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.

Planning Stages 6V6 6G4a Super

Discussion in 'Shock Brother's DIY Amps' started by jmp81sc, Oct 22, 2020.

  1. jmp81sc

    jmp81sc Tele-Meister

    Posts:
    314
    Joined:
    Feb 6, 2015
    Location:
    Orange County
    Hi Everyone,
    I am planning my next project. This will be my 5th tube amp build and my most ambitious to date. I am hoping to plan this one with all of the collective knowledge here. I am trying to become more knowledgeable about design etc., and only order parts one time, ha ha.

    I want to build a 6G4a Super with the harmonic tremolo. I will use 6v6 power tubes because I already have the PT and OT and 20 watts or so is all that I need.

    I am going to use a Hammond S6 chord organ as the donor. PT and OT support 2 6V6 tubes in a fixed bias with a bunch of 12Au7's so it will support the tube compliment for a 6G4a Super, and it has a 4ohm OT with 2 8ohm 10" speakers. I used these same transformer types for a 5E3 that turned out good.

    A poster at TGP, LeonC, posted his 6G4 super that has a 6G3 bright channel instead of the 6G4 normal channel and it sounds really good to me so I am going to copy his idea. The 6G4a has an unused 12Ax7 triode so I am planning on adding a switchable gain stage to the 6G3 side using the left over 1/2 of the tube. Unfortunately his build photos do not show up, nor did he post a schematic.

    Attached is a schematic I cobbled together using the original Fender schematics. The Hammond schematic for the PT I am using shows 285 -0 - 285 secondaries. I plugged it in with no rectifier tube and measured 325-0-325 so with a load I'm thinking it will be about 300-0-300 with 120VAC from the wall. Does that make sense?

    This just happens to be the same as the 5G9 original schematic, so I used that for the power supply. I added negative feedback and a presence control from the 6G4a.

    My other question is how Hammond got -30v from the B- side (schematic also attached). That would be great for the bias supplly, but when I measure the B- it acts just like a conventional PT center tap like the schematic shows. On my 5E3 build I grounded that B- and it works great like that.

    My initial idea is to use all of the stock resister values and see where the voltages end up. The harmonic trem uses some pretty hi voltages. I have no idea how sensitive it is to voltage changes.

    Thanks
    John

    Figure6 (2).jpg Figure6 (3).jpg

    super 5g9 6g3.gif.png
     
  2. tubegeek

    tubegeek Friend of Leo's

    Age:
    60
    Posts:
    3,879
    Joined:
    Jan 31, 2020
    Location:
    Brooklyn, NY
    The Hammond bias scheme, with a resistor between the (-) end of the 1st filter cap and ground, won't seem to work unless you have something drawing current from the power supply. If you're testing without anything connected to the power supply it'll look like the bias supply's not working. I saw this on a AO-15 I've started messing with.

    At first it made no sense to me but the idea is that the current drawn by the tubes that are connected to the B- supply develops a voltage drop across the bias resistor (actually, a string of resistors in your schematic but no difference.) Since one end of the resistor is grounded, and the other has a voltage drop, it is negative with respect to ground.

    Weird but kinda cool, really.
     
    Last edited: Oct 22, 2020
  3. jmp81sc

    jmp81sc Tele-Meister

    Posts:
    314
    Joined:
    Feb 6, 2015
    Location:
    Orange County
    Hi Tubegeek
    Thanks for the explanation, I will have to get my head around this and figure out how to implement my bias circuit. This is definitely going to be a learning experience.
     
    tubegeek likes this.
  4. tubegeek

    tubegeek Friend of Leo's

    Age:
    60
    Posts:
    3,879
    Joined:
    Jan 31, 2020
    Location:
    Brooklyn, NY
    The one drawback I see with this scheme is, it lowers the B+ available for the preamp circuit compared to other bias schemes. Cathode bias would basically work out the same as this for the power amp, for any particular power transformer. But the high voltage node is kind of offset by the bias voltage, relative to ground, so the following power supply taps are based off the dropped voltage so they're starting off lower. The dropper resistors will be smaller than if the (-) end of the power supply was at 0 V the way we are used to seeing.
     
  5. elpico

    elpico Tele-Afflicted

    Posts:
    1,148
    Joined:
    Sep 14, 2011
    Location:
    Vancouver BC
    Just use a fender style bias circuit. The Hammond one is a bad design.
     
  6. SnidelyWhiplash

    SnidelyWhiplash Friend of Leo's

    Posts:
    2,673
    Joined:
    Jun 18, 2009
    Location:
    Hoggtown, KY.
    You are a much braver soul than I, OP. Hope everything goes well.
     
  7. jmp81sc

    jmp81sc Tele-Meister

    Posts:
    314
    Joined:
    Feb 6, 2015
    Location:
    Orange County
    I think it will be easiest to use a tap off of the high voltage for the bias circuit. That is what is shown on the 6G3. I'll work up another schematic showing this.
    Thanks for your input everyone.
    John
     
    tubegeek likes this.
  8. tubegeek

    tubegeek Friend of Leo's

    Age:
    60
    Posts:
    3,879
    Joined:
    Jan 31, 2020
    Location:
    Brooklyn, NY
    It seems to me that there is indeed a potential reliability problem: when the filaments are warming up, there's no negative bias on the power tubes' grids.

    On the other hand, Hammonds are built like tanks and seem to work forever, so I am not in a hurry to call their designs flawed - I want to know more about this, myself.

    I do have my doubts, though. @elpico was that the reason you don't like the Hammond scheme?
     
  9. tubegeek

    tubegeek Friend of Leo's

    Age:
    60
    Posts:
    3,879
    Joined:
    Jan 31, 2020
    Location:
    Brooklyn, NY
    But that power transformer doesn't have the dedicated bias winding the 5G3 does. So you'd have to drop down a higher voltage and burn some heat in a resistor to get a "normal" bias voltage using the 5G3 setup - I'm not saying one is better or worse, just that the 5G3 has a specific winding that is optimized for that setup and the Hammond doesn't.

    I don't want to seem like I'm advocating for Hammond's system: I've never seen it in any amp other than this one and the AO-15, and never on a guitar amp. I'd just like to know more about it, based on the fact that Hammond amps are generally reliable and high quality.
     
  10. elpico

    elpico Tele-Afflicted

    Posts:
    1,148
    Joined:
    Sep 14, 2011
    Location:
    Vancouver BC
    The bias supply should be the most stable, independent, bulletproof supply in the amp. It should be derived in a way that ensures it's always present when the amp is on, and just as importantly, always stays at the correct voltage.

    The bias voltage in this circuit is dependent on how much current the power tubes are drawing. If a tube fails, or the user removes one, the bias voltage is reduced and the remaining tube burns a little hotter. To be fair it would only go up a watt or two because it's somewhat self correcting like cathode bias, but why accept that when the cost of the few diodes and caps you need to make an proper supply is so insignifficant?

    edit: actually it might be more than a couple watt increase, I haven't done the math but you get what I mean
     
    Last edited: Oct 22, 2020
    D'tar and tubegeek like this.
  11. jmp81sc

    jmp81sc Tele-Meister

    Posts:
    314
    Joined:
    Feb 6, 2015
    Location:
    Orange County
    The 6G3 does not have a dedicated winding for the bias supply, it taps off of the high voltage and has a 100k resister in front of the diode. Rob Robinette also has this on his page attached below is the diagram from Rob's web site.

    It's seems to me it will be easier to do it this way, but I am open to suggestions.
    Thanks
    John

    High_Voltage_Bias_Tap.png
     
  12. tubegeek

    tubegeek Friend of Leo's

    Age:
    60
    Posts:
    3,879
    Joined:
    Jan 31, 2020
    Location:
    Brooklyn, NY
    I do, and your objection is exactly what concerned me. I would loveto know what Hammond was thinking there, it does seem inherently a mistake.

    Aha - I was looking at what I see now was the 5G9 Tremolux portion of the schematic that was a composite of a couple different circuits, not the actual 6G3. My bad.

    The bias supply draws so little current that my quibble about heat in the resistor is a pointless one, forget I said anything. I do agree with everyone who said just use the Fender setup, for all the same reasons.

    And that @robrob layout is nice and tidy, too.
     
  13. NTC

    NTC Tele-Meister

    Posts:
    204
    Joined:
    Mar 17, 2003
    What I don't like about the Hammond setup is that all of the current seems to flow through that resistor chain. It is goofy cathode bias of the output in a way. It will also limit headroom. As for a warm up issue, the voltsges will be more negative as the tubes start to conduct, so the should not be pumping too much current through themselves.

    I like the idea of the project and look forward to updates.
     
  14. jmp81sc

    jmp81sc Tele-Meister

    Posts:
    314
    Joined:
    Feb 6, 2015
    Location:
    Orange County
    Thanks everyone for your comments. Below is a the schematic with the Bias supply as discussed. I labeled the voltage dropping resister at 100K but will probably need to adjust this to get the correct bias supply voltage. I didn't put the adjustable bias pot in the schematic but I will implement it for the build.

    I used the 5G9 power section as a starting point because I think the voltages will be similar to what my PT will put out, and it is similar to the 6G4a - fixed bias, long tail pair phase inverter.

    I'm not sure if I am brave or stupid to try this ha ha. It will be a nightmare to debug if it goes wrong. I will really need to build it in stages and test as I go.

    I am starting to do a DIYLC layout, but it is the first time using that program so I am learning as I go.

    super 5g9 6g3 bias 2.gif.png
     
    tubegeek likes this.
IMPORTANT: Treat everyone here with respect, no matter how difficult!
No sex, drug, political, religion or hate discussion permitted here.