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.

Maximum wattage, 2 6V6 in PP, cathode bias

Discussion in 'Amp Tech Center' started by jhawk, Sep 20, 2017.

  1. Silverface

    Silverface Poster Extraordinaire Platinum Supporter

    Age:
    68
    Posts:
    9,781
    Joined:
    Mar 2, 2003
    Location:
    Lawndale CA
    Exactly. Key phrase being, IMO "..not in a way that matters"!

    Yay! Something that's NOT just math!

    It's truly amazing what results you can "force" in a circuit. It's just as amazing how completely irrelevant those "forced" numbers are.

    It's *marketing*. And marketing has never been tied to technical precision. And output power *is* subjective as it can be changed simply via tube swaps. And if they don't list power *with* a THD number the power "rating" could reflect a nasty, coldly-distorted sound.

    You just can't ever correlate advertising with real-worl, practical usage. The audio world has never worked that way, ever, with the exception of scientific and some high-end audiophile products.

    Guitar amp marketing information is inherently "wrong" by its nature.



    I'm still waiting for someone to explain exactly what "maximum dissipation" of a 6V6 *sounds* like. Errr - well, I guess you'd have to pick a specific one first, right?

    An imaginary, theoretical one has pretty crappy tone. So does an equation. So when working to the "nth" degree to figure the absolute maximum dissipation of a pair of specific 6V6's, what is the practical audio result? How is that value used with a guitar plugged in? Because all I've "heard" are numbers and graphs.

    Joking aside - it's a serious question. In all the years I've been involved with this stuff nobody who fully understands the math has ever been able to provide a usable, "practical application" audio example (using a guitar and amplifier) related to precisely calculating "maximum dissipation".
     
    peteb and Commodore 64 like this.
  2. Old Tele man

    Old Tele man Friend of Leo's

    Posts:
    3,941
    Joined:
    May 10, 2017
    Location:
    Once banned always banned
    Amps are mathematically "designed" first, "prototyped" second, "Beta-tested" third, and finally released to the world where customers decide for themselves how good -- or not -- those original mathematical design equations represented the final assemblage of components.

    Of course, today those designs are done using SPICE simulations, which CAN produce FFT-derived model-simulations of how the "sound" might sound. But, to my knowledge FFT has never produced a 'hit' song or record (wink,wink)!

    Equations produce "...a priori..." estimates, not absolutes because of the "stacking" results of component variabilities. Thus, the equations design for the MEAN of the bell curve while component tolerances create the overall ±6-Sigma 'real world' end-system results we get and hear.
     
    Last edited: Sep 24, 2017
  3. Silverface

    Silverface Poster Extraordinaire Platinum Supporter

    Age:
    68
    Posts:
    9,781
    Joined:
    Mar 2, 2003
    Location:
    Lawndale CA
    Yep - exactly my point. And while it may be surprising to some, tweed Fenders were not "designed" mathematically at all - the "designs" were simply adaptations of stock circuits in the Western Electric manual. BF and many SF derivatives were "designed" in similar ways (many SF designs just minor modifications of existing amps).

    Leo didn't design his amps with a slide rule. He copied public domain circuits and changed some things because he simply "knew" the electronics that would work better for guitar frequencies. CBS engineers were the first Fender amp designers to "calculate" amps designs.

    And they did just a sterling job of improving tone, didn't they?

    :lol:

    (Sorry - didn't mean to hijack the thread, but I couldn't resist making the "audio" points. I won't start any other tangential sub-threads.)
     
  4. Old Tele man

    Old Tele man Friend of Leo's

    Posts:
    3,941
    Joined:
    May 10, 2017
    Location:
    Once banned always banned
    Oops, poor word usage on my part. I was implying "design" as the use of mathematical equations to preview the final assemblage, ie: the math representing the physical (electrical) happenings.

    Agreed 100%, Leo's original amps were merely slightly modified versions of existing Western Electric circuits. But, his later amps evolved through HIS design processes, albeit based upon industry-proven evolutions.
     
