3D Vision-Based Susceptor Alignment for Uniform Chamber Gaps
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for aligning a susceptor within a processing chamber are time-consuming and prone to manual errors, leading to inaccurate positioning and non-uniform processing results due to temperature gradients and misalignment.
Innovation Solution
A three-dimensional (3D) map of the susceptor and ring is generated using camera data and 2D profilometer data, adjusted using a machine learning model and optimization algorithm to achieve precise alignment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual calibration methods are used to align the susceptor, then alignment can be achieved, but the process requires extensive time and is prone to manual errors
Solution Approach 1:
The patent replaces manual mechanical calibration with an automated vision-based system. A camera captures images of the susceptor and ring, and a machine learning model processes these images to automatically determine alignment and generate positioning commands, eliminating manual intervention while maintaining high precision
Solution Approach 2:
The patent creates a digital representation (3D map) of the physical susceptor and ring geometry. This digital copy is then used by the machine learning model to analyze positions and predict optimal alignment, replacing the need for repeated physical measurements and manual adjustments
2Loss of time
If automated alignment using sensor positional data is used, then manual time is reduced, but the representation of susceptor position is not fully accurate
Solution Approach 1:
The patent merges multiple data sources: camera images providing visual geometry, 2D profilometer data providing depth information, and machine learning predictions. This combination creates a comprehensive 3D map that accurately represents the susceptor position and geometry, overcoming the limitations of any single sensor
Solution Approach 2:
The patent transitions from 2D sensor measurements to a 3D representation of the susceptor and ring. By creating a three-dimensional map that includes depth, width, and height information, the system achieves more accurate positional representation than traditional 2D sensor data alone
3Measurement precision
If additional positional data gathering and adjustments are made each time a substrate is inserted, then alignment accuracy can be maintained, but processing delays occur
Solution Approach 1:
The patent performs alignment calibration in advance by creating a 3D map and training the machine learning model before substrate processing. Once trained, the model can quickly predict optimal positioning for subsequent substrates without requiring repeated extensive measurements, maintaining accuracy while improving throughput
Solution Approach 2:
The machine learning model, once trained on the 3D map data, autonomously predicts optimal susceptor positioning for each substrate insertion. The system serves itself by using the trained model to generate positioning commands without requiring manual intervention or extensive repeated calibration, enabling rapid and accurate alignment
Data Source
AI summary
Methods and devices for aligning a susceptor are provided herein. Embodiments include creating a three-dimensional (3D) map of a susceptor and a ring within a substrate processing chamber based on camera data comprising image data associated with the susceptor and the ring. Embodiments further include adjusting a position of the susceptor based on the 3D map to create a gap having a target size between the susceptor and the ring.