ELE Times

Courtesy: Lam Research

• AI’s evolution could be thwarted by electrical resistance in 3D architectures
• Molybdenum, a breakthrough material, reduces resistance, improves performance

Lam Research’s ALTUS Halo is designed to address the unique challenges of molybdenum implementation in leading-edge integrated circuit types.

The AI revolution faces an invisible barrier: electrical resistance. As artificial intelligence (AI) demands ever-increasing compute, the semiconductor industry has responded by building upward, creating dense 3D architectures that pack more computing power into each square millimeter.

But this vertical scaling creates an unprecedented engineering challenge: every electrical connection through these towering structures must be perfect at the atomic scale, or AI performance degrades catastrophically.

Advanced 3D Integration and Metallization Barriers in AI Chip Manufacturing

Traditional metallization approaches are reaching their physical limits.

In conventional chip designs, creating electrical pathways meant depositing metal into dielectric etched features. These methods relied on barrier layers (e.g., titanium nitride, TiN) to prevent unwanted interactions between metals and surrounding materials.

While necessary, these barriers add electrical resistance—acceptable in simpler chips, but a fundamental roadblock in 3D architectures where signals must (soon) travel through up to 1,000 NAND layers of vertical connections.

The narrowing of lines caused by device shrinking drives the need for new materials with shorter mean-free paths—the distance electrons can travel before colliding—that match line length to achieve lower resistance.

The surge in compute demand compounds the challenge. Every suboptimal connection, every additional barrier layer, creates performance bottlenecks and thermal management challenges that can degrade overall AI system capability.

Molybdenum Innovation Provides a Breakthrough Material for Advanced AI Chip Architecture

Lam’s leadership in metallization innovation spans decades of industry inflections. Our pioneering work with tungsten atomic layer deposition (ALD) enabled the revolutionary shift from planar to 3D NAND memory. Now, as device features continue to shrink, we’re driving another fundamental transition with molybdenum—a material uniquely suited for today’s confined spaces.

Molybdenum (Mo) emerges as a transformative material for advanced metallization because its shorter mean-free path makes it uniquely suited for today’s confined spaces.

And unlike tungsten and other metals, Mo doesn’t need an adhesion or barrier layer (like TiN), simplifying the manufacturing process while significantly reducing overall resistance.

The transition to molybdenum echoes another historic industry inflection point: the shift from aluminum to copper interconnects in the early 2000s, which Lam led. Just as that transition fundamentally changed semiconductor manufacturing, today’s move to Mo represents a similar watershed moment.

Advanced ALD Solutions for AI-Era Chips

Material selection alone isn’t enough. Our latest innovation, ALTUS Halo, represents a convergence of atomic-scale engineering and practical manufacturing solutions. The platform brings specific innovations for each critical application:

• For 3D NAND it enables void-free lateral and barrier-less fill through advanced ALD technology and precise wafer temperature control.
• For DRAM applications it drives metallization innovation with selective and conformal fill capabilities.
• For logic it offers both thermal and plasma ALD options with an integrated interface cleaning process.

Atomic-Scale Engineering for AI Computing

The implications extend far beyond material selection and manufacturing processes.

Lam’s advances in deposition technology and grain engineering enable optimal molybdenum integration across all leading-edge applications—from 3D NAND wordlines to advanced logic interconnects and DRAM structures.

Initial atomic layers in ALD are critical for interface engineering and subsequent film growth, serving as the template for the material’s properties. The ALTUS Halo quad station module architecture is ideal for creating the most advanced fill processes with the highest productivity due to its flexibility of running different wafer temperatures, process steps and chemistry at each station.

As the industry pushes toward increasingly complex architectures, this precision engineering at the atomic scale becomes even more critical.

A Semiconductor Industry Transformation in Memory and Logic 

The semiconductor industry stands at a crucial juncture. Data-intensive AI applications demand significant advancements in both memory and logic technologies. These next-generation devices require unprecedented precision in metallization, where even small improvements in resistance and thermal performance can have outsized impacts on overall system capability.

Through innovations like ALTUS Halo, Lam is enabling a fundamental transition across NAND, DRAM, and logic. As the semiconductor industry pushes physics and chemistry to their limits, our manufacturing-ready solutions will help define the future of AI computing.

The post Breaking Through AI’s Invisible Barrier With Molybdenum appeared first on ELE Times.

Courtesy: Rockwell Automation

Introducing Logix SIS, a safety instrumentation system from Rockwell Automation.

“The only way of discovering the limits of the possible is to venture a little way past them into the impossible.”

This quintessential quote by Arthur C. Clarke appropriately foreshadows and encapsulates the spirit of innovation and progress that has driven humanity to push past conceivable boundaries. More than ever, people look to their resources, neighbors and leaders for intuitive and adaptable solutions to their persisting problems. Today, in the realm of safety, Rockwell Automation answers the call, standing at the precipice of a new era, where the limitations of the past are giving way to future possibilities.

Barriers of Traditional Safety Systems

Complexities abound in our dynamic world of modern industry, and safety is paramount. From robust oil refineries and chemical plants to intricate manufacturing facilities, ensuring the well-being of personnel, protecting valuable assets and safeguarding the environment are at the forefront of business priorities.

Traditional safety systems, while essential, have often needed help to keep pace with the evolving demands of industry. Their implicit complexity, inflexibility and integration challenges can create barriers, hinder productivity and even compromise safety itself. The need for a more streamlined, adaptable and cost-effective approach to safety has never been more apparent.

Enter Logix SIS: A New Era of Safety

Rockwell Automation, a pioneer in industrial automation and digital transformation, has risen to this challenge with the introduction of Logix SIS. This cutting-edge safety instrumented system redefines the landscape of industrial safety.

Logix SIS is more than merely an incremental improvement over existing solutions — it’s a paradigm shift, a bold leap into the future of safety that empowers a connected approach to how businesses achieve unprecedented levels of protection, efficiency and productivity.

Unlike traditional safety systems that operate in isolation, Logix SIS seamlessly integrates with Rockwell Automation’s Integrated Architecture®, leveraging a common platform for both safety and process control. This simplification reduces the need for separate engineering and maintenance staff, which in turn reduces complexity and can help accelerate project timelines. With familiar Logix programming tools and a streamlined configuration process, you can easily design, implement and maintain your safety system.

Key Features and Benefits of Logix SIS:

• High Availability & SIL 2 /SIL 3 Certification: Logix SIS is engineered for the most demanding safety applications, achieving Safety Integrity Level (SIL) 2 and SIL 3 certifications, the gold standards in industrial safety. This means you can trust Logix SIS to provide robust protection against even the most critical hazards, achieving the safety of your people, assets and environment.
• Scalability and Flexibility: Logix SIS is built to adapt and grow with your business. Its modular design allows you to quickly expand or modify your system as your needs evolve, confirming your safety infrastructure remains practical and relevant in the face of change. Whether you’re adding new processes, equipment, or safety zones, Logix SIS can scale to meet your requirements, providing the flexibility and agility you must stay ahead of the curve.
• Cost Efficiency and ROI: Logix SIS is not just a safety solution; it’s a strategic investment in your business. By leveraging existing Rockwell Automation hardware and software, eliminating the need for additional licenses, and streamlining engineering processes, Logix SIS helps you reduce the total cost of ownership and achieve a compelling return on investment.
• Improved Productivity and Uptime: Downtime is the enemy of productivity. Logix SIS’s high availability architecture, combined with advanced diagnostics and predictive maintenance capabilities, helps minimize safety-related stoppages and unplanned downtime. This delivers continuous operation, maximizes output and helps protect your bottom line.

Tomorrow’s Safety, Today

Logix SIS represents a bold step forward in industrial safety, offering a comprehensive, integrated and future-ready solution that empowers businesses to achieve their safety and operational goals. It’s a testament to the spirit of innovation and progress that has always defined Rockwell Automation.

The post Transform Your Safety Strategy: Introducing Logix SIS for Enhanced Protection and Efficiency appeared first on ELE Times.

Courtesy: Cadence Systems

The demand for semiconductors is surging due to AI growth, data centers, and digital transformation, but the environmental cost is significant. Energy-intensive manufacturing and waste generation pose challenges that must be addressed for long-term sustainability.

Semiconductors power modern innovation, from smartphones to AI systems, yet their production impacts the environment. By integrating sustainability into design and manufacturing, the industry can reduce energy consumption, minimize waste, and conserve resources. Introducing environmental impact (E) to the traditional power, performance, area, and cost (PPAC) model during the early stages of development is key to sustainable progress.

Efforts to lower energy use, optimize fabrication processes, and utilize recycled materials are gaining traction. Innovations like energy-efficient architectures and adaptive power management deliver high performance while reducing carbon footprints, advancing a greener future for the industry.

