Search Results

Introduction – ESP32-C61 Redefines Wireless IoT Connectivity Espressif Systems has introduced the ESP32-C61, a cutting-edge system-on-chip (SoC) that brings IEEE 802.11ax Wi-Fi 6 and Bluetooth 5 (LE) connectivity to resource-constrained IoT applications. As wireless standards evolve and security requirements intensify, the ESP32-C61 addresses modern design challenges whilst maintaining the cost-effectiveness and ease-of-use that engineers have come to expect from the ESP32 family. The ESP32-C61 integrates a 32-bit RISC-V single-core processor running at up to 160 MHz, delivering a CoreMark score of 553.78 at maximum frequency. With multiple variants offering different in-package flash and PSRAM configurations, the chip provides flexibility for applications ranging from basic sensor nodes to sophisticated edge AI implementations. Available in a compact QFN40 (5×5 mm) package with 30 programmable GPIOs, the ESP32-C61 supports operating temperatures from –40°C to 105°C, making it suitable for demanding industrial environments whilst remaining accessible for consumer electronics applications. Espressif's ESP32-C61 Functional Block Diagram Features of ESP32-C61 Addressing Modern IoT Requirements Wi-Fi 6 (IEEE 802.11ax) Connectivity The ESP32-C61 brings Wi-Fi 6 capabilities to the IoT market through a 1T1R configuration operating in the 2.4 GHz band (2412 – 2484 MHz). The implementation includes: OFDMA Support : Uplink and downlink Orthogonal Frequency-Division Multiple Access enhances connectivity in congested environments, allowing multiple devices to share channels more efficiently – critical for dense IoT deployments. Downlink MU-MIMO : Multi-user, multiple input, multiple output technology increases network capacity by enabling simultaneous data transmission to multiple devices. Target Wake Time (TWT) : This power-saving mechanism allows devices to negotiate when and how frequently they wake to send or receive data, significantly reducing power consumption for battery-operated devices. BSS Colouring and Spatial Reuse : These technologies maximise parallel transmissions and minimise interference in crowded wireless environments. Extended Modulation Schemes : Support for MCS0 – MCS9 in 20 MHz-only non-AP mode, with transmit power up to 19.5 dBm for 802.11ax and up to 21 dBm for 802.11b. The chip maintains full compatibility with IEEE 802.11b/g/n protocols, supporting 20 MHz and 40 MHz bandwidth with data rates up to 150 Mbps. This backwards compatibility ensures seamless integration into existing wireless infrastructures. Bluetooth 5 (LE) with Advanced Features Bluetooth Low Energy implementation in the ESP32-C61 is Bluetooth Core 6.0 certified and includes: Extended Range and Data Rates : Support for 125 Kbps, 500 Kbps, 1 Mbps, and 2 Mbps speeds, with coded PHY for extended range applications. Direction Finding (AoA/AoD) : Angle of Arrival and Angle of Departure capabilities enable precise location services for asset tracking and indoor positioning systems. Periodic Advertising with Responses (PAwR) : Enhances efficiency for applications requiring bidirectional communication with multiple peripheral devices. LE Power Control : Dynamically adjusts transmission power to optimise battery life and connection quality. Multiple Role Support : Devices can operate concurrently in Broadcaster, Observer, Central, and Peripheral roles, enabling complex network topologies. Bluetooth LE receiver sensitivity reaches –106 dBm at 125 Kbps, whilst transmit power extends up to 21 dBm, providing robust connectivity over extended distances. High-Performance 32-bit RISC-V Processor The ESP32-C61 features a custom RISC-V single-core processor (HP CPU) with: Five-Stage Pipeline : Operating at up to 160 MHz, delivering 3.46 CoreMark/MHz. RV32IMAC ISA : Base integer (I), multiplication/division (M), atomic (A), and compressed (C) standard extensions, plus Zc extensions (Zcb, Zcmp, Zcmt) for enhanced code density. 32 KB L1 Cache : Reduces latency for instruction and data access, improving overall system performance. Advanced Debug Capabilities : RISC-V trace encoder compatible with Efficient Trace for RISC-V Version 2.0, hardware breakpoints/watchpoints, and JTAG/USB debug support. Privilege Modes : Machine (M) and User (U) modes with Physical Memory Protection (PMP) for up to 16 configurable regions. Memory Architecture and External Storage Internal Memory : 256 KB ROM for bootloader and core functions 320 KB SRAM for data and instructions 4096-bit eFuse (1792 bits user-accessible) In-Package Options : ESP32-C61HF4: 4 MB Quad SPI flash ESP32-C61HR2: 2 MB Quad SPI PSRAM ESP32-C61HR8: 8 MB Quad SPI PSRAM External Memory Support : The chip supports connection to off-package flash and PSRAM via SPI, Dual SPI, Quad SPI, and QPI interfaces. Through the cache system, it can map up to 32 MB of instruction memory space and 32 MB of data memory space. The SPI clock frequency reaches 120 MHz for both in-package and off-package memory. Rich Peripheral Set Connectivity Interfaces : 3× UART (up to 5 Mbaud) General-purpose SPI (1-, 2-, 4-line modes) I2C (standard and fast modes) I2S (master/slave, full/half-duplex, TDM and PDM support) USB Serial/JTAG controller (USB 2.0 full-speed compliant) SDIO 2.0 slave controller (up to 50 MHz) LED PWM controller (6 channels, up to 20-bit resolution) Analog Interfaces : 12-bit SAR ADC with up to 4 channels Temperature sensor (–40°C to 125°C range) Analog voltage comparator System Features : Two 54-bit general-purpose timers 52-bit system timer Multiple watchdog timers GDMA controller with 4 channels Event Task Matrix (ETM) for hardware-level event handling All GPIOs feature flexible routing via the IO MUX and GPIO Matrix, allowing most peripheral signals to connect to any available pin. Detailed Specifications for ESP32-C61 Variants Power Supply and Consumption Operating Voltage : 3.0 V – 3.6 V (3.3 V nominal) on VDDA and VDDPST pins. Current Consumption : Active Mode (RF Working) : Wi-Fi TX: 802.11b @ 21 dBm: 360 mA peak; 802.11ax @ 15 dBm: 240 mA peak Wi-Fi RX: 88 mA (802.11ax, HE20) Bluetooth LE TX @ 18 dBm: 283 mA peak Bluetooth LE RX: 81 mA Modem-Sleep Mode  (160 MHz CPU): WAITI instruction: 11 mA CPU whilst loop: 16 mA Low-Power Modes : Light-sleep: 0.2 mA (all peripherals disabled) Deep-sleep: 10 µA (LP timer and memory powered) RF Performance Characteristics Wi-Fi (2.4 GHz) : RX sensitivity: –94 dBm (802.11ax, HE20, MCS0) to –68 dBm (MCS9) TX power: Up to 21 dBm (802.11b), 19.5 dBm (802.11ax) Adjacent channel rejection: 37 dB (802.11ax, MCS0) Bluetooth LE : RX sensitivity: –106 dBm (125 Kbps) to –94 dBm (2 Mbps) TX power: Up to 21 dBm with configurable output levels EVM performance exceeds Bluetooth Core specification requirements Package and Environmental Specifications Package: QFN40 (5×5 mm) 30 programmable GPIOs Operating temperature: –40°C to 105°C (High temperature variants) Storage temperature: –40°C to 150°C Industry Applications and Use Cases Smart Home Devices The ESP32-C61's Wi-Fi 6 support with TWT makes it ideal for battery-powered sensors, smart locks, and environmental monitors. Bluetooth LE enables commissioning and local control whilst Wi-Fi provides cloud connectivity. Industrial Automation Operating across extended temperature ranges with robust RF performance, the ESP32-C61 suits predictive maintenance sensors, industrial wireless controllers, and condition monitoring systems. SDIO slave functionality enables integration with existing industrial processors. Healthcare and Wearables Direction finding capabilities support asset tracking in hospitals, whilst low power consumption enables long-lasting wearable health monitors. The temperature sensor facilitates body temperature monitoring applications. Smart Agriculture Deep-sleep power consumption of just 10 µA enables multi-year battery life for soil moisture sensors, environmental monitoring stations, and livestock tracking tags deployed across large agricultural areas. POS Machines and Retail Secure wireless connectivity with hardware encryption supports payment terminals and inventory management systems. USB Serial/JTAG simplifies in-field updates and diagnostics. Edge AI and Audio Devices With 320 KB SRAM and PSRAM support, the ESP32-C61 can run lightweight AI inference for keyword spotting and simple pattern recognition. I2S interfaces with PDM support enable high-quality audio applications. Security Features for ESP32-C61 Hardware-Accelerated Cryptography ECC Accelerator : Supports P-192 and P-256 curves with 11 working modes for elliptic curve operations. ECDSA Support : Hardware acceleration for digital signature generation and verification with fixed-duration operations to resist side-channel attacks. SHA Accelerator : Hardware implementation of SHA-1, SHA-224, and SHA-256 with both CPU-based and DMA-based modes. XTS-AES Encryption : Hardware-accelerated flash and PSRAM encryption compliant with IEEE Std 1619-2007, protecting application code and sensitive data in external memory. Secure Boot and Code Protection Secure Boot : Prevents execution of unauthorised firmware through cryptographic signature verification during the boot process. Flash/PSRAM Encryption : Transparent hardware encryption/decryption allows secure storage and execution of code and data from external memory. eFuse-Based Configuration : One-time programmable eFuse bits control security features, preventing unauthorised modification of security policies. Advanced Security Features True Random Number Generator (TRNG) : Generates cryptographically secure random numbers from thermal noise and asynchronous clock sources. Access Permission Management (APM) : Controls access to memory and peripheral address spaces with configurable permissions for different bus masters. Power Glitch Detector : Real-time voltage monitoring triggers immediate chip reset upon detecting glitch attacks, with a default threshold around 2.7 V. Physical Memory Protection (PMP) : Up to 16 configurable regions restrict memory access based on privilege levels. Post-Quantum Cryptography Considerations Whilst the ESP32-C61 itself doesn't include dedicated post-quantum cryptographic accelerators, its security architecture provides a foundation for implementing PQC algorithms in software. For applications requiring quantum-resistant security, engineers should evaluate solutions like the SEALSQ QS7001 (available through Ineltek's portfolio) which includes hardware-accelerated Kyber and Dilithium algorithms. Overhead Pin-Out of Espressif ESP32-C61 Getting Started with ESP32-C61 Development Ecosystem Hardware : ESP32-C61 development boards (check Ineltek's latest offerings) Reference schematics and PCB layouts available from Espressif Software : ESP-IDF (Espressif IoT Development Framework) Arduino IDE support ESP-WIFI-MESH networking TLS 1.0, 1.1, and 1.2 support Comprehensive driver libraries for all peripherals Tools : Espressif KiCad libraries Flash Download Tool RF Test Tool for production calibration USB-JTAG debugging without external hardware Design Considerations Power Supply : Use 3.3 V regulated supply with at least 500 mA current capability. Place 10 µF and 0.1 µF decoupling capacitors close to power pins. RF Matching : The reference design includes a CLCCL matching circuit. Custom antenna implementations require careful impedance matching and RF testing. Pin Selection : Consider strapping pin requirements (GPIO7, GPIO8, GPIO9, MTMS, MTDI) for boot mode control. Review restrictions for pins connected to in-package flash/PSRAM. Thermal Management : The chip's small form factor and low power consumption simplify thermal design, but ensure adequate airflow for sustained high-performance operation. Comparison: ESP32-C61 vs. Competitor Solutions The ESP32-C61 occupies a unique position in the wireless MCU market: vs. ESP32-C3 : Adds Wi-Fi 6 support and enhanced security features whilst maintaining similar pricing and form factor. vs. ESP32-C6 : The C61 offers a cost-optimised alternative with reduced GPIO count but maintains core Wi-Fi 6 and Bluetooth 5 functionality. vs. Nordic nRF5340 : ESP32-C61 provides integrated Wi-Fi 6, eliminating the need for separate connectivity solutions in dual-radio applications. vs. STM32WB Series : Espressif's mature software ecosystem and lower pricing provide advantages for high-volume IoT applications. vs. Silicon Labs MG24 : The ESP32-C61 offers superior Wi-Fi performance and a more comprehensive peripheral set at competitive pricing. Conclusion The ESP32-C61 represents Espressif's commitment to bringing advanced wireless technologies to cost-sensitive IoT applications. By integrating Wi-Fi 6 and Bluetooth 5 (LE) with robust security features, comprehensive peripheral support, and exceptional power efficiency, the chip addresses the evolving requirements of modern connected devices. From smart home products requiring years of battery life to industrial sensors operating in harsh environments, the ESP32-C61 provides engineers with a versatile platform that balances performance, features, and cost-effectiveness. Entry level development board for ESP32-C61 See for Yourself Ready to evaluate the ESP32-C61 for your next design? Visit Ineltek's ESP32-C61 product page for datasheets, development boards, and technical support, or contact the Ineltek team for expert guidance, competitive pricing, and rapid sample delivery. FAQs for the Espressif ESP32-C61 covering Wi-Fi 6 and Bluetooth 5 implementation What Wi-Fi 6 features does the ESP32-C61 support? A. The ESP32-C61 implements key Wi-Fi 6 features including OFDMA (uplink and downlink), downlink MU-MIMO, Target Wake Time for power saving, BSS colouring, spatial reuse, and modulation schemes up to MCS9 in 20 MHz non-AP mode. It maintains backwards compatibility with 802.11b/g/n whilst delivering the efficiency improvements that make Wi-Fi 6 valuable for IoT applications. Q. How does the ESP32-C61 power consumption compare to previous ESP32 variants? A. The ESP32-C61 achieves Deep-sleep current consumption of just 10 µA with the LP system active, and Light-sleep consumption of 200 µA. During active Wi-Fi 6 receive operations, current consumption averages 88 mA, representing significant efficiency improvements over earlier ESP32 variants through both architectural enhancements and Wi-Fi 6 power-saving features like TWT. Q. Can the ESP32-C61 run alongside in-package flash and PSRAM simultaneously? A. The ESP32-C61 variants provide either in-package flash (ESP32-C61HF4 with 4 MB) or in-package PSRAM (ESP32-C61HR2 with 2 MB, or ESP32-C61HR8 with 8 MB), but not both within the same package. However, the chip supports connecting off-package flash and PSRAM simultaneously via its dual SPI controllers, allowing flexible memory configurations up to 32 MB each for instruction and data spaces. Q. What security certifications and compliance does the ESP32-C61 support? A. The ESP32-C61 includes hardware security features compliant with industry standards including secure boot, flash encryption using XTS-AES (IEEE Std 1619-2007), ECDSA digital signatures (FIPS 186-3 curves), and SHA acceleration (FIPS PUB 180-4). The chip supports building systems that meet various regulatory requirements, though specific certifications depend on the complete product implementation and testing. Q. How does Bluetooth LE direction finding work on the ESP32-C61? A. The ESP32-C61 supports Bluetooth LE direction finding through Angle of Arrival (AoA) and Angle of Departure (AoD) methods. These techniques use antenna arrays to determine the direction of incoming or outgoing Bluetooth signals, enabling precise indoor positioning and asset tracking applications with accuracy down to sub-metre levels when properly implemented with appropriate antenna configurations. Q. What development tools and software support are available for ESP32-C61? A. Espressif provides comprehensive development support including the ESP-IDF framework with complete peripheral drivers, Arduino IDE compatibility, extensive documentation, RF calibration and testing tools, USB-JTAG debugging without external hardware, and reference schematics with PCB layouts. Ineltek offers additional technical support, including Field Application Engineer assistance for complex design challenges.

