Retrotechtacular: Tube Amplifiers

retrotechtacular-how-tube-amps-work

It’s hard to beat this vintage reel for learning about how vacuum tube amplifiers work. It was put together by the US Army in 1963 (if we’re reading the MCMLXIII in the title slide correctly). If you have a basic understanding of electronics you’ll appreciate at least the first half of the video, but even the most learned of radio enthusiasts will find something of interest as they make their way through the 30-minute presentation.

The instruction begins with a description of how a carbon microphone works, how that is fed to a transformer, and then into the amplifier. The first stage of the tube amp is a voltage amplifier and you’ll get a very thorough demo of the input voltage swing and how that affects the output. We really like it that the reel discusses getting data from the tube manual, but also shows how to measure cut-off and saturation voltage for yourself. From there it’s off to the races with the different tube applications used to make class A, B, and C amplifiers. This quickly moves onto a discussion of the pros and cons of each amplifier type. See for yourself after the jump.

http://www.youtube.com/watch?v=ryCr4Y7ypLA

[Thanks Elliot]

Retrotechtacular is a weekly column featuring hacks, technology, and kitsch from ages of yore. Help keep it fresh by sending in your ideas for future installments.

28 thoughts on “Retrotechtacular: Tube Amplifiers

    1. Jfets work the same way as tubes. Many people have converted low power tubes circuits to jfets sometimes just by redesigning the power supply lines to the much lower voltages jfets work at. This practice is considered sacrilege by tube fanatics, but technically can be done.

  1. That’s about the time I was learning to layout P.C. boards (for a living). Fiberglass was considered a poor substrate, but would do until something better came along. I recall the old timers referring to tubes as “valves” and a radio armatures handbook ad for a IN34 diode for $50.00

  2. Screw tubes today. You’d be better off learning how to rub two sticks together in order to make fire. The whole tube mystique was born in a time when transistors first came out and were not quite as good as the then mature tube technology was. Surprise! Since then we’ve made a few improvements.

    I’ve heard all of the tube arguments and they’re all based on ignorance.

    1. A portion of the audiophile consumer sector is the only one attributing any “mystique” to tubes, for most everyone else they are only used where there’s a measurable advantage to be doing so. The magnetron remains the highest powered active electronic device used in most homes regardless of income. By posting this I doubt Mike is suggesting that we will be returning to tubes, anymore he would be suggesting a return to mechanically operated service brakes on cars if he found and posted an old video explaining how they work.

      1. Most audiophiles know that tubes have higher distortion, they just like the distortion. It’s “warmer” apparently. Apparently that issue’s just simple enough to avoid the quantum uncertaintly of monomolecular solid gold speaker cables.

    2. Bravo! Nothing stimulates a conversation like stating opinion as fact.

      I’ve got several amps at my disposal. Some solid state amps (Dynaco, Carver, NAD, Yamaha), a homebrew Gainclone, a Class D amp and yes a pair of 50W Class A/B tube mono blocks. I’ve got them switched out at the moment but the tube mono blocks knock every amp I’ve ever listened to right out of the park. It isn’t even close. The music just sounds right and I don’t experience listening fatigue whatsoever which I do experience with most solid state amps. It’s interesting to note that I don’t experience listening fatigue with the Class D amps either.

      My opinion is that the reason tube amps can sound better or different is because transistors are prone to nasty crossover distortion when amplifying in a complimentary pair. They stop conducting as they approach zero (one switches off, the other on). It’s not to say there isn’t crossover distortion with tube amps If your tube amp isn’t lined up (under/over biased) you can get it but it isn’t a switch on/off as it is with transistors.

      That said, it doesn’t really matter if tube amps are sonically superior to solid state amps. Even if there were some way to definitively show they were superior sonically, they were destined to be replaced by transistors. Solid state amps use less power, don’t require fussing around to ensure they are biased and balanced and they’re vastly cheaper to produce. That (in my opinion) is the real reason why tube amplifiers are only appreciated by the dedicated few who can build them, fix them or afford to buy new ones.

      Although the tube amps are my current favorites, I’m open minded about it. I’d like to build a set of JFET amps following a Class A single ended tube topology. Something like Nelson Pass’s 1st Watt: http://www.firstwatt.com/j2.html

        1. Carver’s a genius… We used to carry his line at a little boutique shop in the late 80’s, early 90’s. That said, I can listen to my good tube amps without any fatigue whereas after a few hours of solid state, I have to get a break from it. I think it’s got to be the switching (crossover distortion) that causes it.

          1. You can get class A transistor amps, and class AB is biased so they don’t go into switchover until higher power levels, they’re essentially class A at some reasonable listening level, tho that depends on the power rating of the amp.

            The switchover thing would just be another kind of noise, with a frequency and amplitude, it all goes into the THD. I’d want double-blind results to believe it. Not sure why a valve amp’s switchover would be worse than a transistor’s, tho again, valve amps produce much more noise anyway, perhaps any switchover noise is hidden in that.

