When most people think about vacuum tubes, they picture big glass bottles glowing inside antique radios or early computers. History often treats tubes as a dead-end technology that was suddenly swept away by the transistor in the 1950s. But the reality is much more interesting. Vacuum tube technology did not simply stop evolving when the transistor appeared. In fact, some of the most sophisticated and technically impressive tube designs emerged after the transistor had already been invented.
During the final decades of mainstream tube development, manufacturers pushed the technology in remarkable directions. Tubes became smaller, faster, quieter, more rugged, and more specialized. Designers experimented with exotic geometries, ceramic construction, metal envelopes, ultra-high-frequency operation, and even hybrid tube-semiconductor systems. Devices such as acorn tubes, lighthouse tubes, compactrons, and nuvistors represented a last gasp of thermionic electronics.
Ironically, many of these innovations arrived just as solid-state electronics were becoming commercially practical. Vacuum tubes were improving rapidly right up until the market abandoned them.
The Pressure to Improve
By the 1930s and 1940s, vacuum tubes dominated electronics. Radios, radar systems, military communications, industrial controls, and the first digital computers all depended on them. But everyone was painfully aware of their problems.
Traditional tubes were fragile, generated heat, consumed significant power, and suffered from limitations at high frequencies. Internal lead lengths created parasitic inductance and capacitance. At radio frequencies and especially microwave frequencies, those unwanted effects made design difficult.
Military requirements during World War II accelerated development dramatically. Radar systems needed tubes capable of operating at VHF, UHF, and microwave frequencies. Vehicle equipment required devices that could withstand punishment. Computers with tubes suffered from frequent failures, took up entire rooms, and needed special cooling equipment, often bigger than the computer. These pressures drove tube designers into an intense period of innovation.
Acorn Tubes: Tiny Tubes for High Frequencies
One of the earliest major departures from conventional tube geometry was the acorn tube. Developed in the 1930s by RCA, the acorn tube got its name from its distinctive shape, which resembled an acorn with wire leads protruding from the base and sides. Unlike ordinary tubes, where the internal elements had relatively long leads, the acorn design minimized lead length to reduce parasitic capacitance and inductance. At high frequencies, this reduction was crucial.
One famous example was the 955 acorn triode. These tubes found use in experimental television receivers, military radios, and laboratory equipment. Acorn tubes also reflected an important trend in late tube development: engineers were increasingly treating tubes not merely as amplifying devices, but as microwave structures requiring careful electromagnetic design.
The Lighthouse Tube
If acorn tubes were specialized, lighthouse tubes were positively futuristic. Lighthouse tubes abandoned the classic cylindrical glass form almost entirely. Instead, they used stacked disk-like electrodes arranged in a compact coaxial structure. The resulting geometry minimized transit times and parasitic reactances, allowing operation into microwave frequencies.
The tubes vaguely resembled a lighthouse tower. These tubes became essential in radar systems during World War II and the early Cold War period. Some lighthouse designs could operate in the gigahertz range, something impossible for conventional receiving tubes.
Their construction also introduced new manufacturing techniques. Many used ceramic and metal rather than large glass envelopes. This improved heat resistance and mechanical stability while reducing losses at high frequencies.
In many ways, lighthouse tubes represented the transition from classic vacuum tubes and true microwave devices like klystrons and traveling-wave tubes.
Metal Tubes and Ruggedization
Another path of tube evolution focused on durability and compactness. Early tubes used fragile glass envelopes that were easily broken and susceptible to microphonics and vibration. During the 1930s, manufacturers introduced all-metal tube designs. These tubes replaced the glass envelope with a metal shell, improving shielding and mechanical ruggedness.
Metal tubes were particularly attractive for military and automotive applications. Shielding reduced interference, while the smaller physical size allowed more compact equipment layouts.
Hybrid glass-metal constructions also became common. Engineers experimented constantly with new materials and packaging approaches to reduce noise, improve reliability, and extend tube lifespan.
Subminiature Tubes
One of the most impressive developments was the subminiature tube. These tiny devices often looked more like oversized resistors than conventional tubes. Some were less than an inch long and designed to be soldered directly into circuits rather than plugged into sockets.
Subminiature tubes emerged largely from military demands during and after World War II. Proximity fuzes for artillery shells required electronics small enough to survive being fired from a cannon. Traditional tubes would simply shatter under the acceleration.
The resulting ruggedized miniature tubes were shock-resistant and compact enough for portable military electronics. After the war, subminiature tubes appeared in hearing aids, portable radios, test instruments, and early miniaturized computers.
The Nuvistor: The Ultimate Receiving Tube
One of the most interesting late-stage vacuum tube was the RCA Nuvistor. Introduced by RCA in 1959, the nuvistor represented an attempt to create a truly modern vacuum tube for the transistor age.