Imec’s SSTS Program

Imec, the world’s leading independent nanoelectronics R&D hub, has long been at the forefront of sustainability in semiconductor manufacturing. The Sustainable Semiconductor Technologies and Systems (SSTS) program, spearheaded by imec, brings together an impressive coalition of industry giants as well as government bodies, academia, and key associations, whereby the program aims to provide actionable data and tackle the pressing environmental challenges of advanced semiconductor manufacturing.

Cadence is proud to be the first electronic design automation (EDA) partner to join this groundbreaking program. This partnership marks a pivotal first step toward establishing a secure data-sharing platform that enables access to imec.netzero data through Cadence’s state-of-the-art design tools. With this capability, engineers and designers can make informed decisions early in the design process, integrating environmental considerations from the beginning. The mission is clear—to weave sustainability into the very foundation of semiconductor innovation.

How Cadence and imec Benefit the Industry

Through this partnership, Cadence and imec will be redefining the design mindset, enabling sustainability to be a core driver during the design phase. Currently, life cycle analysis is typically done after a product is designed, so any required changes to the bill of materials or process will set back the project timeline by months.  With Cadence’s “shift left” process, designers can proactively evaluate the environmental impact of their decisions during the design phase, where it matters most.

“Welcoming Cadence as the first EDA partner in our SSTS program marks a significant milestone in valorizing our data and ensuring sustainability is embedded at the heart of semiconductor innovation,” said Lars-Åke Ragnarsson, program director of Sustainable Semiconductor Technologies and Systems (SSTS) at imec.

Cadence’s Leadership in Sustainable Innovation

This partnership underscores Cadence’s position as a global leader in sustainable innovation. We have pledged to achieve net-zero greenhouse gas emissions by 2040, and we are backing up this promise with bold actions. By developing generative AI and digital twin technologies, strategically acquiring companies that drive sustainable innovation, and offering tool solutions that address both current and future environmental challenges, Cadence is paving the way for a sustainable, efficient, and technologically advanced semiconductor ecosystem.

The post How Cadence Is Energizing Sustainable Semiconductor Design appeared first on ELE Times.

Courtesy: u-blox

LEAP brings pinpoint accuracy and ultra-low power consumption to smartwatches, fitness trackers, and sports wearables.

The wearables conundrum
Smartwatches, fitness trackers, and GPS-enabled sports wearables have become essential tools for millions of users tracking their daily activities and athletic performance. But with these compact devices come big challenges – especially when it comes to delivering highly accurate GNSS positioning without draining the battery.

For device designers and users alike, the demands are increasing. Accurate tracking is expected even in dense cities, deep forests, or open water. And nobody wants to charge their wearable every day. The push to squeeze more performance into ever-smaller packages has created a tension between precision and power.

To meet this challenge head-on, u-blox has introduced LEAP (Low Energy Accurate Positioning), a powerful new GNSS technology built into the u-blox M10 platform. It enables wearables to deliver consistently accurate positioning while extending battery life –
solving one of the most persistent problems in wearable design.

Why low power and accuracy usually don’t mix
Accurate GNSS positioning is hard work, especially in the kinds of environments wearables are typically used in. In open-sky conditions things aren’t so bad. But in a dense forest, an urban canyon, or on the side of a mountain, GNSS signal quality starts to drop and accuracy starts to suffer. Add in dynamic movements like arm swings or vibration, and things get even more complicated. Plus, the GNSS antennas that wearables use aren’t very big to begin with.

The consequence is that the device uses lots of power trying to receive weak signals, and filtering out noise, reducing battery life. This is why traditional low power GNSS solutions tend to compromise on accuracy in order to conserve battery – and why high-accuracy solutions often drain battery quickly. LEAP was designed to avoid this compromise.

The solution: How LEAP works

LEAP is a smart GNSS mode developed by u-blox to deliver optimal performance for wearables while also extending battery life.

• Smart signal selection is at the core of LEAP. Rather than using all available GNSS signals, LEAP selectively uses only those that offer the strongest signal or the most accurate data. It dynamically filters out low-elevation or noisy signals that could introduce errors, and applies multipath mitigation techniques to reduce the impact of reflected signals common in cities or wooded areas.

• External low-noise amplifier (LNA) switching also helps minimise battery usage. LEAP can automatically switch the device’s LNA on or off based on real-time signal conditions. If signal quality is already high, the LNA can be disabled to save power. When signals are weak or noisy, the LNA reactivates to maintain positioning performance.

Together, these innovations ensure that a device using LEAP doesn’t waste power trying to receive poor quality data – and also improve accuracy, giving your device a powerful edge in the wearables space.

To further improve accuracy, LEAP includes activity-aware dynamics: tailored motion models for activities such as running, cycling, and hiking. These models account for specific movement patterns, like arm swings or stride variations, allowing the GNSS system to make smarter assumptions and corrections based on user behaviour. LEAP has even been validated for various sports like running, hiking and cycling – and further enhancements are in development.

What LEAP delivers

In side-by-side tests with standard u-blox M10 GNSS mode, LEAP reduced power consumption by up to 50%, while delivering similar or better positioning accuracy. In forest environments, LEAP delivered a circular error probable (CEP95, which essentially means that the probability of the data being at the stated accuracy is 95%) of 8 metres, compared to 14 metres from competitor products. In opensky tests, the improvement was from 4 metres to 2 metres. These kinds of gains matter. Whether users are running under tree
cover, hiking through gorges, or biking through the city, they can trust that their wearable device is delivering accurate, energy-efficient tracking.

Why it matters for designers

For wearable device designers, LEAP opens up a new range of possibilities. By delivering high-accuracy GNSS positioning at low power, it enables smaller batteries and slimmer form factors without compromising user experience. The chip package is impressively compact at just 2.39 x 2.39 mm, making it ideal for modern wearables, including those with severe
space and weight constraints. Whether you’re designing a rugged GPS sports watch or a lightweight everyday fitness tracker, LEAP will fit.

It’s also a future-proof solution. Firmware upgrades, which can be delivered via external flash or a connected MCU, mean LEAP can continue to evolve after deployment. Future firmware updates could introduce new models for additional activities, further optimise
power savings, or enhance positioning accuracy based on the latest innovations. This ensures that devices built with LEAP can stay competitive and adapt to emerging user needs. It supports Android systems and is fully compatible with u-blox AssistNow, enabling faster positioning and lower startup power draw. Built-in support for protocols like SUPL means LEAP can integrate seamlessly into today’s connected wearable ecosystems.

A LEAP forward for wearable GNSS

With LEAP, u-blox has redefined what’s possible for wearable GNSS. By combining low power consumption with high accuracy, it solves one of the biggest challenges in GNSS for wearables. And by making smartwatches and sports watches more capable, it gives users the freedom to explore further, train harder, and go longer between charges.

The post Maximize positioning accuracy and battery life with LEAP : Low Energy Accurate Positioning for wearables appeared first on ELE Times.

A high-performance DAQ system that meets the latest evaluation and test needs in the automotive, mechatronics, and power electronics fields 

Yokogawa Test & Measurement Corporation announces the release of the SL200 High-Speed Data Acquisition Unit, a ScopeCorder series product with a wide range of data logging functionalities for evaluation and test applications, including high-speed sampling and analysis. The SL2000 is a modular platform that combines the functions of a mixed signal oscilloscope and a data acquisition recorder, and is designed to capture fast signal transients and long-term trends. It is suitable for applications such as R&D, validation, and troubleshooting.

The SL2000 can be used separately or in combination with the DL950 ScopeCorder, depending on the application. No other product family on the market offers this level of flexibility in handling multi-channel measurements. With the ScopeCorder product family, Yokogawa provides a multifaceted, total solution for the high-precision mechatronics and electric power markets that is contributing to the advancement and development of new technologies and applications. 

Development Background
In the four years since the DL950 ScopeCorder was first brought to the market by Yokogawa Test & Measurement, there have been many technical advances in the electric vehicle (EV), renewable energy, and other industrial fields. Today, there is an ever-greater requirement for the capability to simultaneously measure multiple parameters and for the systemization of mechatronic measurements in product development. For example, in the development of motors for industrial and EV systems, one essential test for checking and improving a product is the durability test. This test takes a long time to complete and requires a highly reliable measuring instrument and high sampling rates.

Main Features
1. Enabling both high-speed sampling and multi-channel measurement
The SL2000 performs long-duration multi-channel measurements while precisely analyzing even the most detailed aspects of waveforms. With its dual capture function, the SL2000 can perform durability tests over long periods of time at speeds of up to 200 MS/s.

