Adjustable bias design resources and questions

Discussion in 'Shock Brother's DIY Amps' started by King Fan, Feb 11, 2019.

  1. King Fan

    King Fan Friend of Leo's Ad Free Member

    Posts:
    3,357
    Joined:
    Jan 1, 2013
    Location:
    Salt Lake City
    I've been reading and drawing out bias circuits recently, and wondered if folks can suggest other good sources for design and values as to pots, resistors, caps, and 'features.'

    1) A nice resource for *calculating* values and understanding design features is, not surprisingly, Merlin. He favors a 10kl pot:

    The total resistance of P1+R2 forms part of the power valves’ grid leak resistance, and there will be a maximum allowable value given in the data sheet. Since we will usually want to make the grid leak resistors as large as possible to avoid loading down the driver stage, the resistances used in the bias supply need to be small. A 10k bias pot is ideal.
    But he notes there are a zillion variants out there. One obvious difference arises if we have to borrow HT voltage from the rectifier socket (eg pin 4 or 6) or OTOH if our PT has a bias tap (often 50V).

    He has an elegant and simple method to calculate the maximum (negative) bias voltage you'll need, for each power tube type, based on screen voltage. He discusses the logic (with some trial and error) for sizing the 'tail' resistor (or as he says, 'idiot resistor') that prevents the pot from zeroing out the bias, and he shows a simple and elegant 'emergency' resistor that prevents meltdown if the pot should fail. He goes into helpful detail about how the bias scheme can be modified for different rectification schemes.

    2) Another great resource is of course @robrob -- his generic mods page shows nice examples of several generalized and specialized bias circuits, with his usual elegant diagrams. He includes *two* options for balancing the bias and an elegant 'universal' bias circuit for amps without a bias-tapped PT.

    3) Another good source is Doug Hoffman, whose bias page I've used a lot and seen used by other builders. The drawings there were clearly done in an earlier era of computer graphics, but they're especially helpful for the popular Fender bias-wiggle circuits, and include options for 10kl and 50kl pots. Doug also carries both 10kl and 50kl mini trim pots you can mount on any bias board, and the nice Fender-style CTS 10kl screwdriver pot that allows external access, plus those great test-point sockets. You'll never know what 'easy' is until you adjust bias without dropping the chassis. :)

    I've tried reading Aiken on bias, but although he explains in detail what it does in the tube, I couldn't find much info about designing bias circuits.

    There's still a lot I want to know. How much filtering is enough? One cap or two? What are the pros and cons of 10kl v. 50? What are the constraints on resistor value? Are there advantages of making the pot a variable resistor or a voltage divider? How does bias-wiggle trem change the design?

    So your thoughts and suggestions will be welcome.
     
    Last edited: Feb 11, 2019
    flyswatter and theprofessor like this.
  2. King Fan

    King Fan Friend of Leo's Ad Free Member

    Posts:
    3,357
    Joined:
    Jan 1, 2013
    Location:
    Salt Lake City
    Trying to make my questions a little less vague. Answer as few or as many as you want. :D
    1. What guides modern cap values and how many caps do we want?
    2. What are the pros and cons of 10kl vs. 50kl pots?
    3. Does it matter if you make your pot a variable resistor vs. a voltage divider?
    4. Do we need a different circuit or values to do (power-tube) bias-wiggle trem?
    5. What are sensible upper and lower limits of the total resistance here?
    6. Are there reasons to go with bigger or smaller resistors (once we knock down the HT on a PT without a bias tap)?

    Also, pics or it didn't happen, right? FWIW I'll attach my drawings of a Blencowe and Hoffman bias scheme, plus the stock 5G9 bias circuit and my tentative ideas for how to make it adjustable. Note Merlin's version here is assumed to come from a high-voltage source, eg pin 4 on the rectifier; the others come from bias taps.

    5G9 bias ideas.png
     
    flyswatter likes this.
  3. CoyotesGator

    CoyotesGator Tele-Holic

    Posts:
    506
    Joined:
    Jul 12, 2014
    Location:
    Gulf Coast
    This got me to thinking.

    I honestly can not remember what exactly I did on my last build.

    I looked back through my pictures.

    It is a through chassis 50K.

    In my pictures, I can see the pot with no wires, and I can see the board after it went in covering the wires.

    Great.

    Guess whom is pulling a board tomorrow?

    :lol:o_O:twisted::lol::lol::lol:
     
    King Fan likes this.
  4. robrob

    robrob Poster Extraordinaire Ad Free Member

    Posts:
    7,329
    Joined:
    Dec 29, 2012
    Location:
    United States
     
  5. King Fan

    King Fan Friend of Leo's Ad Free Member

    Posts:
    3,357
    Joined:
    Jan 1, 2013
    Location:
    Salt Lake City
    I'm gonna quote Rob's awesome answers inline. They got cut off in my view of the forum...