Unlike classic glass tubes, nuvistors used a compact metal-and-ceramic construction. They were extremely small, highly reliable, vibration-resistant, and capable of excellent high-frequency performance. They also exhibited very low noise characteristics. At first glance, a nuvistor hardly resembles a traditional tube at all. You could easily mistake these for some other component in a metal can.
Technically, nuvistors were excellent devices. They offered superior performance in many RF applications compared to early transistors, particularly in television tuners, instrumentation, and aerospace electronics.
High-end studio microphones also adopted nuvistors because of their low noise and desirable electrical behavior. Some audiophiles still use nuvistor-based equipment today.
But despite their capabilities, nuvistors arrived too late. Semiconductor technology was improving rapidly. Silicon transistors were becoming cheaper, more reliable, and easier to manufacture in large quantities. Integrated circuits loomed on the horizon. The nuvistor may have been the best small receiving tube ever made, but it was competing against a technology whose economics would soon become overwhelming.
Compactrons
As semiconductor electronics advanced, tube manufacturers attempted another strategy: integration. The Compactron, introduced by General Electric in the early 1960s, combined multiple tube functions into a single envelope. A compactron might contain several triodes, pentodes, or diode sections in one package. This reduced component count, simplified wiring, and lowered manufacturing costs for television sets and other consumer electronics. Of course, tubes with multiple electrodes weren’t new. They dated back to at least 1926. However, GE’s aggressive marketing of the brand was an attempt to prevent designers from defecting to the solid-state camp.
In some sense, compactrons were the vacuum tube answer to integrated circuits. Engineers were trying to achieve greater functional density while keeping tube-based designs economically competitive. GE’s Porta-Color, the first portable color television, used 13 tubes, including 10 Compactrons. They usually have 12-pin bases and an evacuation tip at the bottom of the tube rather than at the top.
Compactrons saw widespread use in televisions, stereos, and industrial electronics during the 1960s and early 1970s. But again, semiconductor integration advanced even faster. The battle was becoming impossible to win.
Specialized Tubes Survived
Even after transistors took over consumer electronics, vacuum tubes remained important in specialized fields. Microwave tubes such as klystrons, magnetrons, and traveling-wave tubes continued to dominate high-power RF applications. Radar systems, satellite communications, particle accelerators, and broadcast transmitters all relied on advanced vacuum devices. In some areas, they still do.
A modern microwave transmitter aboard a communications satellite may still use a traveling-wave tube amplifier because tubes can handle very high frequencies and power levels efficiently.
No Instant Win
One misconception about electronics history is that the transistor immediately rendered tubes obsolete after its invention at Bell Labs in 1947. That is not what happened.
Early transistors had many limitations. They were noisy, temperature-sensitive, low-power, and expensive. Tubes often outperformed them in RF circuits, audio applications, and high-power systems well into the 1960s.
For a significant period, designers genuinely did not know which technology would dominate certain markets. Tube designers were still making substantial advances. Nuvistors and Compactrons were not desperate relics; they were serious engineering efforts intended to compete in a changing world.
Ultimately, however, semiconductors possessed overwhelming long-term advantages. Transistors required less power, generated less heat, occupied less space, and could be manufactured using scalable photolithographic processes. Once integrated circuits became practical, the economics shifted decisively. Vacuum tubes could evolve, but they could not shrink into millions of devices on a silicon chip.
The final years of vacuum tube development are often overlooked because history tends to focus on winners. Yet this period produced some of the most elegant and specialized electronic devices ever created. By the late tube era, vacuum tube manufacturing had become quite refined. Engineers could produce tubes with tightly controlled characteristics and surprisingly long operating lives.
Some early transistorized devices still retained subminiature tubes in certain high-frequency or low-noise stages because transistors had not yet surpassed tube performance in every application. This overlap period is often forgotten today. Electronics did not instantly switch from tubes to semiconductors. For years, many systems used both. For many years, a typical ham radio transmitter, for example, would be all solid-state except for the power amplifier finals, which were often a pair of 6146 tubes.
You can, of course, make your own tubes. If you’ve had enough of making your own tubes, maybe try reproducing some of these advanced models.
As a teenage ham operator in the 80s, all I could afford was old tube equipment. Great memories. Learned a lot from fixing and maintaining that stuff, including the critical practice of probing with only one hand!
Roger that! I forgot just once, in 1979, and became a hand to hand path for an 800 vdc plate supply; I never forgot again
As an audio guy.. God bless the tube!!!
Great overview.
There’s also the Inductive Output Tube (IOT): High power, high efficiency, high bandwidth and reliable tubes used for digital broadcast television transmitters. https://en.wikipedia.org/wiki/Inductive_output_tube
And you can’t ignore the continuing advances in x-ray tube technology: Current tubes operate with astonishing power density: 100 kilowatts in a patch on the anode about a couple of square millimeters in size: A power density a thousand times higher than the surface of the sun. It does this trick by using a dinner-plate-size rotating anode, presenting fresh cool anode material to the electron beam at the rate of 50 meters per second, driven by a motor (and its bearings) running at over 5000 rpm in an almost perfect vacuum. And, oh, by the way, it does this while pulling 12 g rotating at 200 rpm in a computed tomography scanner.