By using the IS8000 integrated software platform, it is easier to perform the long-term measurements required for durability testing, helping to improve the efficiency of product design and evaluation work. In addition, isolation measurement technology ensures the noise resistance required for durability testing in harsh environments.

2. Supporting simultaneous measurement of a wide variety of devices
The SL2000 has eight available slots (with up to 32 channels), for which over 20 types of input modules are available to enable measurements of electrical signals, mechanical performance parameters indicated by sensors, and decoded vehicle serial bus signals. To increase the number of measurement channels, up to five SL2000 and DL950 units can be synchronized.

Major Target Markets
•    Transportation (automotive, rail, aviation, etc.)
•    Power and energy (renewable energy, smart cities/homes, data centers, etc.)
•    Mechatronics, including industrial robots and motors

Applications
•    Durability and reliability testing of components and vehicles that requires high sampling rates and multi-channel simultaneous measurement of analog signals and in-vehicle bus signals such as CAN and CAN FD
•    Simultaneous measurement and evaluation of temperature, vibration, and other mechanical signals that change relatively slowly as well as mechatronic and other such high-speed control signals
•    Electrical analysis and control signal evaluation

The post Yokogawa Test & Measurement Releases SL2000 High-Speed Data Acquisition Unit appeared first on ELE Times.

Industry-first inertial measurement unit (IMU) with dual MEMS accelerometer
and embedded AI measures accurately up to 320g full-scale range

STMicroelectronics, a global semiconductor leader serving customers across the spectrum of electronics applications, has revealed an inertial measurement unit that combines sensors tuned for activity tracking and high-g impact measurement in a single, space-saving package. Devices equipped with this module can allow applications to fully reconstruct any event with high accuracy and so provide more features and superior user experiences. Now that it’s here, markets can expect powerful new capabilities to emerge in mobiles, wearables, and consumer medical products, as well as equipment for smart homes, smart industry, and smart driving.

The new LSM6DSV320X sensor is an industry first in a regular-sized module (3mm x 2.5mm) with embedded AI processing and continuous registration of movements and impacts. Leveraging ST’s sustained investment in micro-electromechanical systems (MEMS) design, the innovative dual-accelerometer device ensures high accuracy for activity tracking up to 16g and impact detection up to 320g.

“We continue to unleash more and more of the potential in our cutting-edge AI MEMS sensors to enhance the performance and energy efficiency of today’s leading smart applications,” said Simone Ferri, APMS Group VP, MEMS Sub-Group General Manager at STMicroelectronics. “Our new inertial module with unique dual-sensing capability enables smarter interactions and brings greater flexibility and precision to devices and applications such as smartphones, wearables, smart tags, asset monitors, event data recorders, and larger infrastructure.” 

The LSM6DSV320X extends the family of sensors that contain ST’s machine-learning core (MLC), the embedded AI processor that handles inference directly in the sensor to lower system power consumption and enhance application performance. It features two accelerometers, designed for coexistence and optimal performance using advanced techniques unique to ST. One of these accelerometers is optimized for best resolution in activity tracking, with maximum range of ±16g, while the other can measure up to ±320g to quantify severe shocks such as collisions or high-impact events.

By covering an extremely wide sensing range with uncompromised accuracy throughout, all in one tiny device, ST’s new AI MEMS sensor will let consumer and IoT devices provide even more features while retaining a stylish or wearable form factor. An activity tracker can provide performance monitoring within normal ranges, as well as measuring high impacts for safety in contact sports, adding value for consumers and professional/semi-pro athletes. Other consumer-market opportunities include gaming controllers, enhancing the user’s experience by detecting rapid movements and impacts, as well as smart tags for attaching to items and recording movement, vibrations, and shocks to ensure their safety, security, and integrity.

With its wide acceleration measurement range, ST’s sensor will also enable new generations of smart devices for sectors such as consumer healthcare and industrial safety. Potential applications include personal protection devices for workers in hazardous environments, assessing the severity of falls or impacts. Other uses include equipment for accurately assessing the health of structures such as buildings and bridges.

The sensor’s high integration simplifies product design and manufacture, enabling advanced monitors to enter their target markets at competitive prices. Designers can create slim, lightweight form factors that are easy to wear or attach to equipment.

The post STMicroelectronics combines activity tracking and high-impact sensing in miniature AI-enabled sensor for personal electronics and IoT appeared first on ELE Times.

Courtesy: Avnet

Myth #1: Demo code is production-ready
AI demos always look impressive but getting that demo into production is an entirely different challenge. Productionizing AI requires effort to ensure it’s secure, optimized for your hardware, and
tailored to meet your specific customer needs.
The gap between a working demonstration and real-world deployment often includes considerations like performance, scalability
and maintainability. One of the biggest hurdles is maintaining AI
models over time, particularly if you need to retrain the application
and update the inference engine across thousands of deployed devices. Ensuring long-term support, handling versioning and managing updates without disrupting service add layers of complexity
that go far beyond an initial demo.
Additionally, the real-world environment for AI applications is dynamic. Data shifts, changing user behavior, and evolving business
needs all require frequent updates and fine-tuning.
Organizations must implement robust pipelines for monitoring
model drift, collecting new data and retraining models in a controlled and scalable way. Without these mechanisms in place, AI
performance can degrade over time, leading to inaccurate or unreliable outputs.
Emerging techniques like federated learning allow decentralized
model updates without sending raw data back to a central server,
helping improve model robustness while maintaining data privacy.

Myth #2: All you need is Python
Python is an excellent tool for rapid prototyping, but its limitations
in embedded systems become apparent when scaling to production.
In resource-constrained environments, languages like C++ or C
often take the lead for their speed, memory efficiency and hardware-level control. While Python has its place in training and experimentation, it rarely powers production systems in embedded
AI applications.
In addition, deploying AI software requires more than just writing
Python scripts. Developers must navigate dependencies, version
mismatches and performance optimizations tailored to the target
hardware.
While Python libraries make development easier, achieving real-time inference or low-latency performance often necessitates
re-implementing critical components in optimized languages like
C++ or even assembly for certain accelerators. ONNX Runtime and
TensorRT provide performance improvements for Python-based AI
models, bridging some of the efficiency gaps without requiring full
rewrites.

Myth #3: Any hardware can run AI
The myth that “any hardware can run AI” is far from reality. The
choice of hardware is deeply intertwined with the software requirements of AI.
High-performance AI algorithms demand specific hardware accelerators, compatibility with toolchains and memory capacity. Choosing mismatched hardware can result in performance bottlenecks or even an inability to deploy your AI model.
For example, deploying deep learning models on edge devices requires selecting chipsets with AI accelerators like GPUs, TPUs or
NPUs. Even with the right hardware, software compatibility issues
can arise, requiring specialized drivers and optimization techniques.
Understanding the balance between processing power, energy consumption, and cost is crucial to building a sustainable AI-powered
solution. While AI is now being optimized for TinyML applications
that run on microcontrollers, these models are significantly scaled
down, requiring frameworks like TensorFlow Lite for Microcontrollers for deployment.

Myth #4: AI is quick to implement
AI frameworks like TensorFlow or PyTorch are powerful, but they
don’t eliminate the steep learning curve or the complexity of real-world applications. If it’s your first AI project, expect delays.
Beyond the framework itself, one of the biggest challenges is creating a toolchain that integrates one of these frameworks with the
IDE for your chosen hardware platform. Ensuring compatibility, optimizing models for edge devices, integrating with legacy systems,
and meeting market-specific requirements all add to the complexity.
For applications outside the smartphone or consumer tech domain,
the lack of pre-existing solutions further increases development
effort.

Myth #5: Any OS can run AI
Operating system choice matters more than you think. Certain AI
platforms work best with specific distributions and can face compatibility issues with others.
The myth that “any OS will do” ignores the complexity of kernel
configurations, driver support and runtime environments. To avoid
costly rework or hardware underutilization, ensure your OS aligns
with both your hardware and AI software stack.
Additionally, real-time AI applications, such as those in automotive
or industrial automation, often require an OS with real-time capabilities. This means selecting an OS that supports deterministic execution, low-latency processing, and security hardening.
Developers must carefully evaluate the trade-offs between flexibility, support, and performance when choosing an OS for AI deployment. Some AI accelerators require specific OS support.

What’s Next for AI at the edge?
We’re already seeing large language models (LLMs) give way to
small language models (SLMs) in constrained devices, putting the
power of generative AI into smaller products. If this is the direction
you’re going, talk to the experts at Witekio.

The post 5 myths about AI from a software standpoint appeared first on ELE Times.

Author: Lisa Trollo, MEMS Sensors Ecosystem and Digital Marketing Manager, STMicroelectronics

Lisa Trollo, MEMS Sensors Ecosystem and Digital Marketing Manager, STMicroelectronics

Biosensors are devices that can monitor physiological states, like heart rate or blood pressure, or detect biological parameters such as glucose levels or the presence of specific proteins in the blood.

