Testing the TK Tar Baby

[TinselKoala]

TK:

I can offer you a theory with moderately high confidence for the super-bright mode.  It was more or less confirmed in the earlier comments.

I am assuming that this was built with the intention that both resonators would have the same frequency.  If course that's not possible without tuning and many types of caps have a tolerance of +/-20%.

Yes, that's right, but the poly roll caps that I used on the transmitter are 5 % tolerance, and measure right on with my silly ProsKit meter. The poly caps in the receiver match, to the accuracy of the meter. But you are right, however I think this circuit is more sensitive to the inductance... probably precise snipping of the loops would be the way to tune. Marko showed some effect of inductors inside the loop but I haven't been able to see any yet.

In normal mode the transmitter is generating the near-field and the receiver is picking it up and lighting the bulb.  You are off of the resonant frequency of the receiver so everything acts like a normal filter.  The Q-factor for the receiver is not so narrow, so you get the bulb lighting up.

When it jumps into super-bright mode something interesting is happening.  The receiver LC resonator wants to resonate at its center frequency and it's response to the external excitation from the transmitter is being "felt" by the transmitter.  This dynamic instantaneous impedance due to the receiver as felt by the transmitter gets stronger as you bring the two loops together.

What eventually happens is that the resonant frequency of the receiving LC resonator is "forced" onto your LC transmitting oscillator and the LC transmitting oscillator starts to slave to the LC receiver.  The LC transmitting oscillator is not running at its natural frequency at all, but the nature of the "load," the instantaneous impedance of the LC receiver, gets the transmitting oscillator to sync up to the load.  Once they are synced up, then you have true resonant near-field energy transfer - the same thing that they have for certain laptops and cell phones where you just put them on top of a special mat.

When you move the LC receiver away from the LC transmitter, the "forcing power" of the synchronization goes away until eventually the main transmitting LC oscillator "breaks free" and goes back to its natural operating frequency.

I think that's the way the cookie crumbles.  The transmitter falls under the influence of the receiver and starts to oscillate in sympathy to the receiver.

MileHigh

I think all the rest of that is right. But the non-inverse square falloff... how do you account for that? If it was simply a matter of being in true resonance, it would seem that you would still get the 1/r² dependence. But what seems to be happening is that the system increases the output power to compensate for the increased distance, and the bulb stays the same brightness. Is this a result of it being somehow "phase locked" and trying to stay in that nice true resonant notch?

[TinselKoala]

 Wait... I'm not sure about that. How could the transmitter not be running at its own natural resonant frequency? I agree that the mutual interaction forces a true resonant condition between the two LCs... but I think it's the receiver that needs tuning, not the transmitter.

ETA: OK, I read it again and I see what you mean now. The xmtr is _normally_ running at its natural resonance, but the interaction with the receiver pulls it down to the receiver's resonant frequency and this drops the impedance of the system and the light gets brighter.

But even if the impedance of the transmitting system is zero, the light should only be as bright as it is under DC, right?

[MileHigh]

TK:

You took the words right out of my mouth; this setup is essentially a form of phase-locked loop.   The "loop" is the influence the receiver has on the transmitter.  So as long as they are locked, you don't really see much 1/r^2 losses.  Also, you are near-field so you won't see 1/r^2 losses in the near field.

When you get far enough away, you "break lock" and the frequency of the transmitter jumps back to its normal operating frequency.  There is no linear change in frequency, by definition it's going to be a jump.

What I would do would be to tune the transmitter LC with a variable capacitor (the loops are perfect, don't touch them) and then tune your MOSFET oscillator (assuming it is separate from the transmitter LC) and then you should be in really great shape.

Based on what you have observed so far, if you assume that the transmitter and receiver have natural frequencies within 1% of each other, you will probably get a "lock" going right away and it should be quite robust.  Whatever you do, you don't want to make major modifications to the transmitter because you want to preserve that "secret sauce" "locking susceptibility."

Looks like fun!

MileHigh

[MileHigh]

TK:

But even if the impedance of the transmitting system is zero, the light should only be as bright as it is under DC, right?

No because think of this thought experiment:  No bulb, and the receiving loop is ideal and the capacitor is ideal.   If you start your transmitter under these conditions then the AC voltage and AC current in the receiver will rise linearly over time until they get so high and eventually stabilize where all of the input energy is being radiated out as radio frequency waves.  Let's assume that would be at a super high voltage.

When you add a bulb we know that "lots of energy" is transmitted from the transmitter to the receiver.  How much is I suppose bottlenecked by the physical space and size of the loops relative to the characteristic impedance of "free space" - some air in a vacuum.   How much energy density can the free space between the loops sustain is the question  (that's very dependent on frequency also of course) .  I doubt the battery and the MOSFETs are the bottleneck.

So when you go back to the bulb, you have "X" power being picked up by the resonating receiver loop.   That power is split between the resistance of the wire in the loop and the resistance of the bulb filament.  Naturally almost all of the power gets dissipated in the filament of the bulb.

If you did the same setup but upped the frequency by 10X, then it's safe to assume that you could make the bulb burn that much brighter.

MileHigh

[MileHigh]

TK:

Disclaimer and a final comment.  Of course I am just flying by the seat of my pants here from the courses I took ummm......32 years ago.

Certainly energy density goes up as frequency goes up.  So the higher the frequency the more power can be transmitted across a medium.  If I recall you fly and you are probably aware that in planes the AC is 400 Hz instead of 50 or 60 Hz.  That way smaller transformers can be used.  So the same principles apply at much lower frequencies also.

Microwaves, gamma rays, part of the nastiness is that they can convey so much power because of the very high frequencies.

MileHigh