Cute Little Amplifier Has A Tube Pre


While you won’t catch us in an argument with an audiophile regarding the sound quality of tube vs. solid state amps, there is a general consensus that tube amplifiers sound much better than their transistorized brethren. Actually building an all-tube amplifier, though, is a bit harder than one built around common ICs – there are transformers to deal with and of course very high voltages. One solution to get the sound of tubes easily but still retaining the simplicity of integrated circuits is a hybrid amp, or a tube preamplifier combined with a solid state power section. They’re easy enough to build as [Danilo] shows us with his hybrid tube amp design (Italian, translation).

[Danilo]’s design uses two ECC86 for the left and right channels powered by a 12 Volt supply. Each channel is sent through a tube and then amplified by a TDA2005 20 Watt power amplifier. After plugging in a CD player, the result is a clear, warm sound that can put a whole lot of power through a speaker.

64 thoughts on “Cute Little Amplifier Has A Tube Pre

  1. “there is a general consensus that tube amplifiers sound much better than their transistorized brethren”

    Don’t think there is. Some people have a preference for tube driven equipment, but many others dont. Tubes produce more noise and distort the sound in ways that their fans like, transistors can be made to be more linear and less noisy which thier fans like better.

    That said, tubes look cool!

    1. Yeah, a lot of people prefer the “warm” sound of tube amps, and that’s totally ok (there’s room in the world for both, after all), but intellectually honest audiophiles will freely admit that the warmth is a form of distortion.

      1. Intellectually honest audiophiles will freely admit that the sonic warmth is a product of their imagination and that the tubes are only there for their visually appealing properties.

        Tests of scientific value, which the average audiophile test doesn’t have because it’s not properly controlled, have been done about this countless times and they all agree: You can’t tell which is which if you don’t get to see them.

        If you see a glowing tube you will hear the tube sound, even if below the tube is hidden a transistor doing the actual amplification.

          1. Except that “warmth” doesn’t sound like warmth at all. Distortion actually sounds very different from how audiophiles think it sounds. Also the human hearing is much less acute than audiophiles think it is.

            Distortion at the levels audiophiles use it is actually too low in level to be audible, the warmth audiophiles hear is only inside their head.

            If you drive the level of the distortion up to the point where it becomes audible, like guitar players do, then you’ll find it doesn’t sound at all like warmth. It sounds like… wait for it… trashy music effects and/or cheap overloading radio.

          2. @anon

            Every single statement in your post is completely subjective. That’s fine, except you seem to think those subjective statements suffice as factual proof.

            If you can prove what you claim in your first two paragraphs, please do so – I’d like to see some evidence either way. If not, you might want to stop putting your opinions forward as facts.

        1. “Tests of scientific value … have been done about this countless times”

          Link(s) please. A lot of people say this, but I’ve never seen any of them provide the actual evidence. I’ve never found such a study, and I’d like to read it if one exists.

          “If you see a glowing tube you will hear the tube sound, even if below the tube is hidden a transistor doing the actual amplification.”

          May well be true, but it’s a psychological effect. It has nothing to do with whether tubes are actually affecting a given signal differently from transistors.

        2. anon and others-

          I would be first in line to agree that a lot of what audiophiles sling around is complete BS. However, the notion of “warmth” in music reproduction is not BS– it’s measurable, and the effect is easily explained with logic.

          Granted, if you’re driving any piece of audio gear to the point of producing square waves, then crap is crap. But in the case of small amounts of distortion, tubes (and other voltage controlled devices like MOSFETs) tend to produce even harmonics. For whatever reason, current controlled bipolar devices (like transistors) tend to produce odd harmonics.

          Music theory then easily explains the apparent “warmth” of the tube sound. Even harmonics are octave-intervals from the fundamental… meaning that the distortion products combine with the fundamental in a musically-pleasing way. It’s like hitting the middle-C on a piano keyboard combined with all of the “C” keys above it. The composite note is still “C”, but now it sounds fuller and more rich.

          Odd harmonics are not musically related to the fundamental, which results in dissonance… lending to the signal a harsh or more “brittle” character.