Introduction – Why Engineers Need to Act Now Electronic engineers face an unprecedented challenge as the memory industry undergoes its most dramatic transformation in decades. The AI (artificial intelligence) boom has created insatiable demand for high-bandwidth memory (HBM), forcing Samsung, SK Hynix, and Micron to dramatically reshape their production priorities. This shift has triggered a supply crisis that's rippling through every segment of the memory market. The implications for design engineers are immediate and severe. Memory components that were once reliable, low-cost BOM items have become volatile commodities. The result: wafer resources are being diverted from mainstream DRAM and NAND to HBM and DDR5, leaving legacy DRAM such as DDR3/DDR4 in short supply . Anecdotally, Ineltek has already seen whole year allocations sold in a single quarter , with prices rising at pace. Projects planned around DDR4 or DDR3 may face procurement crises mid-development. Industrial systems requiring decade-long component availability are being forced into premature redesigns or expensive last-time buys. Understanding these market dynamics isn't just about cost management – it's about ensuring your products can actually reach production. With memory suppliers reporting their entire annual capacity sold within single quarters, and prices rising at unprecedented rates, engineers must act decisively to protect their supply chains. For electronic engineers, the message is clear: secure memory now, or risk redesigns and inflated costs. Ripple Effects on the Memory Supply Chain Tight supply of DDR3/DDR4:  Wafer capacity is redirected to HBM and DDR5. Surging DRAM prices:  DDR4 prices spiked by nearly 50% in a single month in early 2025. Accelerated end-of-life:  DDR4 production may end by 2026, with DDR3 close behind. NAND volatility:  Even NAND flash pricing has begun to climb after years of softness. Increased risk:  Sourcing through the spot market raises concerns about counterfeits and lifecycle mismatch. How Ineltek’s Memory Lines Provide Stability Intelligent Memory ( intelligentmemory.com ) Specialty DRAM and NAND modules. Focus on long lifecycle availability  — ideal for industrial and embedded designs. Drop-in replacements for discontinued legacy memory. Winbond ( winbond.com ) One of Taiwan’s leading memory suppliers. Ramping up DDR4 production with a new 16 nm fab line. Strong roadmap in NOR/NAND Flash and DRAM , offering stability where Tier-1 vendors are exiting. Zentel ( zentel-europe.com ) Specialty DRAM supplier focusing on DDR3/DDR4 continuity . Serves customers requiring consistent supply for long-lived industrial and consumer designs. A critical source for engineers unable to migrate immediately to DDR5. Implications for Engineers Budget impact:  Memory is no longer a commodity BOM line — costs are rising sharply. Design transitions:  Projects based on DDR4 may need redesigns for DDR5 earlier than planned. Last-time buys:  Secure inventory now for legacy systems that cannot be quickly redesigned. Alternative sourcing:  Work with trusted distributors and manufacturers (like IM, Winbond, Zentel) to avoid counterfeit risk. Conclusion - Act Now to secure your DDR3/DDR4 memory supply The HBM gold rush has reshaped the memory market  — with mainstream DRAM and NAND now caught in the supply squeeze. Prices are rising, older standards are being phased out, and long-term planning is more complex than ever. Engineers must act immediately to protect their projects from this supply upheaval. Ineltek’s memory lines — Intelligent Memory, Winbond, and Zentel  — provide engineers with practical alternatives to secure continuity and avoid disruption. Ready to secure your memory supply chain before it's too late? Contact Ineltek today to review your memory requirements and secure supply before prices climb further. Technical FAQs - How to survive the HBM Memory Gold Rush Q: Why are DDR4 prices higher than DDR5 when DDR5 is newer technology? A:  This price inversion reflects severe DDR4 supply constraints as major manufacturers focus on HBM and DDR5 production. Samsung, SK Hynix, and Micron have reduced DDR4 capacity by up to 70%, creating artificial scarcity that pushes older memory prices above newer alternatives. Q: How long will memory suppliers continue producing DDR4 after major manufacturers stop? A:  Alternative suppliers like Winbond and Zentel typically maintain legacy production for 5-10 years beyond major manufacturer EOL dates. Their business models specifically target long-lifecycle applications that require extended availability commitments. Q: What risks should engineers consider when sourcing from alternative memory suppliers? A:  Primary concerns include component authenticity, performance consistency, and long-term support capabilities. Mitigation strategies include thorough supplier qualification, multi-sourcing approaches, and working with authorised distributors who provide traceability and warranty coverage. Q: Can alternative suppliers match the performance specifications of tier-one memory manufacturers? A:  Yes, established alternative suppliers like Winbond use advanced process nodes (16nm and below) and maintain equivalent electrical specifications. Many components are pin-compatible drop-in replacements with identical timing parameters and voltage requirements. Q: How should engineers plan memory procurement in this volatile market? A:  Implement forward-buying strategies for critical components, establish relationships with multiple suppliers including alternative sources, and consider design flexibility to accommodate different memory vendors. Monitor EOL roadmaps closely and execute last-time buys before supply exhaustion. Q: What memory technologies should engineers prioritise for new designs to ensure long-term availability? A:  Focus on DDR5 and LPDDR5 for new designs requiring cutting-edge performance, whilst securing alternative sources for DDR4 and DDR3 in legacy applications. Consider industrial-grade memory variants that typically offer extended lifecycle support compared to consumer-focused components. Q. When will supply relief come? A.  Analysts expect no meaningful relief until 2026 when new fabs and HBM capacity come online.

Introduction – Why Energy Harvesting Matters for IoT Engineers The proliferation of IoT devices has created a maintenance nightmare for engineers designing connected systems. Battery replacement in remote sensors, industrial monitoring equipment, and wearable devices represents both a significant operational cost and environmental burden. EM Microelectronic's latest energy harvesting solutions directly address this challenge by converting ambient energy sources into reliable, continuous power. The company's approach centres on three core energy sources: solar photovoltaic cells optimised for indoor lighting conditions, thermoelectric generators (TEGs) that exploit temperature differentials, and emerging technologies for vibration and AC harvesting. Their silicon solutions have already proven themselves in commercial applications from Tissot solar watches to Urbanista's solar-powered headphones. Features Addressing Energy Autonomy Challenges EM Microelectronic's energy harvesting portfolio tackles the fundamental engineering challenges of ambient energy conversion: Ultra-Low Voltage Operation AT8900 thermal harvesting IC operates from 5mV input Proprietary boost converter architecture maximises efficiency under small temperature differentials Compatible with standard Coilcraft transformers for simplified design integration Intelligent Power Management AT8502 features dual storage capabilities with automatic switching Integrated USB charging for hybrid power scenarios Wake-up timers and power gating to minimise system consumption during low-energy periods Optimised Solar Performance Maximum Power Point Tracking (MPPT) algorithms specifically tuned for indoor lighting conditions Support for both single-cell and multi-cell solar configurations Fast start-up capability even with depleted energy storage Complete Development Ecosystem Comprehensive evaluation boards (EMEVB8900, EMEVB8502) with four different configurations Energy budget simulation tools available on EM's website Application notes covering TEG selection and solar system optimisation The AT8900's standout capability lies in its thermoelectric harvesting performance. Operating from temperature differentials as small as 1°C, it can power wireless sensor nodes using nothing more than the heat differential between a radiator valve and ambient air. This has enabled applications like connected heating controls that never require battery replacement. Product Portfolio Snapshot Part Application Key Features EM8500 Solar (single cell) PMIC, dual storage, ultra-low consumption EM8502 Solar (multi-cell) Software-based DC/DC, MPPT, hybrid light optimised EM8504 DSSC harvesting Designed for dye-sensitised single cell EM8506 Compact solar Ultra-low power, small coil support EM6890 MCU with harvesting Mid-high power range, true MPPT EM8900 Thermal harvesting Ultra-low voltage boost for TEGs Industry Applications and Use Cases Wearables Solar watches (e.g., Tissot T-Touch Connect). Solar headphones (Urbanista). Wearables benefit from EM’s fast cold-start and hybrid indoor light support. Smart Home & Consumer Solar remote controls and toll tags show how everyday devices can avoid battery swaps. BLE advertising mode powered by thermal harvesting enables battery-free smart sensors . Industrial & Automotive Thermal harvesting supports connected valves, condition monitors, and predictive maintenance nodes . EM PMICs integrate with BLE platforms to deliver wireless, battery-independent IoT nodes . Conclusion – Enabling the Next Generation of Autonomous Devices EM Microelectronic's energy harvesting solutions represent a significant step towards truly autonomous IoT systems. The AT8900's ability to extract useful power from minimal thermal gradients, combined with the AT8502's sophisticated solar power management, provides engineers with proven technologies to eliminate battery maintenance across a wide range of applications. The technology has already demonstrated commercial viability in demanding consumer applications, from luxury Swiss watches to premium audio equipment. As the IoT ecosystem continues to expand, particularly in industrial monitoring and smart building applications, energy harvesting will transition from a premium feature to an essential capability. For engineers evaluating energy harvesting for their next project, EM's comprehensive development ecosystem significantly reduces the typical barriers to adoption. The combination of proven silicon solutions, detailed simulation tools, and extensive application support makes it feasible to integrate energy harvesting into products where battery replacement would otherwise represent a significant operational challenge. Ready to eliminate battery replacement from your IoT designs? Contact our technical team  to discuss how EM Microelectronic's energy harvesting solutions can transform your next project into a truly autonomous system. Technical FAQs Q: What's the minimum temperature differential needed for the AT8900 to operate? A:  The AT8900 can begin harvesting energy from temperature differentials as low as 1°C, though practical applications typically see 3-5°C differences. Output power scales with the square of the temperature differential, so even modest thermal gradients can provide sufficient energy for low-duty-cycle wireless applications. Q: How does the AT8502 handle varying solar conditions throughout the day? A:  The AT8502 incorporates intelligent Maximum Power Point Tracking that continuously optimises energy extraction as lighting conditions change. Its dual storage system maintains power availability during extended low-light periods, whilst the wake-up timer system ensures the device remains responsive even when energy storage is depleted. Q: Can these ICs work together in a hybrid energy harvesting system A:  Absolutely. The AT8900 and AT8502 can be combined to create systems that harvest from both thermal and solar sources simultaneously. The switch control functionality allows automatic source selection based on availability, maximising energy capture across different environmental conditions. Q: What solar cell technologies are compatible with EM's harvesting ICs? A:  The portfolio supports conventional silicon solar cells, dye-sensitised solar cells (DSSC) through the AT8504, and Exeger's Powerfoyle technology. Each IC variant is optimised for specific cell characteristics, ensuring maximum energy transfer efficiency. Q: How do these solutions compare to competitors in terms of efficiency? A:  EM's benchmarking data shows their ICs maintain over 85% efficiency across a broader range of input power levels compared to competitive solutions. This is particularly evident in indoor solar applications where competing solutions often drop below 60% efficiency under low-light conditions. Q: What development support is available for engineers evaluating these technologies? A:  EM provides comprehensive development platforms including evaluation boards, energy budget simulation software, and detailed application notes. The company also offers direct technical support for custom transformer design and system optimisation, particularly valuable for high-volume applications requiring bespoke energy harvesting solutions. Q. How do EM PMICs achieve ultra-low start-up? A.  Devices like the EM8900 operate from <10 mV input, enabling energy capture from very small thermal gradients. Q. Can EM controllers manage both supercaps and batteries? A.  Yes, most EM850x devices feature dual-storage paths with switch and LDO management.

Introduction – Why Compact High-Voltage Drivers Are Key to Embedded Safety Feedback In many safety-critical systems—whether medical monitors, smart alarms, or compact personal wearables—reliable, high-voltage actuation is essential for audible alerts, haptic feedback, or atomisation. But the design challenge is clear: engineers must deliver high-energy output from a low-voltage supply , within constrained footprints and power budgets. The MAS2808  from Micro Analog Systems offers a dedicated solution. With an integrated high-efficiency boost converter , short-circuit-protected differential driver outputs, and both pin-based and I²C-selectable output voltages , the MAS2808 is ideally suited to safety sounders , atomisers , and low-power devices that demand strong user feedback —without the bulk or complexity of external driver circuitry. This article explores the MAS2808 in depth and shows how it addresses common design constraints in embedded safety and alerting systems. MAS2808 Key Features – Designed for Safety and Efficiency in Low-Power Systems The MAS2808 is purpose-built to deliver high-voltage actuation in compact embedded systems. Its architecture combines a fully integrated boost converter  with a differential piezo driver stage , supporting both single-ended and BTL configurations  for maximum output swing. High-Voltage Output from Low Supply Operates from 2.6V to 5.5V  input supply Integrated DC/DC boost converter delivers up to 30V output , enabling up to 60Vpp differential drive Eliminates the need for an external Schottky diode or high-voltage power rail Flexible Voltage Control: Pin or I²C 4-level output voltage selection  via simple SCL/SDA pin states for minimal MCU overhead 8-level voltage selection  via I²C interface for dynamic control and calibration I²C device address is 0x76; supports OVC programming for output voltage selection from 9V to 30V Safety-First Power Management Soft-start mechanism  to limit inrush current Piezo output short-circuit protection Thermal shutdown  and undervoltage lockout  to prevent malfunction during fault conditions Extremely low leakage current (<0.5 µA)  when disabled via the EN pin Minimal External Components Internal switching FETs reduce external BOM Recommended with small shielded inductors (2.2–2.7 µH) and ceramic capacitors No high-voltage routing or external driver stage required, simplifying PCB layout and EMI control Together, these features make the MAS2808 an ideal drop-in solution for designers who need robust high-voltage actuation—without compromising board space, cost, or power efficiency. Example Use Cases – Audible and Haptic Feedback for Safety and Alerts The MAS2808’s unique combination of high-voltage drive, compact footprint, and ultra-low standby current makes it ideal for a range of safety-focused embedded applications , particularly in battery-powered or space-constrained devices . 1. Smoke and Gas Alarms – Loud, High-Integrity Alerts In alarm systems, ensuring the user hears a fault condition is paramount. The MAS2808 can drive a piezo sounder at up to 60V pp , generating high sound pressure levels (SPL) from a 3V coin cell or lithium-ion battery. With its soft-start boost converter and thermal protections, the IC maintains safe operation even during long alert cycles or fault conditions like shorted transducers. 2. Medical Wearables – Haptic Feedback in Compact Form Devices such as portable blood glucose meters , pulse oximeters , or medication reminders  often use haptic or audible prompts to assist users with impaired vision or during timed dosing cycles. The MAS2808 supports low-leakage standby (40nA) , simple MCU interfacing , and flexible I²C voltage control , making it suitable for wearables with demanding power budgets. 3. Smart Tags and Trackers – Discreet Alerts, Long Battery Life For asset tracking devices  or personal safety tags , the MAS2808 allows short, sharp vibration bursts or tone playback without constant MCU supervision. Its low quiescent current and flexible enable control (EN pin) help prolong standby life while maintaining rapid wake-up and response. 4. Atomisers and Dispensers – Compact High-Voltage Drive Ultrasonic atomisers for air fresheners , disinfectant sprayers , or portable nebulisers  require high-voltage drive for piezo ceramics. The MAS2808’s integrated boost stage and programmable output voltage mean it can replace discrete multi-IC circuits with a single IC solution , reducing system complexity. Key Specifications and Configuration Options The MAS2808 combines flexible configuration with robust analogue performance. Below is a summary of its core specifications and programmable features: Electrical & Mechanical Characteristics Parameter Value / Range Supply Voltage (V DD ) 2.6V to 5.5V Max Output Voltage (Differential) 60V pp (30V × 2) Output Voltage Settings 4 steps (pin control), 8 steps (I²C) Quiescent Current (EN = LOW) < 0.5 µA Operating Temperature Range -40°C to +85°C Package TSSOP-16 (samples), DFN-12 (planned) Piezo Driver Configuration Single-ended or BTL (V OP /V ON ) Output Voltage Control Pin Control Mode (Default): SCL SDA V OUT Low Low 30V High Low 24V High High 17.5V Low High 15V I²C Control Mode: OVC[2:0] V OUT 0x00 9V 0x01 12V 0x02 15V 0x03 17.5V 0x04 20V 0x05 24V 0x06 27V 0x07 30V I²C address: 0x76 I²C mode enabled when both EN and DIN are HIGH Compatible with 1.2V–5V logic levels This versatile setup allows designers to finely tune the piezo output level based on sounder size, desired SPL, or user feedback intensity. Design Considerations – Efficiency, Protection and Integration Tips While the MAS2808 simplifies high-voltage actuation, optimal implementation requires thoughtful hardware integration. Below are key design guidelines to maximise performance and reliability in safety-critical or battery-powered systems. Inductor and Capacitor Selection The boost converter requires only a 2.2–2.7 µH inductor , depending on the power source. A shielded inductor  is recommended to minimise EMI in compact designs, particularly in wearables or medical devices. C IN recommendations: USB-powered systems: ≥47 µF Coin cell: 100 µF recommended to mitigate ripple C OUT recommendations: Rated ≥35V ceramic capacitors ≥1 µF for general loads; increase for high SPL transducers Protection Against Transducer-Induced Surges In harsh environments, piezo elements can generate reverse voltage spikes  under mechanical or thermal shock. For such applications (e.g. smoke alarms mounted near industrial fans or compressors), external Zener protection  is advised across the V OP and V ON outputs. Use Zener diodes rated just above nominal output (e.g. 32V for 30V V OUT ) to suppress surges without reducing drive voltage. Ultra-Low Power Applications The MAS2808 draws just 40nA in shutdown mode , making it ideal for long-term standby applications. Designers can disable the device via the EN pin , and re-enable it dynamically only when audio or haptic feedback is needed. Supports logic high threshold as low as 0.9V , making it compatible with 1.2V or 1.8V microcontroller logic levels. Compact PCB Layout Keep C IN , C OUT , and the inductor physically close to the IC to reduce trace inductance. Minimise routing to the SW and V OUT  pins, which carry switching currents. V DD can use a narrower trace, as current draw is minimal. These practices ensure clean operation, minimal voltage ripple, and improved efficiency—particularly valuable in safety-critical environments where reliability and system uptime are paramount. Conclusion – Enabling Smart, Safe Feedback with the MAS2808 From high-decibel safety alarms to subtle haptic cues in medical and wearable devices, the MAS2808 enables reliable, high-voltage piezo actuation from compact, low-power systems . Its integrated boost converter, differential driver outputs, and flexible voltage control options make it a strong fit for engineers tackling alerting, atomisation or feedback design challenges—especially where board space, power budget and response time are limited. With support for both pin-mode and I²C programmable output , low leakage current in standby, and integrated protection features, it allows OEMs to embed consistent, responsive feedback mechanisms into their devices without resorting to complex discrete circuits. For developers seeking to integrate audible or tactile safety features  into their next design, the MAS2808 offers a streamlined, high-efficiency path to implementation. Contact Ineltek to request samples, a reference circuit, or a technical briefing on how the MAS2808 could enhance your next safety-critical embedded design.

