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.

AB763 distortion wars

Discussion in 'Shock Brother's DIY Amps' started by RollingBender, Oct 6, 2020.

  1. RollingBender

    RollingBender Tele-Afflicted Vendor Member

    Posts:
    1,408
    Joined:
    May 14, 2011
    Location:
    SW Minnesota
    I built this...
    1C051E20-73B7-41D4-9C10-6714BEB01DD0.jpeg
    Here’s the build thread...
    https://www.tdpri.com/threads/seems-like-i’m-doing-this-backwards.1015555/

    It makes the most glorious clean sounds you can imagine. Cranked, not so much. I have been literally (in the true definition of the word) using every spare minute to read everything about the AB763 circuit and some little known issues that are commonly misdiagnosed. I’ve been sticking my scope probe into this one for a few days to come up with a game plan for fixing the bad sounding distortion without messing with the clean character at the other end of the dial.

    One of the biggest symptoms is this odd shaped scope shot...
    CFD2C9F0-A071-45AF-846E-6A3B0DF7C525.jpeg
    This is at the speaker jack with a 1k sine wave fed in. It gets nastier the farther up you turn the knob. It goes back To a normal looking wave If I pull the (1-tube) reverb tube.

    Before I get as deep into the circuit as the reverb, I want to address what appears to be grid clipping at the second gain stage. The first gain stage keeps a nice looking sine wave all the time. The second stage wave top gets flatter the more the volume is turned up. The top of the wave never moves...the bottom does what is expected.

    To me, it looks like the AB763 uses the volume pot as the stage II grid resistor. As an experiment, I grabbed a couple loose resistors and added a grid stopper and grid leak resistor in front of V1b grid as I scratched onto the schematic below...
    349C476E-BE38-42EE-8DB6-3C3F86B02255.jpeg
    The addition of these parts makes the scope results look much better and it does make the amp’s cranked distortion sound better (departing from the realm of full on blocking distortion) but it does pull down the overall volume a bit.

    Is there a good/better/best combination of values for this mod?
     

    Attached Files:

    Hey_you and ping-ping-clicka like this.
  2. Lowerleftcoast

    Lowerleftcoast Friend of Leo's

    Posts:
    2,653
    Joined:
    Dec 30, 2019
    Location:
    california
    Rob does something similar with a 1M pot for the blocking distortion of a 5E3. You may be able to determine some values if the pot isn't quite getting to a sweet spot.
     
  3. Tom Kamphuys

    Tom Kamphuys Tele-Holic

    Age:
    41
    Posts:
    630
    Joined:
    Sep 18, 2018
    Location:
    The Netherlands
    I'm curious how the reverb tube removal fixing the problem is related to your fix...

    Do you have a (full) schematic for us?
     
    ThermionicScott and D'tar like this.
  4. jsnwhite619

    jsnwhite619 Friend of Leo's

    Age:
    36
    Posts:
    3,353
    Joined:
    Sep 10, 2013
    Location:
    Georgia
    Without going through the build thread, I'll say that a buddy of mine who tech's & restores on the side - but knows way more than I do - said he's learned that the stock NFB values seem to be pretty critical in the Blackface amps. He had a mint Vibrolux in there that he said sounded like crap. Apparently the reason it was so clean was because no one ever wanted to play it.

    He worked through it and found the NFB resistor was way off on the high side and looked like it left the factory that way. Harsh, spiked, top end that was painful all settled out with the stock NFB value. So, maybe double check that area if you haven't.
     
  5. andrewRneumann

    andrewRneumann Tele-Holic

    Posts:
    593
    Joined:
    Mar 22, 2020
    Location:
    Cincinnati, OH, USA
    Very cool and interesting post. I sense an interesting thread coming on. I think you might have similar results by just replacing the 1MA pot with a 250KA pot. A grid stopper is always recommended when trying to tame blocking distortion, but you may notice a drop in highs. You sort of have one foot on the gas with the bright cap on the volume pot and then one foot on the brake with a grid stopper. But if you like the sound, then who is to say what is the right way?

