I am doing a class A pp output stage with cathode feedback. The cathode feedback demands a lot of voltage swing from the drivers, about 110-120Vpp, which puts some attention to this stage.
What I came up with is a choke loaded LTP. The plate choke is a two coil choke so there will be magnetic coupling between the phases (think about it as the primary side of a pp transformer). Hopefully this will neutralize even harmonic. 6FQ7/6CG7 is a quite linear tube and choke loaded with just the load of the grid resistors (2 x 150k) in the output stage the odd harmonics should also be kept modest. I have never seen this done before so I really don’t know what to expect, but I am hoping for THD ~0.5% (or less) at 100Vpp.
I will appreciate comments, advices or experience from other gentlemen or ladies in the DIY community.
Jan E.
What I came up with is a choke loaded LTP. The plate choke is a two coil choke so there will be magnetic coupling between the phases (think about it as the primary side of a pp transformer). Hopefully this will neutralize even harmonic. 6FQ7/6CG7 is a quite linear tube and choke loaded with just the load of the grid resistors (2 x 150k) in the output stage the odd harmonics should also be kept modest. I have never seen this done before so I really don’t know what to expect, but I am hoping for THD ~0.5% (or less) at 100Vpp.
I will appreciate comments, advices or experience from other gentlemen or ladies in the DIY community.
An externally hosted image should be here but it was not working when we last tested it.
Jan E.
Hej Jan E
I know I have seen this somewhere, just don't remember where...
I thought of doing the exact same thing with a 6BX7 driving a 845 pp output stage. approx 300Vpp needed. Never got around implementing it though
My idea was to use a cheap PP OPT with secondary left open as the "balanced plate choke".
I do not see that there should be any problems with this approach compared to ordinary choke loading. The advantage is that DC currents cancel magnetisation of the core and you don't need an air gapped choke.
Cancellation of even harmonics is very effective in a LTP with ccs tail even without this approach.
As I said I nener came around actually doing ths so I an very interested in the outcome of your setup.
/Olof
I know I have seen this somewhere, just don't remember where...
I thought of doing the exact same thing with a 6BX7 driving a 845 pp output stage. approx 300Vpp needed. Never got around implementing it though
My idea was to use a cheap PP OPT with secondary left open as the "balanced plate choke".
I do not see that there should be any problems with this approach compared to ordinary choke loading. The advantage is that DC currents cancel magnetisation of the core and you don't need an air gapped choke.
Cancellation of even harmonics is very effective in a LTP with ccs tail even without this approach.
As I said I nener came around actually doing ths so I an very interested in the outcome of your setup.
/Olof
Putting the coupled choke(s) in the plate side will certainly give you more voltage drive headroom with the bootstrap B+ effect. Putting a choke in the cathodes, like the Boozhound circuit, won't give increased plate swing other than saving some B+ loss across the tail resistor.
Capacitance of the windings at high frequency needs consideration, but with two driver tubes, it would not seem to be out of the ordinary.
There are some SS CCS dual gyrator approaches using two P channel HV Mosfets in place of the coupled plate inductors. Michael Koster has worked on these some I think. But they don't give the B+ boost that you are looking for with the inductors.
Don
Capacitance of the windings at high frequency needs consideration, but with two driver tubes, it would not seem to be out of the ordinary.
There are some SS CCS dual gyrator approaches using two P channel HV Mosfets in place of the coupled plate inductors. Michael Koster has worked on these some I think. But they don't give the B+ boost that you are looking for with the inductors.
Don
Thanks for your comments gentlemen!
A load line for a single choke loaded 6FQ7 (150k load) will look something like this:
This tells me that the 6FQ7 is well suited for this kind of service. 200Vpp voltage swing seems to be no problem. Due to magnetic coupling I suppose that the output impedance (Zout) will be something like rp/2.
Jan E.
A load line for a single choke loaded 6FQ7 (150k load) will look something like this:
An externally hosted image should be here but it was not working when we last tested it.
This tells me that the 6FQ7 is well suited for this kind of service. 200Vpp voltage swing seems to be no problem. Due to magnetic coupling I suppose that the output impedance (Zout) will be something like rp/2.
Jan E.
hey-Hey!!!,
The CT choke anode load is nice indeed. I'd suggest using something like a 5687 at a bit more current, say 15 mA/section. IIRC it was Brook amps that did this commercially, albeit a long time ago....
I've gotten excellent results from a simple wind, as in a dual bay bobbin, one wound recerse direction from the other. That puts the turns in the center next to the core, and also leaves an exactly balanced phase-to-phase capacitance.
cheers,
Douglas
The CT choke anode load is nice indeed. I'd suggest using something like a 5687 at a bit more current, say 15 mA/section. IIRC it was Brook amps that did this commercially, albeit a long time ago....
I've gotten excellent results from a simple wind, as in a dual bay bobbin, one wound recerse direction from the other. That puts the turns in the center next to the core, and also leaves an exactly balanced phase-to-phase capacitance.
cheers,
Douglas
Finished at last!
The amplifier performs very well. But one issue was the KT66s. Tungsol reissues where a bit out off the original specs, more like 6L6 so I had to recalculate the cathode resistor in the output stage. It also turned out to be difficult to find good matched pairs.
500Hz at 1dB below max Po:
Here are some other measurements with Tungsol KT66 as output tubes:
o Max Po: 2 x 13W
o Frequency response: 10Hz - 100kHz (Low power +0/-3dB)
o Output impedance: 0.8 ohm
o THD (1dB below max Po): 0.4%
o THD (1W Po): ~0.05%
o S/N greater than 80dB below rated output.
With 6550 max Po is 18W, the other data is about the same or a bit better.
Schematic:
The 10 kHz square wave response is quite neat.
Here with a 100pF compensation cap in parallel with the 15k feedback resistor.
10 kHz square:
Local cathode feedback is about 4dB and global NFB is 7-8dB. The custom output transformers from Sowter UK behaved excellent, no nasty resonances or other issues.
A picture from the inside -- a mess as always..
All in all a very good sounding amplifier without any major faults.
Jan E Veiset
An externally hosted image should be here but it was not working when we last tested it.
The amplifier performs very well. But one issue was the KT66s. Tungsol reissues where a bit out off the original specs, more like 6L6 so I had to recalculate the cathode resistor in the output stage. It also turned out to be difficult to find good matched pairs.
500Hz at 1dB below max Po:
An externally hosted image should be here but it was not working when we last tested it.
Here are some other measurements with Tungsol KT66 as output tubes:
o Max Po: 2 x 13W
o Frequency response: 10Hz - 100kHz (Low power +0/-3dB)
o Output impedance: 0.8 ohm
o THD (1dB below max Po): 0.4%
o THD (1W Po): ~0.05%
o S/N greater than 80dB below rated output.
With 6550 max Po is 18W, the other data is about the same or a bit better.
Schematic:
An externally hosted image should be here but it was not working when we last tested it.
The 10 kHz square wave response is quite neat.
Here with a 100pF compensation cap in parallel with the 15k feedback resistor.
10 kHz square:
An externally hosted image should be here but it was not working when we last tested it.
Local cathode feedback is about 4dB and global NFB is 7-8dB. The custom output transformers from Sowter UK behaved excellent, no nasty resonances or other issues.
A picture from the inside -- a mess as always..
An externally hosted image should be here but it was not working when we last tested it.
All in all a very good sounding amplifier without any major faults.
Jan E Veiset
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.