Electronic Products announces 2020 Product of the Year Award winners

The annual Product of the Year awards, now in its 45^th year, recognizes outstanding products that represent any of the following qualities: a significant advancement in a technology or its application, an exceptionally innovative design, a substantial achievement in price/performance, improvements in design performance, and a potential for new product designs/opportunities. Here are this year’s winners, selected by the editors of Electronic Products:

• C&K (Electromechanical)
• Cornell Dubilier Electronics (Passives)
• Eta Compute and Maxim Integrated (Digital ICs)
• Hughes Electronics (Interconnects)
• InnoPhase Inc. (RF/Microwave)
• Keysight Technologies (Test & Measurement)
• Maxim Integrated (Analog/Mixed-Signal ICs)
• Navitas Semiconductor Ltd. (Power)
• Seoul Semiconductor Co., Ltd. (Optoelectronics)
• Texas Instruments Inc. (Sensors)

C&K
Electromechanical: Nano Tactile (NanoT) Switch

C&K claims that the Nano Tactile (NanoT) switch series is the smallest solution on the market, taking aim at internet of things devices, including smartwatches and medical wearables such as health-monitoring trackers and hearing aids. The package measures 2.2 × 1.70 × 1.65 mm, giving designers more room to add higher functionality to their new designs or to reduce the size of the printed-circuit board (PCB).

Other switch features include:

• A PIP or SMT on edge-mount configuration to protect wearable and medical products from tough conditions such as vibrations, accidental drops, or exposure to liquid while also improving the devices’ life cycle, as they are used repeatedly, which causes natural wear and tear
• An integrated actuator that lowers cost and integration challenges with full interface buttons
• 100, 160, and 240 gf
• A life cycle up to 300K
• IP67 for sealed-switch compatibility with PCB coatings

“As wearable, medical, and other devices continue to shrink, it’s critical that the switches used to keep them functioning properly fit within emerging size requirements,” said C&K.

“Advancements in technology are leading to consumer demand for smaller products that they can easily take with them wherever they go,” added the company. “As products become smaller, the switches needed to power them on and off also need to become smaller.”

Cornell Dubilier Electronics
Passives: PPC Series of Aluminum Electrolytic Capacitors

Cornell Dubilier Electronics (CDE) introduced a new capacitor form factor with the PPC Series of ultra-thin polymer aluminum electrolytic capacitors. Thinner than a dime, this capacitor family is available in custom values and shapes to accommodate available space, thanks to a versatile packaging technology.

Designed specifically for applications requiring high ripple current and the thinnest-possible profiles, the PPC capacitor opens up new product design options. “These designs have been responsive to market demands for capacitors that provide higher performance while breaking away from traditional cylindrical form factors,” said CDE.

Conventional cylindrical form factors are much thicker, taller, and heavier, and until now, the alternative way around the bulkier packaging was to use a bank of many SMT devices, said the company.

CDE claims that a single PPC capacitor is equivalent to a parallel bank of 50 or more polymer tantalum capacitors and occupies one-fourth the height. It also improves circuit reliability by using a single component versus an entire array of SMT capacitors.

“A single PPC capacitor offers capacitance and ripple current equivalent to dozens of SMT capacitors or a bulky cylindrical device,” said Mario DiPietro, product manager at Cornell Dubilier.

The 1-mm–thin PPC series offers custom values within a capacitance range of 8,000 µF to 20,000 µF, with working voltages ranging from 6.3 to 24 WVDC. The company plans to extend the series to higher operating voltages later in the year. Operating life is 2,000 hours @ 125°C.
The PPC is rated for 10 g peak for vibration and withstands shocks up to 100 g (MIL-STD-202, Method 213, Condition I).

Eta Compute
Digital ICs: ECM3532 AI Multicore Processor

This year, there is a tie for Digital IC Product of the Year between two artificial-intelligence processors: Eta Compute’s ECM3532 edge AI processor and Maxim Integrated’s MAX78000 neural network accelerator chip.

