Help with resistor on SE *parallel* power tubes

Discussion in 'Amp Central Station' started by King Fan, Jul 19, 2019.

  1. King Fan

    King Fan Friend of Leo's Ad Free Member

    Posts:
    3,787
    Joined:
    Jan 1, 2013
    Location:
    Salt Lake City
    I'll attach the schematic for the 1955 Gibson GA-9 I’m restoring. Most of the resistors test within 10%, but the 470 has drifted up to 750. It's sitting on a nice hot power tube socket, so some cook-age is not unexpected. But: Is it a grid stopper? If so, maybe a bit higher might not hurt? But these are single-ended *parallel* 6V6s -- so should their setup be as symmetrical as possible? If so, would it be bad if the drift made them more asymmetrical? Way too many things I don’t understand here. :confused:

    FWIW I’ve been marking up the scheme to increase readability.

    IMG_2615.jpeg

    ga9 markup.png
     
  2. D'tar

    D'tar Friend of Leo's

    Posts:
    2,090
    Joined:
    Jan 11, 2013
    Location:
    WNY
    KF, that looks 47k to me? Whats it gozinta and comoutta?

    Is it R10 maybe?

    Edit: I wouldn't worry about +/- 10%. Anything grossly out of spec would be considered for replacement. As far as symmetry goes, close is good enough. Too close may be not as good:)
     
    Last edited: Jul 19, 2019
  3. FenderLover

    FenderLover Friend of Leo's

    Posts:
    4,730
    Joined:
    Jun 11, 2009
    Location:
    Minnesota
    I would wonder if the schematic is correct, and if it is really wired that way. Gibson did some odd things. Note the 220K to ground at the input jacks. That's what they did. My 1948 GA-50 had 100K on 3 of the inputs for the second 6SJ7. The first stage had 1M.
     
  4. King Fan

    King Fan Friend of Leo's Ad Free Member

    Posts:
    3,787
    Joined:
    Jan 1, 2013
    Location:
    Salt Lake City
    Sorry, D'tar. I shoulda marked up the schematic. I see what you mean about the colors. Without the flash, I saw it as yellow/violet/faded brown (not orange), and it measures 0.750k, *and* there's a big 1w 47K to be R10 on the left 6V6 socket (the one nearer the preamp). So I*think* it's R9, and it seems to go from pin 6 to pin 5 on the right 6V6. But as far as I can tell there's no lead to the pin 6 end. It's hard to see down under the buttload of wires there. Maybe it's put together wrong, or a lead was removed, or...

    As you say, I'm comfortable with ± 10%, but was less so with 750/470 (or 750/47,000 for that matter). :) I do like your note that if it's meant to be 470 and is 750, it's gotta go. I just wonder what it's doing there.

    Thanks, FL, excellent. That did occur to me. Gibson may not have followed (or updated, same diff) their schematic. I guess, as you know better than me, they were infamous for that. This is my first dive into PTP, and the topology that results from using 3 octal sockets (with 2 different pinouts) to replace every eyelet in an entire amp makes decoding the 'gozinta and comsouta' *slightly* challenging.

    I'm outta town for a few days, and will investigate more when I get home. Given the location on a hot socket, I'll replace it -- and maybe with a fireproof MO. But I wish I knew what it did.
     
  5. dan40

    dan40 Tele-Afflicted Silver Supporter

    Posts:
    1,236
    Joined:
    Aug 19, 2015
    Location:
    Richmond Va
    Pin 5 would be a control grid stopper. Pin 6 is a convenient spot to attach the resistor but it would need a lead from r7 (500k volume pot) for the amp to function properly.
     
  6. Mexitele Blues

    Mexitele Blues Tele-Meister

    Age:
    41
    Posts:
    359
    Joined:
    Mar 7, 2018
    Location:
    Westminster, CO
    My Gibsonette GA-8, also parallel SE, has a similar 470 ohm grid stopper on the 2nd 6V6 only. It is mounted over the v2 socket, then wired to the v3 grid.
     
    FenderLover likes this.
  7. King Fan

    King Fan Friend of Leo's Ad Free Member

    Posts:
    3,787
    Joined:
    Jan 1, 2013
    Location:
    Salt Lake City
    Thanks, Dan, that makes sense and it's how I was reading the scheme. But I really appreciate having it confirmed. There are a scad of yellow wires (it looks like at one time these were checked with white, green, black, orange, etc.) all running together between the sockets, and where they're attached, they form big piles of wire and solder that tend to hide the connections to the lonelier lugs. And it doesn't help that the Gibson builders were excellent at soldering the ends of wires to the lugs nowhere near the hole that I lazily use. No doubt this helped them fit six wires or more on a single lug!

    When I get home I'm gonna do a better job of confirming those are pins 6 and 5 -- and that there is really no sneaky wire to the 'dead-end' pin.
     
  8. King Fan

    King Fan Friend of Leo's Ad Free Member

    Posts:
    3,787
    Joined:
    Jan 1, 2013
    Location:
    Salt Lake City
    Thanks, Mexitele, that's very helpful. I had heard the GA-8 did that. As I say, I'm gonna get home and use my ohmmeter and a chopstick and a bright light and some magnification and figure out if and how this one is wired.

    By V2 in these amps do we mean the one next to the rectifier? I'm used to that, working R to L in Fenders, and that's where we see the resistor in this amp. But if we go preamp -> power -> rectifier, then I guess V2 would be the one next to the 6sj7.
     
    Last edited: Jul 19, 2019
  9. King Fan

    King Fan Friend of Leo's Ad Free Member

    Posts:
    3,787
    Joined:
    Jan 1, 2013
    Location:
    Salt Lake City
    My amp-sensei brother found this (post 2) in a forum where 'Mark B.' noted that power tubes in parallel can create Miller capacitance problems. Although they're talking about a 3x6V6 circuit (!) he says, "if you don't have grid stopper resistors fitted, it may be wise to fit them to at least the 2nd & definitely the 3rd 6V6 (1.5K - 4.7K)."

    So *if* I can confirm that 470 --> 750R is actually wired into the circuit correctly, I thought maybe I'd leave it in there and see how it works. If Mark B. is correct, it may be there to tame Miller effects... and maybe asymmetry is not an issue.

    But the internet is so full of a number of things. I then found this on another forum by 'Ken Gilbert' suggesting Miller capacitance isn't a big deal in parallel power tubes but that grid interactions are the problem and may require grid stoppers.

    So we have at least two possible rationales for grid stoppers here. This is getting kinda fun... :)
     
  10. Mexitele Blues

    Mexitele Blues Tele-Meister

    Age:
    41
    Posts:
    359
    Joined:
    Mar 7, 2018
    Location:
    Westminster, CO
    I'm numbering from preamp to rectifier.
     
    King Fan likes this.
  11. King Fan

    King Fan Friend of Leo's Ad Free Member

    Posts:
    3,787
    Joined:
    Jan 1, 2013
    Location:
    Salt Lake City
    Of course, duh. A night's sleep helps. And double duh, it must be pin *5* that has no lead. So I'll just replace it with metal oxide. Luckily I think I can almost reach that resistor; the ones on V2 :) are walled off behind wires, caps, and a huge cathode resistor.

    But I'm gonna try to learn more about parallel SE power tubes and grid stoppers. And I have a new appreciation for boards, tag strips, eyelets, and turrets.
     
    Last edited: Jul 20, 2019
    dan40 likes this.
IMPORTANT: Treat everyone here with respect, no matter how difficult!
No sex, drug, political, religion or hate discussion permitted here.


  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.