I’m not easily impressed, usually only semi-impressed, but being both a mechanic and an electronicist, i have to say, THAT’S impressive!
Thanks, it’s nice to know what all of the whirring noises were. I just envisioned some sort of imaging sensor just casually rotating around inside that ring while I was laying there wondering which planet they had dialed up for me.
Glad to hear that it’s way more impressive and dangerous than I always thought. :-)
Was 200 rpm a typo?
That would need to be a really small yard arm.
But thanks for leading me down a rabbit hole.
Designed 100kw plus RF systems for shortwave transmitters. Guys at UHF and above are nutso.
The whole rotating platform in a clinical x-ray computed tomography scanner rotates at 60-214 rpm, depending on model and operating mode. The bigger ones have around a ton of material whipping around at that speed.
Some components at the periphery see in excess of 15 gravities of acceleration. This includes the power supply (“generator”) that converts 480 Vac 3-phase or 600 Vdc to the 120 kV at 100 kW going into the tube.
Slip rings send power into the rotating ring, and data comes off wirelessly, some running at 40 Gbps.
Mind boggling engineering when you dive into it.
When I was a young child I was given one of my grandmother’s old hearing aids to play with. It used two batteries – a large rectangular one at 45 volts for the B+ supply, and a long cylindrical 1.5V battery for the filaments. The tubes were reminiscent of an NE2 bulb, but bigger and of course with more wires sticking out. Before that, the present for my 5th birthday was an old tube AM radio, which I immediately started taking apart. (I was hooked on electricity and electronics before I could even walk reliably).
My first job out of college was with a Canadian military repair subcontractor. Most of the radar and nav equipment I worked on contained at least some tubes. I almost killed myself on one of those sub-assemblies; I forgot to turn off the 450V B+ supply before I reached in and grabbed the heat radiating fins, (AKA the “plate”) to swap the tube. Put my head on the bench and rested for a bit after that…
Tubes are not gone. Every microwave oven uses a magnetron.
As a kid, I maintained a lot of tube TV sets for family members. I still have a tube tester.
I got a TV for free because it had an intermittent problem and the shop couldn’t find it. Eventually I found the mechanical ground for the 3rd IF stage filament (no wire to it – mechanical inside the socket to the chassis) was flaky. Soldered a wire to where the pin fit to a convenient ground, and used the set for years. 18″ “portable” Zenith with Space Command ultrasonic remote control.
We had an ultrasonic remote as well. Every time the dog scratched its neck, his chain collar would jangle and change the channel
Mine had the same issue. The most accurate way to get it to change was to make a loud ‘kissing’ sound… which was hard to do as a preteen when your friends were rolling around on the floor laughing and mocking you! :D
Wow! You must be the only person I know of that also had a Zenith TV with that ancient 2-button remote, one button for the volume, one for the channels. It must have been amazing tech when it was released but by the time my parent upgraded TVs and i got the handed-down Zenith it took 5 minutes for the tubes to warm up after being turned until you could see the picture. Also only the volume button worked and random clicking sounds would change the volume!
As a musician I still use tubes in modern production amps. 12ax7’s and 6L6’s are the most popular.
A magnetron IS a vac tube.
Tell us youre clueless without saying it.
Don’t pay attention to this. Half asleep.
It’s funny how machines like ion implanters, and the mask exposure tools used in modern semiconductor manufacturing are not considered vacuum tubes, when in fact they are.
Interesting; ironic that it takes vacuum devices to manufacture their successors.
Seriously, why does HA always link Youtube videos? There’s great sites about this kind of stuff.
The first part of the article primed my lizard brain for an old guy talking about extinct animals. Then Al had to include a video (lighthouse tube) of someone who looks like Sir David Attenborough!!!!
Now I can’t unsee it.
I was originally wondering if the author had decided to use some bleeding-edge image format only Chromium based browsers currently support, but nope, even worse, all YT embeds…
Let us not forget all the wonderful tube-powered guitar amps still loved and manufactured to this day by Fender, Marshall, Mesa Boogie, etc…
And I fondly remember the old boat anchor receivers I had in the past — a National NC-57, a modded Wells-Gardner BC-348-Q aircraft receiver, the Hammurlund HQ-129-X — I still like to think of it as some sort of magic even when I know how it works.
In high radiation areas tubes will outlast solid state devices. Even rad hardened semiconductors can’t hold a candle to what tubes can endure.
Which is why the Russians last I knew, were still using tubes, in some of their commo equipment; they are still laboring under the misconception that global nuclear war us somehow survivable.