Introduction – Why This Information Matters The EU Cyber Resilience Act (CRA)  is now law, imposing strict cybersecurity requirements on any product with digital elements sold in Europe. For embedded engineers, this is no longer a regulatory footnote — it directly affects how hardware, firmware, and support processes must be designed and documented. Failure to comply risks blocked market access, fines, and costly redesigns. For design teams already working to tight schedules, understanding and anticipating these requirements is essential. CyberWhiz's service-based approach transforms CRA compliance from a complex multi-vendor challenge into a single-partner solution. Their comprehensive model addresses the three critical compliance phases whilst providing the technical depth required for modern connected product development. CRA Requirements Affecting Embedded Engineers The CRA is broad, but several requirements hit embedded design especially hard: Secure by design:  Products must demonstrate minimisation of attack surfaces and protection against common vulnerabilities. Vulnerability handling:  Vendors must monitor, report, and act on vulnerabilities for the product’s supported lifetime. Patching obligations:  Security updates must be delivered in a timely and secure manner, including mechanisms for verification of authenticity. Transparency of lifecycle:  Customers must be informed of support periods, update mechanisms, and known limitations. Documentation:  Technical documentation demonstrating compliance must be maintained and available for market surveillance authorities. Requirement Impact on Engineering Design Considerations Secure design Early-stage threat modelling Hardened bootloader, code signing Vulnerability handling Ongoing monitoring CVE tracking, incident response plan Patch delivery OTA or secure wired update Firmware signing, rollback prevention Lifecycle support Declared end-of-support Documentation of update policies Compliance evidence Market authority requests Secure records of builds, SBOMs How CyberWhiz delivers complete CRA compliance solutions Holistic CRA Compliance Management CyberWhiz delivers continuous compliance oversight across all three CRA phases: design validation, field deployment management, and incident response coordination. Comprehensive Service Portfolio SBOM (Software Bill of Materials) management and maintenance End-to-end IoT penetration testing covering device, mobile app, and cloud infrastructure Risk assessment and technical documentation services SecOps support and monitoring capabilities RED compliance consultancy integration Notified body partnerships offering 30% discounts for critical products Security Libraries and Edge Protection Specialised security libraries designed for edge devices and mobile applications provide embedded protection without requiring extensive internal security expertise. CyberWhiz Defence Centre 24/7 monitoring and incident response capabilities ensure continuous compliance with CRA's ongoing security requirements. Technical FAQ Q: How does CyberWhiz handle SBOM management for complex connected products? A: CyberWhiz provides automated SBOM generation, maintenance, and vulnerability tracking throughout the product lifecycle. Their system integrates with existing development workflows to ensure compliance documentation remains current without disrupting engineering processes. Q: What makes their IoT penetration testing different from standard security assessments? A: Their testing covers the complete connected product ecosystem—device firmware, mobile applications, and cloud infrastructure—using the same methodology across all components. This unified approach identifies integration vulnerabilities that component-level testing often misses. Q: Can CyberWhiz support products already in development or deployment? A: Yes, their service model accommodates existing products through risk assessment, documentation catch-up, and retrofitting security measures. The September 2026 vulnerability management deadline allows time for systematic compliance implementation. Q: How do they handle the transition period leading to full CRA compliance? A: CyberWhiz provides phased implementation starting with vulnerability management by September 2026, followed by comprehensive compliance by December 2027. Their timeline aligns with the regulation's staged approach. Q: What level of technical integration is required with existing development teams? A: Minimal disruption to current workflows. CyberWhiz operates as an external service provider, integrating through APIs and standard documentation processes rather than requiring internal team restructuring. Q: How does their pricing model work for different production volumes? A: Tiered pricing accommodates production quantities from 10,000 units to over 1 million, with 1-3 year agreement options providing cost predictability for product planning. Call to Action CyberWhiz's service model eliminates the complexity of managing multiple compliance vendors whilst providing the technical depth required for CRA cybersecurity compliance. Their complete CRA compliance solutions address the engineering reality of connected product development where security must be embedded without compromising innovation speed. The December 2027 deadline approaches rapidly, but the September 2026 vulnerability management requirements create immediate action points for engineering teams. Interested in understanding how CyberWhiz's service model applies to your connected product portfolio? Contact Ineltek to arrange a consultation .

Introduction – What is Epson Motion Sensing Technology and Why Does it Matter? Modern industrial systems demand precise motion detection and analysis for everything from predictive maintenance to autonomous navigation. Traditional motion sensors often struggle with accuracy, power consumption, or environmental resilience, particularly in harsh industrial conditions. Epson's motion sensor family addresses these challenges through advanced QMEMS (Quartz Micro-electromechanical systems) technology, offering superior precision and low-power operation across accelerometers, inertial measurement units (IMUs), and vibration sensors. Engineers working on industrial automation, condition monitoring systems, or navigation applications require sensors that maintain accuracy whilst minimising power draw and maximising operational lifespan. The key engineering challenge lies in balancing measurement precision with power efficiency, especially for battery-powered IoT devices or remote monitoring systems. Epson's approach leverages decades of semiconductor fabrication expertise to deliver motion sensors with exceptional bias stability and noise performance. Features of Epson Motion Sensor Modules Addressing Industrial Requirements Epson's motion sensor portfolio delivers several critical advantages for demanding industrial applications: Advanced QMEMS Technology Great bias instability down to 0.5°/h and ultra-low noise for gyroscopes (ARW 0.03°/√h) Exceptional noise density performance at 0.02 µG/√Hz for accelerometers Superior long-term stability through precise microfabrication processes Low-Power Operation Accelerometer modules consume just 13.2mA typical current IMU power consumption as low as 53mW (16mA at 3.3V) Ideal for battery-powered industrial IoT applications Wide Operating Temperature Range Operation from -40°C to +85°C across the product range Suitable for industrial environments and outdoor installations Flexible Interface Options Multiple communication protocols including UART, SPI, CANopen, and RS422 Compact and Robust Design Compact 24×24×10mm form factor for space-constrained applications IP67-rated waterproof and dustproof options available How to Select The ideal Epson Motion Sensor Module IMU Family (M-G Series) - Performance Selection Guide IMU selection requires balancing gyroscope and accelerometer precision with power consumption and space constraints. Epson IMUs are calibrated to high precision over the whole temperature range. Critical Selection Parameters: Bias Instability (°/h)  - The most important IMU specification for Navigation . The higher the value the longer you can trust the gyro data. Values under 1°/h indicate high-precision navigation grade, whilst 3-5°/h suits basic orientation applications. Angular Random Walk (ARW) (°/√h)  - Determines short-term accuracy, especially important for Stabilisation  purposes. Values below 0.1°/√h enable precise attitude determination for robotics and ADAS applications. Power Consumption  - Essential for battery-powered devices. Modern IMUs achieve sub-100mW operation whilst maintaining good performance. Epson IMUs with 53mW typically, are exceptional at power saving. Output Range  - Please consider the detection range according to your specific application. The modules quoted with dual range settings are switchable in software. Feature M‑G330PDG M‑G355QDG0 M‑G366PDG0 M‑G370PDG0 M‑G570PR20 M‑G552XX Bias Instability (Gyro °/h) 3.0 1.2 1.2 0.8 0.5 0.8 Angular Random Walk (°/√h) 0.10 0.08 0.08 0.06 0.04 0.06 Gyro Range (°/s) ±400 ±450 ±450 ±450 ±475 ±450 Accelerometer Range (G) ±8 / ±16 ±8 / ±16 ±8 / ±16 ±8 / ±16 ±15 ±10 Data Output Rate (Hz) 1000 — 1000 2000 2000 2000 (Tilt/Euler 200) Operating Temp (°C) −40 to +85 −40 to +85 −40 to +85 −40 to +85 −30 to +70 −30 to +80 Power Consumption 53 mW (typ.) 53 mW (typ.) 53 mW (typ.) 53 mW (typ.) <1 W (typ.) 384 mW (32 mA @ 12 V) Size (mm) 24×24×10 24×24×10 24×24×10 24×24×10 65×60×30 65×60×30 Weight (g) 10 10 10 10 10150 115 Interface SPI / UART SPI / UART SPI / UART SPI / UART RS-422 RS422 / CANopen Special Features Basic model IEC 61508 SIL1 Standard model Ultra-Low Noise High Bias Stability IP67 unit Accelerometer Family (M-A Series) - Performance Selection Guide When selecting an accelerometer, engineers should evaluate these critical performance parameters: Key Performance Criteria: Noise Density (µG/√Hz)  - Lower values indicate better precision for detecting small movements. High-performance accelerometers achieve sub-0.1 µG/√Hz for seismic and structural monitoring applications, whilst industrial sensors typically range from 0.2-0.5 µG/√Hz. Bias Stability  - Critical for long-term measurements. Look for bias repeatability specifications under ±0.5 mG and temperature coefficients below ±0.1 mG/°C for precision applications. Frequency Range  - Determines measurement bandwidth. Seismic monitoring requires DC to 100Hz response, whilst vibration analysis can need up to 1kHz+ bandwidth. Feature M‑A352AD10 M‑A370AD10 M‑A552AC10 M‑A552AR10 Number of Sensing Axes 3 (XYZ) 3 (XYZ) 3 (XYZ) 3 (XYZ) Output Range (G) ±15 ±10 ±15 ±15 Bandwidth (Hz) DC–460 DC–210 DC–460 DC–460 Noise Density (µG/√Hz) 0.2 (typ.) 0.02 (typ.) 0.5 (typ.) 0.5 (typ.) Max Output Data Rate (Sps) 1000 1000 1000 1000 Interface UART / SPI UART / SPI CANopen RS422 Operating Temp (°C) −30 to +85 −30 to +85 −30 to +70 −30 to +70 Power Supply (V) 3.3 3.3 9–32 9–32 Current Consumption 13.2 mA (typ.) 36.3 mA 35 mA @ 12 V 40 mA @ 12 V Size (mm) 48×24×16 48×24×16 65×60×30 65×60×30 Weight (g) 25 — 128 128 Water & Dust Proof — — IP67 IP67   Vibration Sensor Family (M-A Series) - Performance Selection Guide Vibration sensors require careful frequency range selection based on monitored equipment: Selection Criteria: Frequency Range Capability  - Dual-range sensors offer 1-100Hz for low-speed rotating equipment and 10-1,000Hz for high-speed machinery. Programmable switching provides maximum flexibility. Velocity Range  - Higher velocity ranges (±200mm/s) suit heavy industrial equipment, whilst ±100mm/s covers most standard applications. ISO Compliance  - ISO10816/ISO20816 compliance ensures compatibility with international vibration monitoring standards. Environmental Protection  - IP67 rating essential for harsh industrial environments with dust, moisture, and temperature extremes. Output Format  - Displacement + velocity output provides comprehensive vibration analysis capabilities. Feature M‑A342VD10 M‑A542VR10 Number of Sensing Axes 3 (XYZ) 3 (XYZ) Frequency Ranges (Hz) 1–100 / 10–1,000 1–100 / 10–1,000 Output Modes Raw / RMS / p‑p Raw / RMS / p‑p Velocity Range (mm/s) ±100 ±100 Displacement Range (mm) ±200 ±200 Interface UART / SPI RS422 Operating Temp (°C) −30 to +85 −30 to +70 Power Supply (V) 3.15–3.45 9–32 Current Consumption 29 mA @ 3.3 V 51 mA @ 12 V Size (mm) 48×24×16 65×60×30 Weight (g) 25 128 Water & Dust Proof — IP67 ISO10816/20816 Compliant Compliant Industry Applications and Use Cases 1. Inertial Measurement Units (IMUs) High-Performance IMUs (e.g., M-G355) The M-G355 IMU stands out due to its extremely low bias instability (1.2°/h Gyro) and low angular random walk (0.08°/√h), indicating exceptional long-term stability and precision. Its ability to handle a wide temperature range and its compact size make it ideal for demanding applications. Suggested Applications: Inertial Navigation Systems:  High precision is critical for drones, unmanned ground vehicles (UGVs), and aerospace systems where GPS may be unavailable or unreliable. The M-G355's low-noise performance ensures accurate position and attitude estimation over time. Platform Stabilization:  Its high stability and wide output range make it perfect for stabilizing cameras, antennas, and robotics, even in environments with significant vibration. Industrial Automation & Robotics:  The M-G355's high-performance gyroscope and accelerometer provide the necessary data for precise control and movement, improving the efficiency and safety of robotic arms and other automated systems. Entry-Level and Mid-Range IMUs (e.g., M-G366 and M-G330) These IMUs are designed for applications where high-end inertial navigation is not the primary goal. They provide reliable and accurate motion sensing for shorter duration tasks or for systems that can be periodically re-calibrated. Suggested Applications: Autonomous Ground Vehicles (AGVs) and Drones:  In AGVs and drones that operate in controlled environments with periodic access to external positioning data (like GPS or a vision system), these IMUs can provide robust attitude and heading information between calibration points. General Purpose Robotics and Industrial Machinery:  For robotic manipulators, industrial carts, and other machinery that requires precise but not ultra-high-end motion control, the M-G366 and M-G330 provide a cost-effective solution. They can handle the dynamic movements of these systems, ensuring stability and control. Platform Control and Stabilization:  These IMUs can be used for stabilizing platforms in less demanding environments, such as consumer drones or camera gimbals, where a slight drift over time is acceptable and can be corrected by the user.   2. Accelerometers Ultra-Low Noise Accelerometers (e.g., M-A370) The M-A370 is specifically designed for high-sensitivity applications. With an ultra-low noise density of 0.02 µG/√Hz, it can detect minute accelerations that other sensors might miss. Suggested Applications: Seismic Monitoring:  The M-A370's high sensitivity allows it to detect subtle ground movements and tremors, making it suitable for earthquake early warning systems and structural health monitoring. Structural Health Monitoring:  It can be used to detect and analyse small vibrations or shifts in bridges, buildings, and other large structures, providing early warnings of potential damage. Precision Measurement:  Any application requiring the measurement of extremely small accelerations, such as in scientific instruments, would benefit from the M-A370's superior noise performance. General Purpose Accelerometers (e.g., M-A352) Other accelerometers, like the M-A352, offer a balanced set of features with a good trade-off between performance and cost, suitable for a wider range of industrial and consumer applications. Suggested Applications: Vibration Analysis:  Basic vibration monitoring in non-critical machinery   Structural Health Monitoring:  It can be used to detect and analyse small vibrations or shifts in bridges, buildings, and other large structures, providing early warnings of potential damage. Tilt and Orientation Sensing:  Measuring the angle or tilt of a device, such as in factory equipment 3. Vibration Sensors Industrial-Grade Vibration Sensors (e.g., M-A342) Epson's vibration sensors are designed to analyse vibrations in industrial environments. They typically offer a broad frequency range and a variety of output modes, making them flexible for different types of machinery. Suggested Applications: Condition Monitoring of Rotating Machinery:  These sensors are excellent for predictive maintenance on motors, pumps, fans, and other industrial equipment. By analysing the vibration data, engineers can detect signs of wear and prevent catastrophic failures. Machine Tool Monitoring:  Monitoring vibrations in CNC machines and other precision tools can help ensure the quality of the manufactured parts and detect issues with the cutting tools. Structural Health Monitoring:  Similar to the high-end accelerometers, these sensors can be used to monitor vibrations in large structures to assess their integrity, especially when focused on specific frequency bands of interest.   Conclusion Epson motion sensor modules deliver exceptional precision and reliability for demanding industrial applications. The advanced MEMS technology provides superior bias stability and noise performance, whilst the low-power design enables extended battery operation in IoT deployments. Whether you need basic motion detection with the M-A352AD10 accelerometer, high-precision navigation with the M-G370PDT IMU, or industrial vibration monitoring with the M-A542VR10, Epson's comprehensive portfolio addresses diverse engineering requirements with proven reliability. For technical support, samples, or application guidance on Epson motion sensors, contact the Ineltek team to discuss your specific requirements and receive expert recommendations for your next project. FAQs - Selecting the Right Epson Motion Sensor for Your Application Q. What advantages do Epson’s QMEMS sensors offer over conventional silicon MEMS? A. Epson’s QMEMS technology delivers significantly lower bias instability (down to 0.8 °/h for gyros and 0.02 µG/√Hz for accelerometers) compared to typical silicon MEMS. This translates into less drift over time, particularly valuable for navigation and structural monitoring where recalibration is costly or impractical. Q. When should I specify an Epson accelerometer versus an IMU or vibration sensor? A. Use an Epson accelerometer when ultra-low noise vibration detection is critical (e.g. seismic or structural health monitoring). Choose an IMU if you need full six-axis inertial data for navigation or stabilisation. Vibration sensors are optimised for condition monitoring with ISO-compliant velocity / displacement outputs, making them simpler to integrate for predictive maintenance systems. Q. How much in-system calibration do these modules require? A. Epson sensors are designed for industrial deployments with stable bias over years of use. Product longevity support is typically 10 years minimum, which aligns with long-life infrastructure projects and aerospace requirements. Q. What is the long-term stability and lifetime support of Epson MSMs? A. Epson MEMS sensors offer superior bias stability, lower noise density, and exceptional long-term accuracy. The advanced microfabrication technology provides better temperature stability and reduced drift compared to conventional sensor designs, making them ideal for precision applications. Q. Do Epson vibration sensors add value compared to using an accelerometer plus software? A. Yes - vibration sensors provide velocity and displacement directly, selectable via registers. This avoids the need for computational integration of acceleration signals, reducing processor load and eliminating cumulative integration errors in long-term monitoring. Q. In practice, what difference does the ultra-low noise performance of Epson accelerometers and IMUs make to my design? A. Ultra-low noise directly improves resolution and accuracy of small motion detection. For accelerometers like the M-A370, with noise density as low as 0.02 µG/√Hz, you can resolve subtle seismic or structural vibrations that would otherwise be lost in sensor noise. For IMUs, reduced angular random walk allows much longer dead-reckoning periods without drift, critical in navigation when GNSS is unavailable. The net benefit is higher confidence in data, less reliance on post-processing or frequent calibration, and system designs that can achieve precision levels normally reserved for laboratory-grade instruments but in compact, low-power modules.

