-
The influence of artificial intelligence (AI) is driving up the energy demand of data centers across the globe. This growing demand underscores the need for efficient and reliable energy supply for servers. Infineon Technologies AG opens a new chapter in the energy supply domain for AI systems and unveils a roadmap of energy efficient power supply units (PSU) specifically designed to address the current and future energy need of AI data centers.
By introducing unprecedented PSU performance classes, Infineon enables cloud data center and AI server operators to reduce their energy consumption for system cooling. The innovative PSUs reduce power consumption and CO 2 emissions, resulting in lower lifetime operating costs. The powerful PSUs are not only used in future data centers but can also replace existing power supply units in servers and increase efficiency.
In addition to the current PSUs with an output of 3 kW and 3.3 kW available today, the new 8 kW and 12 kW PSUs will contribute to further increasing energy efficiency in future AI data centers. With the 12 kW reference board, Infineon will offer the world’s first power supply unit that achieves this level of performance and supplies future data centers with power.
“At Infineon, we power AI. We are addressing a critical question of our era – how to efficiently meet the escalating energy demands of data centers,” says Adam White, Division President Power & Sensor Systems at Infineon. “It’s a development that was only possible by Infineon’s expertise in integrating the three semiconductor materials silicon (Si), silicon carbide (SiC), and gallium nitride (GaN) into a single module. Our PSU portfolio is therefore not only an example of Infineon’s innovative strength, which leads to first-class results in terms of performance, efficiency and reliability for data centers and the AI ecosystem. It also reinforces Infineon’s market leadership in power semiconductors.”
Infineon is responding to the requirements of data center operators for higher system efficiency and lower downtimes. The growth of server and data center applications has led to an increase in power requirements, necessitating the development of power supplies with higher power ratings from 800 W up to 5.5 kW and beyond. This increase is driven by the growing power requirements of Graphic Process Units (GPU) on which AI applications are computed.
High-level GPUs now require up to 1 kW per chip reaching 2 kW and beyond by the end of the decade. This will lead to higher overall energy demand for data centers. Depending on the scenario, data centers will account for up to seven percent of global electricity consumption by 2030; this is an order of magnitude comparable to India’s current electricity consumption.
Infineon’s new PSUs contribute to the efforts to limit the CO 2 footprint of AI data centers despite the rapidly growing energy requirements. This is made possible by a particularly high level of efficiency that minimizes power losses. Infineon’s new generation PSUs achieve an efficiency of 97.5 percent and meet the most stringent performance requirements. The new 8 kW PSU is capable of supporting AI racks with an output of up to 300 kW and more. Efficiency and power density is increased to 100 watts per in³ compared to 32 W/in³ in the available 3 kW PSU, providing further benefits for the system size and cost savings for operators.
From a technical perspective, this is made possible by the unique combination of the three semiconductor materials Si, SiC and GaN. These technologies contribute to the sustainability and reliability of AI server and data center systems. Innovative semiconductors based on wide-bandgap materials such as SiC and GaN are the key to a conscious and efficient use of energy to drive decarbonization.
The 8 kW Power Supply Unit will be available in Q1 2025. For more information about the PSU roadmap, please click here.
Infineon at the PCIM Europe 2024
PCIM Europe will take place in Nuremberg, Germany, from 11 to 13 June 2024. Infineon will present its products and solutions for decarbonization and digitalization in hall 7, booths #470 and #169. Company representatives will also be giving several presentations at the accompanying PCIM Conference and Forums, followed by discussions with the speakers. Information about Infineon’s PCIM 2024 show highlights is available at www.infineon.com/pcim.
Original – Infineon Technologies
-
PANJIT introduced its latest 60, 100, and 150V AEC-Q101 qualified MOSFETs, engineered with advanced trench technology to set new standards in performance and efficiency. Designed for both automotive and industrial power systems, these MOSFETs offer unparalleled figure of merit (FOM), significantly lower RDS(ON), and reduced capacitance. This ensures minimal conduction and switching losses, resulting in enhanced overall electrical performance.
The new MOSFET series is available in various packages, including DFN3333-8L, DFN5060-8L, DFN5060B-8L, TO-252AA and TO-220AB-L. These compact packages facilitate efficient design solutions for modern electronic systems. With an operating junction temperature of up to 175°C, these MOSFETs are robust and reliable, further evidenced by their AEC-Q101 qualification.
These MOSFETs are ideal for various automotive applications, including wireless charging transmitters, battery management systems, front and rear lighting systems, DC/DC converters, infotainment systems and more. Their low on-resistance and high efficiency enhance the performance and reliability of these systems. Additionally, their versatility extends to industrial power systems, broadening their range of applicability and utility.
