Main parameter for frequency response is the leakage inductance.
Capacitance can shift the roll-off up and down in frequency and also the smoothness (or peaking) by forming a resonance together with the effective R as well. But leakage inductance is the primary factor.
4nF is massive capacitance in my book, unless we are talking about really BIG transformers. At typical EL84 SE transformer (at least mines), properly done, will barely have some 0.5-0.7 nF. I have never had a transformer with more than 1.5 nF, in the worst case. I remember clearly that the SE transformer I designed for Ale to work with the 845 and 814 triode-strapped SE amp had some calculated 1.2 nF and he actually measured less, around 1 nF, despite its simplicity.
PP transformers also benefit of the topology which halves the effective capacitance respect to SE.
Any reasonably engineered amplifier, pentode or triode, MUST work without any global feedback, IME. This is not an OP-amp.....
Feedback can be connected/tested late, after checking that grounding is as it should be. As others said, grounding is necessary for use of global feedback.
Capacitance can shift the roll-off up and down in frequency and also the smoothness (or peaking) by forming a resonance together with the effective R as well. But leakage inductance is the primary factor.
4nF is massive capacitance in my book, unless we are talking about really BIG transformers. At typical EL84 SE transformer (at least mines), properly done, will barely have some 0.5-0.7 nF. I have never had a transformer with more than 1.5 nF, in the worst case. I remember clearly that the SE transformer I designed for Ale to work with the 845 and 814 triode-strapped SE amp had some calculated 1.2 nF and he actually measured less, around 1 nF, despite its simplicity.
PP transformers also benefit of the topology which halves the effective capacitance respect to SE.
Any reasonably engineered amplifier, pentode or triode, MUST work without any global feedback, IME. This is not an OP-amp.....
Feedback can be connected/tested late, after checking that grounding is as it should be. As others said, grounding is necessary for use of global feedback.
Last edited:
4nF is massive yes, although it exists.
Leakage inductance is the better roll-off factor. Let's not forget that typical speaker impedance rises with frequency, so leakage becomes less of a factor vs frequency.
IIRC, push pull transformer capacitance can decrease from 2 to 4 times. The factor significance increases with the interleaving factor. Crowhurst and RDH4 give such info, but you can also calculate them by hand for each specific configuration.
I designed a 100W SE 5k OPT for Ale Moglia, which has 1.5nF input capacitance. Although the original idea was a 35W OPT for an 845 tube.
Leakage inductance is the better roll-off factor. Let's not forget that typical speaker impedance rises with frequency, so leakage becomes less of a factor vs frequency.
IIRC, push pull transformer capacitance can decrease from 2 to 4 times. The factor significance increases with the interleaving factor. Crowhurst and RDH4 give such info, but you can also calculate them by hand for each specific configuration.
I designed a 100W SE 5k OPT for Ale Moglia, which has 1.5nF input capacitance. Although the original idea was a 35W OPT for an 845 tube.
Last edited:
Yeah that's a big SE transformer but not the BIG I was thinking about! 😀
And that confirms what I was saying, actually.
And that confirms what I was saying, actually.
One end of the OPT secondary may be grounded in order to enable taking negative feedback from one of the other connections.
This is an interesting thread. 26 Posts.
It is all about the theoretical, with a few experiences thrown in.
I am glad it did not start with: "My amplifier high frequency response and high frequency distortion sounds horrible".
If it did, hopefully we would have solved the problem by now.
I think we have covered all the theory and experience that is needed.
Just read all 26 posts Very Carefully.
It is all about the theoretical, with a few experiences thrown in.
I am glad it did not start with: "My amplifier high frequency response and high frequency distortion sounds horrible".
If it did, hopefully we would have solved the problem by now.
I think we have covered all the theory and experience that is needed.
Just read all 26 posts Very Carefully.