Introduction – Why Powerline Communication Matters As energy systems converge around electric mobility and digital grids, reliable data exchange across power lines has become a cornerstone of modern infrastructure. Powerline Communication (PLC) allows equipment to transmit high-frequency data over existing electrical wiring, eliminating the need for separate data cabling in environments where space, cost, or retrofitting constraints exist. In the EV charging ecosystem , PLC isn’t optional. It underpins compliance with ISO 15118-3, the standard governing vehicle-to-charger communication for authentication, billing, and energy management. Smart grid operators, meanwhile, depend on PLC to extend visibility and control deep into the distribution network without deploying costly new communications infrastructure. The engineering challenge is clear: how do you move clean, reliable data across a medium designed for 50/60 Hz power delivery  - one that is inherently noisy, subject to transients, and carrying voltages that demand galvanic isolation? The answer lies in the transformer. More than a passive component, the PLC transformer defines insertion loss, bandwidth, isolation performance, and chipset compatibility. In practice, it is the single element that determines whether your design will pass compliance testing and survive long-term deployment in the field. Features of E&E Magnetics PLC Transformers Designing for PLC means balancing signal integrity, safety isolation, and long-term reliability. This is where E&E Magnetics distinguish themselves with a portfolio built specifically for EVSE, smart grid and industrial deployments. ISO 15118-3 compliance  – Ensures seamless communication between EVs and charging infrastructure using HomePlug Green PHY, eliminating interoperability risks. AEC-Q200 qualified  – Components are tested against automotive thermal, vibration and humidity cycles, providing confidence for both EV and harsh industrial environments. High isolation up to 4000 V  – Industrial-grade safety insulation protects sensitive chipsets and meets stringent utility and automotive safety standards. Broad frequency coverage (1.5–30 MHz)  – Fully supports HomePlug Green PHY and other PLC protocols used across smart grid and charging applications. Wide chipset compatibility  – Multiple turns ratios are available to match the input requirements of major PLC chipsets, including those from Qualcomm and Maxim. Extended operating temperature (−40 °C to +125 °C)  – Guarantees performance across the extremes of automotive under-hood and outdoor metering deployments. For engineers, these features translate into reduced design risk  and faster certification cycles. Rather than tuning a generic signal transformer, the E&E PLC family provides a proven, standards-aligned foundation for any PLC-based design. Industry Applications and Use Cases Image credit: Qualcomm Inc. EV Charging & Vehicle-to-Grid (V2G) Powerline Communication is central to ISO 15118-3, which governs how EVs and charging stations exchange data for authentication, load management, and bidirectional energy transfer. A PLC transformer that fails compliance can delay or derail certification. E&E’s AEC-Q200 qualified transformers are already tuned for HomePlug Green PHY, giving designers confidence that their EVSE hardware will interoperate across the ecosystem. Smart Grid Communication Utilities increasingly depend on PLC to control and monitor distribution assets and meters without deploying new communication cabling. Here, the transformer must guarantee both signal bandwidth  and robust isolation  against line surges and noise. E&E’s range offers 4000 V isolation and wide temperature capability, ensuring reliable operation in substations, outdoor cabinets, and embedded meter modules. Industrial Automation Factories use PLC to retrofit intelligent monitoring and control onto legacy wiring. The challenge is EMI from heavy machinery and long cable runs. Transformers with the wrong leakage inductance or insertion loss will result in marginal links. E&E’s portfolio, designed specifically for PLC, provides engineers with predictable EMC behaviour and long-term reliability in electrically noisy environments. Energy Storage and Microgrids Battery systems and distributed energy resources also rely on PLC for integration with grid operators. With multiple turns ratios and chipset compatibility, E&E transformers make it straightforward to scale designs across different platforms and suppliers, while maintaining compliance and safety. Part Number Turns Ratio OCL (µH min) Insertion Loss (dB max) Leakage Inductance (µH max) Isolation (Hi-pot) Operating Temp (°C) Notes / Typical Use Case A21V00042F 1:1 10 @ 1 MHz 1 (1.5–30 MHz) 0.25 @ 1 MHz 3000 V −40 to +125 Automotive-grade, EVSE/EV charger PLC (HPGP) A21V00043F 1:1:1 5 @ 100 kHz 1 (1.5–30 MHz) 0.5 @ 100 kHz 3000 V −40 to +125 Broad compatibility, Qualcomm/Maxim chipsets 821-02110F 1:2CT:1 12 @ 500 kHz 1.2 — 3000 V −40 to +125 Flexible chipset interface, smart grid meters 821-02111F 1:1:2 7 @ 500 kHz 2.25 — 3000 V −40 to +85 Compact industrial PLC node 821-02113F 1:1:1 20 @ 100 kHz 2.0 — 3000 V −40 to +125 Ruggedised outdoor PLC devices 821-02114F 1:1:1 8 @ 100 kHz 2.5 — 3000 V −40 to +125 Extended industrial range 821-02115F 1:1:1 2 @ 10 kHz 2.5 — 3000 V −40 to +125 Specialised low-frequency PLC Conclusion / Call to Action Powerline Communication is now central to EV charging, smart grid, and industrial control but only when the signal path is designed for reliability. The transformer is the decisive element: it must provide the right turns ratio for chipset compatibility, deliver low insertion loss for clean data transfer, and guarantee long-term isolation in demanding environments. E&E Magnetics’ PLC transformers are engineered precisely for this role. With ISO 15118-3 compliance, AEC-Q200 qualification, and isolation ratings up to 4000 V, they take the uncertainty out of design and certification. Whether you are building EVSE hardware, rolling out smart meters, or integrating PLC into industrial automation, the portfolio offers the breadth and robustness to match your requirements. To discuss your design needs or request samples from the PLC transformer range, contact the Ineltek team for technical support and tailored recommendations. Engineer FAQs – Powerline Communication Transformers Q. What frequency range do E&E PLC transformers support, and why does it matter? A. They are designed for the full 1.5–30 MHz band required by PLC standards, ensuring clean signal transfer across the spectrum without bandwidth limitations. This coverage is critical for both smart grid and EVSE applications. Q. How low is the insertion loss across that band? A. Depending on model, maximum insertion loss is 1–2.25 dB. Lower loss improves link margins, giving engineers confidence that their design will pass EMC and interoperability testing on the first attempt. Q. Are E&E's transformers optimised specifically for HomePlug Green PHY (HPGP) and ISO 15118-3? A. Yes. E&E’s portfolio includes parts tuned for HPGP, directly supporting EV-to-EVSE communication mandated by ISO 15118-3. This avoids the uncertainty of adapting generic magnetics. Q. What level of galvanic isolation do your PLC transformers provide? A. Isolation ratings are up to 4000 V, protecting chipsets from high-voltage transients and ensuring compliance with industrial and automotive safety standards. Q. Do you offer AEC-Q200 qualified parts for automotive and harsh environments? A. Absolutely. AEC-Q200 qualification guarantees reliability under automotive thermal cycling, vibration, and humidity — equally valuable in EVSE, grid, and outdoor industrial deployments. Q. How do you ensure compatibility with major PLC chipsets? A. E&E provides multiple turns ratios — 1:1, 1:2CT:1, 1:1:2, 1:1:1 — and even centre-tap options. This matches the input requirements of chipsets from Qualcomm, Maxim, and others, reducing integration risk. Q. What package formats are available to suit different board densities? A. The range spans compact SMT parts for space-constrained modules through to larger through-hole packages for higher isolation or thermal mass. This flexibility simplifies layout decisions in both EVSE controllers and smart meters.

Understanding Edge AI Computing in Embedded Systems For embedded system engineers, Edge AI computing  refers to processing artificial intelligence workloads—such as computer vision, speech recognition, anomaly detection, or predictive analytics—directly on the device, without relying on continuous cloud connectivity. This approach brings three major advantages compared with traditional embedded processing or cloud-based AI: Real-time decision-making  – Local inference eliminates round-trip latency to cloud servers, enabling instant reactions in safety-critical or high-speed environments (e.g., industrial robotics, autonomous vehicles). Reduced data transfer and cost  – Only processed results or event triggers need to be sent over networks, significantly lowering bandwidth requirements and operational costs. Improved privacy and resilience  – Sensitive data can be analysed locally and discarded after processing, reducing exposure to interception and allowing systems to operate even with intermittent connectivity. By integrating an AI-capable compute module into an embedded design, engineers can move from basic control and monitoring to autonomous, context-aware systems . Tasks that once required a dedicated server—like recognising product defects on a conveyor, optimising motor performance based on sensor fusion, or authenticating users via facial recognition—can now run entirely at the edge. The transition to edge AI does require careful hardware selection. Processing requirements, thermal constraints, industrial interfaces, and security features all differ widely between module classes, making an informed choice essential. Why Selecting the Right Edge AI Solution Matters Not all edge AI compute solutions are created equal. Some excel at high-throughput multimedia processing, others are designed for secure, low-power control with AI acceleration, while others prioritise industrial connectivity or compatibility with existing OS and development stacks. Selecting the right solution early in the design phase ensures: Adequate AI performance  for the intended inference tasks Compatibility  with required displays, sensors, and connectivity Thermal and power budget alignment  for the deployment environment Long-term availability  for production stability This article compares four flagship options from different segments: SIMCom SIM9650  – Multimedia and IoT-focused AI smart module Espressif ESP32-P4  – Secure, low-power MCU-class AI controller Advantech AOM-2721  – High-performance embedded vision platform Nuvoton MA35D1  – Industrial Linux-capable HMI and AI SoC What is TOPS - and what does it mean in practice? TOPS  stands for Tera Operations Per Second  — essentially a measure of how many trillion operations a processor (often an AI accelerator or neural processing unit) can perform in one second. In the context of edge AI compute solutions , TOPS is used to express AI inference performance , typically for operations like multiply–accumulate (MAC) used in neural networks. A few points engineers should know about TOPS: It’s architecture-dependent  – Different chips count “operations” differently, so a 14 TOPS rating on one module may not be directly comparable to another unless the test methodology is identical. It’s usually measured at INT8 precision  – Many edge AI workloads are quantised to 8-bit integers for efficiency. Higher-precision (FP16, FP32) processing usually yields lower TOPS numbers. It’s a peak figure  – Real-world performance can be lower due to memory bandwidth limits, model structure, or other system bottlenecks. In short: TOPS is a useful headline metric for AI acceleration capability , but engineers should also look at actual benchmark results for their specific models before finalising a module choice. Breaking down 14 TOPS 14 TOPS  = 14 trillion operations per second For AI accelerators, an “operation” usually refers to a basic multiply–accumulate (MAC) used in neural network layers. Example – Running an object detection model Say you have a MobileNet-SSD  type model for real-time object detection: It requires about 3 GMACs  (3 billion MAC operations) per inference on a 300×300 image. If the module’s AI engine sustains 14 TOPS  (14,000 GMACs per second at INT8): 14,000 GMAC/s ÷ 3 GMAC/inference = ~4,666 inferences/sec  (theoretical max). Reality check Real-world inferences per second will be lower — perhaps 25–50% of peak  — due to memory bandwidth, pipeline stalls, and software overhead. Even at 25% efficiency, 14 TOPS  could still deliver over 1,100 real-time inferences/sec  for that model, far exceeding most edge needs. Engineers would typically use that surplus to: Run multiple models in parallel Process higher-resolution images Increase model complexity for better accuracy Why this matters When comparing modules, TOPS tells you how much “AI headroom” you have . Low-TOPS modules  (e.g., MCU-based with 0.1–0.5 TOPS) are great for keyword spotting or sensor anomaly detection. High-TOPS modules  (10+ TOPS) open the door to multi-camera vision, real-time video analytics, or simultaneous AI workloads. How to Select the Right Edge AI Compute Solution – An Engineer’s Step-by-Step Process Selecting an edge AI compute solution is not about picking the most powerful device on paper — it’s about aligning the module’s capabilities with the specific functional, environmental, and lifecycle needs  of your project. Below is a logical framework engineers can follow: Step 1 – Define the AI Workload Model size and complexity:  Will you run lightweight models (keyword spotting, anomaly detection) or heavy CNNs for vision? Performance requirement:  Determine whether you need TOPS-heavy accelerators (e.g., SIMCom SIM9650) or a lower-power MCU-based approach (e.g., ESP32-P4). Inference vs. training:  Edge modules typically handle inference only — but some SoCs can run lightweight on-device training if required. Step 2 – Identify Sensor and Interface Needs Camera count and resolution:  Multi-camera AI vision needs dedicated MIPI-CSI lanes and powerful ISP pipelines. Industrial I/O:  For automation, ensure support for CAN-FD, UARTs, isolated GPIO, or fieldbus standards. Other sensors:  Lidar, radar, microphones — check I²S, SPI, and high-speed interfaces are available. Step 3 – Match the Connectivity Profile Local comms:  Wi-Fi 6E, Bluetooth, or wired Ethernet for LAN-based processing. Wide-area comms:  LTE/5G for remote AI nodes (SIMCom modules excel here). No connectivity?  Prioritise modules optimised for full offline operation. Step 4 – Evaluate OS and Software Ecosystem Application stack:  Need Android for app development? Ubuntu for AI frameworks? Bare metal for deterministic control? Ecosystem maturity:  Established SDKs, community support, and driver availability can reduce integration time. Step 5 – Consider Power, Thermal, and Form Factor Constraints Power budget:  Is it mains, PoE, battery, or energy harvesting? Thermal profile:  Higher TOPS usually means higher thermal output — assess heatsinking and airflow requirements. Size:  From tiny LGA MCUs to full OSM modules, ensure fit within enclosure and PCB footprint. Step 6 – Validate Long-Term Availability and Reliability Production lifecycle:  For industrial deployments, look for 7–10 years availability (Nuvoton, Advantech). Temperature rating:  Ensure the module is qualified for the environment (-20°C to +70°C or more). Regulatory certifications:  CE/FCC, RoHS, and application-specific standards (e.g., EN50155 for rail). Step 7 – Prototype and Benchmark Early Before committing to volume, test representative workloads on candidate modules to verify real-world inference speed, latency, and thermal stability. SIMCom SIM9650 – AI Smart Module for Multimedia & IoT Processor:  Octa-core ARM v8 up to 2.7GHz, Adreno 643 GPU AI performance:  >14 TOPS via Hexagon Tensor Accelerator OS:  Android 14 Connectivity:  LTE Cat 4, Wi-Fi 6E (2x2 MU-MIMO), Bluetooth 5.2, GNSS Memory:  4GB/8GB LPDDR4X + 64GB/128GB UFS Displays:  Dual independent display (4K60 DP + FHD MIPI-DSI) Cameras:  Up to 36MP multi-camera input with triple ISP I/O:  PCIe Gen3, USB 3.1 Type-C, multiple UART/I2C/SPI/GPIO Applications:  Smart POS, industrial handhelds, AI-enabled cameras, VR/AR, intelligent cockpits Espressif ESP32-P4 – Secure MCU with AI Acceleration Processor:  Dual-core RISC-V at up to 400MHz with AI instructions Security:  Hardware cryptography, secure boot, trusted execution Memory:  Integrated SRAM + external flash interface Connectivity:  USB OTG, SDIO, Ethernet MAC, multiple SPI/I2C/UART AI role:  Suitable for lightweight inference models, control logic with sensor fusion Low-power focus:  Optimised for battery or energy-harvesting systems Applications:  Secure IoT nodes, low-power AI gateways, portable AI devices Advantech AOM-2721 – High-Performance Qualcomm QCS6490 Platform Processor:  Cortex® Gold+ @ 2.7GHz + 3x Cortex® cores @ 2.4GHz Memory:  Onboard 8GB LPDDR5 @ 8533MT/s GPU/VPU:  Adreno GPU 643, VPU 633 (4K30 encode/decode) Displays:  MIPI-DSI, eDP, DP outputs OS Support:  Windows 11 IoT, Ubuntu, Yocto I/O:  PCIe Gen3, USB 3.2, Ethernet, MIPI-CSI for cameras Form factor:  OSM 1.1, 45 x 45mm Applications:  Embedded vision, industrial AI gateways, high-resolution HMI systems Nuvoton MA35D1 – Linux-Ready Industrial HMI and AI Control SoC Processor:  Dual-core Cortex-A35 (Armv8-A) Memory:  DDR interface for external RAM Security:  Secure boot, TrustZone, hardware crypto I/O:  CAN-FD, multiple UART, SPI, I2C, Ethernet AI role:  Runs AI inference via external accelerators or optimised CPU instructions OS Support:  Linux-based industrial applications Applications:  Factory automation, transportation control, secure industrial gateways Specification Comparison Table Feature SIMCom SIM9650 Espressif ESP32-P4 Advantech AOM-2721 Nuvoton MA35D1 CPU Octa-core ARM v8, 2.7GHz Dual-core RISC-V, 400MHz Cortex Gold+ 2.7GHz + 3x2.4GHz Dual-core Cortex-A35 AI Performance >14 TOPS Lightweight inference GPU/VPU acceleration CPU / external accelerator Memory 4–8GB LPDDR4X + UFS Integrated SRAM 8GB LPDDR5 External DDR OS Android 14 Bare metal/RTOS Win 11 IoT, Ubuntu Linux Connectivity LTE, Wi-Fi 6E, BT, GNSS USB, Ethernet MAC Ethernet, PCIe Ethernet, CAN-FD Displays 4K60 DP + FHD MIPI Basic LCD via SPI/parallel MIPI-DSI, eDP, DP External controller Camera Up to 36MP multi-cam External modules via SPI/I2C Dual MIPI-CSI External Target Use Multimedia IoT Low-power control High-end vision Industrial HMI Data Sheet Application Guidance – Which Module Fits Which Project? High-end multimedia AI & connectivity:  SIMCom SIM9650 Low-power secure AI controllers:  Espressif ESP32-P4 Embedded vision & compute-intensive AI:  Advantech AOM-2721 Industrial HMI & control with AI hooks:  Nuvoton MA35D1 Conclusion / Call to Action Choosing the right edge AI compute solution starts with understanding the processing, connectivity, and application priorities of your project. Whether your priority is high-resolution multimedia AI, secure low-power control, or industrial Linux integration, one of these flagship options will align with your needs. For full specifications, evaluation kits, and engineering samples, contact Ineltek’s technical team to discuss your edge AI requirements. FAQs - Choosing the right EDGE AI Compute module for your embedded system Q. What is the main benefit of edge AI in embedded systems? A. It enables real-time AI processing locally, reducing latency, bandwidth costs, and privacy risks. Q. Which module is best for multi-camera AI vision? A. The SIMCom SIM9650 and Advantech AOM-2721 are strongest for high-resolution camera input and processing. Q. Which option suits harsh industrial environments? A. Nuvoton MA35D1 offers industrial interfaces and long-term Linux support. Q. Can low-power devices still run AI models effectively? A. Yes, with optimised lightweight models, the ESP32-P4 can run local inference on constrained power budgets. Q. How do I estimate the AI performance I actually need? A. Profile your model on a desktop environment, then scale to the module’s architecture. If the module runs your model with at least 30% performance headroom, it’s a safe choice. Q. Can I run multiple AI workloads in parallel? A. Yes, if the module supports multi-core processing and adequate memory bandwidth — the Advantech AOM-2721 and SIMCom SIM9650 are well-suited for concurrent inference and application logic. Q. What if my AI model changes during product life? A. Choose a module with firmware/OTA upgrade support and enough processing headroom to handle heavier models without redesign. Q. Is there a trade-off between AI power and battery life? A. Yes — higher TOPS modules consume more power. For portable devices, balance model complexity with available energy budget. Q. How important is hardware security in edge AI? A. For systems handling sensitive data, features like secure boot, encryption engines, and trusted execution environments (as in Nuvoton MA35D1 and ESP32-P4) are critical to prevent tampering and protect inference data.

