Below are two errors that seem to occur on a regular basis, yet I have seen no one yet correct them.

These are fundamental faux pas and should not be propagated any further. They are:

1) Square waves contain all frequencies.

2) Transformers do not faithfully pass square waves. A 50% duty-cycle square wave applied to the primary of a transformer, will only yield a "short pulse" on its secondary. This is because most of the square wave time is spent on a steady voltage, either +V, or 0V... i.e. the wave form is hardly ever "changing" over time.

Both notions are absolutely false!

1) True square waves contain only odd order (integer) harmonics. A sawtooth wave form does contain all integer harmonics (one example).

2) All transformers have a finite bandwidth. Any wave form can be disassembled into its constituent sine wave components, a la Fourier. A square wave can be reasonably constructed using the first 5 sine waves, i.e 1, 3, 5, 7, and the 9th harmonics. The more harmonics applied, the more exact the resulting square wave shape will be. It comes down to transformer bandwidth, and what square wave frequency is applied to it. Too high an applied frequency, and the corners will start to become more rounded. In addition, the flat portions will begin to show "ripple" until finally, only the fundamental sine wave will appear. Too low a square wave frequency, and the flat portions will begin to tilt. At no time however between these two extremes, does the secondary exhibit a "short pulse-like" output. The transformer does not arbitrarily convert the square wave duty cycle to something other than what it is at the primary. Transformers therefore can pass square waves, and in fact any wave form faithfully, as long as the input frequency lies reasonably within the bandwidth of the transformer. If this were not the case, Audiophile tube amplifiers would not be possible.

Excellent!

What would be a correct description of rotating magnetic fields? This is something that irks me every time it is mentioned but I've learned to shut up about it.

Thanks!

As you probably already know. I can't agree with that description but that isn't the point or important at this time. For most practical purposes that is the way it should be considered.  What is important is that it is here and available for those who do not know. They can build upon that idea as they experiment.
The two things related here that I see:

1. The creator of the TPU has not used the term 'rotating magnetic field'.

EDIT>>> I stand corrected... There is comentary from the inventor and those around him that indicate a rotation of magnetic field is thought to be involved. Perhaps as ZPE described or maybe beyond my comprehension.

2. Alternating or switching a magnetic pole from one coil to another is not the same as the rotating bar magnet you have described, at any speed.

EDIT>>> Yes - pseudo would be a correct description. It works quite well with motors and generators. Indeed, this may be exactly how a TPU should work. I'm not sure at this point. However, using the two loops mentioned with a gradual shift of polarization should be very interesting. I'll try it.

EDIT>>>My intention was to create another reference point - not to start an argument. Such should be opened on another thread.

Entropy,

I'm not sure what your post has to do with mine actually.

I was simply stating in it's purest form, how I envision a RMF. It had nothing to do with Faraday's paradox. You are talking about rotating a bar magnet on it's own axis a la Faraday's disk dynamo.

I was talking about rotating a bar magnet on an axis formed through the center point between the poles.... i.e. with a 6 cm long bar magnet, the axis of rotation is at (and perpendicular to) the 3 cm mark.

z_p_e

Thank you from everyone here that needs such help (especially myself hehehe) for this great effort. I have questions galore but let's start with what's on the table already.

I read Faraday's Paradox can see it has no bearing on the turning of the magnet on its width central axis, but on the length central axis. It is obvious that on its length the magnet is only turning into itself so the same effect is always maintained as it was when it is stationary. I can also understand why the magnet would produce a magnetic ball field or close to a ball field on it width central axis.

BEP says that switching between two coils would not produce such a ball field and this is also understandable since you are now creating two north/south poles and not one turning.

Now what if you simply had like the open TPU either four single wound (collector) coils in series or four double wound coils both in series to make two separate and parallel loops, and what if you simply changed polarities on either one set or on two sets in opposition. Would this be considered a ball field?

If not, can we presume that an artificial ball field with coils is impossible if the coil or coils do not physically rotate?

"One thing is an absolute fact, and that is that the electron transit time of tubes is very-very fast
compared to transistors"......Marco

Transit time is defined as:

1: The time taken for a charge carrier to cross a given path
2: The average time a minority carrier takes to diffuse from emitter to collector in a junction transistor
3: The time an electron takes to cross the distance between the cathode and the anode

Switching time is defined as:

1: The interval between the reference time and the last instant at which the instantaneous-voltage responseof a magnetic cell reaches a stated fraction of its peak value.
2:The interval between the reference time and the first instant at which the instantaneous integrated-voltage response reaches a stated fraction of its peak value.

sources: Modern Dictionary of Electronics, Third Edition
   Principles of Electron Tubes, Bell Telephone Laboratories Series, D. Van Nostrand 1966

There seems to be some confusion on this forum of "transit time" vs. "switching time". With regard to vacuum tubes  and "electron transit time", the spacing between the electrodes creates a pure transport delay of information because the carrier electrons require some time to traverse the physical distance. This is not to be confused with "switching time" (which is more effected by interelectrode capacitance) as the information bandwith is not affected, just delayed.

As electrons start with near zero velocity at the cathode and are accelerated towards the anode based on the anode applied voltage, we then have at least two parameters that govern transit time, electrode spacing and applied voltage.

I understand (to some degree) the mechanism of electron transfer in a crystal lattice with doner atoms, but am baffled why this must be slower in all regards than tubes.

Transit time oscillators using semiconductors easily achieve gHz performance.

Perhaps someone can clarify why the claimed transit time of transistors is slower than tubes. Since I can adjust voltage and spacing, this seems to be a generalization. Your help appreciated.

Regards......V.

@Vortex1

Could it be that tranistors have to turn on and off whereas a tube is continuously on and continuously pushing electrons from the filament to the plate. Transitors will produce a more uniform controlled speed, whereas tubes will produce speeds that are inherent to the actual randomness of nature itself. This would imply that the actual speed of a tube is not steady but often times much much faster and sometimes slower than a transistor but in the end, if the plate is considered to be a tank, it is getting filled faster.

The question you ask would be like asking why is a capacitor discharge faster then a straight DC pulse, in its ability to travel through coil windings. Geez, or is it?

@Sparks

I don't know about the black hole as a magnetic ball, but I have thought about this when considering how gravity works. I visioned a tiny black hole in the center of the Earths core and everything is falling into this hole, but the Earths mass, rotation and magnetic field is holding all or most of the mass out of the hole - for now. lol