Surface position detection method of silicon melt in CZ furnace

In the Czochralski mono-crystalline silicon crystal growth system, silicon melt surface position detection is a key problem. For this problem, this paper proposes a triangulation method based on the line laser and area image sensor. The method uses the image sensor to capture the line laser reflected by the silicon melt surface. To improve the signal to noise ratio and reliability in each image, the position of the line laser in each image is extracted by using digital image processing method. Then the model-based filtering method is used to remove the impulsive noise which is caused by to the complex condition inside of CZ furnace. Finally, the calculation of melt surface position is based on line laser triangulation method. The experimental results show that this method can achieve detection accuracy of <;±0.2mm in the detection range of 100mm and has a good real-time performance and robustness.