The information biosensors collect can be used to support a medical diagnosis (for instance, a specific infection) or to provide feedback to the user on parameters of interest (for instance, the number of calories burned in a workout).

Originally developed in the 1960s for medical diagnostics, biosensors are now used by a diverse range of people – including medical patients, healthcare professionals, athletes, industrial workers, and even everyday consumers – to track their health, improve performance, and enhance safety.

Where biosensors are found

Biosensors are becoming an indispensable part of modern life. They are integrated into smartphones, smartwatches, and other wearable tech. From rings and earbuds to headsets, smart patches, and even clothing, biosensors make it easy to track health data in real-time.

In the consumer healthcare sector, wearable biosensors focus on detecting physiological signals for personalized health tracking – like monitoring athletic performance through smart watches, chest-bands or other accessories. Through these devices, they offer personalized health tracking, helping people monitor sleep quality, fitness progress, and overall wellbeing.

Biosensors are also a key component of medical devices like cardio or smart patches. They enable real-time monitoring of heart activity, glucose and various metrics, like sodium, potassium or calcium levels. These are used mainly for management of diabetes and to ensure timely medical interventions and personalized healthcare.

The science of detection: how do biosensors work?

Biosensors measure various biological levels and changes in the body, including heart rate, respiration, muscle activity, and blood oxygen levels. They use various technologies to convert these changes into electrical signals that can be used to provide real-time data for the users.

One of key characteristics of many biosensors is that they are non-invasive – meaning that they measure what is going on in the body from outside the body. This has been key for the proliferation of biosensors in consumer devices.

Biosensors take this measurement in a variety of ways.

Heart rate monitoring can be done by EKG (electrocardiogram) and augmented through a context aware analysis done with the fusion of motion signal captured by an accelerometer, and even by shining a light on the skin and collecting the reflected or transmitted wave of the light with a photodetector. Human body temperature can also be measured using infrared light to measure the temperature of the skin.

Hydration monitoring sensors, typically found in smartwatches or fitness bands, monitor hydration levels through bioimpedance or sweat analysis. In this case, biosensors aim to measure more electrolytes for single tests, which provides users with real-time analysis of their hydration and concentration levels of analytes like sodium and potassium.

It is not yet possible for all measurements to be made non-invasively. For diabetics, the current continuous glucose monitoring devices still require a small sensor wire to be inserted under the skin to measure the glucose levels in interstitial fluid. This is big improvement versus the multiple daily finger pricks needed before.

Biosensors and semiconductors

The semiconductor industry has played a crucial role in their evolution by enabling more precision, functionality, and miniaturization of biosensor devices. Advancements in biosensor technology have enabled them to be connected to IoT devices for seamless data sharing between devices; to become more sophisticated data-processors; and to be integrated into biocompatible materials to enable them to be worn close to the skin without causing discomfort – therefore enhancing the quality of data that can be captured from the human body.

Data protection and privacy

The growing use of biosensors has also raised a number of questions about issues of privacy. These devices collect vast amounts of personal data. Manufacturers are ensuring this data is encrypted, and protected by privacy laws like GDPR in Europe and HIPAA in the U.S. The future of biosensors must balance technological advancement with stringent data security to maintain user trust.

The growing biosensors market

The biosensors market is growing fast. Industry intelligence company Yole says the wearable biosensor sector has a growth rate of over 8%. While the technology wave in the 2010s featured fitness trackers and smartwatches, technology progression has advanced to so-called “hearable” devices, such as wireless earbuds, that can also track health data. Yole also expects biosensors to be used in augmented reality (AR) technology, furthering its use beyond its original application.

The future of biosensors: what’s next?

While smartwatches and fitness trackers have paved the way, upcoming innovations in hearables (earbuds that monitor health), augmented reality glasses, smart patches and smart clothing will push the boundaries of what biosensors can do. As demand for these devices increases, the focus will shift to making them more energy-efficient, secure, and even more embedded in daily life.

Expect biosensors to become an essential tool for tracking health and wellness in the years to come.

The post Proof of Life: The rapid evolution of biosensors for fitness, health, and wellness appeared first on ELE Times.

Bosch Digital Fuel Twin documents the use of renewable synthetic fuels

• Digital Fuel Twin enables digital documentation and provides evidence of climate-friendly fleet operation
• The Bosch solution makes all important data on fuel properties and quantities used available via the cloud.
• Digital Fuel Twin paves the way for combustion vehicles to run on carbon-neutral fuels.

Vehicle fleets are a driver of carbon dioxide emissions, particularly for freight forwarders and transport companies. Opting to use renewable synthetic fuels can greatly reduce their carbon footprint – but documenting this, say for sustainability reports, is a challenge. That’s precisely where Bosch’s Digital Fuel Twin comes in: this software solution, integrated into the vehicle, records the use of climate-friendly fuels and documents the reduced carbon emissions. “Bosch’s Digital Fuel Twin makes it easy for companies to prove that they’re using renewable synthetic fuels,” says Thomas Pauer, the president of Bosch’s Power Solutions division. “It gives them auditable proof of the quantities and the carbon footprint of the fuel used per vehicle, which they can then use in their reporting.” In this way, companies not only comply with ever increasing reporting obligations, but can also document their environmental awareness. The Digital Fuel Twin is currently being used on the Tour d’Europe for the first time, which will also stop off at Bosch in Feuerbach on May 28. This rally to Brussels will see a fleet of cars and trucks with combustion engines refueling exclusively with renewable synthetic fuels at public filling stations as they make their way across Europe

“Bosch’s Digital Fuel Twin makes it easy for companies to prove that they’re using renewable synthetic fuels. It gives them auditable proof of the quantities and the carbon footprint of the fuel used per vehicle, which they can then use in their reporting.“
Thomas Pauer, the president of Bosch’s Power Solutions division

A further field of application for the Digital Fuel Twin would open up in the event that it becomes possible to reclassify vehicles with combustion engines as zero-emission vehicles if they use only renewable synthetic fuels. The EU intends to review this option this year. Its current plan starting in 2035 is to fine all manufacturers of combustion vehicles at such a high level as to make it no longer economically viable to sell them. “Renewable synthetic fuels should be a part of the solution. That’s the only way to achieve the climate targets in the transport sector,” Pauer says. “If the EU decides in favor of reclassification, the Digital Fuel Twin can be an important tool in implementing that.”

Purely digital records, plausibility checks, and documentation

The new Bosch software enables the reliable tracking of all a fuel’s climate-relevant properties: from production through all stages of the supply chain to the filling station and into the vehicle. To begin with, manufacturers of renewable synthetic fuels report to Bosch how much fuel they have sold, to whom, and what the fuel’s carbon footprint is. Transport companies in turn report how much fuel they purchased and when. The Digital Fuel Twin compares this data. If the time and quantity match both in the respective company books and with the recorded pump and sensor data of the transfer interfaces, the fuel properties – the type of fuel, its CO2 content, and reduction potential – are passed on in the supply chain. Any carbon emitted during further transportation is reassigned to the fuel – meaning the shorter the distances, the better for the climate. Finally, at the filling station, a “digital handshake” – an exchange of data between the filling station, vehicle, and cloud – documents exactly how much and what kind of fuel was purchased. Identification is carried out using, for example, a fleet management system. This database provides users of the Digital Fuel Twin with reliable information about the CO2 values of the fuel used as well as auditable proof of use. The fuel data is always mapped digitally as a virtual twin in a protected data room in the cloud. Bosch’s software solution can be used in cars, trucks, and buses, but also in construction vehicles and even ships.

The Digital Fuel Twin is currently undergoing testing in collaboration with many participants along the entire fuel supply chain. The system’s reliability and safety is being tested together with them and with vehicle manufacturers. To date, the Digital Fuel Twin has been retrofitted into vehicles. In the future, however, the plan is to integrate it into the vehicle’s own electronics as a pure software module, thereby ensuring the tamper-proof use of renewable synthetic fuels at the individual vehicle level. “We expect the Digital Fuel Twin to feature in production vehicles as early as 2026,” Pauer says

Renewable synthetic fuels have been available for many years

Renewable synthetic fuels are produced either from plant-based materials or with the help of renewable electricity. In contrast to fuels based on crude oil, they do not release any additional carbon dioxide into the atmosphere. Some of these fuels have been available for years. The most widely used is HVO100 (100 % recycled hydrotreated vegetable oils), which is obtained from waste oils and plant residues. Overall – taking into account the carbon emissions of the fuel itself plus the carbon emitted during its production (“well-to-wheel”) – this diesel fuel offers a CO2 advantage of up to 90 percent compared to its crude oil counterpart. Sales of this fuel have been freely permitted in Germany since 2024, but it has been available for much longer in countries such as Sweden and the Netherlands. For gasoline engines, there is also the ethanol-based fuel E85. Both fuels, HVO100 and E85, are each already available at more than 5,000 filling stations across Europe.

