Mysterious Resonant Circuit

[Peterae]

My torroid is totaly different dimensions to EM's, it may be important to build on plug in dev boards due to the in built capacitance between the pin rows.
Also note so far we have both used batteries, although mine are nicad.

[EMdevices]

I second what Peterae said,  the transistor is critical.   I think it's that high hfe constant, which for me it's 200.   I also tried my more robust transistor for high power and it did not work (it has a lower hfe)

@xee, thanks for the diagram, certainly what I had in mind.    Regarding the flux and the phases, I would add that we need to specify what quantity were measuring.    When I say, phase has to be the same, I'm refering to the voltage phase.  so if V = N dphi/dt, the voltage is wholy dependent, or driven by the flux,  but the CURENT will have a different phase depending on the components we hook up to the windings.    Since my VOLTAGES are out of phase, this can mean a few things,  like Mannix said  1)  ROTATION,  or like I suggested 2)  Electrostatic couplings, or perhaps another explanation.


A slight variation:

I hooked up another toroid to my circuit to help isolate the phenomena.   This new toroid has only two windings, that were wound at opposite sides of the toroid, so they DO NOT overlap.  I wanted to make sure there was no electrostatic coupling, that's why.  One winding has 40 turns of fine wire and is hooked to the transistor Colector and the filtered (+) rail,  the other coil has 8 turns and drives the Base.    Well guess what?  It still workes!  Scoping at the Collector, I see the same waveform, with the typical short drop into the negative voltage range.  The voltage swings up to 17 or 15 volts peak just like before.   I hooked up a variable plate capacitor and tried to varry some capacitances to see what has a profound effect.    It seems the base circuit is what realy sets the oscillation frequency and bringing my hand close to the 8 turn winding increases the capacitance and the frequency goes down  (it's around 5 MHz in this case and droping) This by far is the most sensitive way to affect frequency.  The capacitor that hooks the 100kohm resistor and 8 turn coil to ground, needs to be a low ESR capacitor so the RF signal can have a good path to ground and over to the transistor  (base emiter junction).     This capacitor does not affect the frequency that much beyond a certain point, so the resonant mode seems to be dictated by only the 8 turn coil and it's interwinding capacitance (and maybe junction capacitance in the transistor), much like the TESLA secondary coil.     The many turn collector coil just pumps this tank circuit.   This is a very close setup to an Armstrong oscillator, except for this odd negative dip in the collector voltage waveform.   Since the collector driven winding has 40 turns or so,  and there is no output resistor to suck energy, I measued the input amperage to this oscillator and it was below 1 mA, and my scope just showed zero on the screet to 3 decimal places.  (don't worry I calibrated it with a 1 kohm resistor to make sure it works)

Anyway,  my 3 coil torroid is a bit different since it has all the capacitors and resonating modes, but at least now I'm begining to understand something more about the dynamics of this oscillator.

EM

[Yucca]

You also need a right toroidal winding - yours does not look right. (white wire is roughly wound, spaced far apart, and it is not synced to 3 base windings)

Thanks for the heads up on that.

[Yucca]

@ Yucca
Your documentation is good but I am still confused. Your scope shot seems to show 32 MHz across the power supply resistor. Is your DVM rated for reading AC current at 32 MHz? If not that could be why the DVM reading was bad. EMdevices was filtering out the AC and just reading the DC current.

You're right, I wouldn't trust my DVM on anything but pure sin < 1khz. The 32MHz signal only accounts for 10% or less of the signal the rest is DC underneath, so I thought that even if the DVM didn't see any of the AC it wouldn't matter too much.

I'm gonna put some more time on this when I get the right transistor. As Peterae said it seems that this particular transistor is needed.

Fraser.

[eldarion]

Hi Everyone,

OK, I got round to building and testing the circuit, however I have used a 2N3904 transistor, so this isn't a replication per-se. Also my supply is a not a semi-depleted 9V battery but it's a bench supply set at 9V. Also I have 100 ohm over output instead of 150 ohm.

Thus far does not show cop>1 for me.
I need to get the right tranny.

Fraser.

@Yucca,

Your waveforms are very close to what I got, as is your amperage.  Get rid of that iron powder core!