TK,

Scholasticism in this instance is appropriate, I would think. Considering you initiated a debate about the definition of an abstract term (alternating current). Your theatricality in making a mountain out of a mole hill seems to be the real issue. I've repeatedly agreed with you that any DC-offset is a real concern. I've also pointed out that AC measurements ignore its presence and the fact that magnetic circuits become non-linear due to it causing saturation. Which means you would see its affects on the oscilloscope as distortion--regardless of coupling. Yet you pull another AC-coupling argument out of your magicians hat, that doesn't compare with the actual argument (Alek's transformer), and demand that I address it and say that I'm some how wrong on all my points. Pretty pathetic if you ask me.

Tinman,

As for the DC current and ripple on the voltage topic. The magamp/saturable reactor produces this. Its basically the effect of a voltage source or current source producing a "counter wave" that subtractively adds to the ripple produced by a load. For instance, take a reluctance motor and connect it to a DC power supply with a CC current limit of 1A. Since the supply is shorted by a few hundred milli ohms it goes straight into CC mode of 1A. Now rotate the rotor as fast as you can and observe both the voltage across the motor and the current through the motor. You will see that the voltage is proportional to I_cc*dL/dt or the derivative of the change in inductance with time (there is no generator action taking place). Current remains a constant 1A. What's happening, is the current source is raising and lowering its voltage to maintain a constant current against a changing load impedance, in this instance a reactive one. So the effect, as far as I'm aware can only be seen when a constant voltage source or current source is used to power a time-variant impedance.

Quote from: TheCell on August 07, 2014, 11:01:51 AM
If the output from the amplifier is coupled by a capacitor (in most cases it is, cause you don't want a bias dc current through your load ; for example a speaker box) what's wrong with measuring  input power having a ac - coupling setting in the scope. The ac frequency in the kHz-Range, that's way beyond the cut-off frequency of the scope. A scope can render mains frequency with no problem using AC-coupling .
If you lower the   amplifiers frequency under the cut off freq. of the scope the internal cap in the scope attenuates the signal, but with a frequency in the khz Range this effect is negligable.
A leakage current from the secondary to the primary due to bad insulation results in additional power consumption on the meter.
In fact in the kHz Range there could be a inductive component in the Load Resistor , but Bill says there were no phase shift . So it's our choice to believe him or not.

A valid measurement is one that responds to all significant features of the quantity being measured.  A scope that is AC coupled with a cut-off frequency that is much lower than the signal being measured captures the excursions well.  If there is signal information that is significant that is outside the pass band, then as in TK's bulb experiments the measurements will miss that information.  Bill Alek's choice to AC couple was likely one of convenience.  I do not see high risk of an error due to AC coupling.

Quote from: G4RR3ττ on August 07, 2014, 02:40:16 PM
TK,

Scholasticism in this instance is appropriate, I would think. Considering you initiated a debate about the definition of an abstract term (alternating current). Your theatricality in making a mountain out of a mole hill seems to be the real issue. I've repeatedly agreed with you that any DC-offset is a real concern. I've also pointed out that AC measurements ignore its presence and the fact that magnetic circuits become non-linear due to it causing saturation. Which means you would see its affects on the oscilloscope as distortion--regardless of coupling. Yet you pull another AC-coupling argument out of your magicians hat, that doesn't compare with the actual argument (Alek's transformer), and demand that I address it and say that I'm some how wrong on all my points. Pretty pathetic if you ask me.

Tinman,

As for the DC current and ripple on the voltage topic. The magamp/saturable reactor produces this. Its basically the effect of a voltage source or current source producing a "counter wave" that subtractively adds to the ripple produced by a load. For instance, take a reluctance motor and connect it to a DC power supply with a CC current limit of 1A. Since the supply is shorted by a few hundred milli ohms it goes straight into CC mode of 1A. Now rotate the rotor as fast as you can and observe both the voltage across the motor and the current through the motor. You will see that the voltage is proportional to I_cc*dL/dt or the derivative of the change in inductance with time (there is no generator action taking place). Current remains a constant 1A. What's happening, is the current source is raising and lowering its voltage to maintain a constant current against a changing load impedance, in this instance a reactive one. So the effect, as far as I'm aware can only be seen when a constant voltage source or current source is used to power a time-variant impedance.