The post Providing seamless proof of sustainability appeared first on ELE Times.

Delta, a global leader in power management and smart green solutions, today unveiled its comprehensive solutions for the AI era with a focus on sustainability under the theme “Artificial Intelligence x Greening Intelligence.” The showcase features the newly launched AI containerized data center solution designed for edge computing. This 20-foot container, which integrates power, cooling, and IT equipment, is on display at Delta’s booth and drawing significant visitor interest. Delta is also announcing new certification for the in-rack CDU solution for NVIDIA GB200 NVL72, delivering more solutions for in-rack cooling for NVIDIA’s customers.  In response to the growing power demands of AI computing, Delta also introduces an innovative 800V High Voltage Direct Current (HVDC) power architecture solutions for AI data centers, along with a microgrid solution that enhances grid resilience. With its comprehensive developments in grid-to-chip power and thermal management solutions, Delta aims to optimize energy efficiency in the AI era and enable a sustainable AI future.

Ping Cheng, Delta’s Chairman and CEO, said, “With the rapid expansion of AI applications, industries worldwide are facing the dual challenge of meeting computing demands while maintaining sustainability. As a global leader in power and thermal management, Delta strives to enhance the energy efficiency of its products and optimize power architectures to reduce the stage of energy conversion and minimize total energy loss. For enterprise users looking to adopt AI, we also address the need for rapid and simplified deployment by offering a highly integrated containerized data center solution, including for NVIDIA GB200 NVL72. Through innovative technology, Delta is helping drive the development of sustainable AI.”

Power Solutions and Thermal Management for AI Data Centers

Benjamin Lin, President, Delta Electronics India said, “As India rapidly advances toward becoming a global technology and data hub, the demand for energy-efficient, AI-ready infrastructure is accelerating. Delta’s containerized data center and HVDC solutions represent our commitment to driving digital innovation while ensuring sustainability at scale. These next-generation technologies not only empower faster deployment and lower operational costs, but also align with India’s green data center and Digital India missions. We are proud to contribute to building a resilient digital future, where high-performance computing and clean energy solutions go hand in hand.”

As part of its HVDC solution, Delta showcases its Core Shell Liquid-Cooled Busbar and HVDC Air-Cooled Busbar, supporting up to 50VDC/8000A and 800VDC/1000A power capacity to ensure stable system operation. In advanced liquid cooling, Delta’s liquid-to-liquid cooling systems can provide up to 1,500 kW of cooling capacity. Delta also features rack-level coolant distribution units (CDUs) with cooling capacity up to 200kW, along with liquid-cooled cold plate modules designed for GPUs and CPUs, delivering robust thermal support for next-generation chips.

ICT and Energy Infrastructure Solutions for AI Data Centers

Kelvin Huang, VP and General Manager of Delta’s ICT Infrastructure Business Group, said, “In response to the high power consumption and high-density computing demands of AI servers, we showcase our AI containerized data center solution. Compared to traditional data centers, it can be deployed within weeks, significantly shortening construction time and reducing costs. It allows flexible deployment in remote areas, making it ideal for AI computing, enterprise edge nodes, and telecom facilities.”

Additionally, Delta also highlights that data centers are rapidly adopting microgrids and renewable energy solutions. With key technologies such as hydrogen energy and energy storage, Delta can integrate diverse power sources and dynamic load demands. Through intelligent energy dispatching, Delta’s solutions enable optimal energy allocation and stable power supply.

The post Delta Presents Comprehensive Solutions for AI Data Center with Containerized Data Center & HVDC Power Solution at COMPUTEX 2025 appeared first on ELE Times.

While autonomous technology, the electric vehicle (EV) and increasing expectations of environmental sustainability and fuel economy change the nature of the automotive industry, CFD (Computational Fluid Dynamics) remains the heart of these technologies: an instrument of great power that allows an engineer to model, analyse and optimise fluid flows inside and outside of automobiles. This article discusses, in detail, the latest developments in automotive CFD, where CFD is applied, and in what new ways the scope is being stretched, and also delves on the key contribution Cadence has made toward breaking the frontiers of automotive simulation and design Cadence being the world leader in computational software.

Why CFD is Important in Automotive Design

CFD involves the numerical simulation of fluids flows by employing mathematical models and computational algorithms. In the automotive sector, CFD is used extensively in the modeling of airflow over the vehicle body, thermal management systems, engine cooling, aerodynamics, underhood airflow, HVAC etc. Being able to recreate complex fluid behaviors virtually offers a lot of advantages. It can help cut costs on building physical prototypes, speed up the design process and lead to better vehicle performance and efficiency. Plus, it can make rides more-comfortable and safe for passengers, all while shortening the development time.

Major Uses of CFD in the Automotive Sector

1. Aerodynamic Optimization

Shaping the vehicle to minimize drag and save fuel.

Reducing wind noise and lift forces for better ride quality and handling.

Optimization components like spoilers, grilles and underbody panels.

2. Thermal Management

Modeling airflow in the engine compartment to achieve best cooling.

Regulating battery temperatures for electric vehicles.

Improving cabin climate control and occupant comfort.

3. Combustion and Powertrain Simulation

This involves the modeling of air-fuel mixture and vehicles using I/C (Internal            Combustion) engines.

Simulating and analyzing lubrication and heat-dissipation in engine components.

4. EV and Battery Cooling

Cooling lithium-ion battery packs in a uniform manner to avoid further degradation and ensure safety.

Cooling inverter and electric motor loadings.

5. HVAC and Cabin Comfort

Modeling air distribution and temperature control within the cabin.

6. Water and Contaminant Management

Predicting the interaction between rain, snow or dirt with a car exterior.

Ensuring visibility to drivers and protection of sensitive components.

Technological Developments in Automotive CFD

With greater computational power comes the need for punishingly extended computations with highly detailed simulations, the domain of CFD is quickly changing with time. Few trends are:

High-Fidelity Simulation Modern solvers can tackle huge simulations with high mesh resolutions to give detailed insights into complex mechanisms such as turbulence, transient flows.

Machine Learning and AI Integration AI algorithms have been extended to predict fluid performance, optimize design parameters and automate parts of CFD workflow, thus reducing turnaround time.

Cloud Computing and HPC (High Performance Computing) Cloud-based simulation platform rendered CFD scalable and cost effective while speeding-up iterations and collaborative work among dispersed teams.

Digital Twins CFD plays a critical role in creating digital twins- virtual representations of physical systems that monitor, simulate and optimize performance dynamically.

Multiphysics and System-Level Simulation Integrating CFD with other physics disciplines such as structural, thermal and electromagnetic simulation allows for comprehensive system-level optimization.

With the automotive sector under transformation due to electrification, autonomy and sustainability, very high-fidelity fluid simulations are now in demand. Several big names like ANSYS, Altair, Cadence Design Systems have moved into leadership in computational fluid dynamics (CFD), each with a set of capabilities all their own.

How Cadence is shaping Automotive CFD

Cadence Design Systems, once considered and electronics design automation (EDA) powerhouse, now includes system-level modeling, including CFD, with acquisition of NUMECA and Pointwise. Since then, they have continuously expanded their CFD offerings into a powerful suite, disrupting how automotive engineers think fluid simulation.

Key Offerings of Cadence Automotive CFD

• Fidelity CFD Platform The Cadence proprietary CFD platform – Cadence Fidelity CFD, gives a complete and accurate solution for simulating complex fluid dynamics in automotive and aerospace systems. Included are the following capabilities:

-Unstructured and structured meshing (via Pointwise)

-Advanced turbulence models (RANS, LES DES)

-High-order numerical solvers

-Multi-domain and multi-physics simulation

-Automated workflows for design exploration

• Omnis Simulation Environment Cadence’s Omnis environment binds different simulation technologies into a single platform. It supports aerodynamics, acoustics, combustion and multiphase flows with AI-based mesh generation and user-friendly automation.
• Superior Meshing Using Pointwise Quality of meshing is vital to obtaining a good CFD solution. Pointwise offers the best meshing tools in the industry, with high-quality hexahedral and hybrid meshes to guarantee accuracy even for the most challenging automotive geometries.
• Faster Experiments with HPC and Cloud Link using Cadence CFD tools, big tests can run very fast because these tools are made to work well with HPC systems and easily use extra space in the cloud. This helps learn new things quickly while making designs.
• Works with ECAD for Electronic Cooling as cars get more electric wiring it is very important to keep parts like ECUs and power electronics from getting too hot. Cadence helps link ECAD and CFD which makes it easier to check electronic cooling.