            Crossover usually refers to the things in the speakers (or sometimes before if you’re really set on throwing money away) that divide up the frequencies to feed to the different speaker drivers. I think switchover is the correct word, feel free if I’ve got it wrong. Referring to when one transistor hands over to it’s complementary one as the signal crosses zero, in a class B, right?

      1. I’ve recently been doing research on guitar amps and effects and found several sources on the “tube sound”. The reason they are liked is the “Warmth” but what does it mean? it boils down to clipping and harmonics. The tube amp will saturate and clip far before a well designed transistor amp but they will clip with a different set of harmonics. Tubes will tend to have lower order harmonic content. Transistor amps tend to have higher order harmonics than tube amps. Higher order harmonics are unpleasant to the ear. Crossover distortion of a push pull amp have higher order harmonics. Amps have been designed utilizing FET’s that have tube like distortion and have a very pleasing sound. I wish I could find the paper I read that went into detail about this subject.
        http://en.wikipedia.org/wiki/Tube_sound

  3. I’ve worked on tube circuits for years and still have dozens of old consumer tube audio units to refub. Tube circuits are simple. They’re highly tolerant of operating point and parts tolerances. They’re easy to make work. Like the lawn mower of the audio world, they’re inefficient but they get the job done. You can make a much better audio circuit with a tube and 6 other parts than with a transistor and 6 other parts.
    Pentodes have a transfer function that’s a lot like a JFET. A triode’s transfer function is like a JFET with feedback from drain to gate.

    1. “My opinion is that the reason tube amps can sound better or different is because transistors are prone to nasty crossover distortion when amplifying in a complimentary pair.”

      “Bravo! Nothing stimulates a conversation like stating opinion as fact.”

  4. Tubes live on in a number of applications. Chief among them is musical instrument amplifiers. A lot of guitarists would give up anything before switching to a transistor amp.

    1. My guess is the reverse would be true if solid state was the first active electronic device technology. I always thought it’s interesting that guitarists are willing products that use solid state devices to add even more distortion to tube amplifiers. What is characteristic as the “warmth” of tubes has long been known to be distortion.

      1. I’m not sure if I read your post correctly but as a long time guitarist, I’ve found that most of us would rather have a tube overdrive than solid state but a commercial solid state overdrive is much less costly than a tube one. My experience has been that solid state distortion generally sounds thin when compared to overdriven tubes.

        I think most guitarists would also rather have tube amps than solid state but again cost is a real factor. I’ve had a small solid state Fender amp (Princeton Chorus) for about 25 years that has a really sweet clean channel. The distortion is pretty poor but the clean channel sounds really nice. My main amp for the past 4 or 5 years, a Fender Twin is a tube amp and sounds absolutely fantastic when overdriven, or clean for that matter but it cost a bundle.

          1. Yeah, pretty much… When I was in a band, I used to have a Boss ME-10 pedal board that cost me a fortune when I bought it. I used to run it straight into the PA. About ten years ago, I got a Digitech for a fraction of the cost of the ME-10 and then let the ME-10 go for a fraction of the cost of the Digitech. The Digitech is smaller and IMO, actually sounds better.

            For me the real advances are in computers, software and input devices… I can put an M-Audio device between my guitar and my computer, pick an amplifier and effects from the GUI and start recording. That is amazing!

      2. I think you have to be a bit more specific than “guitarists.” We live in a world where entire genres of music (or maybe just black metal) were built on doing stuff like using a $40 stereo’s line in instead of a guitar amp and headphones as microphones.

        1. Good point… I think there are some geniuses who’d run a guitar through a pair of tin cans with string to get the sound they’re looking for but a lot of musicians are just using what they can get their hands on. If I were starting today, I’d probably be happy running my guitar into a used M-Audio A/D box and USB into my computer with a set of Logitech 3.1 speakers + sub.

  5. You guys are all thinking in terms of audio amplifiers. In the high power RF world such as radar systems, satcoms, x-ray systems and military applications vacuum tubes are still very much king. Solid state is still years away from being able to produce the high power at high enough frequencies for those kinds of applications that are critical to our existance in the modern world. “My smart phone doesn’t have tubes, so they must not be used anymore”. It’s funny how something is really still the way things are done, but people get this idea that they’ve been totally replaced when it’s only been in a few instances that the new thing was actually applicapable and the old way is still totally relevant. Even electronics and engineering schools are guilty of skipping the chapter on tubes because the profs think tubes don’t exist anymore or aren’t worth teaching, then the students end up finding out the hard way that they are still very much around. Sad indeed.

    1. I suppose there’s no limit to how big you can make a tube, and the elements are obvious, metal, in front of you, rather than the dissipated atoms of a transistor.

      The handy thing of starting with tubes, is that they make the operation of an amplifier obvious, again with literal chunks of metal rather than theoretical N and P regions. Funny how following the history of a subject, and it’s developments, is often a good way of teaching it. The decisions and ideas follow on well, and logically, with the benefit of hindsight.

  6. Why do I still use tubes? Because they’re there. And they are more fun to use than to argue about. I enjoy how gingerly I work on 100+ Vdc circuits…

    I still use Ge transistors too. Forces to think about temperature even in non-critical applications. Sue me. ;)

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.