Survivors of both Hiroshima and Nagasaki are proof that nuclear war is quite survivable.
omg, why are people still saying this? I suppose if the world stuck to “atomic” weapons of 15 to 21 kilotons of TNT explosive power, but long ago the world’s powers switched to weapons that use nuclear fusion and fission (so-called “hydrogen bombs”) that are hundreds to thousands of times more powerful. Megatons rather than kilotons of TNT. This is a huge subplot in Oppenheimer, who refused to build Teller’s hydrogen bomb, and aftereards the government ruined his family’s life for it. If you don’t understand that, I don’t think you can appreciate the film.
Certainly weapons have been built -“tactical” weapons of less than 50 kilotons (which is over 3 and a third times larger than Little Boy on Hiroshima) but still, exchanging weapons of that size (remember Japan couldn’t fire back) would be very bad, and even if modern warheads of intercontinental missiles aren’t used, it could still lead to a global nuclear winter. One where nearly everyone dies, as remote from the exchange as they may be. I’d say survivors of the bombs in Japan are evidence that “nuclear war” IS NOT survivable. Even a totally one-sided use of nuclear weapons threatens everyone. If you don’t get that, you’re living in a dangerous fantasy world, and I’m totally sincere I’m not trolling or attempting one-upsmanship in a debate. Nuclear war will kill us all. Painfully. Very painfully. Everyone’s votes and political views should reflect that.
Your interpretation is not based on scientific facts. Yes blast the blast radius of weapons have increased but this does not equate to extermination of the human race. Not everywhere is a target and radiation exposure below a certain point shows no ill effects.
You may not be trolling, but FYI you are insulting to others, and very pedantic about how you are right and everyone else is delusional.
From the POV of military communications its never everyone dies – even the biggest nuke only kills everyone quickly a relatively small radius compared to the whole nation, especially if you have bunkers of some sort. Sure many of these people may die from it in a few decades or even just weeks time, but till the effects really kick in they are able to co-ordinate in the recover efforts or fight back etc.
The only way everyone really dies is in a very very very complete nuclear war that destroys the ecosystem we rely on entirely, which is a war way beyond the likely pain threshold of any nation as the oceans are so vital and won’t be directly targeted.
It would be horrible, but people are very capable of surviving horrors long enough to make new people and ensure the species survives in general as long as the planet remains remotely alive.
I see gain-of-function work on human pathogens for biodefense which is actually a back door to bioweapons research as FAR greater threat than nuclear war. The number of BSL-4 labs (the “space suit” labs for the most dangerous pathogens) in the world is rapidly increasing, leaks are not uncommon.
“The number of BSL-4 (maximum containment) labs has been increasing substantially.
According to the Global BioLabs Report 2023 (from King’s College London and partners):
~51 operational BSL-4 labs across 27 countries as of early 2023.
Additional labs under construction and ~15 planned, pushing totals toward ~69 (operational + planned/under construction).
This is roughly double the number from about a decade earlier. Growth accelerated post-2001 anthrax attacks and 2003 SARS for biodefense and preparedness reasons.
Many are in urban areas, and expansion includes countries with varying oversight. Concerns in reports focus on biorisk management gaps, proliferation, and elevated accident potential as numbers grow.”
“Laboratory Escapes and ‘Self-fulfilling prophecy’ Epidemics”
by Martin Furmanski, MD (February 2014)
Scientists’ Working Group on Chemical and Biologic Weapons at the Center for Arms Control and Nonproliferation.
https://armscontrolcenter.org/wp-content/uploads/2016/02/Escaped-Viruses-final-2-17-14-copy.pdf
War. Not a single nuclear blast.
Even a nuclear war is survivable. Targets aren’t likely to get nuked twice. Very little area is ground zero.
@ DerAxeman… that is a dangerous statement. Hiroshima and Nagasaki were 15 to 20 Kilo Ton devices. They would be considered a trigger for current day multi Mega Ton devices. When people believe they can survive nuclear war, it becomes more likely to be tried.
Modern day Global Nuclear war is NOT survivable. Those who live past the initial attacks…will ENVY the dead.
A brain the size of a planet, and I’m shaking this idiot’s chair.
Interesting and well-written, historically accurate article. I was born around the time of the invention of the transistor, and grew up during its development, and the development of sime if the more advanced tubes; my favorite equipment combines tubes and transistors, such as my Drake “C line” radios and my Tektronix 465 oscilloscope. I don’t live in the past, but some of my more modern equipment equals it, but doesn’t beat it. My Heathkit 6 meter rig uses Nuvistors, and Millen’s first grid dip.meter used a 955 acorn triode. I have a couple acorn tubes laying around, someday I’ll build something to use them in; first I’ll have to build the sockets, lol.
