1. Win a Broadcaster or one of 3 Teles! The annual Supporting Member Giveaway is on. To enter Click Here. To see all the prizes and full details Click Here. To view the thread about the giveaway Click Here.

Reverb squeal / feedback

Discussion in 'Amp Tech Center' started by timbraun, Jul 10, 2013.

  1. timbraun

    timbraun TDPRI Member

    Posts:
    88
    Joined:
    Apr 5, 2011
    Location:
    Canada
    I had a chance to do a little work on this yesterday, cleaned the pots and sockets, checked the bias caps. Bias caps are in correctly, and the bias voltages on the triodes all look good - 1.47 V on the amp stages, 8.7 V on the reverb driver.

    The squeal is still there. I'm going to put a 12au7 in the reverb drive, see if that helps.
     
  2. timbraun

    timbraun TDPRI Member

    Posts:
    88
    Joined:
    Apr 5, 2011
    Location:
    Canada
    Nope, lower gain tubes didn't fix it. It did extend the usable range on the knob up to 5.

    Now here's something. The side of the reverb tank with a red paint blotch measures 11 ohms like a reverb input should. The other side measures 180 ohms which seems low, according to available replacements. What should the DC impedance measure?
     
  3. timbraun

    timbraun TDPRI Member

    Posts:
    88
    Joined:
    Apr 5, 2011
    Location:
    Canada
    The DC resistance on the reverb tank in my Traynor ysr-1 measures about the same as the tank in the DR, so it doesn't look like the tank is busted. I took the tank out of the DR and it looks clean.

    I'll try a tank swap but I don't have much hope for that.

    Lead dress looks good in the DR. I'll whack around with a chop stick and see if there are cold solder joints.
     
  4. timbraun

    timbraun TDPRI Member

    Posts:
    88
    Joined:
    Apr 5, 2011
    Location:
    Canada
    jguitarman mentioned a couple of threads in another forum which proved interesting:

    http://music-electronics-forum.com/t5928/
    http://music-electronics-forum.com/t11676/

    Last night I did the reverb tank swap. The squeal was still there at the same reverb setting.

    I then scrounged up a way to plug a switch into the RCA jack and turned reverb off, as suggested in one of the threads above. The squeal is still there.

    Well, that was surprising, and does eliminate a lot of circuitry as the problem source.

    While I was into switches, I tested the tremolo circuit. It works, though the slow setting is still a little fast, IMHO. A question of taste.

    Back to Reverb! I checked the bias voltage on the shared 820 Ohm resistor, 1.47 V. The cap is the right way around, and it's new. The impedance of the resistor is 890 Ohms, a little high but within 10%.

    Other than the resistor / cap on the cathode of the 7025 V4, what could be... what if the 220K resistor to ground didn't go to ground?

    Just ramblin' out loud, making notes for myself. :)
     
  5. timbraun

    timbraun TDPRI Member

    Posts:
    88
    Joined:
    Apr 5, 2011
    Location:
    Canada
    I found a resistor that was out of spec, that increased the maximum impedance seen by input of the gain stage that oscillates. And the bias on the tube is on the high side. So I replaced the resistors and the oscillation didn't start until the reverb setting was 5. That's an improvement.

    I swapped a 12au7 in place of the 12ax7 and that allowed full range on the reverb knob without squealing! Time to give it a listen.

    I did try a 12at7 but that was unstable, too.

    Anyway, after bolting the reverb tank back into the cabinet and putting the chassis back in, I was able to play the amp for a bit. Actually I had to practice. :)

    The amp sounds pretty good. Normal channel is okay and Vibrato channel has some groovy reverb available. All clean up to 5 or so, after which the special channel gets an ugly bit of distortion - more like a speaker breaking up than tube overdrive. The normal channel doesn't have that problem. I guess more carbon comp resistors are due for replacement.

    This is more time consuming than I want right now.
     
  6. timbraun

    timbraun TDPRI Member

    Posts:
    88
    Joined:
    Apr 5, 2011
    Location:
    Canada
    I swapped the 12au7 back out for a new 12at7 and the vib channel sounds clear of that nasty noise it had. Played the amp for a while, with a tele, then a bit with a Les Paul. It sounds good. It doesn't break up like my 69 Bandmaster, though.

    Between gain questions and the lingering reverb problem, I do believe there is still an issue with the amp.
     
  7. bwacke

    bwacke Tele-Meister Gold Supporter

    Posts:
    368
    Joined:
    Oct 28, 2007
    Location:
    Toledo, OH
    Tim,
    If you put a 12AU7 or 12AT7 in the circuit set up for a 12AX7 you won't get anywhere near the overdrive voltage the amp is capable of when tubed properly. Try putting a 12AX7 back in the reverb recovery/final preamp spot and see what happens.

    Which resistor did you find out of spec? 470K? 220K?

    Bob
     
  8. timbraun

    timbraun TDPRI Member

    Posts:
    88
    Joined:
    Apr 5, 2011
    Location:
    Canada
    It was the 220K in parallel with the reverb control that was measuring about 260K. That allowed too much impedance on the grid of V4B when the reverb was turned up.

    Last night Don Voth looked at it and he split the cathode biasing for the reverb recovery amp and reverb mixer stages (V4 A and B). That allowed stable full range of reverb control with a 12ax7 in place. Whoo hoo!

