (5), R is not necessarily negative when X0>X_load, but only when tan(θ) * (X0-X_load) is negative because it also depends on tan(θ), so at certain frequencies relative to the line length. Is it possible to choose X_load such that tan(θ) * (X₀-X_load) < 0? Of course, since X_load can be chosen arbitrarily, so the error is not there.
So where does the error come from?

It is in the sentence "We are interested in the case where Z0 is purely reactive (say jX0)". In the absence of losses, a characteristic impedance is always resistive since it depends on the ratio of the impedance of an inductor to that of a capacitor (or the reverse if C and L are swapped), so omega disappears. With losses (R series in L and R // in C) the characteristic impedance can be reactive, therefore frequency dependent, but never "purely reactive" and it will always remain to be demonstrated that R<0 at the input will not be cancelled out by this resistive part of the characteristic impedance, due to losses.

Concerning the physics part of your development, I have no opinion for the moment, it is a lot of ideas to see but without experimental proposal, difficult to know if the tracks are valid.

[Smudge]

..........So where does the error come from?

It is in the sentence "We are interested in the case where Z0 is purely reactive (say jX0)". In the absence of losses, a characteristic impedance is always resistive since it depends on the ratio of the impedance of an inductor to that of a capacitor (or the reverse if C and L are swapped), so omega disappears. With losses (R series in L and R // in C) the characteristic impedance can be reactive, therefore frequency dependent, but never "purely reactive" and it will always remain to be demonstrated that R<0 at the input will not be cancelled out by this resistive part of the characteristic impedance, due to losses.

There you go again, you are fixated on the magnetic transmission line acting like a classical transmission line where the characteristic impedance "depends on the ratio of the impedance of an inductor to that of a capacitor".  Were it so I would agree with your statements.

The magnetic line is not classical.  When you compare the twin wire line and its distributed L and C with the twin ferrous rod line there are significant differences.  Both have E and H fields in the space outside the "conductors" but is I show they are swapped over.  And as I stated the phase between the E and H fields is different.  In the twin wire case not only is Z₀ related to the ratio of L and C but also the propagation velocity is related to the product of L and C, and these also apply to the transverse E.H wave travelling along the line with E and H in phase as in far field radiation.  In the twin ferrous rod case the propagation velocity is primarily determined by the "conductor" properties, its magnetic viscosity.  And with the E and H fields in phase quadrature the wave is like near-field radiation.  In the twin wire case the E x H Poynting vector carries the energy along the line.  In the twin rod line with E and H in phase quadrature what does the Poynting vector do?  It is certainly different.  I am sorry to labor this point but that 90-degree phase between E and H tells me the line has a reactive Z₀, it is not a classical transmission line.

Concerning the physics part of your development, I have no opinion for the moment, it is a lot of ideas to see but without experimental proposal, difficult to know if the tracks are valid.

But surely worth investigating just in case it leads somewhere.  Below is a short paper showing the positive reluctance of a core as the slope of flux v. mmf, where introducing a time delay between the application of mmf and the flux creates a CCW loop as an energy loss.  Also shown is the negative reluctance induced into the magnetic circuit by a capacitively loaded coil.  Of interest is a time delay between the mmf (current) of that coil and flux which creates a CW loop, indicating an energy gain.  If the area of the CW loop could be made larger than the area of the CCW loop then there would be an overall gain.  Since the magnetic viscosity plays its part in both loops then this possibility may seem improbable.  But the magnetic wave propagation can also be influenced by surrounding the ferrous rods with high K dielectric, which increases the magnetic delay.  A series of coils along the rods each one shunted by a capacitor also increases delay.

This argument may also play into the Holcomb claims.  They may have got their arguments wrong with regard to how electron spin accounts for their results, but if they really do have positive results the energy comes from somewhere, and it seems the magnetic steel is the important part.  The Manelos and Sweet devices also obtain their gains from the ferrous material.

Smudge

[F6FLT]

...
The magnetic line is not classical.  When you compare the twin wire line and its distributed L and C with the twin ferrous rod line there are significant differences.
...

I remind you of what you wrote: "The impedance Zin looking into a line of characteristic impedance Z0, of length x(m) and terminated in an impedance Zload is given by ..."

All your development is based on the classical equation of the impedance of a line. This leads to inconsistencies that I have pointed out. You only answer them by claiming that it is not a classical line. Then the use of your starting equation is not justified, which invalidates your conclusions of negative R.

[Smudge]

I remind you of what you wrote: "The impedance Zin looking into a line of characteristic impedance Z0, of length x(m) and terminated in an impedance Zload is given by ..."

All your development is based on the classical equation of the impedance of a line. This leads to inconsistencies that I have pointed out. You only answer them by claiming that it is not a classical line. Then the use of your starting equation is not justified, which invalidates your conclusions of negative R.

You have made the assumption that the classical transmission-line equations I have used can only apply to classical lines.  That is not the case, they are much broader than that.  You and I and the rest of the world have extensive experience of classical transmission lines where the characteristic impedance is non frequency dependent and is resistive.  The line equations of course cover those classical lines, but they also cover non-classical lines where the characteristic impedance is non frequency dependent and is reactive.  I suspect that you and the rest of the world do not knowingly have experience of such non-classical lines and would find it hard to describe a line having that characteristic.   Your criticism is unfounded.

