Iterative and deep studying reconstruction algorithms significantly improve failure analysis throughput and image quality for the Zeiss Xradia Versa and Context microCT systems.
Zeiss has developed a machine learning AI algorithms to enhance its 3D X-ray imaging for failure analysis of semiconductor packages. 3D X-ray analysis is increasingly important for 3D packaging of a number of chips in a single bundle with vias and interconnect.
The Advanced Reconstruction Toolbox for the Zeiss Xradia Versa sequence of non-destructive 3D X-ray microscopes (XRM) and its Xradia Context 3D X-ray micro-computed tomography (microCT) systems is predicated around two modules: OptiRecon for iterative reconstruction, and DeepRecon, the primary commercially obtainable deep learning reconstruction technology for microscopy purposes.
3D XRM is used for imaging defects to aid root trigger investigation of package failures because it permits visualization of options that aren’t visible in 2D X-ray projection images. In package failure evaluation, both quick results and high success charges are important. Consequently, reducing imaging time whereas sustaining image quality is of very excessive worth. Sometimes, Feldkamp-Davis-Kress (FDK) filtered back-projection algorithms are used to reconstruct the 3D dataset from many 2D projections acquired at different pattern rotation angles. When image exposure times or numbers of projections are lowered in an effort to enhance throughput, the FDK techniques usually result in degraded image quality.
The OptiRecon and DeepRecon engines allow increased scanning speeds whereas sustaining and even rising image quality with improved contrast-to-noise ratios for semiconductor superior packaging failure and structural evaluation. Along with electronics and semiconductor packaging, the Superior Reconstruction Toolbox can be utilized for supplies analysis, life sciences, and advanced battery development.
OptiRecon is aimed toward analyzing a broad vary of semiconductor packages and is appropriate for both analysis and development purposes, as well as FA. It makes use of iterative reconstruction, where variations are calculated between actual and modeled projections by multiple iterations till convergence. This implies fewer projections are needed with and less acquisition time in comparison with FDK. Scanning speeds as twice as quick with comparable or higher image quality could be achieved for semiconductor packages.
Computrol, Inc., a world-class supplier of mid- to low-volume, high-mix digital manufacturing providers to OEMs, introduced that it has bought an X7056-II computerized 3D X-ray inspection system from Viscom. Newly installed on the company’s Meridian facility, the award-winning system features extremely high throughput and excellent image quality for the necessities in high-end electronics manufacturing.
The X7056-II AXI in-line system can guarantee the exact inspection of hidden solder joints and components in high-volume manufacturing. It’s equipped with the xFastFlow transport module, which cuts printed circuit board altering instances. With this function, Computrol can process up to three boards at the similar time. The handling design also implies that significantly larger board sizes might be inspected and the scope of angled radiation is prolonged.
Computrol continues to assemble on its core suppliers and capabilities by investing in versatile, high-speed manufacturing equipment and technology, in addition to continuing training packages for employees.
The company regularly updates its state-of-the-art services with new advanced methods to maintain up with its clients’ needs. This addition is one other milestone that can assist guarantee product quality and reliability in actual time.