ot- why are p-p tube output transformers unbalanced?
Close
Login to Your Account
Results 1 to 18 of 18
  1. #1
    Join Date
    Jul 2004
    Location
    N. GA- 33.992N , -83.72W usa
    Posts
    3,821
    Post Thanks / Like
    Likes (Given)
    70
    Likes (Received)
    891

    Default ot- why are p-p tube output transformers unbalanced?

    why are p-p tube output transformers unbalanced?

    i have 8-10 of them , and they all typically measure

    250 ohm/100/150 , 300 ohm/180/120 resistance on the primary windings . i know
    that ac impedance and inductance , and center
    tapped 350-400volts , and an output load come into
    the equation...... but is this intentional , or
    does it just exist by default? does it just
    "balance out" in practical application?

    it would seem that one 6l6 or el34 would see more
    current on the lower resistance?

  2. #2
    Join Date
    Jul 2007
    Country
    UNITED STATES
    State/Province
    Florida
    Posts
    792
    Post Thanks / Like
    Likes (Given)
    94
    Likes (Received)
    401

    Default

    Quote Originally Posted by tnmgcarbide View Post
    why are p-p tube output transformers unbalanced?

    i have 8-10 of them , and they all typically measure

    250 ohm/100/150 , 300 ohm/180/120 resistance. i know
    that ac impedance and inductance , and center
    tapped 350-400volts , and an output load come into
    the equation...... but is this intentional , or
    does it just exist by default? does it just
    "balance out" in practical application?
    SWAG unbalanced so you get 3 choices of impedance, center tapped would only give you 2.
    Tubes outputs are much higher impedance than transistors and impedance matching is critical for max power transfer.

    CarlBoyd

  3. #3
    Join Date
    Jul 2004
    Location
    N. GA- 33.992N , -83.72W usa
    Posts
    3,821
    Post Thanks / Like
    Likes (Given)
    70
    Likes (Received)
    891

    Default

    so , @ a 60/40 differential, with 400v input to the
    center .. and a balanced phase splitter tube or
    intermediate center tapped transformer- the load on
    the output tubes would be equal from the input and
    not the plate voltage?

  4. #4
    Join Date
    Feb 2004
    Location
    peekskill, NY
    Posts
    25,080
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    4997

    Default

    Quote Originally Posted by tnmgcarbide View Post
    why are p-p tube output transformers unbalanced?
    Can you post a photo of the transformer label, or at least a manufacturer and model number?

  5. #5
    Join Date
    Jul 2004
    Location
    N. GA- 33.992N , -83.72W usa
    Posts
    3,821
    Post Thanks / Like
    Likes (Given)
    70
    Likes (Received)
    891

    Default

    hmm. i could,
    but it was a question in general.
    ALL PP output transformers that i've encountered are
    not even close to balnced . i just don't understand
    why. windings on a fixed ohm 4/8/or 16 secondary
    wouldn't possibly be standard?

    a single ended, class-a only has one path .

    who cares.


    Can you post a photo of the transformer label, or at least a manufacturer and model number?.

    all different from 50L6 to eL34 .

    no point really .... if they aren't balanced?
    some from hifi,guitar, pa, lab stuff . like i said
    all similar. i dont know that it matters unless there's more than one tube ?

    but two tubes push/pull have to be close.

  6. #6
    Join Date
    Nov 2004
    Location
    Webster Groves, MO
    Posts
    7,555
    Post Thanks / Like
    Likes (Given)
    1914
    Likes (Received)
    3609

    Default

    The answer is simple as can be. The winding are done in one shot, pausing only to attach the center or other taps. Naturally, the outer half of the windings will have more feet of wire on it than the inner half. It doesn't make much difference since the number of turns on each is the same and the voltage on each will also be close to the same, although leakage reactance will be somewhat diffeent. The DC resistance is not high enough to have much effect.

    I have made transformers where this factor was important, solved by winding both strands of wire together. Naturally, this will only work if the voltages are low. Another solution is to wind one layer of one winding, then a layer of the other. You can also have each occupy half of the available length on the core.

    The hardest I ever did like that was 6 strands of #6 square to maximize coupling between windings. You wouldn't pay for one like that.

    Bill

  7. Likes JST, tnmgcarbide, Hopefuldave liked this post
  8. #7
    Join Date
    Jul 2004
    Location
    N. GA- 33.992N , -83.72W usa
    Posts
    3,821
    Post Thanks / Like
    Likes (Given)
    70
    Likes (Received)
    891

    Default

    thank you bill, that was close to the answer i was
    asking here . what about ULTRALINEAR ot's ? they have
    a ct and 4 output taps .i don't have one to measure
    today , but could they be any different( and does it matter in sound) ?

    it would seem that in the real world, the AC output
    transformer issue doesn't exist outside of my thoughts....

  9. #8
    Join Date
    Feb 2014
    Location
    Southern Illinois University
    Posts
    29
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    6

    Default

    9100 is right. Ultralinear hooks a tap at like 80% to the screen grid to impart some nfb or something like that, and it improved thd. Usually, they end up making more power like that too. Been a long time since I worked with the stuff.