RTK and UWB - Complementary Technologies for Precision Tracking In the pursuit of safer, more efficient, and more automated operations, accurate location tracking has become a critical enabler across sectors from construction and logistics to healthcare and public infrastructure. Two technologies now dominate the high-precision positioning landscape: Real-Time Kinematic (RTK) GNSS  and Ultra-Wideband (UWB) . Though both deliver centimetre-level accuracy , they excel in different environments and offer distinct advantages: RTK GNSS  enhances satellite navigation (GPS, Galileo, BeiDou, etc.) using real-time correction data from a base station or network. This enables absolute positioning accuracy of 1-2 cm in open-sky outdoor conditions - ideal for wide-area applications such as surveying, fleet tracking, or railway infrastructure. UWB , in contrast, operates over short ranges (typically up to 20 m line-of-sight) using rapid, low-power radio pulses to calculate the distance between tags and fixed anchors. It functions reliably indoors or in GNSS-denied environments , offering precise relative location in complex, cluttered, or metal-dense settings such as warehouses, factories, or hospitals. Where RTK GNSS provides absolute global coordinates , UWB shines in local, real-time movement tracking  with ultra-low latency. This makes them highly complementary: A rail worker may be geo-fenced using RTK while on open track, then seamlessly transition to UWB coverage while entering a tunnel. A delivery robot might use GNSS for navigation across a depot yard, then switch to UWB for exact positioning inside a fulfilment centre. A wearable tag could use RTK for construction site safety outdoors and UWB for indoor floor-level worker tracking during building fit-out. As the following sections show, industries across Europe are adopting hybrid RTK and UWB systems  to extend safety coverage, boost asset visibility, and improve operational control. With falling costs and maturing standards, these technologies are no longer experimental - they are fast becoming foundational. Workplace Safety: Preventing Accidents with Real-Time Tracking High-precision location technologies are being rapidly deployed to reduce risks and protect workers in industrial environments. From factories and logistics centres to construction sites and energy infrastructure, UWB and RTK positioning systems  enable new forms of real-time awareness, proximity warnings, and geo-fencing that were previously impossible with traditional methods. UWB for Forklift and Vehicle Collision Avoidance One of the most prominent applications of UWB in workplace safety is forklift-pedestrian interaction prevention . In warehouses and production plants across Europe, UWB tags worn by staff and anchors installed on vehicles create dynamic "safety bubbles". If a worker comes too close to a moving forklift, the system can trigger instant alerts  for both parties - visual, haptic, or audible, or even slow down the vehicle automatically. SPARK Microsystems ’ SR1020 transceivers, operating in beaconing mode , are ideally suited to such proximity warning systems. They offer adjustable detection zones from 25 cm to 20 m , operate at less than 5 µW active power , and enable battery-powered tags with multi-year lifespans  on standard coin cells, crucial for deployment at scale. These UWB RTLS deployments are proving effective. At Velux’s Danish factory, for example, a UWB system with 12 anchors and 59 tags not only improved worker safety but also boosted productivity by 10% through better logistics coordination. RTK GNSS for Geo-Fenced Safety Zones For large outdoor worksites such as construction yards, mines, and heavy industry, RTK-enabled GNSS wearables  are used to establish precise virtual safety zones. Workers crossing into restricted or hazardous areas receive alerts in real time. The centimetre-level accuracy of RTK  ensures minimal false alarms, enabling confident separation of personnel from operating machinery. Neoway ’s GN07-A1  and GN10-A1  GNSS modules support integrated RTK , with sub-10 second convergence and low current consumption. The GN07-A1 offers dual-band L1/L5 tracking and supports GPS, BeiDou, Galileo and GLONASS, while consuming as little as 22 mA  during L1-only tracking. These modules are suitable for integration into ruggedised wearable devices, helmets, or asset trackers. RTK-based geo-fencing is already in use across UK and EU industries. In rail maintenance, Network Rail’s geofencing solution has been successfully trialled to protect workers on live tracks. Wearables incorporating RTK GNSS warn the wearer and notify supervisors if boundaries are breached - with 1 cm accuracy  enabling precise zone definition. Emergency Mustering and Safety Analytics Both UWB and GNSS systems also support emergency response . In the event of an evacuation or incident, UWB RTLS can pinpoint each worker’s last known position and track real-time movement to ensure no one is left behind. Outdoors, RTK GNSS modules provide the same visibility at scale. European chemical plants and energy facilities are using these capabilities to streamline drills and enhance rescue operations. Furthermore, tracking near-misses and safety incidents , such as how often a worker came within a metre of a vehicle, enables organisations to analyse patterns and adapt training or layout  accordingly. SPARK’s SR1020 tags, with low-latency detection and location logging, support this kind of data-driven improvement. Healthcare: UWB RTLS for Patient, Staff and Asset Monitoring In hospital environments, precise location data can be the difference between timely care and costly delays. Real-Time Location Systems (RTLS) built on UWB are being increasingly adopted across healthcare facilities in the UK and Europe to track patients, staff and critical assets with sub-metre accuracy. Asset Tracking for Operational Efficiency Medical equipment is often highly mobile; infusion pumps, ultrasound machines, defibrillators and wheelchairs are constantly on the move between wards, often disappearing just when they’re most needed. UWB-enabled RTLS tags allow these assets to be located instantly via a digital map, reducing the time staff spend searching and improving equipment utilisation. Hospitals in Germany have already seen success implementing such systems, with measurable reductions in search times and improved care delivery. By tagging equipment with compact, low-power UWB modules such as SPARK’s SR1020, which consumes just microamps in beaconing mode, hospitals can track thousands of assets in real time with minimal impact on battery life or infrastructure cost. Enhancing Patient and Staff Safety UWB also plays a growing role in protecting vulnerable patients and frontline staff. In dementia care units, wearable tags establish digital safe zones. If a patient attempts to leave or enters a restricted area, alerts are triggered automatically. Similarly, infant security systems can detect unauthorised movement of new-borns from maternity wards. For staff, wearable UWB duress buttons allow nurses and clinicians to request emergency assistance. Unlike Wi-Fi or BLE, UWB provides room-level or even bed-level location accuracy with far fewer false alarms - even in complex hospital environments with heavy radio interference. SPARK’s presence detection platform enables these use cases with tags that can be integrated into ID badges or uniforms. With support for motion sensors and bidirectional communication, these tags can relay both location and status, enabling faster response and improving overall site safety. Future Integration with Smartphones and Hybrid Systems UWB radios are now appearing in smartphones - and the NHS has acknowledged their potential for patient-facing applications. In the near future, a patient’s own device could help guide them to the right clinic or diagnostic room via UWB-powered indoor wayfinding. Similarly, RTLS software platforms may begin to integrate hybrid architectures combining UWB, BLE and GNSS for seamless tracking indoors and out. With trusts under pressure to digitise operations and reduce inefficiencies, UWB is emerging as the next logical step beyond traditional Wi-Fi or BLE-based systems. It offers the precision, reliability and low-latency tracking that modern healthcare workflows increasingly demand. Rail: Trackside Worker Safety and Next-Gen Train Positioning The rail industry faces a dual challenge: ensuring the safety of maintenance crews operating on or near live tracks, while also modernising train control systems for greater efficiency. Across Europe, both RTK GNSS and UWB are being deployed to meet these needs, often working in tandem. Geo-Fencing for Trackside Worker Protection Working near live rail lines poses obvious risks, particularly when visibility is low or train schedules are tight. To address this, rail operators are implementing RTK-based geofencing systems  that create precise virtual boundaries around danger zones. Workers equipped with RTK-enabled wearables receive real-time alerts if they approach a hazardous area, such as an active track or restricted section. In the UK, Network Rail has trialled such systems with promising results, using 1 cm-accurate GNSS receivers  embedded in smart vests or helmet-mounted units. These systems are now being expanded across regions as part of broader safety initiatives. The ability to designate and update safe zones dynamically through software is a major leap forward compared to traditional signage or physical barriers. UWB Coverage for Yards, Tunnels and Underground Sections In areas where satellite reception is poor or unavailable such as tunnels, depots or underpasses, UWB-based local positioning systems  provide the missing link. (For higher value assets - Epson 's IMUs provide a phenomenal solution for high accuracy non-GPS conditions) Belgium’s Infrabel, for instance, has deployed a hybrid system combining RTK GNSS and UWB. Workers wear UWB tags that are tracked by anchors placed throughout the infrastructure. The system continues to function with 10 cm accuracy even when out of GNSS range. This indoor-outdoor handoff ensures continuous tracking of all personnel, whether they are working along open tracks or inside enclosed maintenance facilities. UWB’s resistance to multipath and metal interference is particularly valuable in the steel-heavy environments typical of railway infrastructure. High-Precision Train Localisation for Signalling and Traffic Control Beyond worker safety, rail operators are also exploring how RTK GNSS can enhance train tracking for European Rail Traffic Management System (ERTMS)  upgrades. Traditional signalling systems rely on trackside infrastructure like balises and circuits. With RTK GNSS and sensor fusion, trains can calculate their exact position relative to digital maps, potentially replacing much of the physical signalling network. Trials across Italy, Spain, France and Germany have demonstrated that dual-band GNSS receivers with RTK or PPP corrections can meet the stringent safety requirements for mainline rail operations. This could enable moving block signalling , reduce infrastructure costs, and improve timetable accuracy across entire networks. Modules such as Neoway’s GN10-A1 or SIMCom ’s SIM66MD - both offering compact dual-band RTK capabilities, are ideal candidates for on-board train positioning units, remote sensing equipment, or railway maintenance drones. Logistics: Seamless Visibility from Yard to Warehouse In modern logistics, knowing exactly where goods, vehicles and equipment are, at all times, is essential for operational efficiency and traceability. Both RTK GNSS and UWB are now central to building fully visible supply chains, with each technology filling a unique role in tracking across indoor and outdoor environments. UWB for Indoor Asset Tracking and Warehouse Automation Inside warehouses and distribution centres, UWB is becoming the standard for real-time indoor location tracking . With dense racking, steel structures and constant movement of goods and personnel, traditional positioning systems struggle to offer the required precision. UWB systems, by contrast, enable centimetre-level accuracy and high refresh rates, making them ideal for: Locating inventory on pallets, shelves or trolleys Monitoring forklift usage and traffic flow Navigating autonomous guided vehicles (AGVs) Preventing collisions between workers and machinery SPARK Microsystems’ SR1020-based tags are widely suited to these environments. Their ultra-low power profile allows them to run for years on a coin cell , making them viable for tagging thousands of items or vehicles. Once deployed, a grid of UWB anchors can support hundreds of tags simultaneously, providing not just live location data but also movement analytics  to optimise warehouse layout and routing. Some European automotive plants using UWB in intralogistics have reported up to 10% productivity gains , with better coordination of stock flow and fewer delays in locating key components. RTK GNSS for Fleet, Yard and Container Tracking When goods leave the warehouse, RTK GNSS takes over. Delivery vehicles, trailers, yard tractors and shipping containers are increasingly equipped with RTK-enabled GNSS receivers  to ensure precise routing, reduce theft, and enforce security perimeters. This is particularly important in large facilities such as logistics parks or ports, where incorrect parking or unauthorised movement can have knock-on effects. Modules such as SIMCom’s SIM66MD provide dual-band L1 + L5 RTK  in a compact LCC form factor with 2 cm positioning accuracy. Supporting GPS, Galileo, BeiDou, and QZSS, it offers reliable coverage across Europe, and at just 40 mA typical consumption , it is suitable for battery-powered trackers, mobile gateways or vehicle-mounted navigation units. Geofences can be drawn with high confidence using RTK triggering alerts as trucks arrive at or depart from precise loading bays, gates or staging areas. This is increasingly being integrated into yard automation software to schedule arrivals, assign docks dynamically, and prevent bottlenecks. Hybrid Handover for Multimodal Tracking As logistics networks become more integrated, tags and tracking systems must operate across indoor and outdoor transitions . This has led to hybrid systems where the same asset, such as a pallet or cart, is tracked via GNSS while in transit, then handed over to UWB for precise location inside the warehouse. This continuity supports accurate in-transit visibility , allowing logistics operators to follow high-value or time-sensitive goods from supplier to shelf without blind spots. It also improves traceability for audits, recalls or customer service. Across Europe, this dual-technology