    Also, if you want to delay the onset of grid current clipping, bias the tube a little colder. Try a 1K8 or 2K2 on that V1B cathode and see if you like it. I would take some voltages and plot a load line before doing this though.

    Finally, a partially bypassed cathode can really make a big difference in distortion tone. Instead of fiddling with the grid leak and stoppers, try something smaller for the bypass cap. Start with 1uF and work your way up.

    Good luck, hope this helps.
     
  6. ping-ping-clicka

    ping-ping-clicka Friend of Leo's Ad Free Member

    Age:
    72
    Posts:
    4,267
    Joined:
    Jun 28, 2019
    Location:
    left coast
    good on you ,looks great sounds great you might try a dirt box, however I appreciate the dignity of your quest.
     
  7. RollingBender

    RollingBender Tele-Afflicted Vendor Member

    Posts:
    1,408
    Joined:
    May 14, 2011
    Location:
    SW Minnesota
    Two separate problems with the same symptoms...blocking distortion. The odd thing with the reverb shows up on the scope with or without the reverb on or turned up.

    I just recently got the scope but it sure is a great tool for problem diagnosis. Rather than throwing parts and values at the problem, the real problem(s) can be diagnosed and solved.


    @jsnwhite619 NFB is also on the list.
     
    Last edited: Oct 6, 2020
    jsnwhite619 likes this.
  8. andrewRneumann

    andrewRneumann Tele-Holic

    Posts:
    593
    Joined:
    Mar 22, 2020
    Location:
    Cincinnati, OH, USA
    Can you remember where your probe was when you observed this?
     
  9. RollingBender

    RollingBender Tele-Afflicted Vendor Member

    Posts:
    1,408
    Joined:
    May 14, 2011
    Location:
    SW Minnesota
    Grid of V1b. Right after the tone stack.

    edit...scratch that...grid of V2a (reverb driver)

    maybe
     
    andrewRneumann likes this.
  10. andrewRneumann

    andrewRneumann Tele-Holic

    Posts:
    593
    Joined:
    Mar 22, 2020
    Location:
    Cincinnati, OH, USA
    Try swapping tubes. Different tubes can experience grid current onset at different points. Some as early as -1V Vgk, others close to 0V Vgk, from what I’ve read.
     
  11. andrewRneumann

    andrewRneumann Tele-Holic

    Posts:
    593
    Joined:
    Mar 22, 2020
    Location:
    Cincinnati, OH, USA
    Ahh so this might be cut-off clipping since this signal is inverted? In that case, we might bias the tube warmer with a smaller cathode resistor.

    I’m not so sure you can detect blocking distortion with a constant source signal. I’d keep an open mind as to what you are observing.
     
  12. RollingBender

    RollingBender Tele-Afflicted Vendor Member

    Posts:
    1,408
    Joined:
    May 14, 2011
    Location:
    SW Minnesota
    Did some messing around with a resistor decade box and found a nice balance with the 68k grid stopper and a 680k grid leak. Tamed the distortion at that point without taking the volume down much.

    Onward through the circuit.

    ***”Seeing” things going on with the scope makes me wonder how many amps are out there with some sort of oscillation outside of the hearing range that is robbing tone???
     
  13. Tom Kamphuys

    Tom Kamphuys Tele-Holic

    Age:
    41
    Posts:
    630
    Joined:
    Sep 18, 2018
    Location:
    The Netherlands
    I've read that blocking distortion happens at the end of a high signal burst. During the burst you get bias excursion, which starts recovering after the burst. During recovery you can get blocking distortion.
    The scope image does not look anywhere close to the text book examples of blocking distortion I've seen.
     