So why do audio enthusiasts always use the less reliable glass tubes? The same effect could be achieved with meatl+ceramic tubes, no?
i have no idea about the technical aspect but in the audio world it’s clearly aesthetic / bragging. these days, they place the tubes so you can see them from the outside. not enough to ‘sound like tubes’, has to look like them too
Well, a interesting exception is the Nu-Vista line from Musical Fidelity, which deploy nuvistor tubes out of sight. Not cheap, though.
Not true…tube powered audio amplifiers have a very distinctive sound, which is more 3 dimensional than solid state amplifiers can achieve. Digital amps are considered better than classic solid state, with much less harmonic distortion at high volume, but discerning audiophiles prefer the “warmer” characteristics of tube amps.
And to throw a completely different wrinkle into the debate? There’s another, smaller subsector of audio amplifiers that harbor what’s probably the most fanatical tube aficionados known…guitar players. Among their ranks, there’s a fierce debate about the finer points of glassy clean 6L6 vs ferociously overdriven EL34 power, not to mention pockets of 6V6 and EL84 motivated small amp fans…and don’t even get me started about the subtleties involved in preamp tube selection!
So as much as you’d like to believe it’s pure aesthetic arrogance? There’s actually an aural aspect to this…but I’m guessing you’re the sort who thinks the speakers in your phone sound pretty good, huh? 😉
My friend Phil played a part in the renaissance of tube amps for hi fi, with Mr Manley, they built the first hi fi tube amp in years, in a shed. Met an American at the Heathrow hi fi show,. Immediately moved to the US to make and sell them, early 1980s
Did you know aesthetics also apply to sound?
It’s because an overdriven tube produces even harmonics which are pleasing and musical, compared to the harsh grating harmonics of transistors.
Modern transistor amps have next to no issues with harmonics, “harsh and grating was 50 years ago or very, very cheap equipment. They are in fact so faithful that the “warm” harmonics audiophiles like can be easily added if necessary.
Why would that be the case? Do you know where that misconception comes from?
Quite funny from a tube lover who could give a damn about aesthetics!
Taste is subjective, and nobody’s taste is ever wrong.
… even if I despise it immensely :D
On the other hand, the measure for whether someone is discerning is actually quite simple. Do they agree with me? If not, they have no discernment at all, and can be safely ignored
;)
Ypu do realize that NOT driven to distortion there is zero difference between solid and hollow state, right?
Get an amplifier big enough in the first place and stop telling people driving an amplifier into distortion is somehow pleasing.
It isnt.
There’s an technical aspect, too. I once read that the amplification of a tube amp can be different compared to a solid-state counterpart.
It’s related to the signal characteristics, the phase, the type of distortion, the way a tube goes into saturation et cetera.
In simple words, the tube can create a “warmer” sounding signal depending on the circuit and the tube specs.
It’s not pure esotheric nonsense like using golden connectors for better sound, in short*.
Though in practice, someone has to be a real audiophile to hear/feel that difference, I suppose.
A laymen might not have a hearing that will notice instantly.
So yes, tube amps are perhaps often bought for reasons of prestige.
(*though gold doesn’t rust, so connections using gold-plated connectors might be more reliable. That’s why PC components used to have gold connections, too.)
Amplifier circuits using vacuum tubes, when driven to saturation (aka clipping), tend to generate even – order harmonics (2nd, 4th etc.), whereas solid state amplifier circuits tend toward stronger odd harmonics (3rd, 5th, etc.). The even – order harmonica are perceived by humans as “octave intervals” and we perceive them as more musical. This is why they are the preferred weapon I. Electric guitar amplifiers. This article fails to mention that a huge consumer of vacuum tubes in today’s world is the music industry. Audiophiles tend to love music playback amplification with vacuum tube amplifiers for the same reason.
Metal + ceramic tubes, I have no experience with, but they sound more expensive. I apologize for not directly addressing your question in the previous paragraph.
The Russell O. Hamm even order harmonics nonsense had been debunked 50 years ago. Did you stop to think why that would be the case or are you just parroting things you don’t understand?
A single-ended output stage without feedback, common in small tube amps, will tend to make asymmetrical distortion, which implies more even order harmonics than symmetrical clipping. If I recall correctly, the lowest harmonic sensed as unpleasant is the 5th, which is odd. So the hypothesis has some basis in some conditions.
In general practice, distortion is unpleasant, and a distinction between odd and even harmonic distortion is negligible.
Cost. Glass is cheap. It is also reliable if you don’t break it, or get it too hot.
4CX150 was a 150W tube using glass, the exact same mechanical design but using ceramic for the higher power 4CX250.
Oh, makes sense. Thanks!
4×150 has always had a 250W dissipation spec. Always. The CX had slightly higher seal temperature limits AND mfrs wanted to go to ceramic for those reasons and lighthouse techniques. The datasheets show 250W for both tubes.
Indeed they can. I have recently built a hi fi amplifier based on the 1950s Williamson ultralinear circuit, using metal bodied 6v6 beam tetrodes.
Glass tubes aren’t really less reliable in consumer use. The glass envelope is rarely the thing that breaks when a tube goes bad.