approach is being piloted in sectors including automotive, retail, e-commerce fulfilment, and cold-chain logistics driven by increasing pressure to optimise lead times and maintain full chain-of-custody transparency. Construction: Machine Control and Geo-Fenced Worker Protection Construction sites present some of the most challenging environments for location technologies; dynamic, irregular, and often partially covered or obstructed. Yet across the UK and Europe, both RTK GNSS and UWB are proving indispensable for improving safety, increasing productivity, and supporting automation in the sector. RTK GNSS for Machine Automation and Surveying RTK GNSS is now widely adopted for machine control . Earthmoving equipment such as dozers, excavators, and graders can operate with centimetre precision using RTK correction data, reducing reliance on ground personnel and minimising rework. The machines align to digital terrain models in real time, executing cuts and fills to exact design levels. RTK is also a staple of site surveying . Whether mounted on rovers, drones, or handheld devices, GNSS modules with RTK accuracy allow survey teams to mark out coordinates, verify build tolerances, and monitor progress live. Contractors using this tech report significant gains in both speed and quality control. Neoway’s dual-band GN07-A1 module, with support for L1 and L5 signals across GPS, Galileo, BeiDou and GLONASS, delivers horizontal accuracy of 1 cm + 1 ppm with under 10 second convergence. Its compact size and power efficiency make it well suited to portable instruments and field devices - even in harsh outdoor conditions. Geo-Fenced Safety Zones for Workers Construction zones are full of moving machinery, elevated platforms, and restricted areas. To protect workers, many firms are introducing geo-fenced wearables  that trigger alarms when personnel enter danger zones. For open-air sections of a site, RTK GNSS modules are used to maintain accurate position data, ensuring that alerts only occur when genuinely warranted. These systems are particularly valuable in infrastructure and highway projects, where workers may operate near live lanes or railway lines. Alerts can be configured for both the wearer and the machine operator, enabling quick action before an incident occurs. UWB for Enclosed Spaces and Tool Tracking In interior fit-outs or dense scaffolding areas where GNSS signals degrade, UWB tracking systems  provide reliable coverage. UWB tags can be used to locate workers within partially completed buildings or to monitor the usage and movement of tools and equipment. SPARK’s SR1120 in Time-of-Flight mode enables fine-grained ranging between tags and anchors, with power consumption low enough to support all-day operation on compact batteries. UWB is also being trialled in collision avoidance systems  for small plant and powered access equipment, alerting drivers or operators when a person is nearby in a blind zone. In long-duration or high-value construction projects, such as data centres or industrial plants, UWB infrastructure can be semi-permanently installed to provide reliable tracking throughout the build. Combined with RTK coverage outdoors, this hybrid model enables full-site visibility even as the physical layout evolves over time. While adoption is still emerging in small to mid-sized projects, regulators and industry bodies are increasingly recognising the value of digital safety monitoring and UWB and RTK are at the heart of this transformation. Choosing the Right Technology: RTK or UWB or Both? While both RTK GNSS and UWB deliver centimetre-level accuracy, they are designed for different environments and use cases. Selecting the right technology, or knowing when to combine them, is critical to balancing cost, coverage and complexity in real-world deployments. RTK GNSS: Best for Outdoor, Wide-Area Tracking RTK is the clear choice for applications that require: Absolute global positioning  (latitude/longitude) Wide-area coverage  across open sites or transport routes Minimal infrastructure (corrections delivered via network or base station) Compatibility with digital mapping, surveying, or machine control software Construction, agriculture, transport, rail and drone operations all rely on RTK for its accuracy and reliability in open environments. Once a device is equipped with a GNSS receiver and correction service, it can operate anywhere with sky visibility, making it highly scalable across fleets and outdoor sites. UWB: Best for Localised, High-Density, Indoor Scenarios UWB excels in scenarios that require: Indoor or GNSS-denied tracking  (factories, warehouses, tunnels, enclosed sites) High-resolution positioning in constrained environments Real-time interaction with moving objects or people Minimal latency for split-second decision-making (e.g. collision avoidance) Because UWB requires fixed infrastructure (anchors) and works over short distances (typically 10-30 m), it is ideal for permanent or semi-permanent facilities where precision and responsiveness are paramount. When to Combine: Hybrid Architectures for Seamless Coverage In many sectors, the optimal solution is a hybrid system : A logistics tag uses RTK GNSS for tracking across a yard or delivery route, then switches to UWB upon entering a warehouse A construction worker is geo-fenced outdoors using RTK and tracked indoors using UWB during fit-out stages A rail technician moves from an open track to a tunnel, seamlessly transitioning between GNSS and UWB coverage Hybrid systems are already being deployed in logistics, rail and industrial safety, supported by evolving software platforms that can merge indoor and outdoor data into a unified view. This convergence allows companies to track assets or personnel continuously , regardless of environment. Infrastructure and Cost Considerations RTK GNSS requires receivers, antennas, and access to correction data (via local base stations or subscription services), but no fixed infrastructure  on site. UWB requires installation of anchors and calibration , but offers higher relative accuracy  and responsiveness in localised environments. The decision often comes down to the nature of the site  (fixed or mobile), required accuracy , latency tolerance , and budget constraints . In general: Use RTK  for scalable, mobile outdoor deployments Use UWB  for high-precision local tracking indoors Use both  when seamless end-to-end visibility is critical Conclusion: From Pilots to Deployment, RTK and UWB Enter the Mainstream RTK GNSS and UWB are no longer niche innovations. They are now proven technologies driving tangible safety and efficiency gains across construction, healthcare, logistics, and transport. With compact, low-power modules from SPARK Microsystems, Neoway and SIMCom, it's never been easier to integrate high-precision positioning into your product or system. You can view the full specifications for each module mentioned in this article in our technical document library following the links below. Neoway GNSS Modules SIMCom SIM66MD module Spark Micro SR1020 Contact Ineltek UK  to discuss your tracking application, request samples, or book a technical consultation.

Introduction – Nuvoton's Expansive Audio Portfolio Engineers building next-generation embedded systems, consumer electronics, or automotive applications often face challenges around balancing performance, power efficiency, and integration complexity in audio designs. Nuvoton’s audio portfolio addresses these head-on, offering a deep and scalable range of solutions—from ADCs and CODECs to full-featured audio SoCs and smart amplifiers. With mature silicon and production-ready devices, Nuvoton provides robust building blocks for everything from compact smart speakers and surveillance systems to automotive safety features like eCall and AVAS. Their roadmap supports real-world deployments today, while premium features such as DSP integration and third-party audio enhancement technology set Nuvoton apart. Feature-Rich Audio Component Categories where Nuvoton excel Nuvoton’s offering spans the full signal chain of audio design: 1. Audio Converters (ADC/DAC) NAU7802 : Precision 24-bit single-channel ADC with ENOB 23-bit and on-chip calibration NAU8502 / NAU85L20B / NAU85L40B : Ultra-low power stereo and quad ADCs for 8–96kHz SR NAU8421 : High-end stereo DAC with 128dB SNR and low latency 2. Audio CODECs NAU88L21 / NAU88L25B : Class-G headphone drivers, up to 192kHz SR, with headset detection NAU88L20 : ULP stereo CODEC with Class-AB amp for power-sensitive designs NAU88S26 : Supports SoundWire for digital audio interfaces in high-performance platforms 3. Amplifiers & Smart Amps NAU83G10 / NAU83G20 / NAU83G60 : Class-D audio amps with Klippel Control Sound (KCS), PEQ, DRC, and ALC features NAU8315 / NAU83110 : Compact mono amps for mobile and wearable designs AEC-Q100 options  are available or in qualification for automotive use 4. Audio SoCs with Integrated DSPs ISD94124 Series : Cortex®-M4F at 200 MHz, USB 1.1 FS, integrated DMICs, AEC, NR, and beamforming for voice capture ISD933H3 : Hybrid architecture (Cortex®-M33 + Tensilica HiFi3), 1.5 MB SRAM, USB 2.0 HS and low-latency ADC/DAC interfaces 5. Voice ICs for Playback & OTA Updates NSP2340A / ISD2130 : Embedded flash voice ICs for announcements, intercoms, or appliances NSP Series : OTA-enabled updates, voice overlay/playback, tiny SOP packages, and ultra-low standby current for appliance and safety systems Features of Nuvoton Audio Solutions Addressing the Challenge Multi-Channel Codec Support Stereo to 8-channel I²S/TDM (NAU88C10, NAU88C20, NAU88C30) Sampling rates: 44.1–192 kHz, 16–32 bit resolution Integrated DSP Engine Acoustic echo cancellation (AEC) Noise suppression and automatic gain control (AGC) Parametric EQ and dynamic range compression Low-Power Operation <3 mA codec-active current; <1 µA shutdown mode Dynamic power gating on unused channels Flexible Interfaces I²S, PCM, TDM master/slave modes On-chip PLL for jitter-free clocking Voice-IA-Ready MCUs Arm Cortex-M55 with Ethos-U55 NPU accelerator On-device wake-word detection and keyword-spotting Up to 256 kB SRAM for audio buffers and ML models Seamless Integration Single-chip solutions reduce PCB area Full software support: HAL, middleware and reference apps Detailed Specifications Parameter NAU88C10 Stereo Codec NAU88C30 8-Channel Codec M55M1 Voice MCU Audio Channels 2 8 N/A Sampling Rate 44.1–192 kHz 44.1–192 kHz N/A Resolution 16–32 bit 16–32 bit 12–24 bit ADC/DAC DSP Functions AEC, AGC, EQ AEC, AGC, EQ Wake-word, noise suppression I²S/TDM Ports 1 × I²S / TDM 2 × I²S / TDM I²S master/slave Active Current 2.8 mA (typ.) 4.5 mA (typ.) 5 mA (core + DSP active) Shutdown Current <1 µA <1 µA <2 µA Package QFN32, LQFP48 QFN48, LQFP64 LQFP100, BGA100 Supply Voltage 2.7–3.6 V 2.7–3.6 V 1.7–3.6 V Highlighting Nuvoton’s Unique Audio Capabilities What sets Nuvoton apart is not just the breadth of its offering, but the integration of advanced audio enhancement and smart functionality. These USPs include: MaxxAudio® DSP Integration Select devices like NPCP215F  and NPCA110 Series  integrate Waves MaxxAudio®, delivering: Virtual bass enhancement Dialog clarity Intelligent volume control Ideal for conference speakerphones, monitors, and soundbars Bongiovi DPS Available on NPCA120D / NPCA121D , this technology enables premium sound tuning through real-time audio remastering Used in VR gaming headsets and portable speaker applications Automotive-Ready Designs Multiple parts (e.g. ISD941B24 , NAU83U15 , NAU88U10 ) are AEC-Q100 qualified or in progress, supporting eCall, AVAS, T-Box and infotainment systems eCall and AVAS reference designs provided to meet ITU and VDA standards for automotive audio Ultra-Low Power Consumption ULP ADCs and CODECs (like NAU85L20B , NAU88L20 ) target battery-powered or wearable devices Voice ICs with OTA updates and <1 µA standby support ultra-efficient consumer and medical devices Product Highlights with Real-World Applications Industry Applications and Use Cases Automotive Infotainment Nuvoton codecs feed DSP-tuned audio into head-units, supporting 5.1 and 7.1 surround, while M55 MCUs handle voice commands for navigation, calls and media control—all under AEC-certified conditions. Embedded Consumer Audio Devices Portable speakers, soundbars and smart home hubs leverage NAU88C low-power DSP to deliver clear voice pickup and rich playback without external DSP chips. Voice-Enabled Edge AI Edge-AI modules using M55M1 process wake-word detection and noise-robust ASR locally, reducing cloud dependency and ensuring privacy for smart assistants and medical monitors. Industrial and IoT Gateways Factory-floor gateways use multi-channel audio analytics (vibration, leak detection), all processed in-chip, minimising latency and simplifying system BOM. Summary table of Nuvoton Audio applications: Application Nuvoton Components Smart Intercom & Doorbell ISD94124S + NAU83G10 Smart Speaker & USB Microphones ISD933H3 + NAU88L21C VR Gaming Headset NPCA121D + 2x NAU8315 Conference Soundbar NPCP215F (DSP + AMP with MaxxAudio®) Automotive eCall ISD941B24 + NAU88U10 + NAU83U15 Embedded Appliance Voice Feedback NSP2340A Series Surveillance/Smart Camera ISD94124C + NAU88C10 + NAU8315 Conclusion – Why Choose Nuvoton Audio Solutions? Nuvoton offers one of the most comprehensive and capable audio portfolios on the market. With everything from discrete ADCs and CODECs to advanced DSP-enabled SoCs and automotive-qualified smart amplifiers, the platform scales across application types and complexity levels. Engineers can confidently design with production-ready parts and benefit from integrated features like MaxxAudio®, Bongiovi DSP, far-field voice pickup, and low-power operation. Looking for support selecting the right audio ICs for your design? Contact our team for recommendations, datasheets, or evaluation boards. FAQs for Nuvoton's Comprehensive Audio Solutions: Q: Which Nuvoton codec is best for stereo vs multi-channel audio? A: Use NAU88C10 for stereo applications; choose NAU88C30 (8-channel) or NAU88C20 (4-channel) when you need surround or multiple mic arrays. Q: How do I implement echo cancellation and noise suppression? A: All NAU88C codecs include built-in DSP blocks for AEC and noise suppression—simply enable them via I²C registers and load the reference coefficients. Q: Can the Cortex-M55 MCU run custom voice-AI models? A: Yes—its Ethos-U55 accelerator supports TensorFlow Lite Micro; you can deploy custom wake-words or command classifiers within the 256 kB SRAM. Q: What software support does Nuvoton provide? A: Nuvoton offers a full C-API HAL, board support packages, middleware (FreeRTOS, TFLM), as well as their own suite of evaluation kits with demo firmware, GUI tools and their NSP Playlist Editor.