          The other big difference between traditional tube and transistor amps has to do with gain and feedback (or lack thereof.) If you want to design a good solid-state amp, you start with a gain block that provides far more gain than you really need… and you add a negative feedback path to tame it down. This stabilizes the amp and broadens its response. The problem is that it takes a finite amount of time for any feedback loop to respond to a transient event at the output… say the attack on the crash of a cymbal. This results in a type of distortion that is unique to amplifiers with feedback. Discriminating ears can perceive this as non-musical or unpleasant in nature.

          Traditional tube amps don’t have as high a gain, so the feedback loop, if any exists at all, doesn’t introduce this type of distortion.

          In the end, if the purpose of a home entertainment amplifier is truly “accurate” reproduction, a high-quality solid-state amp is ultimately the best choice. In order to realize the benefit of what you’ve paid for, you will need top-shelf speakers, equalizer/delay lines to account for room acoustics, and a room whose interior treatment is optimized for audio. That means a room with non-parallel walls, and an interior surface that is likely to be peppered with patches of sound-absorbent material. Your favorite couch or chairs may have to go. Maybe the tile, too.

          Who goes to these extremes? Nobody. I’ll bet that in the end, 99.9% of people with even high-end gear end up adjusting tone controls to positions that “sound good,” not settings that assure “accurate reproduction.” Speakers and furniture are placed where they are convenient or where they look good. In effect, the amplifier becomes a furnishing and a musical instrument in its own right, tuned or tweaked by the owner to optimize their own pleasure.

          Given that reality, it is understandable why some people would gravitate toward tube amps. The tube amp, as a musical instrument, can modify program material in a pleasing way.

          By the way… in years past, it was common practice in recording studios to run levels on a tape machine a little on the hot side… why? Because the distortion produced by the magnetization of the tape itself contained even harmonics which “warmed up” the sound. BASF, if I remember correctly, actually sold a formulation of tape designed to optimize this effect. Predictably, this habit caused endless grief for engineers during the transition from tape to digital recording, because if the signal into an A/D is hot, you don’t get warmth, you get hard clipping… which sounds horrible.

          1. Don’t forget, transistors in a conventional circuit switch when they go from +/- like a relay and that introduces noise (switching distortion). You won’t hear it but it will cause listenning fatigue over time. Tubes don’t switch when they go from +/- which is one of the reasons they are known for “smooth” sound.

            I’ve had a lot of great stuff in the past (worked for a boutique audio shop) but I’m converted now… I now run a set of home built push pull Class A/B monoblocks. I can run a quad set of 6L6s in each or pairs of EL34s. Right now they’re running Groove Tube EL34 tubes and sound amazing.

            There is a transitor circuit I’d like to try called a Le Monstre amplifier. It’s mosfets but the circuit looks like a tube amp design.

        3. I call horse-feathers on your statements. I’ve done hearing and listening tests, and in my personal experience I can confirm that there is a difference and that some high frequencies and harmonics are audible under some conditions. I have no reason to believe that other people (though certainly not all people) can’t hear the same things I can hear. Perhaps in your experience such things are not audible/detectable, but to project your experience upon everyone is just as large of a logical fallacy as that which you attempt to confuse others of making.

    2. From what i’ve read the “warmth” comes mostly from tubes not having an even frequency response, being made mostly from much larger parts than solid state semiconductors they will have a worse response at higher frequencies.

      Most of the tube sound can be replicated with a simple analog filter.

      I do agree tubes look good though, and that retro tech is fun. But i will never claim anything old is better than something new without proper proof…

      1. I just found this interesting two-part article:

        (second part is linked from this one)

        Part 2 concludes in an interesting fashion:

        “You can easily approach the “tube-amp sound” by simply placing a substantial resistance in series with the output of a modern voltage-source audio amplifier. This decreases the damping factor and allows the bell modes on the transducer cone and dust cap to radiate as opposed to being clamped by an output impedance that is effectively zero.”

        I haven’t read the whole thing, but it’s quite detailed and seems reasonably rigorous.