In alpha order, let’s first look at Eta Compute’s ECM3532, claimed as the industry’s most energy-efficient edge AI processor. The ECM3532 is a neural sensor processor (NSP) for always-on image and sensor applications, bringing AI to edge devices and translating sensor data into actionable information for a range of applications including voice, activity, gesture, sound, image, temperature, pressure, and biometrics applications. The standalone AI platform includes a multicore processor that includes flash memory, SRAM, I/O, peripherals, and a machine-learning software development platform.

Eta Compute also claims that the ECM3532 is the industry’s first AI multicore processor for embedded sensor applications. It features the company’s patented continuous voltage frequency scaling (CVFS) with dynamic voltage and frequency scaling and near-threshold voltage operation, delivering power consumption of microwatts (as low as 100-μW active power consumption in always-on applications) for many sensing applications.

Using a hybrid multicore architecture, the ECM3532 NSP, housed in a 5 × 5-mm, 81-ball BGA, combines an MCU and a DSP, both with CVFS, to optimize execution for the best efficiency with procedural programming, signal processing, and neural networks, making it suited for IoT sensor nodes. “Because power is a factor of frequency × V2, CVFS brings a quadratic gain in efficiency,” said Eta Compute.

The company’s patented silicon-proven solutions solve challenges related to CVFS. These include:

• Voltage tuning depending on frequency (55 ULP –0.54 V to 1.2 V)
• Integrated PMIC determines the lowest voltage for a given frequency
• No transition time between power states
• Easy integration with optimizing OS

The CVFS substantially increases performance and efficiency for edge devices, said the company, and the self-timed CVFS architecture automatically and continuously adjusts internal clock rate and supply voltage to maximize energy efficiency for the given workload.

The NSP targets ultra-low–power edge AI, energy harvesting, and IoT applications including audio AI, wearable/hearable devices, asset-tracking modules, people detection/counting, object detection, industrial IoT, preventive maintenance, energy-harvesting sensor nodes, and BLE low-power beacons.

Maxim Integrated
Digital ICs: MAX78000 Neural Network Accelerator Chip

Maxim Integrated’s MAX78000 low-power neural network accelerated microcontroller moves AI to the edge without compromising on performance in battery-powered IoT devices.

The neural network accelerator chip executes AI inferences at less than 1/100 the energy of software solutions, which dramatically improves runtime for battery-powered AI applications, while enabling complex new AI use cases previously considered impossible, said Maxim Integrated.

These power improvements come with no compromise in latency or cost, executing inferences 100× faster than software solutions running on low-power microcontrollers at a fraction of the cost of FPGA or GPU solutions, added the company.

“AI technology allows machines to see and hear, making sense of the world in ways that were previously impractical,” said Maxim. “In the past, bringing AI inferences to the edge meant gathering data from sensors, cameras, and microphones; sending that data to the cloud to execute an inference; and then sending an answer back to the edge. This architecture works but is very challenging for edge applications due to poor latency and energy performance. As an alternative, low-power microcontrollers can be used to implement simple neural networks; however, latency suffers and only simple tasks can be run at the edge.”

Maxim’s solution is to integrate a dedicated neural network accelerator with a pair of microcontroller cores, which overcomes the above-mentioned limitations, enabling machines to see and hear complex patterns with local, low-power AI processing that executes in real time. So applications such as machine vision, audio, and facial recognition can be more efficient, as the MAX78000 can execute inferences at less than 1/100 the energy required by a microcontroller.

The MAX78000 uses specialized hardware that is designed to minimize the energy consumption and latency of convolutional neural networks (CNNs). This hardware runs with minimal intervention from any microcontroller core for streamlined operation.

Hughes Electronics
Interconnects: OptiMod Fiber Optic and DC Power Cabling

Hughes Electronics launched OptiMod, an advanced fiber optic and DC power cabling concept for cellular networks. Designed to meet the harsh environments found in the wireless communications industry with high flexibility, the OptiMod delivers high-performance connections for hybrid, separate (DC and fiber), or mixed cabling layout requirements.

