Ad Widget

Collapse

Announcement

Collapse
No announcement yet.

Overdrive and Reverb...Incompatible?

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

  • Overdrive and Reverb...Incompatible?

    After months of attempting to design an amp with a nice tweedy overdrive, touch sensitivity, and Fender style reverb I've hit a wall where it seems that getting enough signal to overdrive tubes only causes a buzzy distortion tone that sounds terrible. Unfortunately I don't have a program capable of drawing schematics to show you my design, but I can at least explain the topology to you so you may have an idea of what I'm doing wrong, or if this is some other problem I haven't found yet.

    v1a- Preamp (center biased)
    v1b- Cathode Follower
    into typical tweed Volume and Tone control
    v2a- Gain stage (center biased)
    v2b- Cathode follower
    into Fender 6G type tone stack w/second Volume control

    After this the signal splits with the typical blackface 3.3M carrying the dry signal and the 500pf cap carrying some signal to the reverb driver which is a paralleled 12at7 at high voltage off of the screen supply rail, and the typical blackface recovery design into a long-tailed pair phase inverter.

    This was my first design, and at first I thought that maybe I had poorly designed the gain stage in front of the tone stack because it was only caused at high settings of the first volume control while the second volume control acted much more stable, so to limit the distortion I removed the volume and tone control and put the tone stack in it's place between the preamp and following gain stage. That lessened some of the fuzzier distortion that was previously there, but the buzzy sudden distortion was still left. So, thinking that perhaps somehow I was overdriving reverb driver I removed the cathode follower to leave the output impedance higher which didn't help, and I placed a 390k grid stopper on the reverb driver.

    I have noticed now that the buzzy distortion is not really noticeable with the volume knob up high if the reverb knob is off, though it still seems that you can faintly hear it in the extreme upper registers of notes (possibly from the extremely high notes being passed to the next stage without much hindrance thanks to the 10pf cap and 3.3M resistor?) This now leads me to believe that distortion is coming from the second reverb recovery stage where the signals are mixed. However, I've been chasing this problem for awhile now, and am getting tired of attempting to come up with remedies...it especially seems odd that previously the first volume control caused this buzziness even at low settings of the second volume control...if I can make a schematic, and sound clip to demonstrate my problem I will, but for now I just would like to hear ideas from some more knowledgable people about guitar amps than myself to see if I'm doing something wrong (which I know may be impossible to tell until I get a schematic up), or if this design in general was a bad idea.

  • #2
    This is just my own personal opinion, but if you want reverb and overdrive, you should use an overdrive preamp and a master volume. You want reverberated distortion, not distorted reverberation.

    Or with one of these blackface type amps, maybe don't run the amp balls out, use a pedal to get the distortion instead.

    Or build a channel switching amp so you can have the clean channel drenched in reverb, and a much smaller amount on the dirty channel.

    I'm sure someone will contradict me, but that's half the point of this post, to motivate them.
    "Enzo, I see that you replied parasitic oscillations. Is that a hypothesis? Or is that your amazing metal band I should check out?"

    Comment


    • #3
      Yes, that is a bit tough with out a schem... but also tough without an O'Scope because you might be just experiencing a simple "ultra sonic" oscillation who's products are quite audible with the combo of dry and wet.
      However, myself and many many home brewers have had good luck putting reverb in simple amps like this so don't despair! ha ha.
      Bruce

      Mission Amps
      Denver, CO. 80022
      www.missionamps.com
      303-955-2412

      Comment


      • #4
        Well, it's nice to hear that it's at least possible. I have considered a reverb pedal with a dry amp, but I just love the sound of reverb tanks, and since I'm doing this build in an old silverface Fender I certainly have the space to fit a reverb circuit in, so I figure why not. However, since the amp is put in an old fender with the original circuit board I am limited with layout options, and I suppose a parasitic oscillation is entirely possible...

        This brings me to another question about layout. Do you guys have any tips on where parts should be laid. I have some basic knowledge of things like grid should be as short as possible and shielded if possible, plate wires should be left longer than grid wires at least if possible, perhaps a little local nfb will kill an osciallation. Does the placement of cathode resistors and cbc's make much of a difference other than on cathode followers where the signal is taken off the cathode?

        Comment


        • #5
          What do you mean "center biased"? If you are using a 12AX7 with a direct coupled cathode follower, you need to lower the plate voltage of the gain stage because the cathode follower chops about 100V off the headroom of the gain stage. This is why in the 5F6A the cathode resistor is 1K or 820 ohms instead of 1.5K. You need an oscilloscope to see this. What is the B+ in the preamp?
          WARNING! Musical Instrument amplifiers contain lethal voltages and can retain them even when unplugged. Refer service to qualified personnel.
          REMEMBER: Everybody knows that smokin' ain't allowed in school !

