“More than two years ago, Texas Instruments announced its first 600V Gallium Nitride (GaN) power device. The device not only provides engineers with power density and efficiency, but also ease of design, with integrated gate drive and robust device protection. Since then, we’ve focused on using this cutting-edge technology to push the power level up (and down) as much as possible.

“

By Arianna Rajabi, Texas Instruments

More than two years ago, Texas Instruments announced its first 600V Gallium Nitride (GaN) power device. The device not only provides engineers with power density and efficiency, but also ease of design, with integrated gate drive and robust device protection. Since then, we’ve focused on using this cutting-edge technology to push the power level up (and down) as much as possible.

Gallium nitride is critical at any power level. Engineers are working to increase switching speed, efficiency and reliability while reducing size, weight and component count. Historically, you have to weigh at least some of these factors, but Texas Instruments is enabling designs with all of these advantages while saving system-level cost and reducing board component count through complex integration in a single package. From halving the size of PC adapters to creating efficient, compact 10 kW conversions for grid-tied applications, Texas Instruments has GaN solutions for your design. The LMG3410 and LMG3411 series are rated at 600 V and offer a variety of solutions from low power adapters to designs in excess of 2 kW.

Device selection by on-resistance

Internal Gallium Nitride Field Effect Transistor (FET) rated for R_DS(on) – Drain-source or on-resistance – which plays an important role in switching and conduction losses in power converters. These losses affect system-level efficiency and heat dissipation and cooling methods. Therefore, in general, R_DS(on)The lower the rating, the higher the power level that can be achieved while still maintaining high efficiency. but higher R_DS(on)Some applications or topologies may be more suitable, as shown in Figure 1.

Figure 1: 70 and 50mΩ GaN devices using typical power topologies

Overcurrent Protection

The integrated overcurrent protection not only simplifies the user’s layout and design, but high-speed detection is actually necessary for device protection under short-circuit or other fault conditions.Texas Instruments’ GaN device portfolio has

Figure 2: Internal device structure of LMG3410/LMG3411 series products,
Includes FET, internal gate drive, slew rate control and protection functions

TI’s default overcurrent protection method is classified as “current latching” protection; this means that if any overcurrent fault is detected in the device, the FET will safely shut down and remain off until the fault is reset. In our 70mΩ device, the fault triggers at 36 A; for the 50mΩ device, the fault trigger extends to 61 A.

Depending on the application, some engineers may prefer to operate under reasonable transient conditions, for which we provide cycle-by-cycle overcurrent protection. With cycle-by-cycle protection, in the event of an overcurrent fault, the FET is safely turned off and the output fault signal is cleared after the input PWM goes low. The FET can restart on the next cycle and operate under transient conditions while still preventing the device from overheating.

Table 1 shows the key specifications, structures, and typical system power levels of various types of GaN devices from Texas Instruments.

Table 1: Selection of gallium nitride by key parameters

There is no doubt that gallium nitride is leading the semiconductor race for super power switches. As TI’s GaN devices are in volume production and target a wider range of solutions, we will continue to make the technology more scalable and accessible to every member of the power industry.