    1. The tube market these days usually depends strictly on how much stock there is and how badly people want them. I don’t think anyone is doing a new production ECC86 (or ECC88 for that matter), so the prices reflect the fact that whatever has already been made is all that’s left..

    1. My thoughts exactly.
      I’ve seen chinese junk (hybrid) amplifiers that had better designs.

      IMHO It’s a sunday-afternoon-see-what-I-can-make-with-the-contents-of-my-junk-box-project at best. Certainly not worth buying the components to duplicate this so-called ‘design’.

    1. Electrons. They literally fly across a vacuum to another plate with an opposing charge. A wire mesh “Grid” controls the flow of electrons by changing its charge.

      Compare to solid-state electronics, where electrons stay confined within material.

  2. “there is a general consensus that tube amplifiers sound much better than their transistorized brethren” Seriously ? Is this some kind of troll ?

    If you’re doing electronics, you will agree that there’s nothing more to sound than a sum of sine harmonics defined by their phase and amplitude. Transistors are very linear, they multiply the amplitude of all harmonics by the same number and offset all phases by a constant. Tubes are non-linear and distort sound, therefore they perform worse from a technical point of view. Moreover, a circuit distorting the sound exactly like the tubes do can be built from transistors. So what the hell is this tubes vs transistors consensus ? Sounds like a sentense from a bad audiofool magasine rather than serious tech blog !

    1. While there is truth in a few of these comments, I find it a little ridiculous that most of them have the sole objective of shitting on the hack, despite their clearly having only a cursory understanding of audio/electronics.

      More distortion and noise in the comments than anywhere in the amp’s circuitry, IMO.

        1. The “shitting on the hack” wasn’t so much aimed at you, to clarify. Again, I meant to comment, not reply to a specific post (I’m an idiot :) ). Your information about transistors vs. tubes, though, is simplistic at best and simply wrong in places. Tubes and transistors both can be operated in a region in which their response is approximately linear, and can both be used to build extremely clean accurate amplifiers. In the case of such amplifiers I can agree with you that tubes are a bit wasteful in terms of power consumption, heavy transformers, etc. However, I disagree with your idea that you can use transistors to build a circuit that distorts exactly the same as a tube circuit. I’m not saying you can’t get close, but there is a difference in the way a tube and a transistor respond once you get out of that linear region that is very very hard to replicate. That’s why most high end guitar amps are still tube amps. It’s not some magical voodoo like it is with the super clean audiophile stuff, because when you’re not going for super clean (like in a guitar amp) you really do get something different. This is a super dumb internet argument, and actually regarding the amplifier here, I agree with the spirit of your argument, which is that tubes probably aren’t a great choice for a super accurate headphone amplifier. That said, if the goal is good sounding instead of clean, they may actually be better. People don’t like the sound of vinyl because it’s clean. Same goes for analog tape in the recording world. It’s because it’s nonlinear in a pleasant way, and whether the “pleasantness” is the result of nostalgia or something else, it’s very real. I’m not trying to personally attack you (though I know it comes across that way most of the time when you go after someone’s ideas), I just think there’s more to it than you’re seeing.

          I was mostly just bummed out that someone built a cool little amp with his own hands and all people had to offer in the comments was negativity (mostly uninformed at that.)

          1. I agree with you that my description was simplistic, because I was assuming we were talking about a linear amplifier for CD playback. As I said in another comment, guitar amplifiers have indeed a very different sound when they clip sound with tubes.

            I almost never comment on HaD, but I felt like I needed to because there was this huge troll instead of just presenting what the guy made.

        1. It would very much matter if he was using “crappy” .mp3 files.

          First of all it has been demonstrated that with a properly encoded mp3 you can’t tell the difference and second, even if you could, the effects of the playback chain are additive. Playing back a crappy mp3 on a crappy system is going to make the crappy mp3 sound even worse. Playing back a crappy mp3 on a good system is going to make it less crappy and thus more bearable.

        2. …aaaand to toss some fuel on the fire :) it’s my experience that what people generally mean by “the tube sound” is a combination of harmonic distortion, slight compression, and HF rolloff.