          Comment


          • #6
            Originally posted by loudthud View Post
            What do you mean "center biased"? If you are using a 12AX7 with a direct coupled cathode follower, you need to lower the plate voltage of the gain stage because the cathode follower chops about 100V off the headroom of the gain stage. This is why in the 5F6A the cathode resistor is 1K or 820 ohms instead of 1.5K. You need an oscilloscope to see this. What is the B+ in the preamp?
            Oh yeah, that could definitely be a problem considering I forgot about that. The B+ is 350v, 82k plate resistors, I biased to what I thought would be about -2v with a 1.8k resistor. For what's it's worth when the amp was in it's previous state with 2 stages before the tone stack the second stage (which I forgot to mention was a bootstrapped cathode follower) had been biased with many values between 1k and 10k in an attempt to get rid of the buzz...10k gave the least buzz, but it was still present and the gain was of course quite a bit lower.

            Comment


            • #7
              I can't think of a way to record the audio of the sound I'm getting other than through video, and it won't upload. Fortunately the photos of the schematic of the amp will upload...I may have scrawled them too quickly to be legible. If need be I'll take my time draw some more.

              Edti: Oops, I forgot quite a few things on these schematics...mainly the power tube load resistors, and the grid stoppers.

              Load resistors are 220k

              Grid Stoppers are: v1a: 56k
              v2a: 100k
              v3: 330k
              v4b: 390k
              v5a/b: 100k ea.

              These are quite large in my opinion, and have had very little effect the buzz.
              Attached Files
              Last edited by vibrovox7reverb; 01-13-2012, 10:56 PM.

              Comment


              • #8
                Cathode followers typically add a fuzziness, you have 2, one following the other. Try eliminating one or both.

                Your NFB ratio is on the low side for 6V6s into an 8 ohm load, perhaps try a 68K NFB resistor?

                Comment


                • #9
                  The 47K cathode resistors on the cathode followers are trouble. The value needs to be 100K or higher. I know Vox used 56K, but they really cut into the headroom. I know this because I built an amp with two cathode followers such as yours. I ended with 100K pull downs and just tweeked a trim pot in the cathode of the gain stage to get max headroom.
                  WARNING! Musical Instrument amplifiers contain lethal voltages and can retain them even when unplugged. Refer service to qualified personnel.
                  REMEMBER: Everybody knows that smokin' ain't allowed in school !

                  Comment


                  • #10
                    Hmm, interesting...I didn't know that 47k resistors would cause trouble...I knew they would add to compression, but didn't know about them reducing headroom.

                    MWJB, did you mean 6.8k for the nfb loop? 68k would decrease the amount nfb further.

                    This is the original schematic, and I had thought that having a really low impedance go into the second stage without much attenuation from the volume and tone knob might be a source of trouble, so I had removed the volume + tone knob, and put the tone stack in-between the first and second stages which did not solve the problem, and then I removed the second cathode follower going into the reverb driver, which did not work either. If I've lost anything it's been a little bit of preamp tube distortion, but the buzz has remained.

                    I've checked all tubes, and speakers, and cabinets, so I've ruled those factors out. I realize that it's somewhere within this circuit, I'm beginning to think it's a layout issue.

                    I'll post revised schematics soon. Thanks for your help thus far, I've definitely learned some things about cathode followers. Btw, how difficult is it to make a cathode follower distort? I have a hunch that this is another possible problem with the design.

                    Comment


                    • #11
                      You appear to have a 3.9K NFB load resistor and a 22K NFB dropper. This equates to a NFB ratio of 6.6:1...a low ratio and a lot of feedback (~2VAC at max output). It would be more typical to feedback a volt or less (less NFB and a bigger feedback ratio, a bigger NFB dropper), hence I suggest replacing the 22K with 68K. This would give a ratio of 18:1 & feedback around 0.7VAC to the PI at max W RMS (or therabouts).

                      Cathode followers naturally tend towards a grindy, crunch, it comes with the territory, so it's not hard to get them to distort.

                      Comment


                      • #12
                        I wish I knew of a way to get just an audio clip so you guys could hear it. It's not so much a problem with the sound of the distortion because I can hear a pretty pleasing sound, but there is an annoying buzzing and crackling that is very audible beneath it that sounds really bad. I also have had trouble finding a place to elevate the heaters, so whether or not that is the source of the problem I need to do it for the life of my tubes.

                        Comment


                        • #13
                          Bootstrapping the cathode follower gave an unpleasant distortion on signal peaks when I tried it.

                          Comment


                          • #14
                            Agreed, I didn't like the overdriven tone of a bootstrapped cathode follower anywhere near as much as sans bootstrap.

                            Comment


                            • #15
                              Shoot a video of it and upload it to Youtube or a similar video sharing site. Then post a link to it here. You can't upload the video files from your camera straight to the forum: they're too big.

                              If you have a computer with a mic, use Audacity to record sound clips.
                              "Enzo, I see that you replied parasitic oscillations. Is that a hypothesis? Or is that your amazing metal band I should check out?"

                              Comment

                              Working...
                              X