The key to OptiMod’s flexibility is a new modular connector concept called H-Mod. Like all other OptiMod system components, H-Mod is designed to protect fragile glass fiber cores in harsh environments, including rough handling and installation practices. H-Mod connections don’t need tools, special skills, or equipment, said Hughes, and a unique positive lock-latching system enables error-free connections.

In addition, the LC fiber and high-performance DC connection modules are combined in the same robust cast aluminum IP68 housing to form the hybrid H-Mod connectors. This translates into one connection to transmit DC power and fiber to radio equipment, which reduces hundreds of man-hours in deployment, according to the company.

The H-Mod can house discrete fiber or DC connection modules for use with separate power and fiber cables to support up to eight remote radio units (RRUs) per power or fiber connection.

The hybrid cables are pressure-proofed thanks to corrugated aluminum tubing and spiral steel armoring, which provides protection against impact, birds, vermin, and other hazards. Over-sheathing with resilient HDPE provides UV and corrosion resistance.

The exterior-class fiber-only cable features high-grade, low-smoke compounds formulated for harsh environments, and like the hybrid cable, it has flexible steel armoring for crush and rodent resistance. The multi-core DC-only cable incorporates full-braided shielding with optional functional Earth and is available from single to 10 pairs of 10-mm² cores.

The system also includes OneBox, an advanced cabling distribution unit. The H-Mode IP68 cable distribution box for external connectivity reduces installation time by 80%, according to the company, with a black box combi-cable design that is plug-and-play.

OneBox accepts hybrid, separate, or mixed cable layouts for greater flexibility. It can handle up to 400 A of power and 48 cores of fiber input to support up to 10 RRUs. It also incorporates Hughes’s advanced positive lock technology for secure connection.

OneBox and the OptiMod system also incorporate a separate functional Earth and shielding continuity. “These unique but important features address major issues involved in the industry practice of earthing DC power to the structure,” said the company. “Up to 70% of return current may pass via the earthed mounting steelwork unless a suitable FE path is offered. OptiMod provides that path, preventing corrosion and depletion of environmental protective coatings.”

InnoPhase Inc.
RF/Microwave: Talaria TWO INP1010 & INP1011 Modules

Claiming an RF architecture that dramatically reduces Wi-Fi radio current consumption, InnoPhase’s Talaria TWO wireless semiconductor platform is designed to provide a complete system solution for extremely low-power, battery-based wireless IoT applications.

“It addresses the most critical issue holding back the market development and growth of battery-based IoT products: the significant amount of power required by today’s ‘old technology,’ analog-centric radio architectures,” said InnoPhase. “Battery-based IoT products don’t last long enough between recharging or replacement to be acceptable, even to early adopters.”

The underlying issue is the basic radio-signal–processing architectures used by wireless communication products that are based on a 25-year-old RF technique known as IQ radio data representation, explained the company.

InnoPhase’s answer to the problem is PolaRFusion, a digitally intensive radio architecture that significantly reduces the amount of power required to transmit, process, and receive wireless information using industry-standard wireless protocols. The PolaRFusion architecture moves much of the radio-signal processing from power-hungry analog circuits into highly efficient digital logic, said the company.

This translates into many months or years between battery changes for battery-based “edge of the network” IoT products. This will enable “an entirely new category of high-volume, battery-powered consumer, commercial, medical, and industrial applications that are simply not possible using today’s wireless semiconductor technology.”

Specifically, the INP1010/1011 Talaria TWO modules are complete solutions with integrated wireless connectivity plus microcontroller for edge-of-network IoT designs. They support multiple wireless protocols, such as Wi-Fi, Bluetooth/BLE, and Zigbee, through software programming. The products use the company’s Talaria TWO Multi-Protocol Platform with Wi-Fi and BLE for wireless data transfer, an embedded ARM Cortex-M3 for system control, and user applications plus advanced security elements for device safeguards.

InnoPhase claims that the unique digital polar radio architecture makes the modules the industry’s lowest-power Wi-Fi solution. It also provides BLE connectivity for Wi-Fi provisioning, diagnostics, and other local communication. The integrated solution is suited for a range of battery-based, direct-to-cloud devices such as home automation, portable voice-activated speakers, smart door locks, remote security cameras, sensor modules, portable patient-monitoring devices, wearables, and other energy-critical edge-of-network wireless IoT applications.