Real-World Impacts

CFD tools provided by Cadence are being adopted by leading automotive OEMs and suppliers the world over. Some of them are:

• Reducing aerodynamic drag of next-generation electric vehicles
• Optimizing thermal management systems in battery electric buses
• Stimulating underbody airflow for race cars
• Managing cabin comfort and HVAC systems in luxury sedans

Challenges and The Road Ahead

Yet, all the advances still come with challenges:

• CFD simulations remain expensive, in terms of computing requirements and also time-consuming.
• Accurate representation of turbulence and multi-phase flows continues to be one intricate problem.
• Interdisciplinary collaboration between thermal, mechanical and electronic teams needs better integration.

However, companies like Cadence are actively working to address these gaps through Mesh refinement and solver tuning driven by AI, End-to-end automation of CFD workflows, Better user interfaces and learning tools for new users.

Conclusion:

CFD goes far beyond being just a means of simulation; it acts as a stepping stone toward innovation for the automotive industry. The whole new paradigm that CFD helps realize sits downstream of energy efficiency and safety improvements-for growing and realizing a modern-day vehicle design with the right set of parameters.

Cadence is playing a key role in CFD development through its development of next-generation Fidelity CFD, intelligent meshing, and integrated simulation environments. As the industry goes toward a more sustainable and connected form of mobility, CFD will become more inseparable, keeping Cadence ahead of the computational curve.

The post Designing the Drive: CFD as the Engine of Automotive Innovation appeared first on ELE Times.

The embedded systems landscape is evolving rapidly in demand for AI/ML capabilities, real-time data processing and security, coasting microcontroller manufactures down a path to deliver solutions that perform high on price of power and lower cost. Hence Microchip Technology Inc. the global leader in embedded control solutions, had to rise to the challenge with its latest PIC32A series of microcontrollers.

In an exclusive interview with ELE Times, Pramit Nandy, Product Marketing Manager at Microchip’s dsPIC Business Unit, focused on motor control applications and having earned a master’s degree in electrical engineering from Arizona State University, discusses key insights into technological innovation, market strategies, and applications landscape for PIC32A MCUs.

ELE Times: What prompted the launch of the PIC32A series of MCUs? Was it based on specific market demands or technological trends?

Pramit: The launch of the PIC32A series of MCUs was prompted by specific market demands and technological trends, including the need for advanced processing capabilities for complex algorithms and data management, high-speed analog peripherals for precise measurements, robust safety and security features driven by automotive, industrial, and consumer segments, and the ability to handle software complexity with model-based designs.

ELE Times: How does the addition of 64-bit FPU enhance PIC32A MCU applications in AI/ML? Could you share any use cases or test data?

Pramit: The integration of a 64-bit Floating Point Unit (FPU) in the PIC32A MCU, operating at 200MHz, can help enhance AI/ML applications by accelerating complex mathematical operations, improving precision, enabling faster data processing, and supporting more sophisticated neural networks. Additionally, the 72-bit Multiply-Accumulate (MAC) unit in PIC32A MCUs facilitates faster algorithm processing, further boosting the capabilities required to run AI/ML-based algorithms, making it ideal for developing intelligent systems. Potential use cases include real-time data analysis, gesture recognition, predictive maintenance, and autonomous systems.

ELE Times: The PIC32A MCUs integrate a 40 Msps 12-bit ADC, a 5-nanosecond high-speed comparator and a 100 MHz GBWP operational amplifier. How do these devices work together to improve efficiency? How do you balance high-performance analog peripherals with power consumption during design time?

Pramit: The integrated 40 Msps 12-bit ADC, 5-nanosecond high-speed comparator, and 100 MHz GBWP operational amplifier in PIC32A MCUs work together to improve efficiency by enabling fast and accurate signal processing, reducing latency, and enhancing performance. The ADCs support up to 22 channels, enabling rapid and accurate data collection crucial for real-time applications. The DACs with high-speed comparators provide fast and precise signal generation, while the high-frequency op-amps ensure stable and efficient signal processing. The PIC32A family of MCUs are optimized for performance and not for low power consumption. Balancing high-performance analog peripherals with power consumption is achieved by optimizing the use of analog peripherals based on application requirements, leveraging power-saving modes, and efficient design practices.

ELE Times: How do hardware security features such as ECC, MBIST, and I/O integrity monitoring ensure the security of embedded systems? In which application scenarios are these functions particularly critical?

Pramit: Features such as ECC (Error-Correcting Code), MBIST (Memory Built-In Self-Test), and I/O integrity monitoring ensure the security of embedded systems by providing robust mechanisms to detect and correct errors, verify memory integrity, and monitor input/output operations for anomalies respectively. These functions are particularly critical in applications such as automotive systems, medical devices, and industrial control systems, where safety, security, reliability and data integrity are paramount.

ELE Times: The PIC32A MCUs are priced at less than $1 per unit, making them highly competitive in the 32-bit MCU market. How does Microchip achieve such a low cost while maintaining high performance?

Pramit: Microchip achieves low costs for PIC32A MCUs through efficient manufacturing processes, economies of scale, and optimized design that balances performance with cost-effective production.

ELE Times: How do you view the market prospects of the PIC32A MCUs in the coming years? What are Microchip’s market strategies for sectors in intelligent edge sensing, AI/ML and automotive electronics?

Pramit: The market prospects for PIC32A MCUs are promising, driven by increasing demand in intelligent edge sensing, AI/ML, and automotive electronics. Microchip’s strategies include enhancing product performance, expanding software and development tools, and forming strategic partnerships to address these growing sectors.

ELE Times: With the rapid development of IoT, AI/ML and intelligent edge computing, what new technological trends do you foresee in the MCU market, and how is Microchip responding to these trends?

Pramit: With the rapid advancement of IoT, AI/ML, and intelligent edge computing, the MCU market is experiencing new technological trends, including increased processing power, enhanced connectivity, and improved energy efficiency. In response, Microchip is developing MCUs that offer superior performance for efficiently running AI/ML algorithms, robust security features, and support for edge computing as well as IoT applications.

The post Microchip’s PIC32A Microcontrollers Empowering the Next Generation of Embedded Systems appeared first on ELE Times.

Expansion of the Multi Line platform with high-resolution color camera module and oblique view cameras for the inspection of protective coatings (CCI)

Protective coatings ensure the reliability of electronic assemblies. Errors in the coating can therefore have serious consequences for their function. Malfunctions or field failures cannot be tolerated, especially when used under changing climatic conditions and in safety-relevant automotive, military, avionics and telecommunications applications. Painting defects and deviations from the painting plan can now be detected automatically and reliably with the new CCI inspection system from GOEPEL electronic – for the first time simultaneously on both sides, which significantly reduces the inspection time in production.

The new Multi Line CCI for conformal coating inspection (CCI) extends the Multi Line machine platform from GOEPEL electronic. This fully automatic CCI system can inspect both from above and optionally from below. For example, dip-coated assemblies can be inspected on both sides simultaneously without having to turn the assembly over.  Fully automatic return of the assemblies below the inspection level is also possible. The CCI camera module of the Multi Line CCI is equipped with a high-resolution color camera (resolution 8 µm/17 µm), telecentric optics and high-power UV LEDs for illumination from several directions with a wavelength of 365 nm. The camera module can also be equipped with four or eight angled-view cameras in order to inspect connector pins reliably from multiple viewing directions.

The upper component clearance is a comfortable 120 mm; assemblies weighing up to 15 kg can be transported. Inspection can therefore also be carried out directly in product carriers. For seamless protective coating inspection, fluorescent coatings are illuminated by the UV LEDs in the Multi Line CCI. In conjunction with the color camera, the high-resolution, telecentric optics provide high-contrast images and enable inspection within seconds.

Programming the CCI inspection is designed to be extremely user-friendly: The system is set up within a few minutes using CAD data and painting plans. Both the painted and unpainted areas of the assembly are inspected. The new system is based on the Multi Line platform with an operating system and housing optimized for professional production applications. The tried-and-tested Pilot 7.1 operating software now features the new CCI inspection function. This covers all requirements for an all-encompassing protective coating inspection and also enables the import of a painting plan. The overarching hardware and software platform character enables flexible employee scheduling and optimizes the exchange of knowledge within the company, as the AOI is operated centrally and almost identically on all Multi Line systems, without the need for extensive training.

The post Fully automatic protective coating inspection now also on both sides and simultaneously appeared first on ELE Times.

