This tutorial is designed to enable the user to determine the actual operating impedance of an RF powered plasma process chamber using commonly available test equipment and hardware.
The objective is to use the described empirical method to determine the chamber impedance by measuring the output of the impedance matching network (fixed in the tuned position) when the input is terminated by a 50 ohm resistive load. The conjugate of the measured value is the typical chamber impedance.
The following items are required to perform the test:
Impedance matching network – (manually adjusted is best)
Impedance meter – HP Model 4815 (useful up to 108 MHz) or other impedance meter models. There are various Amateur Radio quality meters offered by AEA or MFJ Enterprises that will also work for this measurement.
High quality 50 Ohm low power load (1>2 watts)
Coaxial adapters to connect the 50 ohm load in place of the RF power generator.
Well working plasma process system with:
Typical process gases and chamber pressure
RF power generator operational at typical process power
Impedance matching network properly mounted and is able to reduce reflected power to zero watts
Typical coaxial cable (between RF generator output & matching network input) installed
Set up all operating parameters such as RF power, chamber pressure, gas flow and chamber temperature.
Ignite the plasma discharge and tune the plasma impedance (reduce reflected power to zero). If an automatic network is used, simply switch the servo control into the manual mode after the process stabilizes. The impedance matching network capacitor positions must not move once the RF power is turned off.
Turn the RF power to OFF
Turn off all process gases and heaters
Remove the AC mains power from the RF generator and the impedance matching network
Disconnect the coax cable from the RF generator output and then connect this end to the 50 ohm dummy load. Ensure that all connectors and adapters are properly tightened.
Remove the output conductor from the plasma electrode (atmospheric side only). Leave this conductor connected to the output of the matching network as it is part of the impedance.
Power on the impedance meter and set the measurement frequency to the same frequency as the typical RF generator (i.e. 13.56, 27.12 & 40.68MHz).
Connect the probe between chassis ground (on the matching network) and the output strap.
Adjust the meter’s range so that it properly displays a value.
The HP4815 will read in a vector format (magnitude (Z) and phase (degrees)) – i.e. 20 ohms @ -76 degrees. Other meters will display the complex impedance as series resistance and reactance (R+jX)
The conjugate of the measured value is the typical chamber impedance. For example; if the chamber has capacitive electrodes we know that the complex impedance will be R-jX with the minus (-) sign indicating capacitive reactance. The measured value will display plus (+) as we are looking back into the output of the matching network. Simply replace the plus with a minus and you have the proper measurement.
This test can be performed with various plasma process conditions to obtain a “fingerprint” of your plasma chamber.
Remember that measurements taken at low frequencies (13-27MHz) will not be as sensitive to stray capacitance and/or inductances. As the frequency of operation is increased (40MHz and higher), all output conductors and ground paths will add their value to the measurement.
Also keep in mind that some models of vacuum capacitors used in impedance matching networks operating above 50 MHz may exhibit a transition from being capacitive to inductive. This is called transitioning through the “series resonance” point. Consult the manufacturer’s data sheet before selecting these parts for a new design.