Angstrom wrote ...

I think he meant that in general the impedance of a capacitor goes down with increasing operating frequency (think high pass filter)... but in our special case (DRSSTC resonant circuit) you are altering the properties of the capacitor (capacitance) which changes the resonant frequency which, due to feedback, changes the operating frequency in the opposite direction. The impedance also goes in the opposite direction.

|Zc| = 1/(2pi*f*C)

The general case would be true if we were changing the frequency elsewhere (such as primary coil).

Steve McConner wrote ...


No, this is also wrong. The impedance of a tank circuit is sqrt(L/C), therefore increasing C will lower the impedance as well as lowering the resonant frequency. Increasing L will increase the impedance and lower the frequency, and so on.

Steve McConner wrote ...

Yes, you are going crazy. When a tank circuit is operating at its resonant frequency, the impedance of the capacitor and the impedance of the inductor are the same, so we call this value the characteristic impedance of the tank circuit.

That's one definition of resonance, in fact: the frequency where the inductive and capacitive reactances, being the same magnitude but opposite signs, cancel each other out leaving only the resistance of the conductors.

To be sure, a DRSSTC's primary tank doesn't operate quite at its resonant frequency, because of frequency splitting from the two coupled tank circuits. But the difference is only 5-10 percent.

Steve McConner wrote ...

Yes, you are going crazy. When a tank circuit is operating at its resonant frequency, the impedance of the capacitor and the impedance of the inductor are the same, so we call this value the characteristic impedance of the tank circuit.

That's one definition of resonance, in fact: the frequency where the inductive and capacitive reactances, being the same magnitude but opposite signs, cancel each other out leaving only the resistance of the conductors.

To be sure, a DRSSTC's primary tank doesn't operate quite at its resonant frequency, because of frequency splitting from the two coupled tank circuits. But the difference is only 5-10 percent.

SteveC wrote ...

Still sorta confused...
the capacitive reactance (measured in ohms) decreases with increased capacitance correct?

Yes, if you only change the cap value and don't mess with the AC frequency then it would decrease.

Changing the capacitance of the MMC does change the resonant frequency of the primary but since the secondary might be considered "fixed", the only thing left to change is the physical dimensions of the primary wire. no big deal right?

Well, if you just change the mmc value, it will change the resonant frequency of the entire tank circuit, so you'll have to then lose some primary turns to maintain the same frequency.

As for the rest of what you wrote, I didn't see anything that looked wrong as far as the numbers went. (I hope so) However, one problem that would probably arise from the "Lets have a huge mmc value to save money because we'll need less caps because there will be less voltage" idea would be that it would make your surge impedance incredibly low and you'd be hitting very large currents way to fast in the tank circuit.

Steve McConner wrote ...

Unfortunately there's no definitive answer to that question yet. There never will be until someone comes up with a computer model of the electrical behaviour of Tesla coil streamers.

I used to recommend a loaded Q of 10, which means you choose the impedance of your cap such that it drops 10 times the inverter's output voltage at the resonant frequency of your secondary.

But Steve Ward has lately reported that his big coils work somewhat better with considerably more capacitance than this. He claims that the larger capacitor can have a lower voltage rating, so the overall spend on capacitors is less for the same spark output, and also that the system gives about 10% more spark output for the same kW input.

Which by the way is how you measure the efficiency: spark length in inches divided by the square root of power consumption in watts gives you the "Freau factor", and the number to beat is something like 1.7. You might prefer some other metric, like Youtube views per hundred dollars spent on plexiglass, or whatever.

He uses a special tuning method, tuning primary and secondary to the same frequency and taking advantage of the resulting notches to achieve complete energy transfer in as little as 5 cycles. That's all I know unfortunately.

My own experience is that DRSSTCs work reasonably well with a wide range of tank capacitances, but bigger seems to be a bit better. I've tried everything from 0.01uF to 0.1uF in my own tabletop-sized coils, and 0.1 was definitely best.