Keysight Technologies
Test & Measurement: CX3300A Anomalous Waveform Analytics

Keysight Technologies has embedded an innovative feature — Anomalous Waveform Analytics (AWA) — into the Keysight CX3300A Series of device current waveform analyzers. This feature enables automatic extraction of rare anomalous (current and voltage) signals in a large volume of waveform data exceeding a terabyte that are not possible to be performed by manual measurement analysis, said the company. Enabling users to identify any anomalous signals quickly to rectify the problem, it also significantly reduces the processing time and manpower operations in a product development cycle, said Keysight.

Until now, there have been no comparable single hardware solutions able to meet these test and measurement requirements, which are essential to resolve any big data measurement applications, according to the company.

“Although data loggers can capture large amounts of data, they have relatively low bandwidth limitation and are prone to overlooking high-frequency signal components,” said Keysight. “Conventional oscilloscopes are suitable in capturing high-bandwidth signals, but they have limited data storage capacity. Even a high-performance oscilloscope with large memory depths are unable to capture data at a high sampling rate over a prolonged time of hours or days. In addition, current probes used in an oscilloscope do not have sufficient range to capture dynamic range of high- and low-level current measurements.”

AWA incorporates an advanced clustering algorithm, data analytics with machine learning, and a proprietary database-compression technique. It can cluster large data volumes, enabling users to build and navigate from a large waveform database (up to 100-hour acquisition), simultaneously analyzing the data. It provides a high-resolution visual display of each type of waveform and enables users to drill down into a selected portion of the waveform type for deeper analysis.

Maxim Integrated
Analog ICs: MAX16160

Added to the company’s Essential Analog portfolio of high-performance, single-function analog ICs, Maxim Integrated’s MAX16160, a four-channel voltage monitor and reset IC, delivers precision measurement for consumer, communications, industrial, and medical applications. In addition to improving measurement accuracy, it can also reduce power consumption and solution size in a range of applications, including cloud infrastructure, IoT, intelligence at the edge, on-device AI, and smart and emerging applications in consumer, communications, industrial, and medical markets.

The MAX16160 is the only four-channel voltage monitor and reset IC that uniquely asserts reset low without requiring a VCC rail if any of the rails are above 1 V, said Maxim.

“It features the industry’s most reliable startup and continuous operation of multi-supply systems by asserting output low,” added the company. “This ‘no-power power-up’ eliminates unknown system states and significantly reduces the burden on design engineers to accurately sequence their power rail startup. With a voltage-monitoring accuracy of ±1% for all inputs, it is up to 50% more accurate than competitive offerings, commonly ±1.50% for all inputs.”

Size is another benefit: Housed in a six-bump WLP (1.408 × 0.848 mm), it is 85% smaller than the closest competitor, according to Maxim. It is available in a small, six-pin SOT23 package and six-bump WLP.

Navitas Semiconductor Ltd.
Power: GaNFast Power ICs NV612x Series

Navitas introduced a new range of 650-/750-V–rated GaNFast power ICs with a proprietary, integrated cooling pad for high-efficiency, high-density power systems, housed in 6 × 8-mm PQFN packaging, enabling faster charging in a smaller size.

The new power ICs are built on gallium nitride (GaN) technology that runs up to 20× faster than silicon (Si) and enables 3× more power or 3× faster charging in half the size and weight. The GaNFast power ICs enable next-generation upgrades across diverse markets, said Navitas. Applications include 25- to 100-W consumer and mobile USB-C fast chargers and adapters for smartphones and laptops, 200- to 800-W TVs and all-in-one computers, and up to multi-kilowatt EV, industrial, and data center power supplies.

The new range of 650-V–rated power ICs includes complete gate drive and protection circuits plus GaN power FETs in 6 × 8-mm surface-mount PQFN, low-inductance (high-speed) packaging.
The devices are the NV6123 (650/750 V, 300 mΩ), NV6125 (650/750 V, 175 mΩ), and NV6127 (650/750 V, 125 mΩ).