SUMMARY " Secure flash memory is now a regulatory necessity under the EU Radio Equipment Directive Delegated Act (RED DA) and the forthcoming Cyber Resilience Act (CRA). Winbond’s W77Q and W77F devices deliver CC EAL5+ certified hardware root-of-trust, secure boot, encrypted SPI channels and rollback protection in drop-in SPI-NOR form factors." 5 min read, 944 words Introduction – What Is Secure Flash Memory and Why Does It Matter for RED DA and CRA Compliance? The new Radio Equipment Directive Delegated Act (RED DA)  and the upcoming Cyber Resilience Act (CRA)  are driving sweeping changes to how embedded systems handle data protection, firmware updates, and device authentication. Under these EU regulations, manufacturers of wireless-enabled and connected devices will soon be required to: Implement secure boot and firmware integrity verification Ensure only authorised updates are installed Prevent unauthorised access to communication services or sensitive data While much of the focus has been on cryptographic software or secure elements, the often-overlooked vulnerability is the external SPI flash memory  where code and sensitive assets are stored. Winbond’s W77Q  and W77F  series, part of the TrustME® secure memory platform, address this gap with drop-in compatible secure flash memory. These devices provide hardware-level protection  with CC EAL5+ certified security , root-of-trust integration , and robust mechanisms to detect, resist, and recover from attacks . Ineltek supports both the W77Q (substantial security) and W77F (high security) series enabling engineers to adopt RED DA and CRA-ready memory designs without re-architecting the main MCU or SoC. Features of Winbond W77Q and W77F Addressing the Challenge The W77Q  and W77F  secure flash families from Winbond are engineered to mitigate common attack surfaces in embedded systems, particularly those involving unauthorised access to external memory. They build on Winbond’s proven SPI NOR flash platform, adding security features without requiring host-side cryptographic redesign. W77Q – Substantial Security for Connected Devices Built on W77Q capabilities, plus: Advanced tamper resistance for physical attack mitigation Cryptographic isolation for high-value credential storage Stronger compliance profile for eID, V2X, Android Strongbox and automotive security Same SPI command set and pin-out for seamless upgrade This makes the W77Q ideal for designs that require enhanced but cost-sensitive cybersecurity integration. W77F - High Security for Critical Infrastructure The W77F  targets systems that need the highest assurance levels including eID, V2X modules, Android Strongbox applications, and smart access systems. It builds upon the W77Q’s capabilities by adding: Advanced asset protection  with tamper resistance Higher cryptographic isolation  for credential storage Stronger compliance profile  for use in national ID, secure mobile, or automotive security domains Where the RED DA demands secure firmware update paths and the CRA expects built-in resilience and secure data lifecycle management, the W77Q and W77F offer a direct hardware route to compliance. Both series simplify adoption by maintaining SPI command compatibility and pinout with standard NOR flash making them an easy upgrade for existing designs preparing for the new legal requirements. Secure Flash Memory Specifications – Winbond W77Q and W77F Feature W77Q Series W77F Series Target Security Level Substantial High Recommended Applications Smart Home, Industrial, Automotive eID, Car Key, Strongbox, V2X Security Certification CC EAL5+ CC EAL5+ Hardware Root-of-Trust Yes Yes Secure Boot and Firmware Validation Yes Yes Secure SPI Channel Encrypted and authenticated SPI Encrypted and authenticated SPI Resilience Features Rollback, OTA version control Advanced tamper protection Drop-in NOR Flash Replacement Yes (SOIC8, WSON8) Yes (same footprints) Typical Density Range 16 Mb to 128 Mb 64 Mb to 128 Mb OTA Firmware Update Support Supported with signature checks Supported with signature checks Host MCU Requirements No changes to SPI protocol No changes to SPI protocol Supply Voltage 3.0 V (typical) 3.0 V (typical) These specifications make the W77Q a practical choice for general-purpose secure memory upgrades, while the W77F suits applications with elevated security and regulatory requirements, such as national ID or secure mobile applications. Use Cases and Industry Applications The regulatory landscape across Europe and beyond is making secure flash an essential component for embedded systems, not just a high-end feature. Winbond’s W77Q and W77F are being adopted across a growing number of sectors where RED DA and CRA compliance is either required or anticipated. Smart Meters and Grid Infrastructure Smart meters are specifically targeted under the RED Delegated Act due to their wireless interfaces and remote update functions. The W77Q enables secure firmware storage, protects update authenticity, and helps grid device vendors meet both resilience and secure communication provisions — all without redesigning their SPI-based memory interface. Industrial Controllers and IoT Edge Devices From factory automation to building controls, many IEC 62443-compliant systems now require integrity-checked boot sequences and resistance to memory-based attacks. W77Q's secure SPI channel and rollback protections ensure edge nodes cannot be tampered with or rolled back to vulnerable firmware states, making them ideal for CRA-governed products. Automotive ECUs and Connected Mobility The W77Q is suitable for non-critical ECUs requiring OTA updates or secure data logs. For applications with higher security expectations — such as telematics control units or digital keys — the W77F provides added cryptographic protection and tamper detection aligned with automotive-grade cybersecurity frameworks. eID, Smart Access and Secure Mobile The W77F is designed for high-assurance identity systems, supporting secure element-style features without changing the system architecture. It is suitable for use cases such as e-passports, smart door locks, or Android Strongbox storage extensions. Complementary Secure Elements For engineers evaluating broader secure storage strategies, the W77 series may be used alongside or as a lighter-weight alternative to full secure elements. For completeness, Ineltek also supports SEALSQ’s VaultIC292 , a certified secure element suitable for PSTI, RED DA and CRA mandates. Read more . Conclusion – Prepare for RED DA and CRA with Secure Flash Memory With cybersecurity legislation now influencing hardware design, embedded engineers must think beyond the host processor when securing their systems. The upcoming RED Delegated Act  and Cyber Resilience Act  are clear about protecting update processes, enforcing integrity checks, and embedding resilience into connected devices. Winbond’s W77Q  and W77F  series offer a practical route to compliance — without changing your SPI interface or re-architecting your memory layout. These secure flash devices bring root-of-trust, rollback protection, secure SPI channels, and EAL5+ certified storage directly into the flash layer. Whether you're updating a smart meter, launching an industrial IoT controller, or securing an automotive ECU, these drop-in secure memories deliver trusted performance with minimal integration effort. To discuss your project or get access to datasheets and samples, contact Ineltek today . FAQs - Secure Flash Memory for Red DA & CRA Compliance Q: What security certifications do the W77Q and W77F hold? A: Both series are CC EAL5+ certified under Common Criteria, offering formal hardware evaluation for secure boot, key storage and attack resilience. Q: How does the hardware root-of-trust in these devices work? A: At power-on, the root-of-trust verifies firmware integrity via a hardware-anchored key, blocking unauthorised code before execution. Q: Can W77Q and W77F be drop-in replacements for standard SPI NOR flash? A: Yes—both maintain full SPI command compatibility and identical pin-outs, requiring no firmware or PCB changes. Q: Which applications most benefit from W77F’s high-security features? A: eID systems, V2X modules, smart access controls and Android Strongbox integrations where tamper resistance and cryptographic isolation are paramount.

SUMMARY Silicon carbide (SiC) and gallium nitride (GaN) are wide-bandgap semiconductors that deliver exceptional efficiency, switching speed, and thermal performance compared to silicon. This guide breaks down their key differences—voltage rating, frequency capability, cost, and thermal management—and shows you which technology best fits your industrial power application. 18 min read, 4771 words Introduction: How SiC and GaN are transforming Power Design Silicon carbide (SiC) and gallium nitride (GaN) have emerged as game-changing wide-bandgap (WBG) semiconductors in power electronics. Compared to traditional silicon devices, they offer significant advantages in efficiency, switching speed, and thermal performance. WBG transistors can switch faster and at much higher frequencies than silicon MOSFETs, with far lower losses due to lower capacitances and negligible reverse-recovery charge. This enables smaller, lighter, and more efficient power conversion systems, whereas silicon approaches its physical limits at high power and high frequency. As a result, SiC and GaN technologies are playing a growing role in industrial power applications – from electric vehicles and renewable energy to server power supplies – delivering performance that silicon simply cannot achieve. Power-frequency map illustrating the typical application domains of Si, silicon super-junction (Si SJ), SiC, GaN, and legacy IGBT devices. GaN devices dominate in high-frequency (hundreds of kHz to MHz) and lower-to-medium power ranges, enabling extremely compact converters. SiC excels in high-voltage, high-power applications (tens to hundreds of kW) at moderate switching frequencies. In short, SiC and GaN are enabling a new generation of power electronics with higher efficiency and power density. They can handle higher voltages and temperatures than silicon, allowing simpler thermal management and improved reliability. With these advantages, industries are rapidly adopting WBG devices: automotive inverters are replacing silicon IGBTs with SiC MOSFETs to run at higher frequency and efficiency, and data centre power supplies are turning to GaN transistors to shrink size and losses. Technology Comparison: SiC vs GaN Let's compare SiC vs GaN across several key technical parameters: Key Technical Features Bandgap Energy: SiC: ~3.2 eV GaN: ~3.4 eV Silicon: ~1.1 eV Breakdown Voltage: SiC: 650 V–1.7 kV+ (ultra-high voltage) GaN: up to 650 V (mid-voltage) Switching Frequency: SiC: reliably up to ~1 MHz GaN: up to 10 MHz+ Thermal Conductivity: SiC: ~3.7 W/m·K GaN: ~1.2 W/m·K Efficiency: Both enable >98% converter efficiency (e.g., Titanium-level 80 PLUS® specifications demand ≥96% efficiency 80 PLUS Titanium Spec) Cost Trends: GaN-on-Si fabs lower cost per wafer in high volume SiC wafer costs dropping with 6″→8″ expansion Bandgap Energy SiC and GaN are both wide-bandgap materials, with bandgap energies about three times that of silicon. 4H-SiC has a bandgap ≈3.2 eV while GaN is about 3.4 eV, versus silicon's ~1.1 eV. This wide bandgap translates to a much higher critical electric field for breakdown and a lower intrinsic carrier concentration, enabling devices that tolerate higher voltages and temperatures before the material conducts. In practice, this means SiC and GaN can handle far higher voltages  and run hotter than silicon devices without thermal runaway. Breakdown Voltage Thanks to their wide bandgaps, both SiC and GaN support very high breakdown voltages. SiC is typically used for the highest voltage devices – today SiC MOSFETs and diodes are readily available at 650 V, 1200 V, and even 1700+ V ratings, and laboratory devices extend into the kV range. GaN, by contrast, is currently targeted at the mid-voltage range – most GaN power transistors are rated up to 600–650 V (sufficient for offline AC mains applications). This is because most GaN power devices are lateral HEMTs on silicon substrates, which face difficulty beyond ~650 V. In summary, SiC is preferred for ultra-high voltage (>650 V) applications  such as traction inverters or grid-tied systems, whereas GaN excels in the sub-650 V domain . Both easily outperform silicon MOSFETs and IGBTs in breakdown strength. Switching Frequency GaN devices generally offer faster switching capabilities than SiC. GaN HEMTs have exceptionally low gate and output capacitances and high electron mobility, allowing switching frequencies in the MHz range – GaN power transistors can operate at >1 MHz and even up to ~10 MHz in some designs. This enables designers to drastically shrink the size of inductors and transformers. SiC MOSFETs can also switch faster than traditional silicon (in the hundreds of kHz range reliably, vs tens of kHz for silicon IGBTs) but typically top out around ~1 MHz for practical power levels. For example, SiC devices have been demonstrated switching >100 kHz in inverter applications (much higher than silicon IGBTs ~20 kHz). Thus, for very high frequency or fast-switching applications, GaN is often the choice , whereas SiC is more than sufficient for medium-frequency (tens to hundreds of kHz) needs at high power . Power Efficiency Both SiC and GaN enable substantially higher efficiency in power conversion. Their low conduction losses (due to lower R_DS(on) for a given die size, especially at high voltage) and dramatically lower switching losses (thanks to minimal charge storage and the absence of a sluggish body diode) mean that converters can reach efficiency levels unattainable with silicon alone. For example, SiC MOSFETs replacing silicon IGBTs in EV inverters eliminate significant switching loss and allow >98% inverter efficiency. GaN FETs have virtually zero reverse-recovery loss (no body diode), which is ideal for hard-switched bridge circuits (like totem-pole PFC). As a real-world illustration, a state-of-the-art 8.5 kW power supply that leverages GaN and SiC achieves ≈98% efficiency for data center applications. Both technologies enable meeting the highest efficiency standards (Titanium-level 80 PLUS® specifications demand 96%+ efficiency) with margin. In short, WBG devices slash conduction and switching losses , yielding higher efficiency and lower heat dissipation than silicon solutions. Thermal Management SiC and GaN can handle higher temperatures, but SiC is particularly outstanding in thermal properties. SiC has about 3× the thermal conductivity of silicon (approximately 3.7 W/m·K for SiC vs ~1.3 W/m·K for silicon), which means SiC devices can more easily spread and conduct heat away from the junction. This allows higher power density and operation at junction temperatures of 175°C or even 200°C in some cases, benefiting high-power industrial designs. GaN's thermal conductivity is lower than SiC (GaN is roughly 1/3 of SiC's thermal conductivity), so GaN devices in high-power applications often rely on good packaging (e.g., thermal vias, copper spreading) to handle heat. Even so, GaN can operate at high junction temperatures; for instance, automotive-grade GaN FETs are now rated for T_j up to 175°C (matching typical SiC and silicon ratings). Overall, both WBG materials tolerate heat better than silicon, which often must be kept below 150°C. In summary, SiC is often chosen for its robust high-temperature, high-power operation , while GaN, although it runs slightly "hotter", still outperforms silicon  and remains manageable with proper thermal design. Cost Considerations Traditionally, the cost of WBG devices has been higher than equivalent silicon parts, due to newer fabrication processes and smaller production scale. SiC wafers are expensive (the material is hard and was limited to smaller diameters, though 6-inch SiC is common now and 8-inch is forthcoming), and GaN device manufacturing (especially GaN-on-Si or GaN-on-SiC epitaxy) is complex. However, costs are rapidly decreasing as volumes grow and manufacturing matures. Notably, GaN devices tend to be cheaper to produce than SiC at a given voltage – GaN can be made on silicon wafers in standard fabs, making it relatively easier and potentially lower-cost in high volume. SiC devices involve costly crystal growth and processing, but those costs are falling as more suppliers (and larger wafer sizes) come online. As of today, silicon MOSFETs remain the most inexpensive option for low-to-moderate performance needs, but when stringent efficiency or high-voltage requirements arise, the improved performance of SiC/GaN can offset their higher device cost (e.g., by enabling simpler cooling or smaller magnetics, which reduces overall system cost). Market trends indicate a steep growth in WBG adoption, which will further drive down cost through economies of scale. In fact, forecasts project the SiC/GaN power semiconductor market to grow from about $1.6B in 2024 to over $40B by 2034, a clear indicator that cost barriers are coming down and the technology is becoming mainstream. Parameter Typical SiC MOSFET (1200 V) Typical GaN HEMT (650 V) RDS(on) 25 mΩ @ 25 °C 30 mΩ @ 25 °C Max Junction Temperature 175 °C 175 °C Switching Frequency 100 kHz–1 MHz 1 MHz–10 MHz Package TO-247, D2PAK, 6 mm² CoolSiC® QFN, LGA Typical Efficiency ≥98% (in EV inverter) ≥98% (in totem-pole PFC) Price (1 k qty) ~$15 ~$8 Industrial Use Cases for SiC and GaN Wide-bandgap devices are being leveraged in many industrial and commercial power electronics. Some key applications where SiC and GaN are making a significant impact include: Industrial IoT Power Management In the Industrial IoT realm, there are countless distributed sensors, controllers, and devices that need efficient, compact power supplies. GaN, in particular, allows for ultra-small point-of-load converters and adapters with high efficiency, reducing wasted power in always-on embedded systems. For example, GaN-based regulators can operate at high frequency to shrink inductors, which is ideal for space-constrained IoT modules. SiC and GaN are also used in higher-power IoT infrastructure – e.g., in smart building energy managers or factory automation hubs – to achieve better efficiency and lower heat. The result is greener, cooler-running IoT devices that can be deployed in tight spaces without bulky heatsinks, aligning with the energy-conscious goals of modern IoT deployments. Motor Drives and Automation SiC in high-power VFDs for large factory motors at 800 V+ DC bus. GaN in precision servo drives and robotics for ultra-fast PWM control. Industrial motor drives (from small servo drives to large factory motors) benefit greatly from WBG semiconductors. SiC MOSFETs are being adopted in variable frequency drives and industrial inverters to handle higher bus voltages (such as 800 V or more DC) with lower losses, which improves efficiency and reduces cooling in motor control centres. GaN transistors, while typically lower voltage, find use in servo drives and robotics: their fast switching allows very precise control and high PWM frequencies for smoother motor operation. In automation equipment (conveyors, machine tools, robot arms, etc.), using SiC/GaN in the inverter stages can increase system efficiency and power density. For instance, servo amplifiers and CNC drives can use GaN to achieve faster current loops and reduce the size of the drive unit. Similarly, robotic and CNC motor drivers are employing SiC/GaN to handle rapid load changes with less loss. In summary, SiC is ideal for high-voltage/high-power industrial motors (e.g., large 3-phase motors in processing plants), while GaN suits lower-voltage, high-frequency drives (e.g., precision servo motors), each enabling better efficiency and performance in automated systems. High-Efficiency Power Conversion (DC-DC, AC-DC) SiC and GaN are transforming industrial power supplies, UPS systems, and converters: AC-DC Front Ends (PFC)  – Traditional silicon-based PFC circuits are limited in switching frequency and incur significant losses. Replacing the silicon diode+MOSFET with a SiC Schottky diode and SiC or GaN transistor in a totem-pole PFC can achieve >99% power factor correction efficiency. This is critical for data centre PSUs, industrial UPS, and telecom rectifiers. In fact, GaN FETs have enabled the totem-pole PFC topology to become practical by virtually eliminating diode reverse-recovery losses. Isolated DC-DC converters  in industrial contexts (48 V to 12 V converters, fork-lift or AGV battery converters, etc.) are using SiC and GaN to push switching frequencies into the MHz range, greatly reducing transformer size. For example, in server and telecom power supplies, GaN is often preferred for up to ~400–600 V input stages – it allows very high frequency PWM operation, which shrinks magnetics and improves transient response. Meanwhile, SiC excels in high-power converters such as those in renewable energy systems or large industrial SMPS where the bus voltage is high (800–1500 V) and efficiency is paramount. In short, for any application demanding high-efficiency AC-DC or DC-DC conversion – from compact 100 W power modules to 50 kW warehouse UPS – SiC and GaN offer solutions. GaN tends to dominate in medium-power, high-frequency converters, while SiC is common in high-power, high-voltage converters. Both contribute to achieving industry-leading efficiency levels, often meeting or exceeding regulatory targets for energy efficiency. Renewable Energy Systems SiC MOSFETs and diodes in solar inverters (600–1500 V DC) improve yield and reliability. GaN in auxiliary and storage-side DC-DC converters for high-frequency operation. The push for renewable energy (solar PV, wind, energy storage) heavily relies on advanced power electronics, and WBG devices are key enablers here. SiC devices are a natural fit for solar inverters – many modern PV inverters (from residential string inverters to large solar farm inverters) use SiC MOSFETs and SiC diodes to handle high DC bus voltages (typically 600–1000 V DC for strings, up to 1500 V in utility-scale) with minimal loss. SiC's high voltage capability and low switching losses directly translate to higher conversion efficiency from DC to AC, which improves the energy yield of solar installations. Additionally, SiC Schottky diodes are commonly used in the boost stages of PV inverters to minimise switching losses and improve efficiency, or as freewheeling diodes in inverter legs. On the energy storage and battery management side, SiC enables efficient high-power DC-DC converters for battery charging/discharging in large systems. GaN is also finding roles in renewable systems, particularly in smaller-scale or auxiliary converters. Overall, renewables benefit from SiC for its high-voltage, high-efficiency operation, and GaN for any high-frequency, lower-voltage tasks. The result is higher efficiency power conversion in solar/wind systems and reduced cooling requirements, helping meet strict efficiency mandates for green energy. Product Offerings from Key Manufacturers Next, we examine how four semiconductor manufacturers – Nuvoton , Novosense , Magnachip , and Bruckewell – are contributing to the SiC/GaN ecosystem with their product offerings. Each brings unique solutions for industrial power engineers: Nuvoton Nuvoton Technology (a Taiwanese semiconductor company known for microcontrollers) has expanded into power semiconductors, including wide-bandgap devices. Nuvoton's key focus is on GaN technology. Through its foundry services and process development, Nuvoton has established a special process flow to fabricate GaN power devices to meet the needs of high-efficiency, high-power systems. While Nuvoton's publicly released product info on SiC/GaN is limited, they have demonstrated capability in GaN through other domains: for example, Nuvoton offers an RF GaN Power Amplifier module for 5G base stations. On the silicon carbide side, Nuvoton does not yet market SiC MOSFETs or diodes; however, they do provide many supporting technologies for power electronics. Nuvoton's line of Power ICs and microcontrollers can complement WBG devices – for instance, their motor-control MCUs and isolated gate driver ICs can drive SiC MOSFETs in an inverter. In summary, Nuvoton's contribution lies in GaN device fabrication capability and system ICs. Engineers can look to Nuvoton for GaN technology especially, and for controllers that manage SiC/GaN-based power stages. Novosense Novosense Microelectronics (China) has positioned itself as a specialist in analogue and mixed-signal ICs for sensing and driving, and they have developed a robust portfolio to support SiC and GaN power devices. Rather than manufacturing the transistors, Novosense focuses on the interface and control – notably gate driver ICs and integrated power stages for WBG transistors. A standout offering from Novosense is their line of isolated gate drivers that are tailored for SiC MOSFETs and GaN HEMTs. For example, Novosense's NSi68515 is a single-channel intelligent isolated driver specifically designed to drive SiC MOSFETs (and IGBTs) in high-voltage systems up to 2121 V DC bus. For GaN, Novosense has developed dedicated solutions as well. They offer the NSD2621, a high-voltage isolated half-bridge driver specifically meant for enhancement-mode GaN FETs. This driver addresses GaN's unique needs: it provides a very high common-mode noise rejection (up to 150 V/ns CMTI) and can tolerate the negative transient voltages (~– 700 V) that can occur at the switch node in half-bridge GaN circuits. Perhaps most impressively, Novosense has an integrated GaN power stage product: the NSG65N15K. This device combines a half-bridge driver (the NSD2621) and two 650 V GaN transistors (each 150 mΩ) into a single compact package. Essentially, it is a half-bridge GaN module in a 9×9 mm QFN that can handle up to 20 A, with both high-side and low-side GaN FETs internally driven. In summary, Novosense's SiC/GaN offerings are in the realm of high-performance driver ICs and integrated power stages. They do not sell SiC/GaN transistors themselves, but they make the chips that drive those transistors to their full potential. Magnachip Magnachip Semiconductor, based in South Korea, is a well-established player in power MOSFET technology (particularly in the trench MOSFET space). While Magnachip's portfolio today is largely focused on advanced silicon power MOSFETs rather than SiC or GaN, their products are highly relevant in the context of wide-bandgap adoption. Magnachip produces Super Junction (SJ) MOSFETs in the 600 V, 700 V, and 800 V classes, which are used in offline AC-DC power converters (similar application space as GaN or SiC up to 800 V). These MOSFETs feature embedded gate-source ESD zener diodes for robustness against surges and have ~30% lower total gate charge compared to previous generations, which directly improves switching efficiency. When it comes to SiC/GaN specifically, Magnachip has not yet released public products in those categories (as of now). However, Magnachip's high-voltage silicon devices often complement WBG adoption. For example, a power supply designer might use Magnachip's 800 V MOSFET for an input flyback stage if GaN is not necessary, or for a cost-sensitive design where silicon suffices. In summary, Magnachip provides high-quality silicon MOSFET solutions that cover many industrial needs up to 800 V. While they do not directly offer SiC/GaN parts as of now, their MOSFETs can be seen as complementary or interim solutions in the WBG journey. Bruckewell Bruckewell Semiconductor specialises in power discretes and notably offers products in silicon, SiC, and GaN technologies. They provide a comprehensive line-up of components that allow engineers to upgrade designs from silicon to wide-bandgap within the same brand. SiC Schottky Diodes:  Bruckewell produces SiC Schottky barrier diodes aimed at high-efficiency rectification and freewheeling applications. Their SiC diode family covers 650 V and 1200 V ratings with current options from 4 A up to 40 A, offered in popular power packages (TO-220, TO-247, DPAK, DFN, etc.). SiC MOSFETs:  Bruckwell also has SiC MOSFETs up to 1200 V in its portfolio. By offering SiC MOSFETs, Bruckewell enables designers to implement full SiC half-bridges (using their MOSFETs and diodes together) for applications like solar inverters, motor drives, etc. GaN HEMTs and Cascodes:  Bruckewell is somewhat unique in that it uses a GaN-on-Sapphire process for its GaN devices. Most industry GaN is on silicon, but Bruckewell chose sapphire substrates, which yields excellent electrical isolation and very low leakage currents (at the expense of thermal conductivity). Bruckewell's GaN device portfolio highlights multiple approaches: A cascode GaN (combining a D-mode GaN with a low-voltage Si MOSFET for normally-off operation) An enhancement-mode GaN HEMT A GaN "IC" that integrates a GaN cascode with a gate driver Overall, Bruckewell distinguishes itself by providing WBG power semiconductors in discrete form that are accessible and flexible. An industrial power designer can obtain SiC diodes/MOSFETs from Bruckewell to instantly upgrade an existing design's efficiency (often as drop-in replacements for legacy diodes/MOSFETs). Design Considerations: Selecting the Right Technology Choosing between SiC and GaN (or determining how to mix them in a design) requires careful consideration of the application requirements. Engineers should evaluate several key factors to select the optimal device: Voltage and Current Ratings Perhaps the first decision point is the DC-link or bus voltage of the application and the required current. SiC devices excel at high voltage and high current – they are available for 1200 V, 1700 V and beyond, and can handle tens to hundreds of amperes per device (or thousands in modules). If your design involves, say, an 800 V DC bus (common in EV or industrial drives) or a 1500 V solar string, SiC is the natural choice; GaN in those voltage ranges is not yet mainstream. GaN devices are generally favoured up to ~600 V applications. For instance, for a 400 V bus (typical of telecom or datacentre power) or offline 240 VAC input, GaN FETs (650 V rated) can be used and will offer excellent performance. In terms of current, SiC MOSFETs in TO-247 can handle >50 A, and power modules even more (with parallel dies), whereas GaN transistors, being smaller die typically, might handle on the order of 10–30 A each (though you can parallel GaN devices too). So, for very high power (kW-level with high voltage), SiC is usually a better fit, while for medium power or lower voltage, GaN is compelling. It's common to see a combination: e.g., SiC diodes or MOSFETs used on a PFC front-end (handling 400–800 V), with GaN used on a subsequent DC-DC stage at 400 V or less. Switching Speed Requirements Determine how fast and at what frequency the devices need to switch. GaN is preferable for the highest switching frequencies – if you need to switch in the MHz range or require extremely fast edges (e.g., for very low switching loss or special modulation schemes), GaN can deliver where SiC might be hard pressed. For example, in a MHz-class resonant converter or a very fast on-off pulsed system, GaN's ability to switch >5–10 MHz (in lower-power cases) and its low gate charge make it ideal. SiC, on the other hand, is quite capable up to hundreds of kHz – many SiC MOSFET-based inverters run at 50–200 kHz with great success (compared to ~20 kHz typical for IGBTs). If your application can achieve its goals at, say, 100 kHz, SiC might be sufficient and offers the high-voltage robustness. If pushing to, say, 500 kHz or 1 MHz to shrink magnetics, GaN may make that easier. Thermal and Packaging Constraints Consider the thermal environment of the design and the form factor. If the design must operate at high ambient temperatures or with limited cooling, SiC's ability to accommodate a higher junction temperature and better thermal conductivity can offer more margin. For example, in a sealed industrial motor drive with minimal airflow, using SiC devices that can run safely at 150–175°C junction (and have lower losses) might prevent overheating better than a cluster of GaN devices (which individually might run hotter due to lower thermal conductivity and possibly slightly higher switching loss at full load). Packaging is related: SiC MOSFETs often come in larger through-hole packages (TO-247, TO-263, etc.) or even modules, which have good thermal paths and can be bolted to heatsinks. GaN devices commonly come in compact SMT packages (QFN, DFN, or even chip-scale BGA packages) to minimise inductance. If your design is very space-constrained, GaN's small packages are a plus, but you must ensure the PCB can dissipate the heat (using copper planes, thermal vias, etc.). If your design can accommodate module packaging, SiC power modules might greatly simplify thermal management for high-power designs by integrating multiple MOSFETs/diodes on a substrate with good cooling. Application-Specific Requirements Finally, consider the nuances of your particular application: Short-circuit and surge robustness:  SiC MOSFETs typically have better short-circuit withstand time than GaN HEMTs (which are smaller geometry and can be more sensitive to overload). If your application (like motor drives) demands the ability to survive short circuits for, say, 5–10 µs while protection kicks in, SiC might be safer. Efficiency vs frequency trade-offs:  If absolute peak efficiency is required at a given power level, you might choose one over the other. For instance, at ~5 kW, a SiC-based design switching at 50 kHz might achieve 98% efficiency, whereas a GaN design switching at 200 kHz to reduce size might have slightly lower efficiency (maybe 96–97%) due to higher switching frequency. EMI and noise considerations:  GaN's very fast switching can lead to more high-frequency EMI (electromagnetic interference), which might complicate compliance with EMC standards unless mitigated. SiC switches fast as well, but typically a bit slower than GaN, possibly easing EMI filtering. Gate drive complexity:  GaN HEMTs have unique gate drive needs – typically a gate voltage of 6 V (for enhancement-mode) and a very strict limit (often max 7 or 8 V, and negative voltage must be avoided unless specified). SiC MOSFETs require higher gate voltage (usually +15 V gate drive and often –3 to –5 V turn-off), so you'll need a driver that can output that swing, and perhaps negative bias for turn-off to prevent the Miller effect. Market Trends & Future Outlook The adoption of SiC and GaN in industrial power electronics is not only well underway – it's accelerating. Several market and technology trends are worth noting: Rapid Market Growth and Investment The WBG power semiconductor market is experiencing robust growth, driven by high demand in automotive, renewable energy, and data centres. Analysts project the global SiC/GaN market to grow from around $1–2 billion in the early 2020s to tens of billions by the early 2030s. SiC devices, in particular, are being produced in greater volumes as multiple companies ramp up 6-inch SiC fabs, and GaN device shipments (especially for consumer and IT electronics) are climbing steeply. Asia-Pacific is a leading region in adoption (over half of the market by revenue), with China investing heavily in both SiC and GaN for electric vehicles and 5G/industrial needs. Automotive Electrification as a Catalyst The electric vehicle (EV) boom is a massive driver for SiC, and to a growing extent, GaN. Automakers have found that SiC devices in the drivetrain inverter can increase range ( by improving efficiency ~5% or more ) and reduce cooling needs, which in turn can lower battery costs or extend vehicle range. Many major EV OEMs have adopted SiC for the main traction inverters and onboard chargers. GaN is being eyed for less demanding automotive applications like 48 V systems , lidar, or perhaps future onboard chargers. With EVs and plug-in hybrids rising, the automotive sector's demand for qualified WBG parts is skyrocketing. Industrial Efficiency Standards and Green Initiatives On the industrial side, ever-tightening efficiency regulations for power supplies, motor drives, and appliances are practically mandating the move to WBG. For example, data centres strive for 80 PLUS Titanium PSU efficiency (≥96%), which is very hard to meet with silicon alone – GaN is being adopted in server PSUs to hit those levels. Renewable energy standards demand high efficiency in inverters to maximise use of generated power; using SiC helps meet those targets and is often required to qualify for incentives. Governments and regulatory bodies around the world are also setting CO₂ reduction and energy saving goals, which translate to using the most efficient technology available – again pointing to WBG. Technology Maturation and Ecosystem Growth Both SiC and GaN technologies are maturing, and we're seeing rapid improvements and new product introductions. On the SiC front, new generations of MOSFETs have lower and lower R_DS(on) and higher robustness; package innovations (like dual-sided cooling, advanced modules) are improving performance. Integration is also happening – for instance, SiC MOSFET half-bridge modules with incorporated gate drivers and protections are emerging, simplifying design. GaN technology is advancing with things like GaN ICs (monolithic integration of GaN FETs and drivers or logic). Another trend is GaN moving to higher voltages: while 600 V is standard, we now hear about 900 V GaN or 1200 V GaN prototypes, often by using GaN-on-GaN substrates or advanced device structures. This could eventually encroach on SiC's high-voltage territory, though that's years away for mass production. Future Outlook – Complementary, Not Just Replacement It's clear that SiC and GaN are here to stay and will increasingly dominate high-performance power electronics, but interestingly, they will co-exist with silicon devices for a long time in a complementary way. As one industry pundit put it, "Silicon, GaN, and SiC all have a place" and engineers will mix them as needed. We are already seeing designs where silicon MOSFETs handle low-voltage sections, GaN handles the middle, and SiC handles the high end. Going forward, wide-bandgap devices might enable new power conversion architectures – for example, single-stage AC/DC converters operating at high frequency, or matrix converters that were impractical with slow devices. There's also research in even wider bandgap materials like Gallium Oxide (Ga₂O₃) or Diamond, but those are a bit far out; in the next decade, SiC and GaN will be the workhorses. Conclusion: Making the Right Choice for Your Application SiC and GaN technologies have proven their ability to revolutionise industrial power electronics design. They deliver substantial improvements in efficiency, switching speed, and power density over legacy silicon devices, enabling engineers to meet ambitious performance and efficiency targets. Silicon Carbide shines in high-voltage and high-power scenarios – from hundreds of volts to kilovolts – offering robust operation and efficiency in applications like motor drives, renewable energy inverters, and EV power systems. Gallium Nitride, on the other hand, dominates at high switching frequencies and in compact designs, making it ideal for server supplies, point-of-load converters, and any application where minimizing size or maximizing speed is critical. Each technology has its strengths: SiC handles heavy loads and high temperatures with ease, while GaN enables extreme switching speeds and integration. Rather than one replacing the other, they complement each other – often working together in modern power conversion systems to achieve optimum performance at each stage. For engineers, the challenge is no longer "should I consider SiC or GaN?" – that is a given – but rather "how to select and implement the right SiC/GaN device for my design." As we've discussed, this involves balancing voltage, frequency, thermal, and cost requirements, and utilizing the growing ecosystem of drivers and tools available. With the right choice, a design that once struggled to meet efficiency or size specs with silicon can not only meet but exceed those specs with margin. Wide-bandgap semiconductors are no longer experimental – they are commercially mature and ready to be designed in, with support from many manufacturers. Contact Ineltek for Expert Guidance The time is ripe to leverage SiC and GaN in your own projects. Whether you are upgrading an existing power supply for better efficiency or architecting a next-generation motor drive, embracing WBG devices could be the key to a superior design. If you're considering any of the technologies mentioned in this article, Ineltek can arrange technology introductions with the people who make them. Drop us a line and let's start talking. FAQs - Sic vs GaN for Industrial Power Design Q. What factors determine whether Sic or GaN is best for my design? A:  Choose SiC when you need >650 V or high-power (>100 kW) converters with robust thermal margins; choose GaN for ≤650 V, medium-power (<10 kW) designs demanding very high switching frequency and minimal size. Q:  How does switching frequency impact the choice between SiC and GaN? A:  GaN HEMTs excel at multi-MHz switching, enabling tiny magnetics and capacitors, whereas SiC MOSFETs typically run up to ~1 MHz - still a big leap over silicon but generally used where switching losses at ultra-high frequency are less critical. Q:  How do thermal characteristics compare for SiC vs GaN? A:  SiC’s thermal conductivity (~3.7 W/mK) is roughly three times GaN’s, so SiC often needs less aggressive heatsinking. GaN’s lower conductivity can be managed with advanced packaging (copper spreaders, thermal vias). Q:  What’s the cost trajectory for SiC and GaN devices? A:  GaN on silicon leverages existing fabs, yielding lower unit costs at volume today. SiC costs are falling fast as wafer sizes grow (4″→6″→8″). Expect cost parity in many mid-power segments within 2–3 years.