    1. What guides modern cap values and how many caps do we want?
    One is good enough but two (two fitler nodes) are better. 100uF caps are cheap today and that's plenty (1x100 or 2x50). For high impedance bias circuits like the high voltage tap on my website you don't want to exceed around 100uF because it can take too long to charge the caps and your bias might get too hot during the cap charge. Leo used small filter caps due to cost.

    2. What are the pros and cons of 10kl vs. 50kl pots?
    50kl gives you more bias voltage swing to for more flexible bias setting, especially if you want to run 6V6 and 6L6 in the same amp. The downside is the bias adjustment becomes more touchy but you don't need to set the bias to an exact value anyway. Another pro is the 50kl adds resistance between the filter nodes if you use two caps.

    3. Does it matter if you make your pot a variable resistor vs. a voltage divider?
    Variable resistor is safer. If the voltage divider wiper fails you lose bias voltage and your tubes will fry.

    4. Do we need a different circuit or values to do (power-tube) bias-wiggle trem?
    Yes, a high impedance bias circuit may be too sensitive to the bias wiggle tremolo circuit.

    5. What are sensible upper and lower limits of the total resistance here?
    I don't know.

    6. Are there reasons to go with bigger or smaller resistors (once we knock down the HT on a PT without a bias tap)?
    The first resistor is used to drop the voltage but it also adds to the "R" of RC filtration for the first node. The following resistor or pot form the next RC filter node so larger values help with that but again you don't want the bias circuit to have too much impedance because it starts to increase the filter cap charge times.

    Thanks, Rob, that's super-helpful. My brain does 10^6 better if I understand *why* things are designed the way they are.

    If I'm thinking straight, I see Merlin's 'emergency pulldown' resistor is there to make the voltage divider design safer.

    And I really appreciate your answer to #5 -- an answer I should use more in my life. :) Merlin likes to keep the resistances 'not too high.' In practice, I guess the original resistances found in the amps we build are "not too high," and in a way that approach (what would Leo do) can inform me about question 4 and question 6 too.

    Thanks again. If I actually build a tweed Tremmie I'm gonna steal your 5f6a bias layout with resistances based on Leo's 5g9 values.
     
    Last edited: Feb 13, 2019
  6. D'tar

    D'tar Tele-Afflicted

    Posts:
    1,870
    Joined:
    Jan 11, 2013
    Location:
    WNY
    50k pot

    pro: more bias voltage swing

    con: more bias voltage swing

    10k pot

    pro: less bias voltage swing

    con: less bias voltage swing

    :)
     
  7. King Fan

    King Fan Friend of Leo's Ad Free Member

    Posts:
    3,357
    Joined:
    Jan 1, 2013
    Location:
    Salt Lake City
    LOL, *another* answer style I need to use more often in my life. Virtually always applicable.

    Merlin's "10k is ideal" kinda speaks to me -- I'm a big fan of Mr. Blencowe. Clearly though 50k isn't 'too big' -- Leo used a total resistance of 82K + 56K in the 5g9.

    One other pro for the 10kl is the through-hole screwdriver adjustment on the CTS Fender-style unit. But on a tweed-style chassis that's less of a big deal.
     
  8. King Fan

    King Fan Friend of Leo's Ad Free Member

    Posts:
    3,357
    Joined:
    Jan 1, 2013
    Location:
    Salt Lake City
    More pics? I drew out Doug Hoffman's 'optimized' bias circuit as used in his 5f6a and Plexi boards -- more filtering, and an example of a variable resistor circuit.

    Hoffman better bias copy.png

    And here's a screenshot of Rob's elegant adjustable bias circuit for the 5f6a, also with two filters and a variable resistor, as a nice example of how to convert the older non-adjustable Fender bias circuits in this era of amp... something very similar will go in my 5g9.

    Robrob 5f6a adjustable bias.png
     
    Last edited: Feb 13, 2019
  9. robrob

    robrob Poster Extraordinaire Ad Free Member

    Posts:
    7,329
    Joined:
    Dec 29, 2012
    Location:
    United States
    The Hoffman circuit needs a jumper on the pot's top and middle terminals to prevent a wiper from burning up the power tubes and output transformer.

    My pot is oriented shaft up and as you turn the bias "up" (clockwise) the bias gets hotter. Assuming the Hoffman circuit is shaft down then the bias will get cooler as you turn the pot clockwise.
     