          I’ve personally pumped mp3s through an old tube reel-to-reel recorder and thought “wow, that sounds great!” because of those factors, and no one would (or should!) hold up that 50-year-old, unmaintained, beaten-up, capacitors-dried-up recorder as a good amp of any sort.

  3. “The BEST amplifier is a wire with gain!” – best description I have ever heard when discussing this tube vs amplifiers stuff. An “ideal” amplifier has 0 propagation delay, linear gain across the band, infinite SNR, 0% THD, no clipping, etc.

    1. “The BEST amplifier is a wire with gain”

      Yes. It’s also impossible to create such an amplifier in the real world. And the things you mention that an ideal amp doesn’t have are often musically useful (including in playback of music recordings).

      Some people will spend a lot of money in the pursuit of a “wire with gain” sound system. Others will spend a lot of money to have a system that adds color (i.e., distortion and compression) they find pleasing.

      Neither side is wrong until they start making claims they can’t back up.

  4. > One solution to get the sound of tubes easily but still retaining the simplicity of integrated circuits is a hybrid amp, or a tube preamplifier combined with a solid state power section.

    Noise and distortion are an additive thing. So with this hybrid amp you get the power consumption and noise of 1960’s console radio combined with the distortion characteristics of a 1980’s car-stereo.

    > Each channel is sent through a tube and then amplified by a TDA2005 20 Watt power amplifier.

    Those are pretty old chips, and designed or car stereo applications, not high-end audio. The LM3885/3886 would be a much better match.

    1. Who are you polling for your consensus?

      I think it’s entirely fair to say that *among musicians and audiophiles* there’s a general consensus in favor of tubes. Whether this consensus is based on much beyond hype is the subject of much debate.

  5. Sorry krylenko, unless you can back up that statement with stats, you can’t make that claim. Tubes are popular with these groups, of that there is no doubt, but I have never seen anything to suggest concensus. Some like tubes some don’t, some high end stuff uses them, plenty don’t. That’s not passing judgement on what is better.

    1. General agreement is the definition of consensus. If it’s popular, there’s general agreement. I think all he was trying to say is that he didn’t really want to start a debate about it and that it’s a thing a lot of people think, for whatever that’s worth.

    2. “I have never seen anything to suggest concensus”

      It’s not like you can back up that statement with stats either. Among the people I know, it is reasonable to say there’s a “general” consensus. That might be true among the people the author knows. Sounds like it’s not true for you and others as well.

      But saying Brian’s trolling is a bit much.

  6. @krylenko

    The author here claims a “general” consensus, not only among musicians as you say, unless you think only musicians have a word to say because they have some kind of golden ears.

    There is even more bad faith in your argument because tubes are indeed very popular in guitar amplifiers where they’re overdriven and clip the signal. There is indeed a big difference between the soft clipping of the tubes and sharp clipping of the transistors. (Even though soft clipping transistor circuits do exist !) But for playing CDs nobody cares as the amplifier works in linear mode and does not clip.

    1. “General consensus” may well be an overreach. My point is, it depends on the implied meaning of “general”. Is there a consensus among people who don’t care either way? No, because they don’t care.

      Musicians don’t have special authority, of course. I mentioned them because I know a lot of musicians and many of them do care and do prefer tubes. I’m also aware that it’s an active debate for audiophiles, and in what I’ve seen more people come down on the side of tubes than transistors.

      “But for playing CDs nobody cares as the amplifier works in linear mode and does not clip.”

      It’s certainly not accurate to say nobody cares. The audiophile tube vs. transistor debates center on preamps and power amps that are being used to play back music from CDs (as well as vinyl and higher-res digital files). Generally these amps are designed to work *very* linearly, but that doesn’t mean they all sound identical. Browse the Stereophile forums for a taste of the back-and-forth.

      For the record, I don’t particularly care either way, though I like tubes a lot for distortion and compression as a musical effect. But it’s interesting to see this debate, which I’m familiar with from the musical side, collide with HaD’s readership.

