Field Programmable Gate Array (FPGA) Power Design

Field Programmable Gate Arrays (FPGAs) are used in a variety of signal processing applications, but engineers must ensure high precision to support a wide range of hardware needs. Form factor and thermal performance are two factors to consider when designing a power supply for a modern FGPA.

Gallium nitride (GaN) is a wide-bandgap (WBG) material used to make semiconductor devices such as diodes and transistors, but with a smaller form factor and higher efficiency than silicon. GaN minimizes losses inherent in traditional silicon, so it does not require the stringent thermal management strategies of silicon. The adoption rate of GaN rises as power density requirements increase.

To ensure a high level of reliability and safety, thermal management strategies should be applied to any electronic device that generates heat. Keeping equipment temperatures within the safe operating area (SOA) is critical to ensuring reliable long-term operation; this becomes increasingly challenging as power densities increase.

Thermal management solutions are designed to maximize thermal efficiency and minimize the size, weight and cost of the equipment used. As the content of electronics in many applications increases, the problems associated with power dissipation in the form of heat inevitably multiply. Losses in a typical switching power supply fall into two categories: switching and conduction.

Thermal design should always be considered in conjunction with the overall system concept, application, and whether active or passive cooling methods can be used. First of all, it is necessary to select power discrete components with high efficiency and low power consumption. DC/DC converters are widely used in power circuits for industrial applications to provide high conversion efficiency and reduce power loss.

DC/DC Solutions

The TPS546D24A buck converter from Texas Instruments enhances thermal performance for industrial applications by delivering up to 160A of output current at an ambient temperature of 85°C (Figure 1). The TPS546D24A maximizes the power density of FPGA power supplies and enables engineers to reduce power loss by 1.5 W in high-performance data center and enterprise computing, medical applications, wireless infrastructure, and wired networking applications. According to TI, the device offers a switching frequency of 1.5MHz and a low-side MOSFET of 0.9mΩ, which is 3.5% more efficient than other solutions.

 Figure 1: Simplified schematic of the TPS546D24A (Image: Texas Instruments)

The device offers less than 1% output voltage error and pin-bundled configuration for more accurate current monitoring for fault reporting. Additionally, its integration helps eliminate up to six external compensation components on the board. The PMBus interface of the buck converter provides an optional internal compensation network, allowing engineers to reduce the overall power solution size for high-current FPGAs per application (ASIC) by more than 10%. The TPS546D24A operates 13°C cooler than recent solutions, improving operational reliability in hot and harsh environments.

"PMBus is a serial interface and open standard that is useful in many markets and applications," said Rich Nowakowski, product marketing engineer at Texas Instruments (TI). "The 40-A TPS546D24A allows configuration via pin-strapping or PMBus commands, and it reports current as a simple single address over the PMBus interface, even when multiple devices are stacked up to 160 A."

Figure 2: TPSM53604 evaluation board (Image: Texas Instruments)

At the same time, TI's TPSM53604 enables engineers to reduce the size of power solutions by 30 percent and consume 50 percent less power than earlier devices (Figure 2). Because 42% of the TPSM53604's Quad Flat No-Lead (QFN) footprint is in contact with the circuit board, the package enables more efficient heat transfer than competing ball grid array (BGA) packages. The TPSM53604 is capable of operating at high ambient temperatures (up to 105°C) to support demanding applications in industrial automation, network infrastructure, test and measurement, industrial transportation, and aerospace and defense environments.

"The TPSM53604 is the smallest 4A, 36V input power module on the market," Nowakowski said. "The device has high power density, excellent thermal performance, and low EMI [electromagnetic interference] noise. We were able to achieve these benefits by integrating inductors and other passive components into a 3D module design while addressing several design challenges. "

The TPSM53604 power module addresses thermal design issues by integrating a very efficient IC within a routable leadframe quad fan package.

Gallium Nitride Technology

In addition to improving conduction losses, the use of GaN can significantly reduce switching losses by increasing the switch turn-on speed. Increasing the switching frequency also reduces the size of many large components such as transformers, inductors, and output capacitors. GaN has better thermal conductivity and can withstand higher temperatures than silicon. Both of these attributes help reduce the need for thermal management components such as bulky heat sinks and cooling units, thereby significantly reducing the overall size and weight of the power supply.

"The faster the frequency, the less magnetic, the smaller the capacitor, the higher the density," said Steve Tom, product line manager for GaN technology at Texas Instruments. "The beauty of GaN is that we can switch faster without heat loss, which is why The reason for such high power density and efficiency."

A major concern is the reliability of the equipment. "We've done over 30 million device hours," Tom said. "As of this year, we have processed 3 GW of power conversion to show that our devices have a robust SOA. They are ideal for power supply and inverter applications."

The GaN solution family integrates high-speed gate drivers, EMI control, thermal and overcurrent protection with 100ns response time. Integrated devices provide an optimized layout to minimize parasitic inductance, maximize common-mode transient immunity (CMTI, measured in dV/dt), and reduce board space.

“One of the unique advantages we offer at TI is our GaN supply chain,” Tom said. “TI owns the GaN process and operates the entire manufacturing process from [epitaxial] to packaging and testing. Then we combine GaN FETs with optimized drivers to achieve the highest speed, highest performance, highest reliability systems.” TI Ownership complete product portfolio of 150-mΩ, 70-mΩ and 50-mΩ 600-V GaN FETs. These devices support high-density designs in industrial, telecom, server and personal electronics applications.

Figure 3: GaN applications (Image: Texas Instruments)

To expand the use of GaN in new applications such as automotive, grid-connected storage, and solar (Figure 3), TI is demonstrating its own convection-cooled, 900V, 5kW bidirectional AC/DC platform. The platform can deliver 99.2% peak efficiency without cooling fans, with a scalable multi-stage solution for 5 kW and natural convection. It includes LMG3410R050 GaN and C2000 digital controllers and supports bus voltages up to 1.4 kV.

FPGA Applications

Due to their programmable nature, FPGAs are ideal for many different markets. The following markets and applications are addressed::

Aerospace and military products 

ASIC prototyping 

Automotive 

Broadcast and Professional AV

Consumer Electronics

Data Centers

High Performance Computing and Data Storage - Industrial 

Industrial.

Medical 

Security 

Video & Image Processing

Wired Communications

Wireless Communication