“The 6 × 8-mm range with advanced cooling pad is offered at the same price as the existing 5 × 6-mm GaNFast range, and in some cases, the new low temperature may enable the designer to substitute a smaller-die version to reduce system costs further,” said Navitas.

GaNFast IC enables the smallest fast chargers to achieve power densities as high as 1 W/cc at 65 W and 1.25 W/cc at 300 W, far beyond any other discrete GaN or Si solutions, said the company.

“The NV612x-series delivers a cool 10°C to 15°C reduction in temperature with an enlarged thermal interface to the PCB and a direct thermal and electrical connection to the system ground, enabling the world’s highest power density and passing all thermal specifications and agency approvals,” said Dan Kinzer, CTO/COO and co-founder of Navitas.

Seoul Semiconductor Co., Ltd.
Optoelectronics: WICOP UHL LEDs

Designed for EV headlamps, Seoul Semiconductor Co., Ltd. released its next generation of wafer-integrated-chip-on-PCB (WICOP) ultra-high–luminance (UHL) series LEDs that reduce headlamp power consumption by up to 20%, with a 40% improvement in heat dissipation compared with other LED products, according to the company.

“Battery power consumption is one of the important requirements that determines the driving distance on a single charge, making it critical to reduce the overall weight of the automotive components to reduce power consumption,” said the company.

By using the WICOP UHL technology in the headlights, the weight of the lamp’s heat-sink structure can be reduced by 75%, said Seoul. In addition, the LED-emitting area of the product is extremely small, measuring approximately 0.5 mm², which enables a slim headlamp design.

The WICOP products are the industry’s first patented package-less LED technology developed by Seoul Semiconductor. “Unlike flip-chip technology that must be bonded in the semiconductor process, WICOP LEDs can be easily surface-mounted in the general substrate bonding process,” said the company.

Texas Instruments Inc.
Sensors: TMP61 Linear Thermistor Family

Claiming the industry’s smallest silicon-based linear thermistors (with traditional passive component 0402 and 0603 package options) and the industry’s first to eliminate the cost/performance tradeoff associated with negative-temperature–coefficient (NTC) thermistors, TI’s TMP61 thermistor family provides the accuracy of a linear positive-temperature–coefficient (PTC) thermistor at a similar size and cost of an NTC thermistor.

“This enables their operation closer to the thermal limits of other board-level components, which helps engineers maximize system performance, simplify design, and reduce total system cost in industrial, automotive, and consumer applications in which precise, real-time temperature readings are fundamental to system performance and protection,” said TI.

NTCs dominate the market because they’re smaller and cost-effective, but they come with hidden costs — most notably, degraded performance at temperature extremes, especially above 80°C, said TI. “As a result, they often require complex calibration, which increases design time, or additional components, which increases total solution cost.”

TI’s solution to this cost/performance tradeoff is the TMP61 family, which includes the TMP61, TMP63, and TMP64, similarly sized and priced as NTCs while providing higher system performance, lower component cost, and faster design.

Here’s why: The TMP61 family delivers up to 50% higher accuracy and one-third the response time of NTCs, which enables the devices to operate closer to the thermal limits of other components and the overall system. “This helps engineers maximize system performance, reduce total solution size and bill of materials, and simplify design while delivering reliable, highly accurate thermal measurements across a wide temperature range [–40°C to 170°C],” said TI.

The linearity and high accuracy of the TMP61 family enables less than ±0.5°C accuracy with a single-point room-temperature offset, which maximizes system performance, while the very low typical drift of 0.5% improves the overall reliability of temperature measurements, enabling designers to boost system performance while maintaining safe operation.

In terms of size, the TMP61, TMP63, and TMP64 are available in 0402 and 0603 package sizes, which are one-tenth the size of similar linear PTCs. They can reduce PCB layout size by at least 33% compared with NTCs by eliminating the need for additional linearization circuitry or redundant NTCs, said TI.