Hackaday Prize Entry: A BSTRD Preamp

For this year’s Hackaday Prize, [skrodahl] is building a beautiful tube preamp. It’s a masterpiece of glass and free electrons, it already works, and it sounds great.

This circuit is a modified version of the Bastard, an amp published in the Danish magazine Ny Elektronik nearly 20 years ago. The original amp was a true bastard, with a transistor phono stage, a valve line stage, and an input selector that used relays. [skrodahl]’s version only uses the line stage, but part of the name remains as a nod to the original design.

The design of this amp uses octal 6J5 tubes, a 80 VDC, 0.1 A and 6 VDC, 1.5 A power supply. This is actually two projects in one, with the power supply comprising an another entire project.

[skrodahl]’s BSTRD is built, and it works, but the question remains: how does it sound? Unlike so, so many tube amp projects on the Interwebs, [skrodahl] actually has test and measurement gear to figure out what the frequency response and THD measurements actually are. For the frequency response, this amp is dead flat from 10 Hz to 30 kHz. THD is somewhere between 0.35-0.4%, or more than acceptable.

This is a great little project, and an awesome extension to an already popular Open Source project. It’s also a great entry for the Hackaday Prize, and we’re pleased to see it entered in this year’s contest.

14 thoughts on “Hackaday Prize Entry: A BSTRD Preamp

  1. Good job, Brian.

    I’ve been a ham for 30 years and went to air force electronic school 20 years before. I’ve always liked tubes better, but construction is ore expensive and then you can’t stray far from the plug in. Years ago I built a 1/4 watt transmitter that fit in a 35mm film can. I was recognized and given a signal report 250 miles away using a wire antenna. Putting stuff together is where I find the fun.

  2. If the frequency response is dead flat and there’s low distortion… then you might as well have used transistors? Excellent project anyway for its technical impressiveness, but the point of tubes in the modern world is for their negative characteristics…

    1. 0.35-0.4% THD is not “low”, it’s just meh.

      If it didn’t distort at all 0.5% then it would be worthless.

      You need a sweet spot. The numbers don’t necessarily catch it all. Personally I like my all tube amp for the power supply sag which distorts based on intensity.

    2. ” the point of tubes in the modern world is for their negative characteristics”…
      Nope. That’s the wrong way.
      The point of tubes is the absolute simplicity enabling to manage the tiniest detail. Zen attitude. Second order effects minimized. Everything (quite) under control.

      1. Tubes are not simple. Their complexity pushed the development of solid state electronics, which drove the economy of the mid-20th Century.

        User and designer control over a tube is limited at best. There is a high initial voltage and somewhat lower maintenance voltage. This means all of the supporting electronics need to be resilient over a range that is beyond the entire necessary range for the 3.3v to 5v of TTL.

        The point of a tube for a sound amplifier is get certain response characteristics while building something retro — valve-punk, if you will. It “sounds warm”, which makes me cringe to say. If I wanted control over tiny details, I’d configure an FPGA to emulate the sound response characteristics that make a tube sound a certain way.

    3. THD alone does not tell the full story. Low order harmonics (as produced by tube amps) are more likely to be masked by the original signal, making them less audible than higher order harmonics (as produced by transistor amps). This is one of the reasons many people say tube amps sound better.

      1. In order to make transistor amplifiers produce harmonics don’t you have to drive overdrive them? In that case, wouldn’t it be the problem of design/usage rather than of the technology?

          1. Thanks for the link. I read through it and the author does mention non-linearities due to imperfections and environmental dependent factors causing distortion. The only thing is the author seems to assign 2nd and 3rd order harmonics to the topology of the amplifier rather than the technology. I Googled more and found articles which more plainly state that transistors create higher order harmonics vs vacuum tubes but in all of them they state that the amplifier is overdriven.

            From http://www.tungsol.com/html/faqs14.html

            “The basic cause of the difference in tube and transistor sound is the weighting of harmonic distortion in the amplifier’s overload region.”

            I don’t doubt there is some form of intrinsic distortion even if an amplifier (regardless of technology) is operating well within its linear range it’s just that the articles I found seem more focused on harmonics when the transistor/tube is overdriven. It is my understanding that transistors are not supposed to be overdriven when amplifying for audio and thus we would be comparing oranges to apples. It would be more apt to say tubes sound better in comparison to transistors when overdriven vs just the generic saying that tubes are better than transistors.

    4. “If the frequency response is dead flat and there’s low distortion… then you might as well have used transistors?”

      Indeed, or even better low noise jfets. The point of using tubes -coolness factor aside- is when you want to bring them into distortion or very close to it such as in guitar amplifiers or music preamplifiers/compressors/limiters etc: tubes distort very very very gracefully when the input signal increases, while solid state devices clip the waveform like an axe producing all sorts of harmonics which will make the sound terrible to the ear; overdriving a tube instead produces a clean signal that at first starts losing its dynamics just like a compressor/limiter, then as the input signal keep increasing goes into distortion, but the process is gradual and much more controllable than with solid state devices, including (J/MOS)fets. So it all boils down on what you are going to do with the signal: if linearity is a must then only a fool would get an expensive tube pre/amp instead of a much cheaper solid state one, but if you want to stay safe in case of clipping, or are playing with it, then tubes might be the right answer.
      Of course if you are designing a guitar amp you’re not forced in any way to use tubes: decades ago they used them because there was nothing else to make amplifiers with, but today some careful design practices will render tubes totally unnecessary. Need a 100 Watts amp? then build a 250 Watts one and design the early stages before the final so that they will clip in a graceful way before the final starts clipping (use parallel ladders of resistor+diodes series). This is particularly important with modern class D amps which are wonderfully good and clean, just until someone overdrives them (please don’t!).

      I am not sure about how this project will perform: those tubes were designed to be supplied a higher plate voltage, but admittedly we’re very close to their minimum requirement so it should sound good, unlike those ridiculous 12ax7 preamps supplied with 12V which have no chance to sound as a real tube circuit because at such low voltages there’s no headroom to get that graceful transition which makes a tube sound like a tube.

  3. In 1947 a new valve amplifier design was publishes in the Wireless World magazine. The design was by Williamson and it became a standard for “hifi”. The power output was 15W and the distortion was 0.1% at 15W. It didn’t have ‘the valve sound” unless over-driven. You could build it today:
    (full original instructions http://www.sowter.co.uk/pdf/Williamson%20Amplifier.pdf ).
    It would be expensive. Solid state took over because it is a lot cheaper and by about 1970 (I guess) solid state amplifiers became as good as valve amplifiers (unless over-driven).
    This little bstrd looks great and would sound good enough for my ears, I have a plan to either build or restore a decent valve based amplifier system some time (mono of course) and this might be a start.

Leave a Reply to pseydtonneCancel 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.