Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

My latest crazy amp project photos

  1. Nov 26, 2016 #1
    It takes real machines to build big amps!!..lol

    EEjyszd.jpg

    11 tubes, 2 channels, tube reverb, 100 watts, channel switching, DC tube heaters.....somewhere in all this I must have lost my mind!!!...lol

    yZ6sflv.jpg

    PRf42AC.jpg

    I guess tomorrow I will see if this beast will work. It is by far the most complex amp I have dreamed up. My friend Dave drove down from Michigan to help me crank it up. I think we are about eight hours from lift off.

    Cheers,

    Billy

    Heavy Iron

    EvQPQH9.jpg
     
    Last edited by a moderator: Apr 19, 2017
  2. jcsd
  3. Nov 26, 2016 #2

    davenn

    User Avatar
    Science Advisor
    Gold Member

    sweet !!

    do you do your own metal bending for the case as well ?
     
  4. Nov 26, 2016 #3
    I wish I did but no sheet metal brake. I am also finding that all the different skills needed to build a completed amp is a bit much for me. Some days I wish I had all the tools in the world and other days I am glad I don't.

    I am still struggling with basic electronics and need to stay focused on that, at least for a while. I have sort of jumped into the deep end of the pool. A 100 watt amp of this nature would be a complex project for anyone with years of experience. At my level, who knows.... I know what I want it to do and sound like and I am sure I will get there but it is a bit stressful. Lots of kind people have helped me learn some stuff but nothing replaces years of experience. I have at least made a bit of progress in the few short months since I started all this.

    Some of this is fun and some of this is just plain hard work!!

    Cheers,

    Billy
     
  5. Nov 27, 2016 #4

    Svein

    User Avatar
    Science Advisor

    Using tubes you are in radio transmitter land. The larges tube amplifier I have seen was 35W - and the output transformer was gigantic.
     
  6. Nov 27, 2016 #5
    What a pity, I have no time to try this: a small tube amplifier with its transformer's output amplified by a precise current follower made of two powerful complementary MOSFETs! Such a follower (at least, if improved by an op-amp keeping the source voltage, the transformer's output, at zero) would add absolutely no distortion; therefore, all the distortion would remain "tube-like"; and the transformer could be small ― but with very thick-wired output coil feeding the MOSFETs' sources. The output transistors must of course be included in the feedback loop.
     
    Last edited: Nov 27, 2016
  7. Nov 27, 2016 #6

    Svein

    User Avatar
    Science Advisor

    So why use tubes at all? MOSFETs have the same characteristics as a pentode...

    2_IRF510adj_Dif_In.GIF

    Of course, transistors are much more precise than MOSFETS in the input stage. I would not use an op-amp since it is usually not good at handling large signals. But - here is a schematic for a 600W amplifier (http://www.elcircuit.com/2012/04/600-watt-mosfet-power-amplifier-with.html):
    600+Watt+Mosfet+Power+AMplifier.jpg
     
  8. Nov 27, 2016 #7
    ― but the thing is, it's awfully interesting to make the Class-B output stage to work as a current-follower, rather than classic voltage follower (and to do that, one have to use a current transformer before the output stage) ― because in that case the nasty crossover distortion could be completely eliminated ― because the sum of the drain currents is always practically equal to the input sum of the source currents ― very much unlike the input and output voltages of a classic power voltage-follower.

    I am awfully sorry, have no schematic diagram to show at the moment, it's still in my head.
     
    Last edited: Nov 27, 2016
  9. Nov 27, 2016 #8

    Svein

    User Avatar
    Science Advisor

    A current-follower is not hard to design, but I doubt it will be of any use. Just as a perfect voltage follower has zero output impedance, a perfect current follower has infinite output impedance. Since a loudspeaker does not have a constant impedance (see figure), driving it with a current will accentuate the nonlinearity instead of damping it.
    1224px-Speaker_impedance.svg.png
     
  10. Nov 27, 2016 #9
    So what? The output pentode, in a classic simple tube amplifer, also has practically infinite output impedance; but the feedback (R108, C105 on the picture) from the speaker to the cathode of the before-output triode makes the output impedance of the whole amplifier low enough.
    6T9-Tube-Amp-Schematic.png
    A similar feedback ought to be used in an amplifier with current-following MOSFETs at the output.
     
  11. Nov 27, 2016 #10

    Svein

    User Avatar
    Science Advisor

    No. The purpose of the output transformer is to transform the output impedance of the power tube to an impedance low enough to handle the loudspeaker (if the turns ratio is n:1 , the impedance is transformed as n2:1). The purpose of the feedback circuit is to correct for the non-linearity in the output transformer.

    One of the problems with the amplifier you show is that the "zero" current in the power tube goes through the output transformer all the time. This current creates a basic magnetic field in the output transformer, moving the operating point away from zero. Therefore push-pull stages ere preferred, since the "zero" current is supplied to both output tubes, but the currents act in opposite directions. For a discussion of output transformers, see http://www.turneraudio.com.au/output-trans-theory.htm.
     