An Integrated Tool Tailored for Engineers, Enabling Fast AI Model Deployment to MCU

As AI technology rapidly integrates into various embedded systems, how to efficiently deploy trained AI models onto resource-constrained microcontrollers has become a significant challenge for engineers. To help developers quickly implement AI solutions, Nuvoton Technology has launched the NuML Toolkit. This tool integrates model conversion, project generation, deployment, and debugging processes. It is specially optimized for the NuMicro M55M1 microcontroller platform and has received widespread acclaim from users.

The NuML Toolkit offers comprehensive process support, including:

• Model Conversion and Deployment

Automatically converts Full-INT8 quantized models into .cpp files and integrates them into Arm Keil project templates.

• Toolchain Support

Supports both Keil (μVision 5 / Arm Compiler 6) and GNU Compiler Collection (make / gcc).

• High Integration

Automatically opens the Keil project and performs flashing after deployment, enabling immediate inference result viewing.

• Flexible Expandability

Developers can adjust Tensor Arena, input / output handling, and other parameters based on model needs.

• Cross-Platform Support

Uses Miniforge to set up the environment, simplifying dependency management and quickly introducing toolchains.

Three Key Advantages of NuML Toolkit:

1. Fast and Simple Model Conversion

Through command-line operations, users can generate corresponding AI MCU projects with just a few parameters, without writing complex code manually.

2. Integrated Workflow

From model quantization → project building → microcontroller flashing → result verification, the toolkit provides a fully integrated workflow that is easy to use for both beginners and experienced engineers.

3. High Flexibility and Maintainability

The generated Keil project contains a modular code architecture (such as Application, Model, Device), making it easy for customers to modify and maintain according to actual needs. Model management is flexible—large models can be configured to use SD cards or HyperRAM, with memory usage strategies automatically adjusted based on resource allocation.

The NuML Toolkit is a highly efficient tool designed for AI applications on microcontrollers, helping customers complete AI model deployment and validation in the shortest time. It has already been widely applied in applications such as smart desk lamps, posture recognition, sound recognition, image detection, and other scenarios, becoming one of the preferred tools in the Nuvoton NuMicro AI MCU developer community.

The post Nuvoton Launches Highly Efficient AI MCU Deployment Tool “NuML Toolkit” to Accelerate Embedded Intelligent Application Implementation appeared first on ELE Times.

High-density power modules and 48V power delivery network enable midrange vehicles to adopt luxury feature

Xiamen Hongfa Electroacoustic Co., Ltd (Hongfa) has designed the industry’s highest performance and smallest active suspension power system with a goal to drive what has been a long-time luxury feature into the midrange vehicle class. Overcoming decades of false starts by prominent automotive technology providers, Hongfa struck a balance between the need to manage active suspension system size, weight and transient performance against the requirement for higher efficiency, improved EMI and symmetrical regenerative power capacity.

As an automotive power management and distribution specialist, whose relays are enabling new 48V zonal architectures, Hongfa partnered with Vicor to integrate a fixed-ratio 800V-48V DC-DC converter power module that sits alongside a network of sensors, electromechanical actuators and sophisticated software to adjust vehicle suspension in real time. The collaboration yielded the smallest active suspension system – almost half the size of the nearest competitor – with the industry’s fastest power conversion transient response.

High-voltage batteries and a 48V PDN trigger a revolution for the Hongfa active suspension vision

Whether a vehicle is on a worn highway, a backcountry dirt lane or navigating suburban potholes, active suspension provides better handling, a smoother, safer ride and reduced road noise. Early industry efforts in the 1970s included a complex electromagnetic solution which strained the capabilities of 12V battery power systems and required four 200-pound motors.

Historically, 12VDC has proven to be inadequate for powering active suspension motors without adding size, weight and cost. Additionally, some conventional DC-DC converters can deliver power without the need for an intermediate battery, but the tradeoff is that they are bulky and lack the fast transient response time required to meet the regenerative demands to recoup and store power.

Despite a history of power-hungry, heavy loads, active suspension systems stand to benefit from today’s higher voltage (400V or 800V) primary batteries and the industry’s adoption of 48V power delivery networks (PDNs); a combination is already powering plug-in hybrids and BEVs.

Figure  — The Hongfa active suspension system (Power System Specs HF3661 800V-48V DC-DC System) is liquid cooled and is the most compact on the market, weighing 2.6kg and measuring 197 x 201 x 71mm.

Vicor fixed-ratio 800V to 48V DC-DC BCM bus converters are inherently bidirectional and provide the fastest transient response (8 million amps/second) of any DC-DC converter topology. BCMs also offer symmetrical bidirectional performance with the ability to buck or boost with the same power level. Their advanced planar packaging simplifies thermal management system design, further reducing overall footprint and weight.

Linking active suspension directly to the main battery with a bidirectional power converter enables optimal energy recuperation. Similar to how a spring absorbs and releases energy, active suspension uses regenerative shock absorbers to collect kinetic energy that is returned to the battery.

Hongfa introduces smallest active suspension system using power modules and 48V

The Hongfa solution leverages Vicor high-density power modules to create a compact (197 x 201 x 71mm) 5kW power supply for each actuator. The system can rapidly process up to 6kW of peak power in either direction. The design of the converter is greatly simplified by using a pair of Vicor BCM6135 modules operating in parallel, instead of several hundred discrete components.

The converter is optimized to work with 800V battery systems and has an operating range between 420V to 920V. With liquid cooling it can deliver up to 100A of current with 97.3% efficiency. The system housing volume is under 1.8l and weighs 2.6kg, providing a major reduction over other systems.

“When it comes to active suspension, our OEM customers require a DC-DC converter with a response rate measured in milliseconds, otherwise, additional battery support is needed,” said Peter Li, Director of Research and Development at Hongfa. “Vicor’s BCM6135 power modules not only delivered the performance we need, they also significantly shortened our development time and have made designing this system much easier for us.”

Figure  — The bidirectional BCM6135 rapid current transient response rate of 8 million amps/second is a perfect match for the power profiles of active suspension and power regeneration demands. The BCM6135 is a 95% efficient 3.5kW peak power 800V-48V bus converter that can provide symmetrical power switching performance.

The combination of 48V and high-density power modules is enabling new levels of innovation in automotive electrification – reducing space, weight and providing superior performance. Together, Hongfa and Vicor are leveraging their automotive expertise to develop advanced technology and support the evolution of high-performing electric vehicles.

The post Hongfa compact active suspension power design will transform driving appeared first on ELE Times.

What’s new

Texas Instruments (TI) announced it is working with NVIDIA in the development of power management and sensing technologies for will help enable NVIDIA’s future 800V high-voltage direct current (HVDC) power distribution systems for data center servers. The new power architecture paves the way for more scalable and reliable next-generation AI data centers.

Why it matters

With the growth of AI, the power required per data center rack is predicted to increase from 100kW today to more than 1MW in the near future. To power a 1MW rack, today’s 48V distribution system would require almost 450lbs of copper, making it physically impossible for a 48V system to scale power delivery to support computing needs in the long term.

The new 800V high-voltage DC power-distribution architecture will provide the power density and conversion efficiency that future AI processors require, while minimizing the growth of the power supply’s size, weight and complexity. This 800V architecture will enable engineers to scale power-efficient racks as data-center demand evolves.

“A paradigm shift is happening right in front of our eyes,” said Jeffrey Morroni, director of power management research and development at Kilby Labs and a TI Fellow. “AI data centers are pushing the limits of power to previously unimaginable levels. A few years ago, we faced 48V infrastructures as the next big challenge. Today, TI’s expertise in power conversion combined with NVIDIA’s AI expertise are enabling 800V high-voltage DC architectures to support the unprecedented demand for AI computing.”

“Semiconductor power systems are an important factor in enabling high-performance AI infrastructure,” said Gabriele Gorla, VP of System Engineering of NVIDIA. ” NVIDIA is teaming with suppliers to develop an 800V high-voltage DC architecture that will efficiently support the next generation of powerful, large-scale AI data centers.”

The post TI teams with NVIDIA to bring efficient power distribution to AI infrastructure appeared first on ELE Times.

Nuvoton Technology Corporation Japan (NTCJ) will start a series production of automotive

HMI display IC “Gerda”, 4th generation, which are 3 types (Gerda-4M/ Gerda-4L/ Gerda-4C).

This lineup features image processing technology, security, and display safety functions for the various HMI devices which are progressing under expectation to support the higher level of safety requirement. And it contributes to realizing the HMI solutions which enhance vehicle safety and comfort.