PANJIT’s new automotive-grade MOSFET series delivers superior performance, reliability, and efficiency. These MOSFETs are set to become a cornerstone in the design of next-generation automotive and industrial systems.
Original – PANJIT International
-
Power Integrations announced the launch of the SCALE-iFlex™ XLT family of dual-channel plug-and-play gate drivers for operation of single LV100 (Mitsubishi), XHP™ 2 (Infineon), HPnC (Fuji Electric) and equivalent semiconductor modules up to 2300 V blocking voltage for wind, energy storage and solar renewable energy installations.
This single-board driver enables active thermal management of inverter modules for improved system utilization and reduces the bill-of-material count for increased reliability.
Thorsten Schmidt, product marketing manager at Power Integrations commented: “It’s a real challenge to build a single-board gate driver for these ‘new dual’ style IGBT modules. Our compact new SCALE-iFlex XLT gate drivers fit inside the outline of the module, allowing the drivers to be mounted on the module, which gives converter system designers a high degree of mechanical design freedom.”
SCALE-iFlex XLT dual-channel gate drivers feature Negative Temperature Coefficient (NTC) data reporting – an isolated temperature measurement of the power module – which allows accurate thermal management of converter systems. This enables system designers to optimize thermal design and obtain a 25 to 30 percent converter power increase from the same hardware.
The isolated NTC readout also reduces hardware complexity, eliminating multiple cables, connectors and additional isolation barrier crossing circuits. The new gate drivers employ Power Integrations’ SCALE-2 chip set which minimizes component count, enhancing reliability. The gate driver board also protects the power switches in the event of a short-circuit.
Original – Power Integrations
-
Nexperia announced that it is now offering its industry leading 1200 V silicon carbide (SiC) MOSFETs in D2PAK-7 surface mount device (SMD) packaging, with a choice of 30, 40, 60, and 80 mΩ RDSon values. This announcement follows on from Nexperia’s late-2023 release of two discrete SiC MOSFETs in 3 and 4-pin TO-247 packaging and is the latest offering in a series which will see its SiC MOSFET portfolio swiftly expand to include devices with RDSon values of 17, 30, 40, 60 and 80 mΩ in flexible package options.
With the release of the NSF0xx120D7A0, Nexperia is addressing the growing market demand for high performance SiC switches in SMD packages like D2PAK-7, which is becoming increasingly popular in various industrial applications including electric vehicle (EV) charging (charge pile, offboard charging), uninterruptible power supplies (UPS) and inverters for solar and energy storage systems (ESS).
It is also further testimony to Nexperia’s successful strategic partnership with Mitsubishi Electric Corporation (MELCO), which has seen the two companies join forces to push the energy efficiency and electrical performance of SiC wide bandgap semiconductors to the next level, while additionally future-proofing production capacity for this technology in response to ever growing market demand.
RDSon is a critical performance parameter for SiC MOSFETs because it impacts conduction power losses. However, many manufacturers concentrate on the nominal value, neglecting the fact that it can increase by more than 100% as device operating temperatures rise, resulting in considerable conduction losses.
Nexperia identified this as a limiting factor in the performance of many currently available SiC devices and leveraged the features of its innovative process technology to ensure that its new SiC MOSFETs offer industry-leading temperature stability, with the nominal value of RDSon increasing by only 38% over an operating temperature range from 25 °C to 175 °C.
Tightest threshold voltage, VGS(th) specification, allows these discrete MOSFETs to offer balanced current-carrying performance when connected in parallel. Furthermore, low body diode forward voltage (VSD) is a parameter which increases device robustness and efficiency, while also relaxing the dead-time requirement during freewheeling operation.
Original – Nexperia
-
Qromis, Inc. brings a revolutionary, high performance substrate solution for large-diameter GaN epi and device wafers. Rather than solving incremental scale and performance challenges of conventional substrates for WBG semiconductors, QST™ (Qromis Substrate Technology) has been developed and validated as a disruptive solution, with cross-functional optimization.
In this approach, the key properties of substrate material such as stress, lattice mismatch, thermal stability and shape control are engineered independently for the best fit with GaN epitaxial and device layers, as well as with different device architectures and performance targets. QST™ materials layers are integrated together in the conventional semiconductor fab with a simple manufacturing flow.
The resulting large diameter substrates (6-inch, 8-inch, 12-inch or beyond) are capable of supporting from a few to tens of microns of high-quality GaN epitaxy, from which unique, high performance and low cost GaN device structures are fabricated.Qromis is designing reference devices for device manufacturers to demonstrate the benefits of the groundbreaking QST™ technology. These devices exploit key advantages of QST™ over other substrate solutions:
– High quality and stress-free GaN,
– Ability to deposit thick, bulk-like GaN epi layers on large diameter substrates,
– Low defectivity, eliminating a major limitation to performance and scalability.