Metal tubes are still around but not in production anymore, ceramic tubes are very rare.
I can only speak from the perspective of a guitar player here, but the quality of tone and responsiveness to my playing is far better with tube equipment than solid-state. It’s a situation where broadly undesirable traits (particularly waveform distortion with increased gain) are embraced in a particular application. The physical string vibrates a physical magnet to send an electrical pulse to a sequence of small 12AX7 tubes that then modulate a collection of large 6L6 tubes to feed power to a speaker cabinet. All with 5U4GB rectifier tubes converting the AC to DC inside the device.
The market talks. I just described a coveted $3K Mesa/Boogie amp. Same power and general thing available for a few hundred in solid-state/digital gear capable of recreating thousands of amplifier tones …almost. But not really.
I want my emergency radio transceiver small, solid-state and crystal clear, however. Only guitar amps warm and fuzzy with power sags.
The glass versions of the early metal tubes tend to be less microphonic.i have an old guitar amp with metal preamp tube that someone put a rubber hose wrapped with wire in a attempt to reduce microphonic problems.
When in my 20s around 1970, i could only afford old tube amps to listen to music. I had a Harmon Kardon, then a Stromberg Carlson and they both sounded great. I friend talked me into buying a new transistorized Marantz. It was so flat and boring to listen to compared to my tube amps from the 1960s that I stopped listening to music until I found some of these old Fisher and Scott amps on eBay around 2002. I bought them, fixed them and redescovered great sounding music again. And, im still listening to them today thx to finally getting around to maintaining the myself with an Oscilloscope, Signal Generator and VOM.
RJ
Between YC130’s and 5CX3000 tubes, I’ve had my share. The problem with rebuilt tubes is getting the purity of the metals inside those vaccuums. Many a “soft tube” comes back as a rebuild. Eimac and Svetlana are the big ‘uns. I play nicely with the older Telefunken and Simens Philips Nixdorf kind when working with the grandfather radios. I have the utmost respect for the tube and valve electronics even though I was taught on DTL, TTL and CMOS technologies.
Remember the adage: The children will inherit the items from what we had in the past…including the problems. Now, I rarely get to do much high power transmitter work, but I still get a soft warm feeling when a young technician would hit the plates and the crash of the contactors would give them the willies. All in all, look how far we have come. Oh…Dayton Hamvention is this weekend and I won’t be able to attend because the wife has been in hospital since late March. If there is any good items that I will miss out on, let the group know.
Or as we say in English- valves!
It’s Röhren. Radioröhren. ;)
Even featured on this site, were Korg’s NuTube, which the manufacturer appeared to hack into a VFD envelope.
More of a gimmick than innovative, but still neat.
X-Ray tubes are a common industrial/medical set piece still. I mean it’s just a diode, but it can be an 800kV, 5kW diode with little engineering effort beyond “make it bigger.”
20 year operating life is the norm too, surprisingly. Lifetime is really just limited by operating conditions and thermal cycling of the tungsten target.
How about Hydrogen thyratrons? An HY-1055, a gas-filled triode switch the size of my fist, can handle 16KV on the anode, 325 amps peak! That’s 3.5 MW pulse!
Is there am SCR that could pull that duty?
Yeah… the datasheet warns of X-ray hazards for it, too.
There are many SCRs that handle that kind of power. PowerEx and Infineon both make devices that switch more than 10 MW continuously, 100 MW peak: many kiloamps. They vary, but are about the size of a few stacked teacup saucers, 150mm dia x 40 mm high.
When multiples are stacked into amusingly-named “valves”, these sorts of devices are used in multi-gigawatt HVDC transmission line converter stations.
Then why are thyratrons still in production? Not arguing here… asking.
I note you cited higher powers and higher currents, but did not mention higher voltage. Perhaps it’s high voltage apps where the thyratron still holds its own?
You’d have to ask the folks who still buy them. I doubt they are using in any new designs, except perhaps some special fringe cases.
Thyratrons used to be king of speed, but SCRs afaik have them beat now, at least below 2 kV. Thyratrons probably still are king for robustness though, not being as susceptible to dV/dT false triggering.
But, yes, SCRs do become pretty scarce above 2 kV, which is why they stack them for the 500+ kV HVDC applications. Being able to trigger those big ‘uns with an optical fiber sure helps, I’m sure.
There’s one x-ray tube in our fleet that’s coming up on 40 years. It had a hard early life, has seen pretty light use the last 20 years, but it’s still spinning up and making photons. More amazing is that it is still on its original generator (power supply).
But, yeah, work them too hard and they die young. Lots of ways to kill them. Either the filament goes from asking too much current, or the anode cracks from not conditioning (pre-warming by a series of low power shots) before a hard exposure series, or the bearings go from overheating or just old age, or internal arc tracks are laid down due to not conditioning before high voltage exposures. The arc tracks can “self heal” to some extent, but ultimately limit the maximum voltage possible. Crazy-old tubes also get evaporated tungsten deposited on their output window, reducing output, eventually falling out of spec.