  10. #9
    Join Date
    Jul 2004
    Location
    N. GA- 33.992N , -83.72W usa
    Posts
    3,821
    Post Thanks / Like
    Likes (Given)
    70
    Likes (Received)
    891

    Default

    i think i get the picture...the sum of losses.

    center charged
    hi voltage b+ DC charges the plates from the ends, so shouldn't be critical. the output's AC signal , on the same wires as the DC b+ passes back through the xformer, and is not as critical as the dc resistance
    would imply .

  11. Likes Hopefuldave liked this post
  12. #10
    Join Date
    Jun 2001
    Location
    St Louis
    Posts
    18,337
    Post Thanks / Like
    Likes (Given)
    2007
    Likes (Received)
    3153

    Default

    Bill is of course correct.

    There is a slight effect, unbalancing the half cycles of output a small amount, which is somewhat equivalent to even harmonic addition, so can be a small contributor to "tube sound".

    That can be minimized by winding the transformer "bifilar", both windings together as described, and also by breaking the winding into sections so that the innermost winding also has sections out farther in the layers. That can maintain insulation, and still get a good balance.

    Since the voltage between the ends of the plate winding can have from 500V to over 1200V between them, winding bifilar, or "two in hand", is impractical, because the wire insulation is not good for anything like that voltage.

    It's a great question, though.

  13. Likes tnmgcarbide, Hopefuldave liked this post
  14. #11
    Join Date
    Jul 2004
    Location
    N. GA- 33.992N , -83.72W usa
    Posts
    3,821
    Post Thanks / Like
    Likes (Given)
    70
    Likes (Received)
    891

    Default

    thanks for the insight from everyone.

    sometimes i just
    refuse to shut up and accept the fact that some things
    just work well without explanation. i've a need to question everything .

    it's a machinist thing.

  15. Likes Hopefuldave liked this post
  16. #12
    Join Date
    Feb 2004
    Location
    peekskill, NY
    Posts
    25,080
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    4997

    Default

    Quote Originally Posted by tnmgcarbide View Post
    thanks for the insight from everyone.

    sometimes i just
    refuse to shut up and accept the fact that some things
    just work well without explanation. i've a need to question everything .

    it's a machinist thing.
    The *real* question is, why do you own so many push-pull audio transformers? Never did see you
    around here:

    Antique Radio Forums • Index page

  17. #13
    Join Date
    Jun 2001
    Location
    St Louis
    Posts
    18,337
    Post Thanks / Like
    Likes (Given)
    2007
    Likes (Received)
    3153

    Default

    Quote Originally Posted by jim rozen View Post
    The *real* question is, why do you own so many push-pull audio transformers? Never did see you
    around here:

    Antique Radio Forums • Index page

    Why do I have a bunch of them? And I have never been to that forum either.

  18. #14
    Join Date
    Jul 2004
    Location
    N. GA- 33.992N , -83.72W usa
    Posts
    3,821
    Post Thanks / Like
    Likes (Given)
    70
    Likes (Received)
    891

    Default

    maybe, it's because i'm from yorktown.... we don't need
    no peekskill stinkin' guys fogging' up the joint.
    grab your damm' tubes ...and get back to work..

  19. #15
    Join Date
    Mar 2011
    Location
    Geneva Illinois USA
    Posts
    6,425
    Post Thanks / Like
    Likes (Given)
    2807
    Likes (Received)
    2509

    Default

    As I recall this from MANY years ago, some of the circuits had pots in the finals cathode circuit to balance the tube current.

    Tom

  20. #16
    Join Date
    Jul 2004
    Location
    N. GA- 33.992N , -83.72W usa
    Posts
    3,821
    Post Thanks / Like
    Likes (Given)
    70
    Likes (Received)
    891

    Default

    i think it depends if it is cathode biased or fixed
    variable design . anything could always use a tweak.
    250 ohms is nothin'...

  21. #17
    Join Date
    Feb 2004
    Location
    peekskill, NY
    Posts
    25,080
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    4997

    Default

    Quote Originally Posted by tnmgcarbide View Post
    maybe, it's because i'm from yorktown.... we don't need
    no peekskill stinkin' guys fogging' up the joint.
    grab your damm' tubes ...and get back to work..
    Ha. It's a holiday for once!

  22. Likes tnmgcarbide liked this post
  23. #18
    Join Date
    Jun 2001
    Location
    St Louis
    Posts
    18,337
    Post Thanks / Like
    Likes (Given)
    2007
    Likes (Received)
    3153

    Default

    Quote Originally Posted by TDegenhart View Post
    As I recall this from MANY years ago, some of the circuits had pots in the finals cathode circuit to balance the tube current.

    Tom
    Not uncommon to have an overall grid bias, and a separate balance control. Or even two separate bias controls.

    Cathode bias circuits often had nothing for balance other than the tube characteristics.

    As for balance at max output, the difference in resistance was often considerably less than the variation of tube voltage drop on peaks, which could easily be 30 to 50 volts.

    Most outputs were pentodes, which tend to act as current sources on their plates (that is what the screen grids did), so the difference would only show up as a limit on max peak voltage (clipping point).

    Triode outputs were often so "soft" that the wire resistance imbalance was quite swamped-out under other effects. If the designers were going for extra special performance, they could have the transformers wound with "sectional balance" to even out the resistance as mentioned above.

    Lots of tricks to use when you cannot rely on more than a few dB of feedback to stabilize gain and linearize the amplifier.


Tags for this Thread

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •