Hello all,
I came across a hi fi tube amp schematic that loosely resembled a Williamson type topology but had included a cathode follower between the phase inverter and the power tubes.
I was curious to what you would expect the inclusion of that stage to offer in terms of both advantages and disadvantages, mostly relating to effect on potential output power vs any other characteristic.
Thank you,
Best,
PJD3
I came across a hi fi tube amp schematic that loosely resembled a Williamson type topology but had included a cathode follower between the phase inverter and the power tubes.
I was curious to what you would expect the inclusion of that stage to offer in terms of both advantages and disadvantages, mostly relating to effect on potential output power vs any other characteristic.
Thank you,
Best,
PJD3
Good point about minimizing blocking. DC-coupling the output of the CF to the grid of the output tube means there's no time constant there to recharge after the output tube grid draws current. However, if the CF output is RC-coupled to the output tube grid then there is still that RC time constant to recharge after grid current.
If the cathode followers have a pretty good amount of plate current going through them, and their cathodes are DC-coupled to the output tubes' grids, then cathode followers driving the grids of the output tubes allow some excursion into Class AB2 for a bit more power output. As the output tubes' grids draw current their input impedance goes way down. The cathode followers' low output impedance can cope with that better than if the output tube grids are driven from the driver tubes' (higher output impedance) plates.
Higher plate current in the cathode followers reduces the possibility of slew rate limiting from the driver stage, which could be important if the output tubes are triodes, or pentodes with 'ultralinear' OPT primary. Triodes and pentodes with UL loading have higher input capacitance than pure pentode operation.
If the cathode followers have a pretty good amount of plate current going through them, and their cathodes are DC-coupled to the output tubes' grids, then cathode followers driving the grids of the output tubes allow some excursion into Class AB2 for a bit more power output. As the output tubes' grids draw current their input impedance goes way down. The cathode followers' low output impedance can cope with that better than if the output tube grids are driven from the driver tubes' (higher output impedance) plates.
Higher plate current in the cathode followers reduces the possibility of slew rate limiting from the driver stage, which could be important if the output tubes are triodes, or pentodes with 'ultralinear' OPT primary. Triodes and pentodes with UL loading have higher input capacitance than pure pentode operation.
Member Tubelab has an excellent article describing his PowerDrive circuit, which is a MOSFET source follower used to drive output tubes. It explains the reasons why a low impedance, high current driver is a good thing for driving a tube output stage. It should help you understand why someone would want to add cathode followers in front of the output tube grids in a Williamson amp.
http://tubelab.com/articles/circuits/power-drive/
http://tubelab.com/articles/circuits/power-drive/
cathode follower drivers have one main advantage apart those already mentioned and thats very high input resistance of follower. Normally you need to use reasonably low grounding resistance in output tubes grids to prevent grid-leaking, lets say 220k or so. By using follower you will have very high current drive capability while maintaining high input impedance, which lowers distortion of previous stage = driver/invertor
There is one disadvantage and thats quite big negative voltage requirements for the follower, as its cathode needs to be at -Ug1 of output tube minus signal swing without limitations, so smth like over -100V. I have used this configurations several times... one example see link below, 6N6P as direct coupled drivers of EL34's, this can do 80W in triode AB2:
https://www.elektrolab.eu/blog/vacuum-tube-power-amplifier-va-tube-amp
(you can use google translate to translate the page, should work somehow)
There is one disadvantage and thats quite big negative voltage requirements for the follower, as its cathode needs to be at -Ug1 of output tube minus signal swing without limitations, so smth like over -100V. I have used this configurations several times... one example see link below, 6N6P as direct coupled drivers of EL34's, this can do 80W in triode AB2:
https://www.elektrolab.eu/blog/vacuum-tube-power-amplifier-va-tube-amp
(you can use google translate to translate the page, should work somehow)
Thank you for the informative responses and I will certainly be reading over Tubelabs article on the MOSFET source follower for this application. I knew it may have something to do with the advantages of a lower impedance driver such as preceding a passive tone circuit but, that was about it. One of my next projects is to breadboard up a channel of this topology and see for myself to see what the differences may be from a sonic perspective.
Excellent, thanks,
PJD3
Excellent, thanks,
PJD3
Another good example of driving the output stage with a cathode follower is Norman Koren's "The Emperor's New Amplifier" design from back in 1997. This type of drive is almost mandatory when operating the output stage in class AB2, like in this design, in order to eliminate blocking distortion.
http://www.normankoren.com/Audio/TENA.html
http://www.normankoren.com/Audio/TENA.html