Announcement

Collapse
No announcement yet.

high gain amp attenuating

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • high gain amp attenuating

    Hello,
    I am trying to design a high gain amp with 4 12AX7 gain stages.*My question is do you need to attenuate the signal after every gain stages and how much?

  • #2
    Originally posted by wouter View Post
    Hello,
    I am trying to design a high gain amp with 4 12AX7 gain stages.*My question is do you need to attenuate the signal after every gain stages and how much?
    Without any attenuation, it would be outta control real fast. With a typical gain of 60 per stage using 12AX7 (I know the tube manual says 100, real world pre's usually yield 50 to 70) 60x60x60x60 gets you about 13 million. There's opportunity between each stage to shape the frequency response with filtering, always a lossy process. Then a final treble mid bass EQ and volume control.

    I just had a Crate V18 on the workbench, I think just two 12AX7 triodes for gain. With volume & master only 1/3 of the way up, it's already out of control and so is the noise. Apparently there's not much in the way of tone shaping/attenuation between stages. Not what I'd call a winning design.
    This isn't the future I signed up for.

    Comment


    • #3
      What he said ^^^^

      To put it another way, it's the "black art" part of designing a distortion circuit. Well, black art tempered with some tech. As Leo noted, it doesn't often sound best to simply pile up the gain. A little in some cases, more in others, often with some other considerations of waveform symmetry and/or frequencies among other things. When I started modding for gain there were no resources for learning. That's not true today. Now you can buy books written by well respected tech/designers that offer some answers and direction. Though I still think the way I did it is probably a valid path. And it starts pretty much the way your post is written. Then you try it for yourself and see what it sounds like (with some direction here so as not to hurt yourself or an amp). Once you try a few things you'll have even more questions, but you'll also have some answers. Then you'll start discovering the nuance of the gain structures and how tuning the circuits changes things. By the time you get to this point your question above will not only have been answered, you'll see that answer as obvious. A good project might be to pick an amp you like (That isn't too complicated. Three channel remote switchers with reverb and effects loop are out) and build one. Because you'll be familiar with the board and components you'll be quite comfortable changing some things to idealize the tone for your personal tastes. There's a lot of guidance and advice available here. Hearing the changes first hand is invaluable for learning just what affects the tone and how.

      To answer the original question more directly, the only time unattenuated gain stage are used would be the very first gain stage (it'll have a volume control but you can turn it all the way up if you like ), reverb tank and effects loop recovery stages (because the signal will be tiny there) and stages that operate into a following stage that is tuned to accept a large AC swing, like the input to a cathodyne phase inverter or cathode follower. The idea with the circuits is that there is almost always more gain than we need to get the clipping we want. This allows us to tune the circuits for even better control and tonal refinements.
      "Take two placebos, works twice as well." Enzo

      "Now get off my lawn with your silicooties and boom-chucka speakers and computers masquerading as amplifiers" Justin Thomas

      "If you're not interested in opinions and the experience of others, why even start a thread?
      You can't just expect consent." Helmholtz

      Comment


      • #4
        I recommend that you start off by studying schematics of classic hi-gain amps, such as the Soldano SLO or Mesa Boogie dual rectifier. Pay attention to the attenuation, gain and voicing.

        No sense in trying to re-invent the wheel.

        Comment


        • #5
          Those would have been my choices for benchmarks. I think the high gain thing sort of peaked with those models.
          "Take two placebos, works twice as well." Enzo

          "Now get off my lawn with your silicooties and boom-chucka speakers and computers masquerading as amplifiers" Justin Thomas

          "If you're not interested in opinions and the experience of others, why even start a thread?
          You can't just expect consent." Helmholtz

          Comment


          • #6
            Thank you for all your replays.
            Are there any guidlines to the maximum voltages you can supply to the next grid?

            Comment


            • #7
              Originally posted by wouter View Post
              Thank you for all your replays.
              Are there any guidlines to the maximum voltages you can supply to the next grid?
              Not trying to belittle the question, this is just to put it into perspective but that's about the equivalent to "How much water should you put in a glass?"

              How much AC you should/could present to a tubes grid depends on the operating parameters, the surrounding circuits, the goal in mind and compromises for how the overall circumstances affect the other circuits and goals of the whole amplifier. There are a lot of different kinds of distortion. Straight up "clipping" with no other influence would be almost exclusive to test equipment signals. So the amount of clipping that is ideal depends on A LOT of other things. To try and avoid giving no answer at all I'll elaborate a little...