Just a diode, sure, but they still take a lot of care and attention to live a long time.
In around 1970 I was working for a company making colour TVs. The firm had ‘acquired’ a set from the US which came into the lab to see if there was anything we could learn. It turned out to be based on ‘Compactron’ valves with very few transistors. Unfortunately someone plugged in to the UK 240v mains which promptly blew several valve filaments…
I still have a pretty sizable RF tube amplifier making sure a shelf doesn’t float away. I just don’t need the power right now.
a tube rf amp that fits on a shelf is comfortable fodder for a solid state amplifier using current tech, a tube RF amp that makes your houses foundations regret being laid, and your local electric co rub there hands with glee when you fire it up is the kind of amplifier solid state stuff struggles with.
With the ‘nanoscale vacuum channel transistor’ concept, vacuum tubes might even still be able to compete with transistors in ICs. Those are still experimental, but they might shrink vacuum tubes to the the feature sizes, voltages, and speeds of the transistors in modern ICs, and let them to be produced lithographically.
Kudos to Al Williams for such a nice compendium of vacuum tubes. I was especially pleased to see mention of my old friends: Nuvistors, acorns, and even Compactrons.
Those interested in the real-world development of ruggedized and minaturized vacuum tubes used in WWII shells will enjoy “12 Seconds of Silence” by Jamie Holmes.
Couple of places valves (sorry, I’m a Pom) work that might not be obvious:
Venus. Yep, don’t need to worry about the temperature of your semiconductor junctions anymore.
Nanoscale. If made very, very, very small you Volts/metre can get stupid high really quick from a low voltage source – well below the voltage of a semiconductor junction, and so less waste heat. Fabrication may be challenging. Workin’ on it.
Vik :v) [The RepRapMicron Project]
I worked with 8 in Hammamatsu photomultiplier
Tubes mid 1980s for the high energy Proton Decay experiment located 2000 ft deep in salt mine. Those tubes could detect the photons of
A candle light on the moon.
Thank you for writing this very interesting article. I enjoyed the interspersed vintage videos as enjoyable interludes.
would be remiss not to mention the selectron tube. A radical design closer to a DRAM chip, these things in their limited production run stored 256 bits each, with prototypes that stored up to 1024 or 4096 bits. In the 1940’s. That’s the kind of thing that semiconductor chips wouldn’t achieve for a few more decades. Even core memory was probably ultimately more actual labor to manufacture, though manufacturing difficulties did delay selectrons enough that they weren’t used in the IAS computer and only found use in it’s clones. And prevented (afaik) any mass manufacture of the larger 1024 and 4096 bit variants.
I read that the development of a microvalve would have surpassed the resolution of modern day screen 📺
What you all missed. It touched on studio microphones, but now in an audio world dominated by digital, vacuum tubes are still found everywhere. The best analog compressors still run on tubes. As mentioned, we still love tubes in our microphones. Audiophiles still love tubes. But the market is entirely dominated by the 6L6, 6V6, EL34, EL84, 12AX7, and the 12AU7. Almost every top guitar player in the world serious about his craft, or still using Marshall and Fender amplifiers. It is not uncommon when the digital guys record, that the engineer will reamp the model output through a tube amp.
But I am telling you people right now, if you were using horn in a high fi system or PA, nothing sounds sweeter to the ear than a tube amp driving it. 1% total harmonic distortion can be ugly from a solid state amp. You get up to 3% to harmonics distortion with that beautiful even ordered distortion from the tubes, it’s a lovely tone. A pleasure for the eardrum.
Do i spot llm generated text here?
During the 60s and seventies I remember tube tssters in grocery stores with boxes of replacement tubes. it wasn’t until the 90s that the cathode Ray tube in televisions and displays were replaced with solid state equivalents.
Story was right up my alley just now, as busy restoring a 1950 tube radio.
Astonished to find all the parts, even the tubes, were made in Australia at that time, yet we make absolutely nothing comparable today.
Very 50’s SciFi technology. No zip ties, but neatly laced string. No Philips head screws, only flat slotted ones. Two, sometimes three electrolytic capacitors in the one can. Resistor and capacitor values in unfamiliar, non-standard values. A resistor colour band scheme that I can’t fathom at all. No PCBs, just solder tag rails. A metal chassis, into which many components are mounted. Bakelite!
Amazing what they did back then, and all HAND BUILT!
