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.

Output transformer primary impedance for push pull

Discussion in 'Shock Brother's DIY Amps' started by GeorgeG, Jul 30, 2018.

  1. GeorgeG

    GeorgeG TDPRI Member

    Age:
    66
    Posts:
    59
    Joined:
    Feb 4, 2018
    Location:
    Florida
    My first build was a single ended octal power tube based on an AX84 design. Everything went well and the amp sounds great. I understood the OT selection for SE no problem and I can reverse engineer common topologies and data sheet recommendations, also no problem. It appears that using (plate voltage-cathode voltage)*2/max tube power rating gets you really close almost every time.

    Now I'm working on my first push pull build but I don't understand how to determine the proper primary impedance given tube/power selection and voltages. It appears that the proper calculation involves comprehending the fact that class AB does not run at max power dissipation continuosly but I don't even see a consistent factor across amp manufacturers. If someone is willing to explain this to me, please also get me right on if the impedance is across the plates of both tubes or the impedence from the center tap to one tube plate. I assume that an advertised say 4K OT primary impedance is across the plates, is that correct? I assume in that case that one tube sees half that or 2K in this case?

    Thanks,

    George
     
  2. elpico

    elpico Tele-Afflicted

    Posts:
    1,190
    Joined:
    Sep 14, 2011
    Location:
    Vancouver BC
    With a 4k OT the load looks like 4k to each tube... until one of the tubes goes into cut off. Then it looks like 2K to the one tube that's still conducting. Sounds odd I know, transformers are a little unintuitive.

    The good news is you don't really need to think about that. You don't build amps by calculating some perfect OT impedance and then winding an OT to that number. The exact value doesn't make enough difference to be worth the trouble and the "ideal" value would be different for different sets of tubes anyways. Instead you just choose an off the shelf transformer based on the tube complement you plan to run. two 6V6, two 6L6, four 6L6 etc.
     
    Last edited: Jul 31, 2018
    jam66tn and LudwigvonBirk like this.
  3. GeorgeG

    GeorgeG TDPRI Member

    Age:
    66
    Posts:
    59
    Joined:
    Feb 4, 2018
    Location:
    Florida
    What accounts for the difference is manufacurer primary load choices for the same tubes:?

    Fender 6V6(2)=8K, Valco=5K
    Marshall EL84(2)=8.2K, Vox=6K
    Fender 50W (octalx2)=4.2K, Marshall=3.6K
    Fender 100W (4xoctal)=2K, Marshall=1.7K

    Thanks,

    George
     
  4. Inglese

    Inglese Tele-Meister

    Age:
    41
    Posts:
    107
    Joined:
    Mar 7, 2018
    Location:
    Italy
    The "proper" way to design tube stages is to use charts and draw load lines.
     
  5. elpico

    elpico Tele-Afflicted

    Posts:
    1,190
    Joined:
    Sep 14, 2011
    Location:
    Vancouver BC
    I really don't think it was any more magical than what I just said. Guitar amp manufacturers didn't invent tube amps. They were already in every home in the form of table or console radio and record players. It was already a solved problem with off the shelf solutions. When various people decided to start making a tube amp "for guitar" they did the same thing everyone else did, looked at the catalog of a transformer maker in their country (which was a way bigger business than the guitar amp company at this point) and bought the transformer that company offered for 2x6V6 amps or 2x6L6 amps etc. If you got lucky your transformer company might make a whopping TWO choices of 6L6 transformer and you could pick the one best suited to your project (or probably more often: the cheaper one :lol:).

    I think most of your differences there can be accounted for simply by the companies being in different countries, so they have to buy from different transformer companies and those transformer companies also have tailored their transformers for tubes produced in that country. (6L6 in america, EL34 in UK)

    Like I said, the exact value makes very little difference anyways so they didn't worry much about 8k vs 6k. You'd be hard pressed to detect ANY difference in volume at all between those choices, and if you did detect any difference in tone (due to slightly different amounts of distortion) you'd again be very hard pressed to say either one was "better" or even reliably detect which is which in a blind test.

    On amps with multiple speaker impedance taps you can change the load the tube sees by mismatching the speaker and try this for yourself. Putting your 4ohm speaker on the 8ohm tap turns your 8K transformer into a 4k one. If this did anything exciting like cut the volume in half or make it breakup at a lower level then the attenuator market wouldn't exist. The fact that it does exist tells you a lot about how little difference 4k vs 8k makes.
     
    Last edited: Aug 1, 2018
  6. elpico

    elpico Tele-Afflicted

    Posts:
    1,190
    Joined:
    Sep 14, 2011
    Location:
    Vancouver BC
    What's step two?
     
  7. Inglese

    Inglese Tele-Meister

    Age:
    41
    Posts:
    107
    Joined:
    Mar 7, 2018
    Location:
    Italy
    This may help...
     
  8. elpico

    elpico Tele-Afflicted

    Posts:
    1,190
    Joined:
    Sep 14, 2011
    Location:
    Vancouver BC
    no, not a lot...
     
  9. RLee77

    RLee77 Friend of Leo's

    Posts:
    4,427
    Joined:
    May 15, 2016
    Location:
    Nevada
    There’s some good info here (a little, anyway) on designing with load lines and OT impedances for push-pull stages: http://www.valvewizard.co.uk/pp.html

    The output tubes each see 1/2 the plate-to-plate primary impedance when both tubes are conducting (since the full primary winding is active), but when one tube goes into cutoff (class B part of the cycle), the conducting tube sees 1/4 of the p-p impedance, because current is only flowing in half the primary, which is 1/2 the turns, and impedance is the square of turns ratio, thus 1/4 the impedance.
     
  10. Inglese

    Inglese Tele-Meister

    Age:
    41
    Posts:
    107
    Joined:
    Mar 7, 2018
    Location:
    Italy
    Ops I forgot to paste the link... however I meant the valvewizard page.
     
IMPORTANT: Treat everyone here with respect, no matter how difficult!
No sex, drug, political, religion or hate discussion permitted here.