RF Generator Frequency Tuning Without Sinusoidal Perturbation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing RF generator systems face challenges in precisely controlling power delivery to plasma chambers due to non-linear loads, leading to issues such as intermodulation distortion and inefficient impedance matching, particularly in advanced semiconductor fabrication processes.
Innovation Solution
An extremum-seeking control (ESC) method is employed to adjust the frequency of RF generators based on historical frequency and output response values, eliminating the need for sinusoidal perturbation signals, thereby improving transient tuning time and reducing sensitivity to system harmonics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional RF generator control methods are used with sinusoidal perturbation signals for frequency tuning, then frequency control can be achieved, but intermodulation distortion occurs and transient tuning time is prolonged
Solution Approach 1:
The patent extracts and eliminates the sinusoidal perturbation signal from the frequency tuning process. By using extremum-seeking control with direct gradient estimation from historical frequency and output response values, the system achieves frequency control without the harmful sinusoidal signals that cause intermodulation distortion in the plasma chamber
Solution Approach 2:
The patent implements feedback control by continuously monitoring output response values (such as impedance or power delivery) and using them to adjust frequency. The gradient signal is generated based on historical frequency values and corresponding output response values, creating a closed-loop system that converges to optimal frequency without external perturbation signals
2Measurement precision
If sinusoidal perturbation signals are used for frequency tuning, then frequency control is achieved, but the system becomes sensitive to harmonics and dithering occurs
Solution Approach 1:
The patent removes the sinusoidal perturbation signal that causes sensitivity to harmonics and dithering. By using extremum-seeking control with direct gradient estimation, the system achieves frequency tuning without introducing periodic disturbances that interact with system harmonics and cause unstable dithering behavior
Solution Approach 2:
The patent replaces the traditional mechanical-style sinusoidal perturbation approach with an extremum-seeking control algorithm. This substitution uses computational gradient estimation from historical data instead of physical perturbation signals, eliminating the source of harmonic sensitivity and dithering while maintaining frequency control capability
3Reliability
If conventional impedance matching methods are used, then power delivery can be controlled, but convergence to optimal settings is slow
Solution Approach 1:
The patent performs preliminary action by maintaining historical frequency values and corresponding output response values in memory. This historical data is prepared in advance and used to rapidly calculate gradient signals during frequency tuning, enabling faster convergence to optimal impedance matching settings without lengthy trial-and-error processes
Solution Approach 2:
The patent implements rapid feedback control by continuously using output response values to generate gradient signals for frequency adjustment. This closed-loop extremum-seeking control quickly converges to optimal power delivery settings by constantly adapting frequency based on measured system response, significantly reducing transient tuning time compared to conventional open-loop or slow feedback methods
Data Source
AI summary
A RF generator includes a RF power source configured to generate an output signal at an output frequency, and an extremum-seeking frequency controller configured to generate a frequency control signal. The frequency control signal varies the output frequency of the RF power source, and the frequency control signal is formed from a gradient signal. The RF generator further includes a gradient estimator configured to generate the gradient signal. A frequency of the frequency control signal is adjusted based on the gradient signal. The gradient estimator is configured to receive frequency values of the frequency control signal and corresponding output response values, and the gradient signal is generated based on the frequency values and the output response values. Other example RF generators, methods for extremum-seeking frequency control of a RF generator, and control systems for controlling a RF generator are also disclosed.