    1. I don’t know, it’s a pretty simple project. Looks nice, and it’s simple enough I’m tempted to try it. Is there much else to say about the build itself?

      I actually think it’s more interesting that it taps into a larger ongoing debate. There’s a lot more to discuss there than in the project.

      1. Sure, it is a nice project and simple too. For me, that’s a beauty and I’m glad that it is provoking some interesting debate. But there are thousands of pages devoted to (in my mind) pointless bickering as to the relative merits of tubes versus transistors. I guess it would just be nice for HaD comments to focus on the projects!

        Perhaps I am just cynical and old…

        1. That’s fair, and I agree there’s a lot of hot air on this subject.

          I personally find it interesting to see here the intersection of the music-focused and engineering-focused arguments on either side. Neither approach is inherently better, and there are lots of people in both camps.

  7. While I’m not looking to to replace my mid ’70s era realistic receiver that still suits me just fine, I thought this would be simple for the hell of it fun build. SOB $35 for a pair of tubes delivered on a slow boat from China. My hopes to be an audiophile snob ;) have been crushed.

    Tubes where where first, those that grew up with them,and are of the few that can detect the difference, naturally will prefer tubes, I suspect others that didn’t grow up with them have somehow been socially conditioned to prefer tubes. In the event solid state was on the scene first, they would say solid state sounds terrible.

      1. I’ve heard that question posed before and have wondered about it myself.

        One thing I’ve read is that first-generation solid-state amps *were* pretty bad – low SNR, a small linear region, and harsh inharmonic and cross-modulation distortion outside the linear mode. When those designs were introduced, tubes were a mature technology with highly evolved designs for accurate amps with wide bandwidth. I can see why someone at the time would prefer tubes.

        With the discussion of “general consensus” I do wonder though – how much does the general public actually care about tubes vs. transistors? I’m used to musicians and audiophiles debating it, and clearly some engineers who aren’t either of those groups also have strong opinions. Does anyone else care?

        1. To audiophiles and Musicians alike the only thing that matters is what it sounds like. I would venture to say any true audiophile would not base there opinion of any amplifier solely on how it tested on the bench. Also there is way more to it….speakers, cd, sacd dvda, vinyl, signal cables, speaker cables. As a musician what kind of music are you playing has a lot to do with what kind of amp you want to use, I just think its not black and white.

    1. Possibly, but I think if solid state had happened first tubes never would have existed, since by most engineering measures the transistor is superior. They might have still produced high powered transmitting tubes for radio, and CRTs, but that’s probably about it.

  8. Interesting article and painful debate. To (kinda) get back on track (so to speak), does anyone know why he is using two tubes that each have two triodes inside? it appears that he could do the same thing with a single tube.

    Other than that, quite clever build IMHO.

    1. That’s a good question. The ECC86 is a medium-mu twin-triode, with two identical triodes internally. He might have strung them together to power the filaments from 12V, but that seems wasteful. I probably would’ve added a ballast inline with one tube, like a #47 pilot lamp (they used to use them for this all the time…)

      1. That is what I was thinking. a simple voltage divider to power one filament, and use the two triodes as left/rightchannel. Since the tubes are 1.) rare and therefore 2.) expensive. it would be more cost effective from a build standpoint to use one tube. If the tubes were inexpensive, then the power budget for heating 2 filaments vs one and dissipating energy into a voltage divider would drive design, IMHO.

  9. I’d be interested in seeing this done with an ECC83 (The ever-popular and easy-to-find 12AX7).
    I’ve run ECC83s with low plate voltages (“Starved Cathode”), and it produces some interesting effects. I ended up building a guitar pedal for a studio using four triode stages with a tone stack and variable gain. It made a nice tunable sound, all the way from a nice “Jazz through a vintage amp” sound nearly all the way to a almost-fuzz when using high-output pickups.

    Starved mode tubes are quite cool, but a bit of a guessing game. When your tubes are designed for 200-300V, running them at 12V isn’t something you can find on the datasheet. I also prototyped a regenerative radio using that trick – nice thing is not burning them out with guesstimate part values!

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.