With these distinct features, designers can now break free from the current technological limitations of GaN process technologies and develop advanced device architectures for lateral and vertical device designs of light emitters, switches, rectifiers, as well as monolithic and integrated circuits.Qromis, Inc., established in March 2015 and located in Silicon Valley, California, is a privately held fabless technology innovator focusing on energy efficient and high performance wide bandgap (WBG) semiconductor materials and device solutions which dramatically reduce global energy use and consumption. The Company is poised to become one of the premier players in the rapidly growing, multi-billion dollar WBG industry with its disruptive and validated solutions in substrate and device technologies, which enable an unmatched cost, performance, and application scale. Markets served include power electronics, light emitting diodes (LEDs), advanced displays, and RF electronics, and other emerging high performance and energy efficient applications.
As a fabless company, Qromis is driving the commercialization of its unique solutions in conjunction with its foundry network. The Company and its foundry partner offer WBG device foundry services, as well as access to advanced and novel process technology platforms, for its customers and partners.To learn more about Qromis and its substrate technology, join the International Conference on Compound Semiconductor Manufacturing Technology which takes place May 20-23, 2024 in Tucson, Arizona.
Original – Qromis
-
DISCO Corporation has developed a process for diamond wafer manufacturing that applies the KABRA process, an ingot slicing method using laser technology. This process contributes to increasing the diameter of diamond wafers.
Diamond is often called the “ultimate semiconductor material” due to its superior material characteristics compared to Si, SiC, and GaN. In particular, as diamond has excellent insulation strength and heat conductivity, it is expected to be a good material for power semiconductors, and device development is underway at various research institutes.
On the other hand, as it is an extraordinarily hard material, it came to be known that it is difficult to process mechanically. Therefore, laser became the general method used to slice wafers from a diamond crystalline ingot. However, with the conventional laser slicing method, as the ingot is processed from the sidewall, there was a limitation on the ingot diameter, and it was difficult to increase it.
With KABRA slicing, there is no limitation on the ingot diameter as the laser is irradiated from the upper surface of the ingot, making it possible to increase the ingot diameter.
Figure 1: Conventional slicing method using laser
Figure 2: KABRA for diamond processing
Features:
- Supports large-diameter ingots exceeding Φ50 mm
With the conventional method where the laser is irradiated from the sidewall of an ingot (Figure 1), the largest supported diameter was approx. Φ30 mm, but with the KABRA method (Figure 2), as the laser is irradiated from the upper surface of the ingot, there is no limitation on the ingot diameter. - Splitting wafers with a thickness of 100 µm or less is possible
By irradiating the laser at a shallow depth from the upper surface of the ingot, it is possible to split thin wafers. Therefore, a larger number of split wafers can be expected compared to the conventional method. - Achieves equivalent or higher throughput compared to the conventional method
The splitting speed is equal to or faster than the conventional method, achieving lower processing costs.
In addition, 45 patents related to this technology have been registered. Test cuts are available. Contact a DISCO sales representative.
Original – DISCO
- Supports large-diameter ingots exceeding Φ50 mm
-
Taiwan Semiconductor – a global supplier of discrete power electronics devices, LED drivers, analog ICs and ESD protection devices – announced a family of 650V silicon carbide Schottky barrier diodes which are suitable for high-efficiency AC-DC, DC-DC and DC-AC conversion applications.
Unlike silicon-based fast-recovery rectifiers, these SiC devices have negligible switching losses due to low capacitive charge (QC). This makes them suitable for high-speed switching applications, benefitting circuit designs with increased power density and can reduce overall solution size.
Key Features
- Max. junction temperature 175°C
- High-speed switching
- High frequency operation
- Positive temperature coefficient on VF
- SPICE Models available
- Thermal Models available
Applications
- AD-DC conversion – PFC Boost
- DC-DC, Solar inverters
- Data center and server power
- Telecom – Datacom power
- UPS systems
Circuit Functions
- PFC boost diode
- Free-wheeling diode
- Full wave bridge
- Vienna bridgeless circuit
Original – Taiwan Semiconductor
-
As the “heart” of charging stations, the performance and reliability of charging modules are undeniably crucial. The popularization of the “super charging” concept has made long cruising range and short charging time become the selling points of more and more electric vehicles on the market.
