Where is GaN heading?
Article By : Steve Taranovich
The Efficient Power Conversion team is leading the GaN power element development toward the creation of a growing set of industry apps.
While visiting Efficient Power Conversion’s booth at APEC 2017, I became keenly aware of what CEO and co-founder, Alex Lidow has been leading up to when he hired an immensely talented technical team initially in 2007 and added more talent along the journey to 2017 which would lead GaN power element development toward a process perfection effort and on through to the creation of a growing set of high profile industry applications to performance heights that the power industry has never seen.
Figure 1: Some key members of the EPC team at APEC 2017 including (left-to-right) Joe Engle, Director of Marcom; Alex Lidow, PhD, CEO and co-founder; Dr. Yuanzhe Zhang, director, Applications Engineering; and Michael de Rooij, PhD, vice president, Applications Engineering posed with EDN/Planet Analog editor Steve Taranovich.
LIDAR sees the light to automotive safety with eGaN
John Glaser, PhD, was instrumental in developing the eGaN-enhanced LIDAR system which led EPC to designs in most every automotive LIDAR application in the industry today. Glaser told me that he learned much about the intricacies of LIDAR design to successfully and reliably perform for the safety needs of the demanding automotive sector.
In this video, John Glaser, PH.D, discusses EPC’s LiDAR design using pulsed lasers which quickly create a three-dimensional image/map of any surrounding area with the need for speed and accuracy of the laser and the capability of eGaN FETs to enable ultra-short transition times of < 1ns. This technology provides increased accuracy in applications such as autonomous vehicles and augmented reality systems.
Glaser explained to me that the EPC9126 demo board was the culmination of his development efforts to create the best LASER driver system for the automotive industry as well as for other LASER applications.
Figure 2: The EPC9126, a 100V, high current, pulsed LASER diode driver demo board highlighting the EPC2016C eGaN power FET capable of 75A pulses with a total pulse-width of 5ns (10% of peak). The eGaN FET is driven by a Texas Instruments UCC27611 gate driver. This board has multiple options for mounting laser diodes and can drive these via a discharging capacitor (as shipped) or directly from a power bus.
The EPC9126 demo board was designed to operate as a capacitive discharge laser diode driver and comes with the EPC2016C eGaN FET and a commercially available gate driver, the Texas Instruments UCC27611, although it can accommodate the much larger EPC2001C, a 100V, 7mΩ FET with a 150A pulse current rating
Glaser commented on the two different techniques for driving Laser diodes. In a LIDAR application, the primary factor is device speed, which is determined by gate resistance, gate charge, and gate and source inductance.
Figure 3: Glaser told me about the two most common laser diode driver circuits for pulsed LIDAR applications: (a) A capacitive discharge driver and (b) an FET-controlled driver. (Source: IEEE Power Electronics Magazine)
Simplicity and the ability to accept stray inductance make the Capacitive Discharge approach popular.
In the FET-controlled driver approach, the circuit must switch slowly enough that stray parasitic inductance does not significantly degrade the waveforms. Considering parasitic inductance, this means rise and fall times as much as an order of magnitude slower than the capacitive discharge approach. This will not work well in a LIDAR system.
Glaser chose the capacitive discharge driver technique in EPC’s demo.
So what does eGaN do for LIDAR?:
- Faster switching for more accurate and smaller pulses
- Higher efficiency using a higher pulse repetition rate
- Smaller footprint which enables higher power density, lower inductance, integration with Laser Diode
Figure 4: The LIDAR pulse can be seen here in the demo booth. Fast edge rates in the sub-nanosecond region in LIDAR translate to more accurate measurements of the surrounding area and how close objects are to the vehicle.
Next: Why Gan is the next step in power management growth »