Express Manufacturing Deploys AI-Driven X-Ray Inspection for High-Reliability PCB Assemblies

AI Meets X-Ray: Automated Defect Detection for Complex PCB Assemblies

Express Manufacturing, Inc. (EMI) has expanded its inspection capabilities with a second automated X-ray inspection (AXI) system — the TR7600FB SII from Test Research, Inc. (TRI). Announced June 1, 2026, the system uses AI-driven inspection algorithms to detect solder defects in high-density assemblies that traditional visual inspection cannot reach.

The investment signals a broader industry trend: as PCB assemblies grow denser and more complex, human-dependent inspection becomes insufficient. AI-powered AXI fills the gap by identifying defects inside BGA solder joints, hidden vias, and multi-layer assemblies without human operator judgment calls.

Key Technical Capabilities

The TR7600FB SII brings several advances over previous-generation AXI systems:

• 5 µm resolution: Detects micro-voids and hairline cracks invisible to standard X-ray
• AI-driven void detection: Machine learning algorithms automatically identify and classify solder voids without manual programming of acceptance criteria
• Automated programming: AI reduces setup time for new PCB designs — the system learns acceptable solder joint profiles from reference samples
• 20 fields of view per second: Maintains production throughput while increasing inspection depth
• Board capacity: Handles assemblies up to 850 mm × 520 mm and 12 kg — covering large format boards common in aerospace and industrial applications

Why AI-Powered X-Ray Inspection Matters

Traditional AOI (Automated Optical Inspection) can only see surface-level defects. For modern high-reliability assemblies, the critical defects are hidden:

Defects only X-ray can find:

• BGA solder ball voids (>25% void = reliability risk)
• Head-in-pillow (HiP) defects in fine-pitch BGA
• Insufficient solder under QFN thermal pads
• Cold joints in bottom-terminated components
• Internal layer misregistration in assembled modules

Where AI adds value over conventional X-ray:

• Reduces false calls by 40–60% — AI learns the difference between acceptable process variation and genuine defects
• Consistent judgment — no operator fatigue, no shift-to-shift variability
• Adaptive thresholds — AI adjusts void acceptance criteria based on component criticality and industry standards (IPC-A-610, J-STD-001)
• Inline capability — fast enough for 100% inspection rather than statistical sampling

Industry Standards Integration

The system supports modern factory communication protocols:

• SMEMA: Standard mechanical-electrical interface for line integration
• SECS/GEM: Semiconductor Equipment Communication Standard for equipment management
• IPC-CFX-2591: Connected Factory Exchange for IIoT data sharing
• IPC-HERMES-9852: Board transport and traceability protocol

This connectivity means inspection data integrates directly with Manufacturing Execution Systems (MES), enabling:

• Real-time yield tracking by component, lot, and operator
• Automatic process adjustment triggers when defect trends emerge
• Full traceability linking each board to its inspection results

The Broader AI Inspection Trend

EMI’s investment reflects an acceleration in AI adoption across PCB inspection:

The pattern is clear: inspection equipment is transitioning from rule-based (human-programmed acceptance criteria) to learning-based (AI determines acceptable vs. defective from training data). This shift is essential as assemblies incorporate more hidden joints (0201 passives under shields, package-on-package stacking, embedded components).

What This Means for PCB Design and Manufacturing

For engineers designing high-reliability assemblies:

1. Design for inspectability: Even with AI X-ray, component placement that allows clear X-ray access improves detection accuracy. Avoid stacking components directly above BGAs where X-ray shadows overlap.

2. Void budget planning: With AI providing consistent void measurements, designers can specify tighter void requirements (e.g., <15% per ball, <25% total) knowing that inspection will actually catch violations.

3. First-pass yield data: AI inspection generates rich datasets that enable design-for-manufacturing optimization in future revisions.

At AtlasPCB, we incorporate DFM feedback from inspection systems into our manufacturing process. When customers design boards with BGA, QFN, and other hidden-joint components, our engineering review considers X-ray inspectability alongside electrical and thermal requirements.

Source: PCB Directory, June 1, 2026. Equipment: TRI TR7600FB SII.

Need high-reliability PCB assemblies with full inspection traceability? Get a quote from AtlasPCB — we support X-ray inspectable designs for aerospace, medical, and automotive applications.

Image: ThisisEngineering via Unsplash