Radio-frequency plasma fun: How to characterize a plasma circuit?

[Dever]

Greetings,

I have shifted much of the focus of my research from thin film specifics to the plasma chamber we are using, as I was unfamiliar with the intricacies of plasma and found that a lot of the parameters we are counting on frankly could not be counted upon. I'm still green, but it's certainly restored my enthusiasm in research, as I find it fascinating! So my question:

I have a plasma generator used to power a 'semi-remote' radio-frequency (13.56 MHz) capacitive plasma in a custom-(poorly) built deposition system. I've attached a snapshot of a roughly-drawn schematic of the circuit for visualization of the problem. I'd like to determine the potential of the plasma electrode at a given output power and assuming match. The generator is supposed to see the standard 50 ohm impedance at the output, and it measures the reflected power, but doesn't give me any information about the expected potential past the matching network.

Won't I have to determine the impedance of the matching network in order to know the Vmax potential drop at the rf electrode? How can this be done at this frequency? I have an o-scope that measures the frequencies quite well, but it can't scale to the lowest output power of this generator (5 W). Even so, the models I'm seeing for the determination of impedance (applying a known resistor in series) [A] may fail at this frequency, and have only one source of impedance, not two matched sources.

The vast experience out there would be greatly appreciated.

 

[vk6kro]

It looks like the bottom ends of the variable capacitors should be grounded, making this a pi-coupler matching circuit.

If that was so, then the output on the right should be at the top of the right capacitor.

It doesn't help much, though, because we would still need to know the actual capacitances of the variable capacitors and the impedance of the plasma chamber.

It would be simplest to just measure the voltage.

 

[Dever]

It looks like the bottom ends of the variable capacitors should be grounded, making this a pi-coupler matching circuit.

If that was so, then the output on the right should be at the top of the right capacitor.

It doesn't help much, though, because we would still need to know the actual capacitances of the variable capacitors and the impedance of the plasma chamber.

It would be simplest to just measure the voltage.

You're probably right. I do think it's a Pi network, so that second one might be parallel. I'll check when I get back to work in the morning, and if so, then let me know if I need to tweak the schematic for the viewers here to avoid confusion.

As for the result, yes, I thought that would be fine to measure the voltage, but I can't access the electrode directly, and so I thought if I could determine the impedance characteristics of the match, the impedance of the plasma chamber has to be equivalent, right? (at match) I wonder why the suggestions I've seen have indicated adding a resistor and measuring points along the circuit? Why go to all the trouble if you can just measure with respect to ground and determine ΔV? Also, I don't know if I can trust the potential on the oscilloscope. I measured three different parts of the circuit excluding the plasma chamber and just including the match, but it appeared that the rms voltage actually rose as it was separated by components from the rf source, rather than dropping. How is that even possible? Each of the peaks in the o-scope showed a phase shift and no change in ω, but how is it possible to have ~3 Vrms from ground after the generator and ~18 Vrms at near ground? There's something I'm not considering in the rf world here.

 

[vk6kro]

If the bottom connections of the capacitors are grounded, then it is a Pi circuit and the output has to be taken from the top of the RHS capacitor.

The circuit has to be resonated and, for this, the two capacitors appear in series across the coil.

This frequency is not excessively high, but you could use an RF probe to get a more consistent reading of voltages.
Here is one:

The output is DC, so the oscilloscope could be replaced with a DC voltmeter.

You can place a resistive voltage divider before this RF probe if the voltages were excessive for the diodes used.

If you can't get the diodes specified, Schottky diodes like the BAT85 would be suitable.