When the power supply is in CC mode the terminal voltage across the motor varies directly as the generator voltage.  If the motor spins fast enough in a direction where the BEMF opposes the power supply, then the power supply will drop out of CC mode and go into CV mode.  If the motor is spun fast enough that the generator voltage exceeds the power supply CV setting, then what happens depends on whether the supply is a one quadrant or two quadrant device.

Quote from: MarkE on August 06, 2014, 06:35:48 PM
A CM choke is just a transformer. If the K factor is high, then the effective inductance seen by each side is twice that of either single winding in isolation.  Bill Alek's windings have low coupling coefficients:  KI think that you don't understand them well.  The idea is that the reactance of the magnetizing inductance in each winding is much greater than the impedance of the external circuit.  Current impressed from the dot end towards the non-dot end of one winding induces a voltage across the other winding that will ideally result in a matching current from the non-dot end towards the dot end of that second winding.  In a perfect world this results in nearly equal currents flowing in opposite directions, reducing the net common current in one direction or the other out of the choke to be very small compared to the original individual currents.  A common mode choke that has a low K factor performs badly.I always like to see interesting data.Take an ideal transformer and drive an inductive load.  The inductive load reactance reflects right back at the primary.  Shorting the secondary of a transformer with a low K has the same effect:  the leakage inductance becomes the load.  In a weakly coupled transformer, the primary current phase shift gets very close to 90 degrees for either condition:  open or shorted secondary. Low phase shift is possible with a tightly coupled transformer such that the magnetizing reactance is much greater than the load plus winding resistance and the leakage inductance reactance is much lower than the load resistance. 

Bill Alek's windings are labelled:  120mH / 122mH for the secondaries, and 3.07mH for the primary.  Let's assume that those values were obtained with an LCR bridge one winding at a time with each of the other two windings open.  Using K values of 0.8 and loading with 0.01 Ohms, the phase shift at 3kHz in the primary is 84.6 degrees.  Loading with 1E9 Ohms (open) the phase shift is 88.6 degrees. Up the K to 0.99 and the 0.01 Ohm condition gets much better due to the winding resistance: 28.2 degrees phase shift, while the open circuit case remains unaffected.

I think that what we see is just a combination of weak coupling and poorly conducted measurements.  That seems to be the legacy of over unity transformer claims.

Hi Mark, In the video linked below I use my little resonant setup to show a phase shift in the output coil (tank) from about
30 degrees to almost 90 degrees when the DC output is loaded by a motor which I stop the shaft of with my fingers to load it.
The unloaded input power is small and the loaded input power is larger, when the motor is stopped the "power" the tank
becomes almost all reactive. but before that it must be something else as well when the phase is about 30 degrees.

My arrangement is as you may have seen it in the QEG thread. I must add that I tuned the HV tank so that this would happen
I can also adjust it so that there is maximum power input with no load, but that's another video clip.

Now if I was to show only the HV tank power and the output tank power with the motor shorted and got a 90 degree
phase shift between voltage and current in the output coil 'some' people might think that was OU.  Or some other load might
do it if I make the correct adjustments.  As long as I don't show the input from the wall, only from somewhere after the wall
and maybe after another transformer. I think it can be done, shown and explained why it appears so.

Anyway I agree with you as my experiment tells me too. 

It costs input power to make "reactive power". If we look back far enough we will see the cost.

Phase shift to almost all reactive on loading resonant tank output.
https://www.youtube.com/watch?v=T1_llp5QUSM

..

Quote from: MarkE on August 07, 2014, 02:23:58 PM
I have proposed an experiment that you can try with your materials.

That sounds rather convenient for you, seeing as how you want me to do all the work. Quite one sided, when you think about it. In reality neither of us has any respectable "authority" on the subject of discussion, or do you claim to have written a book, be a professor, or some other position (a working scientist?) that actually possess any sense of mastery of the subject? And I don't buy "I'm a working engineer," they aren't the people who write in journals or publish books that everyone else uses as reference material. Experience is one thing, but it doesn't convey complete authority that you are above doing your own work and showing it. Also, lighten up man, your writing style is borderline Aspergers.