  12. Nov 27, 2016 #11
    That's the best discussion of output transformers I ever read! But look attentively: it says nothing about tube's own output impedance. It tells about the load impedance "seen by the tube" or which "suits the tube". It's all about getting power from that tube. A good pentode can give audio-voltage about 100V or so on its anode, and current about 0.03A or so; therefore, the speaker impedance has to be transformed to 3000 Ohm, to get full power. But that does not mean that the tube's own impedance is 3000 Ohm!

    The output current and voltage of the best voltage follower are limited in the same way; yet the output impedance of that follower is very small and has nothing to do with those limits.

    These are the characteristics of a pentode from Wikipedia: although it seems an awful big pentode (e.g. the big "horizontal" one from a huge ancient TV), its output impedance (at moderate currents) is hardly less than 30 000 Ohm (and I would have made it much greater with a cathode resistor):
    330px-Kt88pent.gif
     
    Last edited: Nov 27, 2016
  13. Nov 27, 2016 #12

    davenn

    User Avatar
    Science Advisor
    Gold Member

    cuz it doesn't sound the same, isn't so romantic,
    if your MOSFETS's are glowing like tubes, you have a serious problem :-p
     
  14. Nov 27, 2016 #13
    First thing, no modern guitar amp is designed to produce a non distorted sound except when used to play jazz and even then the sound is distorted to some degree by design.

    There are a few high end transistor amps that are being used today by pro musicians, but very few. I have played just about every guitar amp made and there is no transistor amp I have played that sounds like a tube amp. The best ones are close but not 100%. There is not a solid state amp in existence that will accurately reproduce sounds made on a tube amps, again sort of close but any pro guitar player can spot it a mile away.

    Every serious musician (electric guitar) with few exceptions and most amature musicians play tube amps. 99% of the transistor amps are cheap practice amps.

    Sound reinforcement on the other hand is all transistor now days. When you go to a large concert the guitar player may be playing through a 20 watt tube amp. The sound you hear is being reinforced by high quality transistor gear which does a very good job of reproducing and amplifying the original sound from the 20 watt tube amp.

    One notable exception is something call a Roland 120 Jazz chorus. It came out in the early seventies and is highly regarded by guitar players who play jazz. It is still being made today but of poorer quality than the old ones.

    There are many reasons that solid state devices do not produce the same sounds as tube devices. One of the main things is the harmonics produced when the MOSFET or other SS device is driven into distortion. I does not have the same characteristics when driven into distortion that a tube does. As I said, the whole idea behind modern guitar amps is to produce a distorted sound. It actually gets even crazier with modern metal music. The amps are biased so cold that there is a good amount of cross over distortion involved on purpose. That does not sound good to me but I don't play or like metal music to begin with.

    I think the best way to think about guitar amps is that they are "sound effects devices" designed to produce very specific types of sounds with a frequency range of less than 5000hz. They are sound production devices of limited use and have not so much in common with sound reproduction devices.

    This whole guitar/ guitar amp subject is very difficult to understand unless one is a fairly accomplished guitar player/musician and has a good understanding of the electronics involved. To add insult to injury there is a ton of misinformation and pure nonsense kicked around by musicians and others.

    Cheers,

    Billy
     
  15. Nov 28, 2016 #14

    jim hardy

    User Avatar
    Science Advisor
    Gold Member
    2016 Award

    I offer this article as a thought on the age old "tube vs transistor sound"
    http://www.edn.com/design/consumer/...periority-of-current-drive-over-voltage-drive

    I have always thought the "Transient Intermodulation Distortion" explanation just more of that pure nonsense, ignoring the basic physics.
     
  16. Nov 28, 2016 #15

    jim hardy

    User Avatar
    Science Advisor
    Gold Member
    2016 Award

    Hey - it's catching on!
    http://www.current-drive.info/9

    Some of the earliest solid state Dynaco's took their feedback from output current. Their old timers figured it out, sans fanfare.
     
  17. Nov 28, 2016 #16
    I love the neat tagboard construction.
     
  18. Nov 28, 2016 #17

    Svein

    User Avatar
    Science Advisor

    Well, it's not nonsense and it refers to a well-known fact in electronics: Saturation. When a transistor in an amplifier circuit is momentarily over-driven and goes into saturation, it usually takes at least 100μs to get out of saturation (high current gain transistors are not meant to be driven into saturation). What Matti Ottala proposed in his famous paper was to introduce negative feedback in every stage in the amplifier and cut back on end-to-end feedback.

    A copy of the original paper is here: https://linearaudio.nl/sites/linearaudio.net/files/otala low tim amp.pdf
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted



Similar Discussions: My latest crazy amp project photos
  1. My amp is broken. (Replies: 11)

  2. My Project (Replies: 16)

  3. Latest op-amp (Replies: 2)

  4. New Amp Build Photos (Replies: 5)

Loading...