Smudge   

[F6FLT]

@Smudge


The only thing I'm saying is that if you use the classical line equation to describe your own, and it results in inconsistencies, then your choice to use it is not justified.
But you don't answer the objections.

The main inconsistencies are :
Negative R only appears in the equations at certain frequencies, but negative R does not allow the generation of an AC current.
The equations of electromagnetism, including those of lines, guarantee the conservation of energy.

[Smudge]

@Smudge

The only thing I'm saying is that if you use the classical line equation to describe your own, and it results in inconsistencies, then your choice to use it is not justified.

Why not justified?  Because of the inconsistencies?  Shouldn't we be looking into what might be causing those inconsistencies?

But you don't answer the objections.

I have even offered an alternative method of looking into the effect of the inconsistency.

The main inconsistencies are :
Negative R only appears in the equations at certain frequencies, but negative R does not allow the generation of an AC current.

Oh really, have you tried solving the time evolution of current in a L C series negative R circuit?  You would find it instructive.

The equations of electromagnetism, including those of lines, guarantee the conservation of energy.

So you keep saying, and if you believe that there is no way of going beyond those equations why are you on this forum?

Smudge

[F6FLT]

...
Oh really, have you tried solving the time evolution of current in a L C series negative R circuit?  You would find it instructive.

I would be curious to see your equations that would show the maintenance of an LC oscillation with only negative R!
No more need for a transmitter, no more need for electronics, because R negative being like a DC generator with R positive, you would just have to connect an antenna via an inductance and a capacitor, to the car battery...

So you keep saying, and if you believe that there is no way of going beyond those equations why are you on this forum?

Seriously, do you believe that in the equations that guarantee the conservation of energy you will find a non-conservation?
You think that by staying in your country you will see what is beyond?

I remind you that science models what we observe. The models guarantee the conservation of energy of a closed system because we have never observed anything else. No matter how you twist the models, since they have an internal mathematical consistency, if you see OU in them, you can be sure that the error is yours.

In case of real OU, you will either have to use new models, or you will have to integrate the discovered energy source into the old models.
The OU will be in the discovery, not in the known equations. That's why I'm here. The known equations are a starting point, to be completed, they do not contain their own beyond.

[Smudge]

I would be curious to see your equations that would show the maintenance of an LC oscillation with only negative R!

Not maintenance of an LC oscillation but an exponential build-up of the LC oscillation (following an e^+at rise in the envelope) rising to infinity.  If there is positive R there as well the oscillations build up to a fixed level.  Is done in any RF oscillator where the negative R is created by positive feedback but the build up is often not of interest.  It is of interest in super-regenerative amplifiers where the oscillations are built up then allowed to decay and this is repeated at the quenching frequency.   

Seriously, do you believe that in the equations that guarantee the conservation of energy you will find a non-conservation?

Depends on what equations you are talking about.  In the case of the transmission line equations, if they apply to non-classical lines that involve propagation that is not related to EM wave equations but to other dynamics within materials such as atomic dipole flips, quantum rules, fermi velocity of conduction electrons etc. etc., then yes I think there is a possibility that there is a connection to an active aether.

I remind you that science models what we observe. The models guarantee the conservation of energy of a closed system because we have never observed anything else.

If an active aether is the source of the energy then our experiment is not a closed system.  But of course you don't believe in an active aether.

In case of real OU, you will either have to use new models, or you will have to integrate the discovered energy source into the old models.
The OU will be in the discovery, not in the known equations. That's why I'm here. The known equations are a starting point, to be completed, they do not contain their own beyond.

I could not agree more.  Why then do you pour cold water on any attempt to discover an energy source?

Smudge

[F6FLT]

Not maintenance of an LC oscillation but an exponential build-up of the LC oscillation (following an e^+at rise in the envelope) rising to infinity.

This is a triviality, like applying a voltage swing to an LC circuit. But this is not the case for the line, where the supposedly negative resistance does not appear instantaneously.

If an active aether is the source of the energy then our experiment is not a closed system.  But of course you don't believe in an active aether.

Your OU, you don't get it from the ether but from a conventional equation that makes no assumption of ether.


Digressions and non sequitur, it becomes nonsense because it has no logical relation with your initial paper, the one I am contesting.

[Smudge]

Your OU, you don't get it from the ether but from a conventional equation that makes no assumption of ether.

But it does make an assumption of a line that can't be made from L and C and R.  So it is possible that another form of line that has the "impossible" characteristic could yield that "impossible" result.  I even demonstrated how the capacitively loaded coil produces a CW flux v. mmf loop in the presence of a time delay and that did not use any equations.  As you know a CW loop represents an energy source that can be modelled as a negative R.  So the magnetic delay line with that "impossible" characteristic is worth investigating and if it did offer OU then the energy source could well be the active aether.       

Digressions and non sequitur, it becomes nonsense because it has no logical relation with your initial paper, the one I am contesting.

But it could clarify why the conventional equations yield the "impossible" result, and surely that is worthwhile?

Smudge

[F6FLT]

But it does make an assumption of a line that can't be made from L and C and R.
...

I never said it was impossible. But what do you answer except to your own subjects?!
Your fallacy: https://en.wikipedia.org/wiki/Straw_man

I said that maintaining an oscillation was impossible with negative resistance.

Once again what you answer is irrelevant, the rhetorical pirouettes are getting to be too much.