  5. Silverface

    Silverface Poster Extraordinaire Platinum Supporter

    Age:
    68
    Posts:
    9,781
    Joined:
    Mar 2, 2003
    Location:
    Lawndale CA
    Right. I was just clarifying that his "design process" didn't involve 5th decimal point math - if any, except for the pretty rough power calculations (which in reality vary). And while I do understand some people's interest in paring down specs to extremes it's also important to keep those things in an audible, musical perspective.

    Sometimes "sound" just seems to get lost in the minutia.
     
  6. printer2

    printer2 Poster Extraordinaire

    Posts:
    6,930
    Joined:
    May 24, 2010
    Location:
    Canada
    Speaking of Leo, you know that he intended for his amps to be played clean? So talk of him designing amps used as they are now blows math based outcomes out of the water. And he did know the math. And the math? Equations thought up to model amplifiers under real world conditions where they would be used. Instrument amp service for R&R is not real world conditions. We live in our own little world (although it is a nice little world).

    Now on to what I was going to post. So you think a cathodyne splitter being second rate to a LTP? Tell that to Ampeg. Can a pair of 6V6's put out 20W? Sure. But what time frame are we talking about? Clean, maybe for a few mS that captures the peak after the pick stroke. Cranked into distortion, sure but it is not clean power. Both 20W, depends on your reference.

    Cathode bias of the output stage will put out less than a fixed bias stage? How about just increasing the supply voltage on the cathode biased design a little higher than the fixed bias voltage? Should make them apples to apples.

    70% is the magic number to bias a fixed bias amp? The 70% sort-of-rule has to do with the resistor on the grid of the power tube. The resistance value given on datasheets and the 70% value go hand in hand. Lower the resistor value the higher you can go on your bias.

    Power out is less that plate dissipation. True on a Class A amp. Not so much on a Class AB, B or C amp. (Class C in audio? Sure, look at the output waveform of the poor little Champ) Want to know how to get the most power out and dissipate the least across the tube? Drive it with a square wave. On, off, on, off... . Don't forget that the tube's impedance is not constant. It goes from high resistance to low resistance, the impedance that is listed on the datasheet is just an average. So the impedance of the plate changes depending on the amplitude of the signal. Thevenin's power theorem gives us the energy transfer for a source and load impedance. From that we should get the dissipation at any point in the signal. All well and good until we get sneaky and cut off the other tube and the reflected load impedance the plate sees drops (anyone hazard a guess by how much?). Then the slope of the straight line equation we were working with gets a lot steeper. Well, not quite straight line equations but it is much easier to design an amp with them and let the amp work out the little squiglies on its own.

    There was something else, can't remember it now. No bother, time to go look at some guitars.
     
  7. Cjl77

    Cjl77 Tele-Meister

    Posts:
    420
    Joined:
    Nov 12, 2013
    Location:
    lake charles ,La
    I think you'd be surprised to find how much "design" went into Leo's amps. Even the early ones. Those "public domain" circuits are vastly different than what he ended up with.
    He was a engineer and designer. If you look at his circuits, particularly from say 1956 to 63 you'll see that they were very different from what was going on at the time in musical instrument amplifiers. I'm not saying he was re inventing the wheel, or revolutionizing the world of electronics. His circuits did set a standard in the genre. I'm positive that he approached his efforts from a technical perspective. Meaning, he did use a slide rule(not that it matters)and was very aware of and kept up with current advancements and practices in the electronics world of his times.
     
  8. Old Tele man

    Old Tele man Friend of Leo's

    Posts:
    3,941
    Joined:
    May 10, 2017
    Location:
    Once banned always banned
    1950-60's slide rule = current day I·Pad "calculator" app.

    And, "yes," I still have (and occasionally) actually use my 'old' slide rulers, both my common 14" K&E and my 3" circular S.I.Co (from Japan)...although I prefer using my computer.

    Leo "assembled" his MUSICAL amps using separate "circuits" from multiple different electronic sources: from Western Electric he got the PP amp but the pre-amp and PI were common knowledge. Most RADIOs of the time only used a single class-A output tube and very few people had record players, which also typically used one class-A tube. The only other industry using similar amps was the fledgling "electronic organ" (Wurlitzer, etc.) from which Leo borrowed his "Vibrato/tremolo" circuitry...and, much later, the reverb.

