There’s something uniquely appealing about walking into a large, darkened field and seeing the glow of fireflies flickering in the sky. It’s almost like a thousand eyes, each blinking with its own rhythm, watching everything on the forest floor below.
That’s probably why a new packaging inspection system for integrated circuits — with its ability to see defects that might otherwise be missed — borrows the name of that luminescent, winged denizen of the night.
The Firefly G3 delivers inspection and metrology process control during high-volume production of panel-level substrates. Source: Onto Innovation The Firefly G3 system, launched recently by Onto Innovation, provides inspection and metrology for automated process control during high-volume production of panel-level substrates, including next generations of glass and copper clad laminate (CCL). Through proprietary feed-forward and feedback software, the system produces high-resolution data to optimize layer-over-layer overlay accuracy throughout all layers on each side of panels as they are processed. This improves current panel performance and yield and enabling process maturation in less time.
The system’s capabilities complement Onto’s JetStep family of panel-level lithography systems. Its multiple imaging modes include Onto’s patented Clearfind Technology, a technique for detecting residue defects on metal and metal defects on organic layers. Integration with Onto Innovation’s Discover Defect and TrueADC software turns defect data into actionable process control, improving defect classification and reducing manual review. Optional 3D metrology sensors can also be used to measure film thicknesses and the height of metal redistribution (RDL) lines.
Applications for the Firefly G3 include:
- Advanced IC substrates (AICS)
- Fan-out panel level packaging (FOPLP)
- 2.5D/interposers
- Embedded die/embedded interposer
- 3DIC
In a nutshell, the Firefly G3 does the job of the multiple tools and human inspectors typically used for defect review and classification. Its high resolution and productivity are a fit for demanding end applications like high-performance computing (HPC), artificial intelligence (AI), cloud computing and machine/deep learning.