It is interesting the quantity of hollow state equipment currently in niche circles, within which certain aesthetics, analog design, utility, history, nostalgia and emotional connection artistically combine in unique ways. Developing SMD-based pcb’s (which I do) is useful, and certainly keeps our lights on. But, the recent redesign and restoration of my 1954 Globe King 500 AM transmitter (out of service since 1993, last used as a linear amp by my brother to reach me when I worked in Antarctica), putting it on the air and chatting, in pure analog signalling, technically real-time, with like minded afficiandos, many of whom contributed to the outcome, utilizing a natural billion year old phenomenon- basically priceless. Impossible to impart, sort of like a gorgeous sunset, or good surf break, has to be experienced.
The Clam’s guitar player built his own tube amp, and it was mind-blowing. Surf music from Mars.
When I was a kid, a neighbor of ours (and an amateur radio operator) was an engineer at RCA, Harrison. He had worked on the Nuvistor Tube, and would say that it worked very well, but was two years too late. The transistor came along, and with its rapid development, like the planar process and epitaxy deposition, it rapidly displaced the Nuvistor. With the RF JFET’s, Dual-Gate Mosfets, RF BJT’s ,GaAs Fets and their economy of scale it was Game Over!
My Fender Hot Rod Feluxe has tubes. Many musicians use tube amps for a desired tone.
In the late 90s and early zeros I worked at a radio station. Our primary and backup transmitters both used tubes for the final output. They looked nothing like the typical glass envelopes though. They were ceramic pucks about the size of a fist with big round aluminium heatsinks on top that looked a lot like today’s round CPU heatsinks.
The backup transmitter was old, I would guess from the 1960s. The main transmitter looked more modern for the times.
Right after I left they replaced the main transmitter with a solid state one. I presume the old one probably replaced the backup transmitter. Hopefully that found a good home!
So.. anyway… in my mind when I was there the old backup transmitter was ancient and the main transmitter had forever left to go. The upgrade they did after I left made sense to me.. but only because it brought them the ability to do HD.
But I recently heard they are about to upgrade again. I heard them speak on the air about it being 20 years old. How can that be long enough? Then I think back and realize that the current main and backup pair aren’t really much younger today than what I remember was when I was there. OMG old age comes fast! Watch out kids, your time is coming!
Not much was said about high power valves such as those used in 100kW+ HF (shortwave) transmitters. I used to babysit one. In these high power applications there are sometimes solid state alternatives but in many applications valves are still preferable or the only practical alternative
No offense, but the reporting for this story left out the most important detail; vacuum tubes became Field Emission Devices, which are now feasible in the nanometer scale and which have an excellent future.
In guitar amplifier sound they are so far unbeatable. The fear is that quality tubes will disappear – GOD forbid!
I wonder why the Vacuum tubes superiority in the audio frequency bandwidth isn’t mentioned in this article? As a lifelong music listener. The single ended 300B , the 212, and even AR3 are all transmit more micro dynamic information than any transistor I’ve ever listened to. I want to preface that statement by saying I have not yet heard Nelson Pass’ single ended transistor designs that he utilizes in his First Watt Amplifiers.
EL34 and 84 tubes produce the best tones in the best guitar amplifiers, many of which sell for over $50,000 (1940’s Fender Deluxe Tweeds) and 1980’s TrainWreck amplifiers). They also offer more bang for the buck in audio reproduction than solid state in PP or SE configurations.
I have yet to hear anything that bests the Reimyo PAT-777 . It’s a 7.5 W amplifier that uses a WE310A driving a 300B . Dynamics are startling . My horn loaded Lowether drivers are 114db efficient. The air they drive is 100 times heavier than the driver effectively dampening any ring or overshoot better than any other type of transducer by 2 orders of magnitude.
Physics can be mitigated by engineering, but when you utilize the advantages of low power (reducing emf’s in all signal and power cabling first and foremost ), the gains made are not subtle.
It’s so weird that this popped up right as I was about to look for replacement parts for my dad’s Philco record player.
I grew up in the town where Ken-Rad, later GE, made tubes. We had a bunch donated to my high school. We also had a tester and stuff to work with them. My physics teacher liked to use them to teach electronics although we actually used transistors and ICs.
One place tubes really shine (or at least used to) is in transmission towers for TV and radio. Not sure if modern transmitters have big honking tubes as final output stage amps but I know they used to.
As mentioned in the very first comment.
Tubes amplify voltage. Transistors except for FET transistors amplify current, which must then be converted to voltage.
The best low noise preamp were FET. And came closer to the warmth of 12au 12at and 12ax7’s.
Built bothe back in the late 60’s and early 70’s.
A tube is any electronic component in which electrons fly across a space gap and does something other than lighting. Furthermore, tubes do not have to be vacuum; they can be gas filled (example: thryatrons) or even open-air (example: the ion chamber of a smoke detector). Thus, some common electrical parts that are not usually considered tubes can be used as such; one example is a neon light bulb, which can be used in relaxation oscillator circuits or as voltage regulator tubes.
How ironic, I told my young doctor that the ringing in my ears was the same pitch as the warm-up on an old television (tube warm-up). She didn’t no what I was talking about. I feel old.