Achievements:

1. By producing the high visibility of interior display utilities such as electronic mirrors, AR-HUD, and cluster meters, “Gerda” assists comfort and fatigue-resistant safe driving by system.
2. By high-speed and large capacity memory built in “Gerda” allows customers to develop the high-functional HMI systems with advanced display capability without external memory at reasonable cost in system.
3. “Gerda” features OTA (Over-The-Air software updating) and secure boot functions, and contributes to the safe and comfort systems by mitigating the risks of on-vehicle cyber security and supporting the functional safety level ASIL-B.

Features:

1. Refined visibility and comfort of HMI equipment is realized with unique image processing engine.

Our video processing technology is ideal for electronic mirrors, AR-HUDs, and cluster meters, which are becoming widely spread due to digitization of equipment installed in cars, to create a safe and comfortable driving environment.

–　Features for electronic mirrors

The issue with conventional e-mirrors had issues with reduced visibility, especially in night time camera images, due to glare caused by headlight light and white blown out/blacked out due to big gap in brightness.

With Gerda-4M, it is possible to improve the visibility by adjusting the picture contrast locally, in response to the brightness distribution by area (“local contrast” function), can be achieved with low-latency processing of 1 frame.

–　Features for AR-HUD

It is essential with AR-HUD that the display naturally kept shown in the driver’s eyesight by harmonizing the displayed contents to the windshield shape.

For this, displaying image needs to be projected in low latency, and flexible deformed to fit the windshield. Image warping engine allows free deformation to fit the screen at lower latency than 1 frame. And it supports “dual projection HUD” with single chip, which both 2 images are projected in near and far focuses.

  –　Features for cluster meters

It is essential with cluster-meter that multiple assisting information is conducted to drivers accurately and intuitively. And its design needs to flexibly match the interior. Nuvoton’s Gerda has 2.5D GFX engine which allows customers to design the screen with cubic effects like depth and shades added to the design by 2D objects. It provides the display with higher visibility and contributes to the accurate decision of drivers. And by combination with the image compression engine, displaying images at as high resolution as WXGA@60FPS which enables the higher-grade cockpit design.

2. Multi-functional HMI system is realized at reduced system cost with built-in high speed/ large size memory.

In general, construction of high-performance HMI systems requires external memories the system. Nuvoton’s lineup of Gen.4 is embedded with the memory of high speed and large capacity, which makes external memory unnecessary, and contributes to reduce the component counts and system cost. And Gerda-4L has built-in the dedicated image compression engine which enables the real-time processing at half memory usage than normal use, and it realizes high resolution image display as WXGA@60FPS only with the built-in memory.

 

3. Supporting Functional Safety ASIL-B, and contributing to the risk mitigation of vehicle cyber security

Gen.4 Gerda features the display safety functions as video signal monitoring, alert display monitoring etc., and contributes to realize the system of Functional Safety ASIL-B by supporting the functions as fail-safe, real-time monitoring. And the built-in HSM which supports EVITA-Full achieves the software update via network (OTA) and secure boot, which allows only trusted software to boot up. It helps to mitigate the risk of vehicle cyber security.

NTCJ developed the 3 types of 4th generation of HMI display IC “Gerda” series which is equipped with a wide range of image processing technologies and advanced security functions and has started its mass production.

The post Nuvoton Releases 4th Generation of “Gerda,” the Automotive HMI Display IC Series in 3 Variants appeared first on ELE Times.

ASMPT presents central platform for data exchange in electronics manufacturing.

“WORKS Integration ensures that all production data is available exactly where it is needed for things like production planning and setup preparation, material flow and process optimization, and quality control,” said Thomas Bliem, Vice President R&D at ASMPT SMT Solutions. “By creating a centrally networked database with high connectivity across all systems, the platform makes fully digitized and highly automated intelligent production possible.”

A central data hub for all systems

Extending across all protocols and versions, WORKS Integration establishes an IIoT communication base to which the sensors on the production hardware as well as all software applications send their data. In addition to connecting the entire hardware and software portfolio of ASMPT SMT Solutions, WORKS Integration makes it possible to integrate third-party machines and programs via industry-standard interfaces. Even complex customer-specific applications can be easily linked via proprietary interfaces.

All connected systems obtain the information they need exclusively via WORKS Integration. Operating in the background, the application enables the continuous exchange of data between different hardware and software entities irrespective of their format or manufacturer.
WORKS Integration supports proven internal machine interfaces as well as industry standards such as IPC-2591, CFX, and SECS/GEM. Adapters convert all incoming information into a shared public data structure.

Centralized, standardized, resilient

This unified system offers numerous advantages for users. Since the various instances do not communicate with each other directly but exclusively through WORKS Integration, the number of interfaces and data channels is reduced considerably, as are potential interdependencies and sources of errors.

Machines and/or programs can be easily replaced, updated and scaled up or down without creating version conflicts or redundant data streams while encrypted communication reliably protects the system from unauthorized access. In addition, a central health monitoring system provides valuable information for technical support and fault analysis.

The post WORKS Integration – the data center for intelligent manufacturing appeared first on ELE Times.

Rohde & Schwarz is hosting its fifth Satellite Industry Day on June 3, 2025. Rohde & Schwarz test and measurement experts and partners from the industry will present topics from 5G Non-Terresterial Network (NTN) and satellite testing to monitoring and regulatory issues

After the keynote speech from Rohde & Schwarz Executive Vice President Test & Measurement Christina Gessner, Reiner Stuhlfauth from Rohde & Schwarz discusses tackling the challenges of NTN evolution on the path to 6G, including technology aspects, challenges, and testing. His colleague Goce Talaganov focuses on ensuring 5G NTN device performance, presenting market trends, use cases, challenges, and test solutions.

Obilor Nwamadi from Viavi Solutions presents an end-to-end emulation system for evaluating NTN connectivity to guarantee return on investment before design or deployment.

Rohde & Schwarz expert Dr. Yvonne Weitsch explores testing and technologies for Electronically Steered Array (ESA) antennas in satellite and NTN integration, discussing their advantages and proposing test solutions.

Prof. Dr. Klaus Schilling from the Center for Telematics discusses the future of space with smarter, smaller, and more cooperative satellites, highlighting the shift to networked small satellites and innovative technologies for global connectivity.

Alexander Spaniol from Terma Technologies GmbH covers the transformative role of Software-Defined Radio (SDR) in advancing space technology, emphasizing its flexibility and role in satellite communications, testing, and ground stations.

Rohde & Schwarz expert Pia Feurstein talks about improving signal quality in satellite communication with antenna combining with the help of the R&S MSR4 multipurpose satellite receiver. Her colleague Jean-Pierre Messmer introduces Space Nexus, a software solution for end-to-end satellite link channel emulation to validate systems in a laboratory environment.

Uwe Baeder from Rohde & Schwarz discusses the regulatory framework for space communication, focusing on the ITU’s role, coordination procedures, evolving regulations, and key issues for WRC-2712.

The post Rohde & Schwarz Satellite Industry Day 2025: Connecting the world with New Space and 5G NTN technologies appeared first on ELE Times.

Space-Saving Device Offers Low Max. RthJC of 0.36 °C/W and Wettable Flanks to Improve Thermal Performance and Solderability in Industrial Applications

To provide higher efficiency for industrial applications, Vishay Intertechnology, Inc. introduced a new 80 V TrenchFET Gen IV n-channel power MOSFET in the PowerPAK 8x8SW bond wireless (BWL) package with best-in-class on-resistance. Compared to competing devices in the same footprint, the Vishay Siliconix SiEH4800EW offers 15 % lower on-resistance while reducing RthJC by 18 %.

With on-resistance down to 0.88 mΩ typical at 10 V, the device released today minimizes power losses from conduction to increase efficiency while improving thermal performance with a low maximum RthJC of 0.36 °C/W. With its 8 mm by 8 mm footprint, the space-saving device occupies 50 % less PCB space than MOSFETs in the TO-263 package while offering an ultra-low profile of 1 mm.

The SiEH4800EW implements a fused lead to increase the source PAD solderable area to 3.35 mm², which is four times larger than a traditional PIN solder area. This decreases the current density between the MOSFET and PCB, reducing the risk of electro-migration risk and enabling a more robust design. In addition, the device’s wettable flanks enhance solderability while making it easier to visually inspect the reliability of solder joints.

The MOSFET is ideal for synchronous rectification and OR-ing functionality. Typical applications will include motor drive controls, power tools, welding equipment, plasma cutting machines, battery management systems, robotics, and 3D printers. In these applications, the device offers high-temperature operation to +175 °C, and its BWL design minimizes parasitic inductance while maximizing current capability.

RoHS-compliant and halogen-free, the MOSFET is 100 % Rg and UIS tested.

The post Vishay Intertechnology 80 V MOSFET in PowerPAK 8x8SW Package Offers Best in Class RDS(ON) of 0.88 mΩ to Increase Efficiency appeared first on ELE Times.