    But, to get ALL those circuits to "work" together, Leo had to RE-design & modify them to "play well" together...which was his genius. And, FWIW, his TMB tone circuit, although similar to others, is totally of his own design...the transition of which can be seen and traced through his schematic changes thru the years.
     
    Last edited: Sep 23, 2017
    Cjl77 likes this.
  9. Silverface

    Silverface Poster Extraordinaire Platinum Supporter

    Age:
    68
    Posts:
    9,781
    Joined:
    Mar 2, 2003
    Location:
    Lawndale CA
    I sat and talked to Leo a few times when G&L was around the corner form my old "office". Also Forrest White and George Fullerton.
     
  10. printer2

    printer2 Poster Extraordinaire

    Posts:
    6,930
    Joined:
    May 24, 2010
    Location:
    Canada
    That is odd, I salvaged a number of P-P radios from before Leo's time. I still have one chassis together with a pair of 6F6G's for output tubes. The radio was a Philco 37-2650 from 1937, slightly different styling than this one.

    [​IMG]

    Most radios were smaller but that does not mean larger and better equipment was uncommon. The 6F6G would put out 11W in P-P.

    Tabletop 1938 with a pair of 6F6's

    [​IMG]
    As far as pre-WWII guitar amplifiers go who could forget the Gibson EH-150 or EH185?
    [​IMG]

    [​IMG]
     
  11. peteb

    peteb Friend of Leo's

    Posts:
    4,208
    Joined:
    Apr 25, 2003
    Location:
    Cascadia

    Thanks

    I'm sorry if I am slow in understanding this. What is the 4?
     
  12. peteb

    peteb Friend of Leo's

    Posts:
    4,208
    Joined:
    Apr 25, 2003
    Location:
    Cascadia
    Jayhawk,

    Thanks for providing the neutral space in which we can express our thoughts.


    It's kind of like the jayhawk state where both sides are present and represented.
     
  13. peteb

    peteb Friend of Leo's

    Posts:
    4,208
    Joined:
    Apr 25, 2003
    Location:
    Cascadia
    Max power out of a class A champ (6W) is about half of the Max plate dissipation of the tube.

    Class Ab gets more power out of the tubes but I think the limiting factor is the max plate dissipation.



    It's pretty well accepted that max plate dissipation of a 6V6 is 12-14W. Look at the max power out of the highest powers dual 6v6 amps. They follow the rule exactly, max clean power approaches the max plate dissipation of the tubes but can't equal or exceed it.




    Suppose someone put super high voltages on 6v6s and squeezed 30 watts out of a pair.



    I would explain this as either they don't have 30 W of clean power or, at higher voltages, maybe 450-500, the max plate dissipation would creep higher, it could reach 15 W per tube or higher.









    In the RCA tube manual, it clearly states for class A that the plate dissipation is equal to the power applied to the tube. They don't say the same thing for class AB. I think it is similar but different:



    Class A runs 100% all of the time, the power out is constant between idle and full power, this is because the grid is hit with a full sin wave, both upper and lower halves, so any signal brings a rise in current Followed an equal reduction in current. So in class A it is clear cut, the power in is constant, it equals the constant plate dissipation, which ends up being about double of the max power out.




    Class AB: one is likely to set the idle bias at 70% of max plate dissipation. This idle bias power would equal the input power at idle, but not the input power at full output. With the addition of positive signal voltages on the grids, the current thru the tubes increases and the plate dissipation increases, possibly reaching 100 %, this would be the situation during full power out. This should help illustrate how the power out is limited by the max plate dissipation.


    Max plate dissipation means different things to different people in different situations.




    In this situation, max plate dissipation means the maximum amount of power that the tube plate can handle being applied to it.



    And in this definition, it makes sense to me that as the plate voltage is raised, the maximum amount of input power that a tube can handle would also increase.
     
    Last edited: Sep 24, 2017
  14. peteb

    peteb Friend of Leo's

    Posts:
    4,208
    Joined:
    Apr 25, 2003
    Location:
    Cascadia
    From the inter web:

    "In a typical voltage amplifier, including power stages of most audio power amplifiers, DC bias voltage is negative relative to cathode potential. Instant grid voltage (sum of DC bias and AC input signal) should never rise above cathode potential to prevent grid-to-cathode currents that overload preceding amplifier stages and may cause severe even-order distortion."







