O'Brien, I've never been responsible for "developing" HiPower Amps. In fact, all my life I found Analog, and specifically RFMW field of EE, quite boring and "not my cup of tea", and design engineers better do something in life they're excellent at, not something they're "drawn into" by a series of events/circumstances. However I became involved in AnalogRFMW & much amplifiers in other ways. My old roots were in theoretical Physics, Chemsitry, some fascination with Astronomy but that died early, then Electronics and really hard-on switch to Digital by mid/late 1980's when it was all the rage & my classmate "converted" me into his religion. That classmate went on to become Vice President of a major transAtlantic Software firm, serving Euro-Canada/U.S.-Russia, all he did was to show me a D-flipflop in a CMOS IC used in a touch-control door bell, and I was hooked! I though, instead of a nunch of transistors sensing exact analog voltages, here's a D-flipflop that holds (registers) a level that's all we need! It was asynchornous (no clock) though, back then. Bulding computers from components, not like today even a monkey can buy a few parts like motherboard and plug one onto another, but I mean down to component level - create REALLY COMPLEX PCBoards, install/solder all components, test, tune, modify Firmware, add Software which was a hard task in those years. Programmable calculators. We were "tech mafia" in 1980's schools, guys asked me to fix their tape recorder (multiple), radio, audio amps, though not digital it was electronics, but I really disliked the insides of my high-end "All-Wave" receiver bought by father at a huge premium (he was a dept. head & engineer so he could afford) - i discovered a bunch of coils inside, tuned elements, weird unpredictable behavior when I experimented with antennas, Gosh... only a FEW components make a HUGE difference, unlike digital a HUGE number of components with fairly predictable behavior. I hated AnalogRF already then. Nowadays there's no border between high-speed digital & AnalogRF, clock frequencies risen so high - EVERYTHING now requires a Analog RFMicrowave undersrtanding, everything is a transmission line, not a lumped conductor, OK where's my tranquilizer pill... I hate Vias, especially burried Vias running 4GHz clock !!! This isn't digital, nor is it Analog, it's sheer Microwaves, the worlds have converged and you better worship Maxwell, like it or not.
Anyways, I became deeply involved in AnalogRFMW since a freak accident (hiring) in year 2005, after drinking cool aid, appearing at a job interview at some VERY very prominent RFMicrowave company, and smashign their expectations so I got hired same day as interview. Their principal engineer was foaming at mouth - "ohh what a deal, we get this guy and only for $50K"
Since than amusingly, surprisingly to myself, I became part of exactly what I wasn't supposed to be - Analog RF/MW. S-parameters? WHAT? I knew X, Y, Z parameters, obviously well-versed in Fourier & Laplacian transforms, Z-transforms in discrete signal DSP, whatever, but umm "S-Parameters" and to worry about reflected waves, hello? Why such complications? If I apply 500Hz triangle signal Iam supposed to get... a 500Hz triangle at the other end. Gosh, these people threw a very thick classic book on Microwave frequency theory, and forced me to study it daily and report, they forced me under a microscope, and even to wirebond in case assemblers fail. That book was sheer evil, it revolted my previous education, it made me vomit but I had to eat it. Over time appettite came during th emeal, i started to bloody love that book, I thought this is it. God wants me to be what i didn't expct to become. And it's betetr than a Murderer!
Here's another fact - in schools & colleges I got terrible grades in ElectroMagnetics once we left realm of theoretical physics, and yet i got perfect, like best in classes grades in Switching/Logic systems, Digital circuits, ALU's & Microprocessors, silicon/chip-level design, anyways...
I never developed HiPower Amplifiers from the scratch, I "created" them, new models, new perfomance, but outside Amplitech and entire different frequencies, and to best summarize it:
On SYSTEM-LEVEL. I don't like that stuff down to components-level, if I should go down into component-level, I'd do digital controls like Micros/MCU's, Firmware.
I frankly could care less about what's inside RFHIC other than my cinviction it's mostly class A-B (mixed) with a push-pull pair at the very output stage, devices are GaN's orig. CREE, but then becamse Qorvo, Wolfspeed, american components used in Korea.
NOW GO SLEEP !!!
