©2019 Impressum | Data Privacy Statement | Sitemap You are here » Outreach » Press/News

Latest News

21.10.2019 00:00 - Polyimid Film Removal of Glass Wafer - Reclaim Process

NQW developed a reclaim process for Glass wafer with multi-layer, structured Polyimid (PI) Films...

21.02.2019 00:00 - 100um Thin Glass and Quartz Wafer available

NQW has developing a process to produce 100um Thin Borofloat33 and Quartz Wafer, and the transfer...

22.10.2018 00:00 - Nano Quarz Wafer GmbH (NQW) successful recertified with latest ISO 9001-2015 revision

In September 2018 NQW showed once again that the Quality Management is in line with the...

Tag Cloud

Für die ordnungsgemäße Darstellung dieser Applikation wird Flash und Javascript benötigt.

18. September 2010 13:49
Category: Events

Innovation: SemiQuarz Wafer developed automated cluster analysis/ macro inspection of scratches and edge defects

cluster analysis macro inspection

SemiQuarz Wafer has developed a system, which identifies very precisely defects, e.g. scratches, marks or faults on the edge, on quartz, glass and sapphire wafers. The automated cluster analysis and macro inspection is based on DBSCAN - Density-Based Spatial Clustering of Applications with Noise and enables SemiQuarz Wafer to analyze each wafer very carefully and ensure its quality on very high levels. The innovative system has many advantages for the industry. First of all there is no knowledge necessary regarding the number of clusters or the domains. Secondly, the method is optimized for scanning scratches and edge defects especially on quartz and glass substrates and can be dynamically adjusted to the area and to the form of the scan area. Thus, it improves the production process as well as the rework efficiency and offers an advanced process control. The new system also allows SemiQuarz Wafer to automatically compare specific criteria on the base of customer requests.

How does the defect identifying system work? It starts off by capturing the data of the image and then goes over to a real time analysis on the base of an algorithm. As a result, the system subsumes all detected defects of the wafers and shows them on error patterns (see images below). Hence the wafers can be classified. Finally the system provides a quantitative analysis of the results and a suggestion on how to proceed further.