GaNFet. For a primer on what's special about these fast-switching parts, we return to my AGD Vivace review whose designer was at International Rectifier now Infineon working with new semiconductor materials and emerging power conversion technologies. He worked on integrated circuits at Vishay Siliconix and acquired 10 patents, two of which are for novel power module topologies, two for GaN driver ICs and GaN power Mosfets: "Originally nobody but International Rectifier dared to start the commercialization specifically in the classical consumer and industrial markets of GaN-based power Mosfets with a simple value proposition. They wanted an ideal switch which could dramatically improve the performance of any existing silicon-based power semiconductor. We looked into different emerging and mature markets like wireless charging, Lidar, envelope tracking, power supplies, motor drives and audio where, by an order of magnitude, the GaN superiority could provide fast or immediate payback. There are no advantages but big disadvantages when GaN Mosfets appear in any classical linear circuit. GaN Mosfets are designed and optimized specifically for switching applications. Their fundamental objective is to minimize power losses and to enhance efficiency as much as feasible.

"The device I use was developed very specifically for class D audio, not for power supplies, low-voltage points of load or RF circuits. Hence the key figures of merit for the device were optimized for best performance within specific voltage ranges, frequencies and current. To simplify, devices for power supplies need voltage capabilities beyond 500-600V, with resistance in the linear region perhaps as low as 30-40mΩ. GaN used in POL apps need 2-3mΩ or lower yet voltage capabilities up to perhaps 40V, typically 30V. For RF applications, resistance can be much higher, voltage can be high too but the key parameter is frequency. That could easily hit GHz bands. For the power range of consumer and high-end audio gear—PA is another story—max voltages are perhaps 200V, resistance is in the 30-50mΩ range and frequency in the low MHz spectrum. We also want lowest possible i/o capacitance. So all these parameters specialize a given GaN part in size, package style and performance parameters such that any use beyond the limits of its intended optimization returns performance below existing tech. A producer of tires will never be able to optimize a Formula 1 racing tire from a tractor tire no matter what. Yet both are fundamentally just rubber tires."

From my AGD Vivace review, the yellow wave form shows the faster rise time of a GaN part, the pink wave an equivalent silicon Mosfet with its far shallower rise/fall times.

"About the speed at which one may switch these parts, this is no trivial issue. It directly involves the characteristics of the PWM generator (with class D usually a ΣΔ circuit), the power stage switching losses, overall parasitics and of course the output filter. Possible scenarios for less than ideal behavior are when power stage efficiency diminishes and excess switching losses manifest in excessive power dissipation hence increased device/amp temperature; when the PWM generator won't operate correctly; when a layout isn't able to transfer the switching current/voltage wave forms without introducing spikes or delays. When moved into the frequency domain, such spikes or delays inevitably mean distortion and/or sonic artifacts. The material makeup and construction of the output filter's inductor and capacitor, although reduced in size with higher switching frequencies, can add other limitations.

"On a simple conceptual basis, a higher switching frequency of the PWM ΣΔ stage is equivalent to transferring to the power stage a more defined signal. That's obvious if you imagine that stage as a 1-bit A/D converter. Now more harmonics of the original signal transfer to the output. However, if that feature is not matched by a power stage that can switch at precisely that very same frequency to deliver a transition that is perfect in rise/fall time to lack any overshoots, then perhaps so high a switching frequency is not a good choice for the actual operating conditions. So it's all about optimizing all elements of a design. With the components, layout and material I use, I can operate the power stage up to 768kHz although in most cases, 400-500kHz is perfectly fine."

Directly from GaN maker Infineon, these parts "offer fundamental advantages over silicon. In particular the higher critical electrical field makes it very attractive for power semiconductor devices with outstanding specific dynamic on-state resistance and smaller capacitances compared to silicon switches. That makes GaN great for high-speed switching."

A PDF by Efficient Power Conversion Corp. explains the science behind these lateral parts which are promoted as replacements for the ubiquitous silicon-based Mosfet. "The lateral structure of the GaN transistor makes it a very low-charge device. It can switch hundreds of volts in nanoseconds, giving it multiple megahertz capability. This will lead to smaller power converters and higher-fidelity class D amplifiers."

This link goes to an AES paper on the subject, this link to Texas Instruments.

Finally from my AGD review, "for his GaN power stage the designer claims virtually zero stray inductance and 15ns fully symmetrical rise and fall times. Even with a 3'700V/μs slew rate, the output waveform at the switching node of our power stage ahead of the LC filter is completely free of oscillation."

Unpacked, our brief tech talk calls GaNFet specialized fast switches. When properly exploited, they can uprate class D precision by raising speed and bandwidth whilst lowering distortion and dead times. Popular myths will predictably claim GaN to be a silver bullet whose mere appearance guarantees superiority over silicon-based parts; and chase ever higher switching speeds as parallel proof of better quality. So there's a new form of techno snobbery in the wings. Rather than ask how powerful your amp is, its faction asks how high it switches to call anything less than 1MHz insufficient. Hey, guys adore their numbers. It's basic school-yard braggadocio. My daddy is bigger than yours. Those who graduated from that psychology long ago know that implementation continues to trump parts choices and shiny specs. That wraps up our basic GaN story. Now we get EJ on the horn for the nitty gritty of his class D execution; and what he's done in the Aura preamp.