Ad Widget

Collapse

Announcement

Collapse
No announcement yet.

Parasitic Oscillation in an otherwise fine amplifier...

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

  • Parasitic Oscillation in an otherwise fine amplifier...

    Hello everyone!

    Nice forum.....seems like you all know what you're doing. However, I don't seem to know what I'm doing, as I just built a screechy-squealy amp that is one side AC30/4 (EF86) and the other side is AC30/6 (12AX7) but with a Master Volume and a full EQ section. Trannies are Mercury Magnetics Woden clones, custom aluminum chassis, etc... I'll post pics when I get home.

    The amp sounds GREAT on the EF86 side....warm, articulate, bright when needed. SUPER COOL!

    HOWEVER, the 12AX7 side has some issues when the preamp volume (gain) is turned up past 3/4 of the way up. Screechy-squealy-phaseshifty nastiness....bummer, as I'm so closed to being finished that I can taste it!

    I'm pretty sure the tubes are not microphonic. I've tried them in other amps, and have had no problems. Also, I have tapped on them and there are no signs of microphonics.... My instinct says that the Tung Sol 12AX7s should be just fine (Top Hat has been using them in their King Royale amp with great success, and my amp is very similar to the King Royale...and the tubes came from a reputable dealer who tests them thoroughly), so I'm inclined to consider the layout to be a more likely suspect. The cathode follower 12AX7 seems to be the most sensitive, as if I install an EH 12AX7 in place of the Tung Sols the oscillation is better. It seems that the oscillation is connected to either higher gain or more top end, so that makes sense that the brighter, gainier Tung Sol 12AX7 would exacerbate the problem.

    I did some work on it to clean up some things that I didn't get to in the initial build phase, such as:

    I ran some shielded wire from the master volume pot to the input cap on the PI, moved/cleaned up some grounds in the tone stack section, re-ran some eq wires, and straighted up some of the shielded input cables. Now, the noise is even more present and obnoxious than before. So, perhaps I'm on the right track but headed in the wrong direction. TOTALLY frustrating...

    So, here is what I've found to be the case now:

    1) The EF86 channel sounds absolutely KILLER. Perfect. Volume and tone knobs, and that's it. No troubles, no nothing, just great sounding tone for days. Crank it or keep it low, it is GOOD GOOD GOOD!

    2) The 12AX7 Master Volume side sounds great, too, until you crank the (500k) volume pot (gain) up past 3/4 open, then it gets phase-shifty and thin, and the squeal-scratch-oscillation starts happening on the top end. Gain down, no issues...

    3) Turning the Cut and Treble knob down helps tame the squeal a bit, but the pot that has the most affect on reducing the squeal is the MIDRANGE pot. Interesting... Turning the mids all the way down almost takes care of the problem, though a little bit is still present. Don't know how to interpret that...

    4) When the amp is in high power (switch in or out a 121 ohm 10w from between the recto and the fuse to either lower or raise the voltages downstream) and high filtering modes (32uf instead of 16uf in various places...), the squeal is more noticeable. Mostly, it is with higher voltages and with higher preamp filtering (1st gain stage...V1) that the squeal is more prominent, as raising the filtering in the power amp section doesn't affect the squeal at all.

    So, it seems like the higher voltages, higher preamp filtering, and higher midrange settings are the straws that break this oscillating camel's back. Personally, I'm considering shooting this camel in the head and burying its body in the back yard, but then I can tell that I'm THIS close to having a truly fine amplifier on my hands....so I'll hold off on that.

    Any suggestions, given the above? I've not tried switching the grids on the PI or the primaries on the OT yet, as my instincts say that the problem lies in the preamp somewhere.... Your help and input is greatly appreciated!

    Sorry about the long first post....THANKS SO MUCH!

  • #2
    Here are some pics I took today, so all the latest re-wirings are represented. I'm considering maybe the wire off the plate of V1 may be the cause of the trouble...I'll get to chop-sticking now and see if moving a wire helps...

    Your input is greatly appreciated!
    Attached Files

    Comment


    • #3
      FIXED IT!!!!!

      It was the fact that I had a turret very close to the input for the second channel - the plate cap from the cathode follower was connected to this turret (you can see it in the pics...), and the the wire from that went off to the treble pot. Moving this wire even 1mm closer to the tip of my guitar cable (in the jack...) made an EXPONENTIAL difference (for the worse...), so I knew that was it. I've temporarily re-wired that cap to another spare lug on the strip (that was way farther away...), and jumpered the wire to the treble pot, and BINGO, no squeal!!!

      However, the gain is a bit edgy still, so I'll have to tweak some more.

      THANKS THANKS THANKS!!!!

      Comment


      • #4
        Nice to see pictures of what is being discussed.
        That's some nice looking workmanship David.
        Regards,
        Tom

        Comment


        • #5
          Originally posted by Tom Phillips View Post
          Nice to see pictures of what is being discussed.
          That's some nice looking workmanship David.
          Regards,
          Tom

          Thanks. I'll have to post updated pics showing what was changed, and why. That should provide a nice resource for others who might face this issue in the future...

          Comment


          • #6
            Maybe ?

            WOW that wiring is a neat as a pin, as well as the layout.

            Great build

            A project build I done some time ago got ugly when the gain was cranked.

            I looked with pain staking detail for wiring mistakes, since the layout
            was split on two gromet boards.

            Eventally I resorted to introducing a feedback loop in 2nd triode O/P
            to the divider stage attunuating 1st to 2nd. Use a 1 meg pot and
            find the setting that works.

            To much feedback will diminish the cascade eventally

            Comment


            • #7
              Yeah, troubleshooting is the worst...but it's worthwhile once all is well and working. So far, the amp is sounding really good. I had the opportunity to compare it to a Vox AC30cc and a couple Top Hats and I was very pleased with the results....

              Comment

              Working...
              X