Recently, the newly released Chinese Standard GB/T20234-2023, which focuses on Part 4: High-Power DC Charging Interfaces for Electric Vehicle Conductive Charging Couplers, has been significantly revised to expand the voltage range to 1500Vdc and the corresponding current range to 1000A. Additionally, the standard now includes new provisions related to liquid-cooled charging stations.
This means that in the future, higher-power megawatt-level charging stations will gradually become a reality. As long as your electric vehicle supports supercharging, it will be as convenient as refueling at a gas station.
Charging Module Technology Analysis
In fact, WeEn Semiconductors has long focused its business on the “low-carbon” track, while acknowledging that charging stations, as crucial supporting infrastructure, will evolve towards directions of higher power, greater efficiency, full liquid cooling, and comprehensive supercharging capabilities.
WeEn’s latest research and development effort, the BYC100MW-600PT2, will enable customers to achieve designs for 40kW+ high-power, high-efficiency charging modules. The WND60P20W will offer customers a higher voltage design margin to meet the demands of more complex and challenging application scenarios, thereby supporting the rapid development of new energy vehicles and the achievement of low-carbon objectives.
A charging module is essentially a power electronics converter that converts alternating current (AC) from power grid into direct current (DC) that can be stored in the battery of an electric vehicle.
Charging module converters typically have a two-stage topology. The first stage is usually a three-phase Power Factor Correction (PFC), most often using the Vienna PFC topology. Its main function is to convert AC to DC and to correct the power factor.
Figure 1: Vienna PFC Topology Architecture
The second stage typically involves a DC-DC conversion, most often using the high-efficiency LLC topology. This stage primarily converts the high voltage DC output from the PFC (800Vdc) into a wide range of adjustable DC voltages from 200Vdc to 1000Vdc, to match the needs of different battery voltage levels. Additionally, the DC-DC stage also achieves electrical isolation from the power grid through a high-frequency transformer.
Since the current charging modules are primarily used for delivering power to electric vehicles, the output rectification in the DC-DC stage commonly employs Fast Recovery Diodes (FRD). Benefiting from the negative temperature coefficient characteristics of Fast Recovery Diodes (FRD), and given that the LLC topology generally does not require stringent reverse recovery performance, FRDs are particularly suitable for use in charging modules that operate under high temperature and high current conditions.
Benefiting from the negative temperature coefficient characteristics of Fast Recovery Diodes (FRD), and given that the LLC topology generally does not require stringent reverse recovery performance, FRDs are particularly suitable for use in charging modules that operate under high temperature and high current conditions.
Figure 2: LLC DC- DC Topology Architecture
Currently, the mainstream power ratings for charging modules on the market range from 20kW to 40kW. Superchargers typically operate by outputting through several charging modules connected in parallel. Therefore, to ensure that each module is not affected by others during startup, Oring diodes are essential. When functioning normally, these diodes are in a conducting state, primarily incurring conduction losses. Consequently, standard rectifier diodes with low forward voltage (VF) are the best choice.
WeEn’s Professional Solutions
WeEn semiconductors, including the BYC75W-600P for LLC secondary-side rectification and the Oring diode WND60P16W, have been mass-produced reliably for years in leading charging module manufacturers’ 30kW models. With the recent trend towards higher power in charging modules and the need to accommodate for the harsh operating environments of charging stations, we have responded to our customers’ actual needs by launching the BYC100MW-600PT2 for 40kW charging modules and the higher voltage-resistant WND60P20W, helping our customers solve practical application issues.
#BYC100MW-600PT2 Features:
- Maximum current up to 100A
- Extremely low reverse leakage current
- Optimal VF-QRR trade-off performance
- Robust Eas capability
#WND60P20W Features:
- Maximum reverse voltage up to 2000 Vdc
- Extremely low forward conduction voltage drop
- Enhanced forward surge current capability
- Robust Eas capability
By comparing the specifications of the BYC100MW-600PT2 and BYC75W-600PT2, we find that the BYC100MW-600PT2 offers significant improvements in forward voltage (VF) while maintaining the same reverse recovery charge. As a result, it is more suitable for applications in 40kW high-power charging modules. Customers using the 40kW modules have observed an actual temperature rise reduction of 8°C to 10°C, which substantially enhances the thermal design of the system.
Figure 3: BYC100MW-600PT2 VF Curve
Figure 4: BYC100MW-600PT2 Qrr Curve
In the context of charging station applications, considering that there is quite a distance from the output of the charging module to the high-voltage power battery, potentially up to 30-40 meters, it is important to note that at the moment the charging module begins outputting, stray inductance in the charging cable and capacitors within the system will oscillate. This causes the diode to endure a spike in reverse voltage. If the voltage exceeds the diode’s avalanche voltage, it will cause avalanche breakdown; if the diode’s avalanche energy is insufficient, it will be damaged.
