In Case You Missed It

PCB Inspection

“MSA-DETR: Multi-scale Alignment Detection Transformer for PCB Defect Detection with Enhanced Cross-scale Feature Fusion”

Authors: Renjie Zhang, Yanjue Gong, Fu Zhao and Jinkai Fan

Abstract: Printed circuit board defects are challenging to detect due to their small size, sparse features and high similarity to background textures. Existing deep learning approaches suffer from limited feature extraction, insufficient multi-scale fusion, and inadequate spatial alignment for small defects. To address these challenges, this paper proposes MSA-DETR, a multi-scale alignment object detection framework. First, an adaptive multi-scale residual (AMSR) block is designed for the backbone network, which enhances cross-regional modeling capability and preserves high-frequency detail information through large selective kernel mechanisms and multi-level residual connections. Second, a small-object alignment and frequency-enhanced pyramid (SAFE-P) is constructed to achieve effective multi-scale feature alignment through spatial feature redistribution and cross-scale multi-directional receptive field fusion. Finally, a context-aware spatial calibration (CASC) mechanism is designed to realize collaborative calibration of semantics and geometry via global context retrieval and learnable deformation alignment. Experimental results on two benchmarking datasets demonstrate significant improvements. On DsPCBSD+, MSA-DETR achieves 86.1% mAP0.5 and 52.4% mAP0.5:0.95, improving 2.6% and 2.7%, respectively, over RT-DETR baseline, while on HRIPCB it reaches 98.4% mAP0.5 and 55.8% mAP0.5:0.95, improving 2.3% and 1.8%, respectively. The model uses 7% fewer parameters than the baseline. These results validate the effectiveness of the multi-scale alignment strategy in improving small object detection accuracy, providing a robust solution for industrial PCB microscopic defect inspection. (Measurement Science and Technology, Nov. 19, 2025, https://iopscience.iop.org/article/10.1088/1361-6501/ae2152)

“Multi-Criteria Selection of Image Acquisition Technologies for Automatic Optical Inspection in Assembly Lines”

Authors: Michele Ronchi, Alberto Regattieri, Mauro Gamberi and Cristian Cafarella

Abstract: Ensuring product conformity in industrial assembly lines is becoming increasingly complex due to high production rates, as well as the growing complexity and customisation of products. Zero-defect manufacturing (ZDM) aims to eliminate production defects by integrating advanced quality assurance technologies, thereby enhancing process efficiency and sustainability. Automatic optical inspection (AOI), powered by computer vision, has emerged as a key enabler of ZDM, automating defect detection and enhancing quality control. Industrial adoption of AOI systems is still in its early stages, however, and a structured methodology is needed to assess their technical feasibility and select the most suitable image-acquisition technology. This paper proposes a novel multi-criteria decision-making (MCDM) methodology for selecting image-acquisition technologies in AOI systems. The approach integrates the analytic hierarchy process (AHP) and the weighted sum model (WSM) to evaluate alternatives based on key factors such as product dimensions, inspection area dispersion, visibility constraints and system flexibility requirements. The methodology is validated through two real-world case studies in the automotive sector, demonstrating its effectiveness in supporting technology selection and guiding AOI implementation in real production environments.  (The International Journal of Advanced Manufacturing Technology, Nov. 23, 2025, https://link.springer.com/article/10.1007/s00170-025-17021-5)

PCB Materials

“Printed Circuit Board Substrates Derived from Lignocellulose Nanofibrils for Sustainable Electronics Applications”

Authors: Yuliia Dudnyk, Pavel Kulha, Václav Procházka, Gustav Nyström and Thomas Geiger

Abstract: This study investigates lignocellulose nanofibrils (LCNF) as a sustainable alternative material for PCB substrates, demonstrating an application through the development of an eco-friendly computer mouse demonstrator. LCNF is derived from lignin-rich cellulose pulp, a side-stream product of biorefinery processes, combining the natural strength of cellulose fibrils with lignin to enhance mechanical and electrochemical properties. The research outlines the process of fibrillating lignin-rich cellulose pulp at 10kW/h per kg into LCNF, followed by thermal and pressure treatment (at Δp = 50 – 1500 kN, ΔT = 30 – 120°C) to achieve a rigid PCB substrate. Comprehensive characterization of the LCNF substrate included assessments of its mechanical properties (flexural and tensile testing), dimensional stability, electrical properties, surface uniformity and thermal conductivity. The LCNF PCB was integrated in a computer mouse demonstrator featuring inkjet printing of circuit layouts and electronic component assembly, while the mouse housing was designed and 3-D-printed using eco-friendly wood-PLA filament. Electrical properties characterization of the printed circuit and resulting functionality of the computer mouse showcases a sustainable approach to eco-electronics using wood-derived materials. This study underscores the potential of wood-derived nanomaterials like LCNF to reduce electronics waste (e-waste) associated with conventional PCB materials and promote the development of more eco-friendly electronics, contributing to sustainable, high-performance ecoPCBs and advancing green technology. (Scientific Reports, vol. 15, 2025, https://www.nature.com/articles/s41598-025-91653-1)

Solder Joint Reliability

“Electromigration in Eutectic Gold-Tin (80Au20Sn-wt%) Solder Joints”

Authors: Whit Vinson and David Huitink

Abstract: Eutectic gold-tin solder joints (80Au20Snwt%) with a diameter of 300μm were subjected to electromigration (EM) conditions of 10,000–20,000A/cm² in the ambient temperature range of 125°–215°C. The present work examines a set of 36 experimental test runs using two different test coupons (18 runs with each coupon type and a total of 90 joints tested). Seven of the 36 coupons survived the maximum test duration period of 500hr. One coupon type created a linear electric current pathway through two lengths of 254μm-diameter copper wire connected by a single gold-tin joint. The other coupon type daisy-chained four gold-tin joints together in series in a flip-chip configuration using two Rogers 4003C printed circuit boards with electroless nickel immersion gold (ENIG)-finished 2oz. (c.56.6g) copper traces and 240μm solder mask-defined (SMD) pads. Time to EM failure (TTEF) was recorded for each experimental run, demonstrating reduced lifetime with increasing current density and/or ambient temperature. To account for the Joule heating associated with the high current densities used in this study, temperature coefficient of resistance measurements (TCR) were carried out for both sample groups. To describe the TTEF data, a lifetime model for 80Au20Sn-wt% joints combining the experimental testing conditions with TCR data has been developed. Post-failure analysis (PFA) on the tested solder joints shows several morphological phenomena occurring in the gold–tin joints under EM conditions, including solder migration into the trace, spinodal decomposition, phase coarsening, and phase separation. (Journal of Electronic Materials, Sept. 9, 2025, https://link.springer.com/article/10.1007/s11664-025-12344-1)