I returned from draft duty in early ’54, and from that time on I’ve been absorbed with high fidelity audio. Of course, the early years meant embracing vacuum tubes. Transistors weren’t ready for prime time, so tubes were the only option. And I soon learned that tube technology was far from perfect. Tubes suffered high incidence of failure, and their abundant heat cooked adjacent parts. But those faults could become my gain if I learned radio/TV repair, so I built (from kits) a tube tester, audio oscillator, oscilloscope, bought a multimeter, and began my career in the industry.
Tubes reflect their Neo-Victorian vintage (1904); they’re just not high precision parts. Why not? Well, to start, the tube manufacturers identify vacuum tube operating parameters only by listing “average” or “typical” characteristics. They never specify tubes by providing precise min./max. limits (as with solid-state devices), so tubes lack uniformity from the git-go. That’s why tubes of the same type often differ so widely. Further, all tubes exhibit random long term drift when put into service; plate current falls, grid bias shifts. These changes reflect a persistent degradation that begins at initial turn-on and ends in cathode depletion failure—barring other modes of premature demise (e.g. open filament, vacuum leaks, gassing, microphonics, atypical distortion, hum/noise). So vacuum tubes are not a wise choice when predictable, stable circuit performance is a serious design goal. Regardless, for some 70 years tubes were all that we had. Creativity got stale toward the end of that era. Tubes were just too big* (and too inefficient) to use more than the functional minimum. Innovation later revived with the debut of fully-complementary solid state technology.
Early angst: In 1963 I bought a “hi-end” Fisher FM-200B tuner, one of the top FM signal sources of the day, but its RF/IF stages exhibited incessant drift due to tube aging. I had to perform very tedious realignments annually. And my 1962 Marantz 8B stereo power amplifier needed quarterly output stage re-biasing to keep the measured IM distortion inside 0.5%, plus I had to install four new EL34s every two years. (CVA, in FL, currently resells Chinese EL34 tubes at $55 per matched pair.) Indeed, I got so anxious to dump vacuum tubes that I built my own solid state power amps in the mid-’70s, just as soon as PNP silicon power transistors became affordable. Free at last!
Vacuum tube commerce has collapsed in the 40+ year lapse since my escape. All of the principal domestic, British, Dutch, and German producers are now either defunct (like Tung-Sol Electric, my employer from ’57 - ’60), or they’ve long since ceased making tubes. The entire world market for (receiving-type) tubes is now confined to a small coterie of audio and guitar buffs, and served only by obscure Russian and Chinese suppliers with no previous market recognition. (There are other minor sources in former Soviet bloc countries; also, perhaps, one in Canada.) The quality and reliability of the tubes made by those arcane foreign suppliers is a subject worthy of concern. And those sources will persist only as long as there’s viable demand, so the outlook for assured access to replacement stock seems dicey. Further, this situation prevails at a time when every instrumented means of evaluating audio quality validates the measurable superiority of modern solid state design. Tube boosters reply that “my ears are more accurate than your instruments”, but their faith is mired in groupthink. There’s no credible A/B/X aural evidence to support the “tubes sound better” cult. Tubes were marching to the casket 40 years ago. Don’t consort with zombies.
*A 12AX7 dual triode tube contains two digital gates. A smart ’phone utilizes > 8 million gates.