The WND60P20W is an enhancement of the existing WND60P16W product from WeEn Semiconductors, with the reverse withstand voltage increased to 2000Vdc while also improving its capability to withstand avalanche breakdown. The WND60P20W can meet the increasingly complex and harsh working environments of charging modules, providing greater safety margins for customer module designs.
Figure 5: Voltage Oscillation Across Oring Diode
Original – WeEn Semiconductors
-
Littelfuse, Inc. announced the launch of the IX4352NE Low-side SiC MOSFET and IGBT Gate Driver. This innovative driver is specifically designed to drive Silicon Carbide (SiC) MOSFETs and high-power Insulated Gate Bipolar Transistors (IGBTs) in industrial applications.
The key differentiator of the IX4352NE lies in its separate 9 A source and sink outputs, which enable tailored turn-on and turn-off timing while minimizing switching losses. An internal negative charge regulator also provides a user-selectable negative gate drive bias for improved dV/dt immunity and faster turn-off. With an operating voltage range (VDD – VSS) of up to 35 V, this driver offers exceptional flexibility and performance.
One of the standout features of the IX4352NE is its internal negative charge pump regulator, which eliminates the need for an external auxiliary power supply or DC/DC converter. This feature is particularly valuable for turning off SiC MOSFETs, saving valuable space typically required for external logic level translator circuitry. The logic input’s compatibility with standard TTL or CMOS logic levels further enhances space-saving capabilities.
The IX4352NE is ideally suited for driving SiC MOSFETs in various industrial applications such as:
- on-board and off-board chargers,
- Power Factor Correction (PFC),
- DC/DC converters,
- motor controllers, and
- industrial power inverters.
It’s superior performance makes it ideal for demanding power electronics applications in the electric vehicle, industrial, alternate energy, smart home, and building automation markets.
With its comprehensive features, the IX4352NE simplifies circuit design and offers a higher level of integration. Built-in protection features such as desaturation detection (DESAT) with soft shutdown sink driver, Under Voltage Lockout (UVLO), and thermal shutdown (TSD) ensure the protection of the power device and the gate driver. The integrated open-drain FAULT output signals a fault condition to the microcontroller, enhancing safety and reliability. Furthermore, the IX4352NE saves valuable PCB space and increases circuit density, contributing to overall system efficiency.
Notable improvements over the existing IX4351NE include:
- A safe DESAT-initiated soft turn-off.
- A thermal shutdown with high threshold accuracy.
- The charge pump’s ability to operate during thermal shutdown.
The new IX4352NE is pin-compatible, allowing for a seamless drop-in replacement in designs that specify the existing Littelfuse IX4351NE, which was released in 2020.
“The IX4352NE extends our broad range of low-side gate drivers with a new 9 A sink/source driver, simplifying the gate drive circuitry needed for SiC MOSFETs,” commented June Zhang, Product Manager, Integrated Circuits Division (SBU) at Littelfuse. “Its various built-in protection features and integrated charge pump provide an adjustable negative gate drive voltage for improved dV/dt immunity and faster turn-off. As a result, it can be used to drive any SiC MOSFET or power IGBT, whether it is a Littelfuse device or any other similar component available on the market.”
Original – Littelfuse
-
DACO Semiconductor has developed new platforms of Power MOSFET products called Standard HV Power MOSFETs and High Performance HV Power MOSFETs that cover 900V and 1200V range. Those HV MOSFETs are designed with a proprietary chip design and undergo process improvements that dramatically enhance the power handling capability and system efficiency.
High Performance HV Power MOSFETs product family integrates a faster body diode which reverse recovery time (trr) is reduced to suite phase-shift bridge motor control and uninterruptible power supply applications (UPS).
Both standard HV Power MOSFETs and High Performance HV Power MOSFETs include a wide product line and bring the benefits of enhanced performance and cost-effectiveness to key market applications in the mid-voltage range.
Example applications are offline switch-mode power supplies of all sizes, UPS and telecommunication applications. DACO Semiconductor offers a superior market standard SOT-227 and HB-9434 (34mm) packages. Others packages like discrete TO-247 and 62mm modules are under development.
Features:
- International Standard Packages
- Dynamic dv/dt Rating
- Avalanche Rated
- Fast Intrinsic Rectifier
- Low reverse recovery time trr
Applications:
- Switch-Mode and Resonant-Mode Power Supplies
- DC-DC Converters
- Battery Chargers
- UPS
- AC Motor Drives
- High Speed Power Switching Applications
Original – DACO Semiconductor
