The important thing - Is the Grid signal at pin 2 V1 identical but exactly complimentary (180 degrees out of phase) to the Grid signal at pin 2 V2. Also important is the amount of distortion at those Grids, that said some distortion is taken out by the negative feedback via R3.
You can simulate part of the circuit in LTspice, just simulate the front end (without negative feedback) and see what the outputs at V1 pin 2 and V2 pin 2 are. Don't put V1 and V2 into your simulation because a lot will depend on T2 and the speaker SP.
The 6N2P is definitely not equivalent to a 12AX7. Mu is slightly lower, and heater pin arrangement is different. It can replace a 12AX7 in plenty of applications though, but it is not plug-n-play.
Non-cathodyn phase splitters need some kind of adjustments to get good balance. This balance is also
affected by the tubes various parameters, thus no circuit tweaks ( less a CCs) can fix this.
Adjustments also needs some kind of measurment equip to obtain symmetry. Thus the cathodyn
is the obvious choice for beginners build.
affected by the tubes various parameters, thus no circuit tweaks ( less a CCs) can fix this.
Adjustments also needs some kind of measurment equip to obtain symmetry. Thus the cathodyn
is the obvious choice for beginners build.
@petertub, a tube equipment designer I respect a lot once explained to me that you need to unbalance R5, R6 in the post # 1 schematic to get a balanced output, because the output impedances of the two sides are unequal. He spoke about a 10% unbalance.
Would that not fix the unbalance you mention?
Jan
Would that not fix the unbalance you mention?
Jan
@6A3sUMMER
I must recognize that I never used this kind of output stage in practice. I had experienced the OTL using 6JN6 tubes with great successfull, or single ended pentodes/tetrodes. But not the common push pull. I am only saying from my learning reading the classic books: Terman, Gray, Langford Smith, Wallman, etc. With regard to current imbalance, unless you are using toros at the output stage, small inbalance may be allowed thanks to the natural gap that always exists on normal laminations in the frontier between a E and a I sections.
And, certainly I listen the music or radio program over the amplifier itself
I must recognize that I never used this kind of output stage in practice. I had experienced the OTL using 6JN6 tubes with great successfull, or single ended pentodes/tetrodes. But not the common push pull. I am only saying from my learning reading the classic books: Terman, Gray, Langford Smith, Wallman, etc. With regard to current imbalance, unless you are using toros at the output stage, small inbalance may be allowed thanks to the natural gap that always exists on normal laminations in the frontier between a E and a I sections.
And, certainly I listen the music or radio program over the amplifier itself
It’s one of the things I’ve wanted to try for a long time.
I remain convinced that these two unbalanced resistors are not there by chance on many old diagrams.
As an aside Jan, I recently learned that you came to France for an informal meeting with JM Plantefeve and Jean Hiraga in the north of France, I saw photos, it was nice to see you in real life among people I like.
I remain convinced that these two unbalanced resistors are not there by chance on many old diagrams.
As an aside Jan, I recently learned that you came to France for an informal meeting with JM Plantefeve and Jean Hiraga in the north of France, I saw photos, it was nice to see you in real life among people I like.
@huggygood Yes they wanted me to demo my paX amplifier.
I attend the yearly European Triode Festival and there is always a very large and strong French group, with great equipment.
Many are members of Melaudia. I think Plantevefe is also a member.
Very serious audiophiles!
Jan
I attend the yearly European Triode Festival and there is always a very large and strong French group, with great equipment.
Many are members of Melaudia. I think Plantevefe is also a member.
Very serious audiophiles!
Jan
Last edited:
The unbalance would change at every tube change. To balance it you need some kind of measuring device which not all have available, or accept dist that may to some degree be reduced by the NFB.
Whats your point ?
R13 and C7 don't do very much.
I think with 6P15 you'd want the screen grids to sit at no higher than 250V. I figure the B+ is 300V to 325V in the above, and R13 isn't going to drop more than just a very few volts. I guess it would work if you could use a 470uF 400V cap for C7. But that would be a big, expensive part. $10 to $22 each (USD) from Mouser. But the 6P15 screen grids would still be up above 300V. Can they take that?
BTW, Chinese 6P15 (or Russian 6P15P) is not an exact equivalent of EL84 or 6BQ5. 6P14 is (Russian version 6P14P). 6P15 is closer to an EL83. Screen grid is wound with thinner wire for higher transconductance, but that reduces the screen grid's power dissipation limit. The above circuit probably red plates a lot of 6P15 tubes...
The point being that you can balance such a stage, if you use slightly different resistors to compensate for the different source impedances presented by the two paths. But yes, you are right, it would have to be measured in some way, and can drift and change with tubes.