    King Fan likes this.
  10. robrob

    robrob Poster Extraordinaire Ad Free Member

    Posts:
    7,329
    Joined:
    Dec 29, 2012
    Location:
    United States
    I'm a big fan of a bias circuit with bias and balance adjustment. This is what I recommend for new builds and mods:

    [​IMG]

    You can add a second filter node by decreasing the bias cap shown to 50uF then add a second 50uF from pot wiper to ground (+ to ground of course).
     
    King Fan and D'tar like this.
  11. sds1

    sds1 Tele-Holic

    Age:
    42
    Posts:
    914
    Joined:
    May 4, 2017
    Location:
    Orlando, FL, USA
    Has anyone here in practice ever experienced a problem with a design due to too high bias trimmer and/or exceeding the maximum grid leak resistance?
     
  12. King Fan

    King Fan Friend of Leo's Ad Free Member

    Posts:
    3,357
    Joined:
    Jan 1, 2013
    Location:
    Salt Lake City
    Excellent points, and exactly the sort of subtle but vital stuff I'm looking for. The whole 'shaft up / down' and CW / CCW thing is like 'god is in the details', and the jumper thing is key.

    And I need to go beyond simply adjustable to level-and-balance adjustable. For those keeping score at home, you'll find Rob already has balance adjusts for example on his generic amp mods page, 5f6a mods, AB763 mods, etc.
     
  13. King Fan

    King Fan Friend of Leo's Ad Free Member

    Posts:
    3,357
    Joined:
    Jan 1, 2013
    Location:
    Salt Lake City
    Yeah, I was thinking about that. Merlin is superb and seldom wrong AFAIK, but from my lowly vantage point, he maybe sometimes looks at 'idealized' design goals -- he doesn't favor the way many big amp firms anchored their grounds, or routed heater wires across sockets, or combined and split grounds among sectors, or... Am I wrong in thinking he just doesn't want us to put in 'very high' resistances here, and that most of us fly well under those values just by experience and example?
     
  14. sds1

    sds1 Tele-Holic

    Age:
    42
    Posts:
    914
    Joined:
    May 4, 2017
    Location:
    Orlando, FL, USA
    Right, I think it's important to differentiate between the theoretical and the practical. There is a lot of fudging in tube amps, you need to know what you can get away with. We know this circuit has been designed 1000 different ways. What if I only have a 100K trimmer on hand -- will it affect the amp in some way I can measure or perceive otherwise? I want to know. :)
     
    King Fan likes this.
  15. King Fan

    King Fan Friend of Leo's Ad Free Member

    Posts:
    3,357
    Joined:
    Jan 1, 2013
    Location:
    Salt Lake City
    LOL, maybe there are two types of builders, those like you, empirical and curious, who want to find out what doesn't work... and those like me, flatfooted and cautious, who just don't. :D
     
    sds1 likes this.
  16. robrob

    robrob Poster Extraordinaire Ad Free Member

    Posts:
    7,329
    Joined:
    Dec 29, 2012
    Location:
    United States
    You could use that 100k trimmer because only the resistance between the wiper and output will be in circuit so the 100k wouldn't push you over the grid leak maximum. If the amp is biased normally then the trimmer resistance will also be a "normal" value.
     
    D'tar likes this.
  17. Ten Over

    Ten Over Tele-Meister

    Posts:
    335
    Joined:
    May 13, 2015
    Location:
    Central California
    The bias will default to its coldest bias upon wiper failure and the tubes will be safely in cut-off. He does need that jumper upon wiper failure, however, to prevent the bias circuit from charging up to the peak HT voltage and frying the 10uF filter caps.
     
    King Fan and robrob like this.
  18. robrob

    robrob Poster Extraordinaire Ad Free Member

    Posts:
    7,329
    Joined:
    Dec 29, 2012
    Location:
    United States
    Yep, Ten Over is correct. I was looking at the layout wrong.
     
    King Fan likes this.
  19. King Fan

    King Fan Friend of Leo's Ad Free Member

    Posts:
    3,357
    Joined:
    Jan 1, 2013
    Location:
    Salt Lake City
    Thanks, I wondered about that a bit -- but always struggle to picture 'directions' inside a pot (as if there were any). But due to my ignorance I'm still confused. By peak HT voltage do we mean the full PT secondary voltage (over 300V) or just the bias tap voltage (like 50V) -- assuming we're using a bias tap as Doug has it drawn?
     
  20. Ten Over

    Ten Over Tele-Meister

    Posts:
    335
    Joined:
    May 13, 2015
    Location:
    Central California
    I went to the Hoffman bias page you linked and I looked at what he said about his 5F6A and Plexi bias that you mentioned. He doesn't specify where he is getting the raw AC or the value of the current-limiting resistor, so I went to his Plexi build section and he is getting it from the HT via a 220K resistor. So I went with that.
     
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.