  14. ThermionicScott

    ThermionicScott Poster Extraordinaire

    Age:
    40
    Posts:
    6,328
    Joined:
    Dec 3, 2005
    Location:
    CID
    Don't forget that the right side of the tone stack also helps act as the grid resistor for V1B. With all the tone controls up, that's 510K in parallel with the volume control, so there's a maximum grid resistance of ~337K, minimum of 0 (short to ground). And if the volume control is at less than 100%, there is some resistance in series with the grid. In combination with all the gain loss from the tone stack, it's hard to believe that blocking distortion would be happening at V1B. Most likely, you're just reducing system gain and that's making the amp sound better due to something else down the chain.

    If you're concerned about lopsided clipping at that stage (which *is* possible), I think Andy Neumann is right that you could just adjust the bias a bit. But some high-gain amps deliberately bias gain stages off-center for tone... ;)


    Left-field thoughts: For the bad distortion, do you ever hear any oscillation with the amp turned up? Any chance that your NFB is hooked up to the wrong side of the primary? Just for debugging/curiosity, you might try disconnecting the feedback resistor and see how the clipping looks without it.

    I'm still mostly suspicious of that reverb driver tube. Maybe double-check all of its connections, ensuring that both grids have a ground path...
     
    Last edited: Oct 7, 2020
    King Fan likes this.
  15. Tom Kamphuys

    Tom Kamphuys Tele-Holic

    Age:
    41
    Posts:
    630
    Joined:
    Sep 18, 2018
    Location:
    The Netherlands
    ... but that also does not look like the scope image.
     
  16. ThermionicScott

    ThermionicScott Poster Extraordinaire

    Age:
    40
    Posts:
    6,328
    Joined:
    Dec 3, 2005
    Location:
    CID
    Yeah, the original post is confusing because he's talking about clipping at V1B, but the scope trace is from the speaker jack. It would have been better to separate those thoughts a little.

    P.S. Signal boost on seeing that full schematic... :)
     
    Last edited: Oct 7, 2020
    Tom Kamphuys likes this.
  17. RollingBender

    RollingBender Tele-Afflicted Vendor Member

    Posts:
    1,408
    Joined:
    May 14, 2011
    Location:
    SW Minnesota
    I haven’t drawn a schematic for this one...I did the layout (attached on post #1) in DIYLC as a cut-and-paste adventure from the Hoffman single channel AB763, and from a few other projects I’ve done.

    I’m going to return to square #1 and remove both V1b grid resistors for now.
    Probe on grid of V1b. 1k signal input for all tests.

    13CA3ECD-6646-4191-B8A8-04A706EA9133.jpeg
    Amp volume at 3/10

    663C50D8-51ED-41AC-BC75-437B608E6447.jpeg
    Amp volume at 5/10

    370FEAB4-A4F8-4956-B1C3-B9B934B1B52A.jpeg
    Amp volume at 8/10

    DD18BD12-00E1-469B-A724-04D13ED04591.jpeg
    Amp Volume at 8/10 with bright cap engaged???

    D0139D64-927F-405D-B051-FE75081C1C37.jpeg
    Amp Volume at 8/10 with reverb tube pulled.
     
    Last edited: Oct 7, 2020
  18. Tom Kamphuys

    Tom Kamphuys Tele-Holic

    Age:
    41
    Posts:
    630
    Joined:
    Sep 18, 2018
    Location:
    The Netherlands
    V2b is the reverb driver in the schematic you posted. Is that where you measured?
     
  19. RollingBender

    RollingBender Tele-Afflicted Vendor Member

    Posts:
    1,408
    Joined:
    May 14, 2011
    Location:
    SW Minnesota
    V2b is reverb recovery, V2a is the driver (1-tube reverb).

    Probe shots in post #17 all taken at V1b grid.
     
  20. andrewRneumann

    andrewRneumann Tele-Holic

    Posts:
    593
    Joined:
    Mar 22, 2020
    Location:
    Cincinnati, OH, USA
    Was it common for these kind of amps to have tremolo, reverb, and power tube screens all powered from the same power supply node?
     
IMPORTANT: Treat everyone here with respect, no matter how difficult!
No sex, drug, political, religion or hate discussion permitted here.