    My personal opinion:


    A -1 volt DC bias voltage being used in the purpose of amplifying of a 20 volt signal voltage is uncharted waters. It is not normal tube operation, not even close to being normal tube operation.
     
  15. Old Tele man

    Old Tele man Friend of Leo's

    Posts:
    3,941
    Joined:
    May 10, 2017
    Location:
    Once banned always banned
    One-half of one-half...the principle BEHIND push-pull operation!
     
    peteb likes this.
  16. Old Tele man

    Old Tele man Friend of Leo's

    Posts:
    3,941
    Joined:
    May 10, 2017
    Location:
    Once banned always banned
    1) Most is not all (look through old Sams Photo-Facts).

    2) 6F6 is a true power pentode, not a beam power tube; it was viewed as RCA's predecessor of their 6L6 and smaller 6V6.

    3) Neither said nor implied that Fender was first or only maker of musical amplifiers.
     
  17. peteb

    peteb Friend of Leo's

    Posts:
    4,208
    Joined:
    Apr 25, 2003
    Location:
    Cascadia
    Why does cathode biased class AB put out less power than fixed bias class AB?

    How much less power does the cathode bias have than the fixed bias?




    Class AB cathode bias:



    You can set the idle plate dissipation at 100% of max plate dissipation.



    But adding signal increases current increasing bias which has the result of reducing the current so you end up at less than 100% plate dissipation at full power.




    Back to fixed bias class AB, to get the absolute most power out, the idle bias is set lower than 100%, but not too low, or too high, to the place where the max signal pushes it right up to 100% of max plate dissipation. Max energy in means max energy out.
     
  18. FenderLover

    FenderLover Poster Extraordinaire

    Posts:
    5,695
    Joined:
    Jun 11, 2009
    Location:
    Minnesota
    You haven't understood what you read here yet (post 27, regarding post 3):

    http://www.tdpri.com/threads/fender-cathodyne-phase-inverters.755580/page-2#post-7818784

    Can you name a guitar amplifier where your comment is true? It is hardly uncharted or not normal, whether you understand it or not. You would be farther ahead to admit that you do not understand it. Re-read the link above. It's very clearly written.
     
  19. Silverface

    Silverface Poster Extraordinaire Platinum Supporter

    Age:
    68
    Posts:
    9,781
    Joined:
    Mar 2, 2003
    Location:
    Lawndale CA
    "Genius" is a good description.

    It was fascinating talking to him and seeing his desk. Very little was on paper except longhand notes describing things he'd done and results. Many of his early Fender notebooks were on display in the Fullerton Museum Center about 15 years ago.

    Nearly all of his "getting things to work together" was intuitive based on experience and electronics knowledge - so he was primarily an "experimenter". But unlike some "scientific" types he didn't do much, if any, in the way of calculations ahead of the actual "experiments". He'd generally make a few idea notes, draw a "thing" up (a physical drawing or partial schematic - but complete enough to be clear. Usually...), physically put something together or change parts. He loved talking about "gadgets" of any kind.

    There are several well-known stories concerning early days at Fender about how he'd give a change to amp assemblers but not tell anyone else. Drove George Fullerton nuts. :oops:

    If you've ever talked to a stereotypical "rocket scientist" type that was friendly but seemed to be speaking (at least partially) in another dimension - well, it was pretty similar. And it always amazed me that you could generally just walk into the place unannounced and see him if he wasn't in a meeting. G&L was a very compact shop in a small industrial park at the time.
     
    D'tar and Bendyha like this.
  20. printer2

    printer2 Poster Extraordinaire

    Posts:
    6,930
    Joined:
    May 24, 2010
    Location:
    Canada
    And in this definition, it makes sense to me that as the plate voltage is raised, the maximum amount of input power that a tube can handle would also increase.[/QUOTE]

    Why?



    1.) Yeah.
    2.) So?
    3.) No but it sounded like you were saying Leo did something with his circuits others were not doing from your post. Leo did two things. He made amps with the same circuits as others. He did two things different. His amps were more reliable, he came out with sexy electric guitars. d
     
IMPORTANT: Treat everyone here with respect, no matter how difficult!
No sex, drug, political, religion or hate discussion permitted here.