previous: lion This interesting issue came up in a... -- 1/10/2000 4:19 PM

Re: Cathode follower drive to output tubes

1/10/2000 5:09 PM

Randall Aiken

Re: Cathode follower drive to output tubes Those "humps" that get worse as you turn up the volume are crossover distortion, but not caused by incorrect bias. They are caused by the grid clamping I mentioned earlier. The peaks of the grid waveform are clamped to a spot just above the cathode voltage, and the rest of the waveform is "pushed down" as the level is increased, so the effect is the same as if you had adjusted the bias more negative, resulting in crossover distortion.     A cathode follower will prevent this, but you must make sure the cathode resistor is driven from a negative supply that is larger than the bias voltage (the best bet is to tap the raw negative bias before it is run through the attenuator/trimpot to get the output tube bias - it may require an additional filtering stage).     An extra 12AX7 will only add a few mA to the B+ supply, but will add an additional 300mA to the filament supply. Whether or not it will be okay depends on how close to the design limit your tranny is.     The physical placement of the tube is important, it should be near the output tubes. Cathode followers can oscillate, so you will need a small grid stopper on the grid of the output tube (5.6K is fine).     The plate-cathode voltage on the CF should be no more than 300V or so. If your raw bias is -50V you can supply it with +250V. The cathode follower goes right in between the "normal" bias feed resistors and the "normal" grid stoppers.   Just break that connection and insert two cathode followers.     Randall Aiken

 

Replies:

R.G. q{Just break that connection and in... -- 1/10/2000 5:22 PM Graywater Folks, No disagree... -- 1/10/2000 6:43 PM