              A popular trick of high gain amps is asymmetry to create more even order harmonics. In this situation the a tube is biased more positive or negative putting it's operating center closer to saturation or cutoff. If this off center biasing is considerable, as it is with some high gain designs, the amount you can drive the grid of that tube without unwanted effects is greatly diminished. In another case a tubes sound and response is affected by power supply resistance. This is significant because the more the tube tries to amplify, the more trouble it has doing it. This causes compression. Too much is a bad thing but none is also bad. This property will largely depend on the power supply and not attenuation between stages. But attenuation between stages can be idealized (by ear, no math) for how the power supply circumstances are making the tube behave. It's really just as I said. Build something and play with the circuits. You will discover what affects what and how and if you ask questions here and read some books you'll learn why also. I only covered two of maybe a dozen or more major considerations that affect how much you should/could attenuate a circuit.

              Please don't feel like this stuff is over your head. It's just a thousand variables but not all that complicated once you know some of the important ones. I'll estimate that you only need to do a few mods before you'll see things starting to clear up and start having fun. And you may experiment and find a good level of attenuation, or lack of, for some circuits you tinker with. But you'll be years learning all the things that affect how much to attenuate a circuit and any one of them might be the important one in a given scenario. There's just no pat answer like "Always attenuate to overdrive no more than 12dB." It just doesn't work like that.
              "Take two placebos, works twice as well." Enzo

              "Now get off my lawn with your silicooties and boom-chucka speakers and computers masquerading as amplifiers" Justin Thomas

              "If you're not interested in opinions and the experience of others, why even start a thread?
              You can't just expect consent." Helmholtz

              Comment


              • #8
                You want "a number?" .
                Nobody can supply you that, not even Soldano and others, because not even them know it as such; the proper gain, attenuation, bass cut, high boost and cut, etc. are all highly interactive, final result is arrived at by endless tweaking and playing/listening tests.

                Go buy those good books suggested, build and experiment.

                Our friend chuck h here has made design work for Dean Markley, none the less, and yet his designs will probably be different from others, each designer has his own goals and ways to reach them.

                So go and do it your own way
                Juan Manuel Fahey

                Comment


                • #9
                  Originally posted by J M Fahey View Post
                  Our friend chuck h here has made design work for Dean Markley
                  I really never meant to name drop with that info. I just thought it was novel and fun to share. Anyone who lived near Saratoga California probably had dealings with Dean

                  And to be clear, none of my designs with him ever went to market (though the reissue CD60 was my suggestion). I presented some stuff, did some circuit adjustments to some other ideas they were kicking around and did one NAMM show with three prototype amps of my design in the Ultrasound booth since the DMS booth is always all strings. Fun ride, short story.
                  "Take two placebos, works twice as well." Enzo

                  "Now get off my lawn with your silicooties and boom-chucka speakers and computers masquerading as amplifiers" Justin Thomas

                  "If you're not interested in opinions and the experience of others, why even start a thread?
                  You can't just expect consent." Helmholtz

                  Comment


                  • #10
                    No sense in trying to re-invent the wheel.
                    Exactly! If you hope to create some revolution in this field I'll have to disappoint you big time.
                    Start studying the well known high gain amps like SLO, 5150, rectos also some ENGL stuff and plenty more. Build one and experiment with every part and that's it.
                    If want to know more theory check out Merlin's book as a start.

                    Comment


                    • #11
                      Originally posted by Gregg View Post
                      Build one and experiment with every part and that's it.
                      Two likes. Sorta what I said, but I like your concise version better. There are so many good books on the subject now too. Some are somewhat accessible to those without any experience. Leaf through a few. Easy to do now with Amazon or there may even be a couple at your local library. Find one that you can somewhat understand (but not entirely ) That'll be one written to your level. Start there. But don't be afraid to build or mod in the meanwhile with any information that looks interesting and checks out with some research. Great fun. Amaze your friends and family! Getting ANY closer to the tone in your head is worth the price of admission.
                      "Take two placebos, works twice as well." Enzo

                      "Now get off my lawn with your silicooties and boom-chucka speakers and computers masquerading as amplifiers" Justin Thomas

                      "If you're not interested in opinions and the experience of others, why even start a thread?
                      You can't just expect consent." Helmholtz

                      Comment


                      • #12
                        The suggestion to start with the SLO and Recto is a good one The early Dual Rectos were more/less a slightly modded SLO circuit. Randall tweaked a couple values and moved the FX loop to after the tone stack (which is where it belongs, IMO). This is a nice stage by stage break down of the SLO:
                        Soldano Super Lead Overdrive - First Stage
                        -Mike

                        Comment

                        Working...
                        X