    It annoys me that we need to mod the circuit to get it stable. We used 1500 Ohm biasing for each stage, with a 25 uF bypass cap each.

    Now... the overall gain in the amp is still not what it should be. What should the voltage gain on the P.I. stage be? The preamp stages are pretty much what you can get from a 12ax7 - about 40 dBV (100x).

    According to http://www.ampbooks.com/home/amplifier-calculators/long-tailed-pair/ we should see 28 - 30 dB. We were seeing more like 3 dB... hmmmm.

    I like our progress. Thanks for the interest, Bob!
     
  9. timbraun

    timbraun TDPRI Member

    Posts:
    88
    Joined:
    Apr 5, 2011
    Location:
    Canada
    The long-tailed-pair calculator I used ignore the negative feedback loop effects. If you add in the 10% NFB that Blackface circuits generally have, you get a voltage gain including the output tubes of g / (1 + NFB*g) = 7.75, for NFB = 0.1 and g = 30.

    But that would be gain for both PI and power tubes. I'm going to open the NFB loop to see whether the PI stage has appropriate gain.
     
  10. timbraun

    timbraun TDPRI Member

    Posts:
    88
    Joined:
    Apr 5, 2011
    Location:
    Canada
    More on the continuing saga... On Monday I took a scope home and opened up the NFB loop. Didn't believe what I saw on the 'scope. Literally. It did not look like a sensible waveform coming out of the PI at all. I thought I'd blown the scope lead, so I reconnected the resistor and shut 'er down for the night.

    Last night, back at iTronix (Don Voth) we tried the same thing, and his scope showed the same unstable wave form. So opening the feedback loop makes the amp unstable. Not too surprising.

    I was able to borrow another BFDR and we compared waveforms at the various gain stages and circuit nodes. My amp starts showing loading issues at the output of the reverb recovery amp. If we lift the vibrato connection the input to the PI looks okay. If we lift the power tubes, the output from the PI looks okay. The capacitors aren't leaking DC, the DC levels are sensible.

    We checked power tube biasing, and it's very nice at 30 mA. (29.8 & 30.0 mA)

    Each stage looks okay on it's own, but together... not as much gain as there should be. By a factor of 10 or so. Quite the unusual problem.

    Another session or two will be required.
     
  11. Chritty

    Chritty Tele-Afflicted

    Posts:
    1,227
    Joined:
    Nov 26, 2012
    Location:
    Newcastle, Australia
    Thanks for the update.
     
  12. bwacke

    bwacke Tele-Meister Gold Supporter

    Posts:
    368
    Joined:
    Oct 28, 2007
    Location:
    Toledo, OH
    Tim,

    "If we lift the vibrato connection the input to the PI looks okay."

    Sounds like the LDR unit is shorted or very low resistance. You could temporarily replace it and the intensity pot with a 47K or 68K resistor to eliminate the LDR and pot to see if levels and operation return to normal. But check the DC level to ground on the high side of the intensity pot first. If it's not zero volts, the 0.1 uF plate blocking cap may have shorted, blowing the LDR and/or pot to a short.

    "My amp starts showing loading issues at the output of the reverb recovery amp."

    In light of the other statement, this doesn't exactly jive. The Reverb Recovery is on the input side of the final stage of Vibrato channel. It wouldn't be "loaded" by the LDR on the output side of the final Vibrato Channel stage. What could affect the recovery level is a failure of a component in the 3.3 meg/10 pf network or the 0.003 uF/100K pot/470K-220K attenuator, which are both connected to the grid of the final preamp.

    What do you think?

    Bob
     
  13. timbraun

    timbraun TDPRI Member

    Posts:
    88
    Joined:
    Apr 5, 2011
    Location:
    Canada
    We got it! It was a bad solder joint on ground connection to the tail off the PI. Effectively increasing the negative feedback ratio and reducing gain in the PI and power stage. Same connection as the cathode bypass for the reverb return.

    The old carbon comp resistors are still within spec. The ceramic disks are okay, too.

    Finally the amp sounds like it should! Tomorrow I get to wind it up and check it out. Maybe I'll use it this Sunday...

    I do still have to get the grommet onto the three prong power cord I put in, but that is very minor. :)

    Thanks for the ideas and encouragement everybody!
     
  14. Chritty

    Chritty Tele-Afflicted

    Posts:
    1,227
    Joined:
    Nov 26, 2012
    Location:
    Newcastle, Australia
    Yay. I love a happy ending!
     
  15. timbraun

    timbraun TDPRI Member

    Posts:
    88
    Joined:
    Apr 5, 2011
    Location:
    Canada
    Ooh, this is a nice amp. The more I play it the more I like it. :) Happy ending indeed.

    The old Utah speaker sounds good, lots of bottom.
     
  16. Lowellg

    Lowellg NEW MEMBER!

    Age:
    41
    Posts:
    1
    Joined:
    Feb 4, 2020
    Location:
    Pensacola, fl
    Foxy229 nailed it... Just fixed this exact issue by replacing the V4 bypass cap.
     
IMPORTANT: Treat everyone here with respect, no matter how difficult!
No sex, drug, political, religion or hate discussion permitted here.