News | The Component Club

As Predicted! Engineers Turn To DDR3 in QSMP-20 Module

As AI continues to put pressure on the memory market, the recent release of the QSMP-20 module, developed by Direct Insight, uses DDR3 to circumvent these shortages. This shift to older technologies raises questions about whether the latest tech is necessary or if its contributing to a bloated industry.

DDR3 memory, QSMP-20 module, AI industry memory demand, DDR4 and DDR5 prices, memory technology shortages, Direct Insight QSMP-20, industrial-grade STM32MP2 MPU, mature memory technologies, modern technology bloat, RAM optimization

Vishay 40 LHE Extends Linear Sensing To 40 mm

Vishay’s 40 LHE linear position sensor brings non-contact Hall effect sensing, 40 mm stroke, 12 µm resolution, IP67 sealing, and true power-on position reporting to harsh industrial, robotic, structural, and vehicle measurement systems.

Vishay, Vishay Intertechnology, Vishay Sfernice, 40 LHE, linear position sensor, Hall effect sensor, displacement measurement, motion control, industrial automation, robotic gripper, structural monitoring, crackmeter, dendrometer, process measurement, e-bike sensor, motorcycle sensor, railway equipment, agricultural machinery, IP67 sensor, PWM output

China Moving Towards Liquid Cooling For New Datacenters

China’s rapid AI expansion is driving server power and thermal densities beyond the limits of air cooling, forcing a nationwide shift to liquid-cooled data centres. This article examines China’s semiconductor constraints, the rise of companies like Envicool and Sanhua, government initiatives, and why liquid cooling is becoming essential for next-generation high-performance computing.

China AI infrastructure, liquid-cooled servers, AI data centres China, server cooling technology, Envicool, Sanhua Intelligent Controls, Nvidia AI cooling, Intel AI data centre, Alibaba cloud infrastructure, Tencent data centres, precision cooling systems, immersion cooling, direct-to-chip cooling, high-performance computing cooling, HPC thermal management, power-dense servers, rack-scale liquid cooling, Eastern Data Western Computing initiative, PUE efficiency, AI training servers, inference server cooling, hyperscale data centres

Samsung HBM4 Ships At 11.7Gbps And 3.3TB/s Per Stack

Samsung begins mass production of HBM4 delivering 11.7Gbps per pin and up to 3.3TB/s per stack with 4nm logic base die, 2,048 I/Os, and improved power efficiency for AI datacenters.

Samsung Electronics, HBM4, high bandwidth memory, 11.7Gbps, 13Gbps headroom, 3.3TB per second, 4nm logic base die, 1c DRAM process, 12-layer stacking, 16-layer HBM, AI datacenter memory, GPU memory bandwidth, TSV technology, power distribution network optimization, HBM3E comparison, hyperscale computing

ROHM Expands Its Nano Cap LDO Line With Higher Current Support

ROHM’s BD9xxN5 LDO regulators deliver stable 500 mA output using sub microfarad capacitors, improving flexibility for compact automotive and industrial designs.