But isn't that what tube aficionados like ? 😎
Jan
Degeneration at cathodes tend to alleviate AC umbalance in tubes. Think a push pull like a specialized form of a differential pair and ubp'd cathode resistor as the ccs.
dubadub,
1. The construction of two tubes can be different, and still have the same specifications.
The Original 6BQ5 is a Beam Power tube (Beam Formers).
The Original EL84 is a Pentode (Suppressor Grid).
Suppressor Grid and Beam Formers are different construction, but these two tubes have essentially the same specs.
Some manufacturers got the idea to Dual Label them as 6BQ5 / EL84. (The dealer only has to stock one tube).
You can interchange these two tubes, tube rolling, no problem.
If the amplifier uses adjustable bias, you have to re-bias them, just like you would if you replaced the tube with the same type.
Plug and Play (Not Plug and Pray).
The EL34 Pentode (Suppressor Grid), and KT77 Beam Power (Beam Formers), have the same specifications, Except . . .
EL34 Suppressor Grid only connects to pin 1, and is not internally connected to the cathode, But the . . .
KT77 Beam Formers are internally connected to the cathode.
Connect pin 1 to pin 8 (to use an EL34), and then you can interchange the EL34 and KT77.
If the amplifier uses adjustable bias, you have to re-bias them, just like you would if you replaced the tube with the same type.
Plug and Play (Not Plug and Pray).
There is the SV-83 tube. Some try and use it as a 6BQ5 / EL84 substitute.
They find out the SV-83 screen grid often burns out when used in 6BQ5 / EL84 amplifiers. Do Not use it as a substitute.
2. I use the parallel cathodes and CCS phase splitter for my phase splitters, because they are intrinsically balanced.
3. I do not want to have to change the resistors in the phase splitter each time you change tubes.
I figure, use it unbalanced, and let the global negative feedback partially fix the effect of the un-balanced phase inverter, and just live with the performance you get.
But here is a much closer analysis of this kind of paraphase splitter (the more you make the two plate signals equal, the stronger this effect is:
For most paraphase splitter circuit topologies, the input tube sends 2nd harmonic distortion to the top (push) output tube;
and the other tube in the phase splitter sends No 2nd harmonic distortion to the bottom (pull) output tube.
This is often overlooked, and often misunderstood.
Push driven by a signal that has some 2nd harmonic distortion.
Pull driven by a signal that has essentially No 2nd harmonic distortion.
Talk about an un-balanced distortion in the push pull output stage!
Have Fun!
1. The construction of two tubes can be different, and still have the same specifications.
The Original 6BQ5 is a Beam Power tube (Beam Formers).
The Original EL84 is a Pentode (Suppressor Grid).
Suppressor Grid and Beam Formers are different construction, but these two tubes have essentially the same specs.
Some manufacturers got the idea to Dual Label them as 6BQ5 / EL84. (The dealer only has to stock one tube).
You can interchange these two tubes, tube rolling, no problem.
If the amplifier uses adjustable bias, you have to re-bias them, just like you would if you replaced the tube with the same type.
Plug and Play (Not Plug and Pray).
The EL34 Pentode (Suppressor Grid), and KT77 Beam Power (Beam Formers), have the same specifications, Except . . .
EL34 Suppressor Grid only connects to pin 1, and is not internally connected to the cathode, But the . . .
KT77 Beam Formers are internally connected to the cathode.
Connect pin 1 to pin 8 (to use an EL34), and then you can interchange the EL34 and KT77.
If the amplifier uses adjustable bias, you have to re-bias them, just like you would if you replaced the tube with the same type.
Plug and Play (Not Plug and Pray).
There is the SV-83 tube. Some try and use it as a 6BQ5 / EL84 substitute.
They find out the SV-83 screen grid often burns out when used in 6BQ5 / EL84 amplifiers. Do Not use it as a substitute.
2. I use the parallel cathodes and CCS phase splitter for my phase splitters, because they are intrinsically balanced.
3. I do not want to have to change the resistors in the phase splitter each time you change tubes.
I figure, use it unbalanced, and let the global negative feedback partially fix the effect of the un-balanced phase inverter, and just live with the performance you get.
But here is a much closer analysis of this kind of paraphase splitter (the more you make the two plate signals equal, the stronger this effect is:
For most paraphase splitter circuit topologies, the input tube sends 2nd harmonic distortion to the top (push) output tube;
and the other tube in the phase splitter sends No 2nd harmonic distortion to the bottom (pull) output tube.
This is often overlooked, and often misunderstood.
Push driven by a signal that has some 2nd harmonic distortion.
Pull driven by a signal that has essentially No 2nd harmonic distortion.
Talk about an un-balanced distortion in the push pull output stage!
Have Fun!
Last edited: