I have a fixed bias output stage with 6V6's. The stage is driven by a typical Fender style long tail pair phase splitter. Output tube plates only have 315V and screens have 310V. Output tube bias is provided by an arrangment similiar to Ampeg's Gemini's. (Capacitive coupled from mains transformer and zener regulated. mains rectifier is a full wave bridge type.) At idle, bias is steady at -22V. When amp is driven, bias heads progressively negative to -40V. Is this normal? What made me check it was that this fixed bias amp sings kinda like a cathode biased amp.
It's not blocking out, just singing and a little spongy. Your thoughts.
It's not blocking out, just singing and a little spongy. Your thoughts.
I assume that you are measuring this at the 6V6 grids
- if not ignore this.
What you are seeing IS "blocking" - just not enough to cause full blocking distortion.
When you drive the amp hard enough such that the signal peaks go above 0V then the 6V6 start conducting grid current and the coupling caps from the phase splitter charge up, thus shifting the bias negatively.
You can "play" with this by:
1) Increasing the 6V6 grid stop resistors so that grid current is reduced and the coupling caps don't charge as much. Try values up to 47K. Go as high as you can without sacrificing brightness.
2) Decreasing the grid 1 to 0V impedance through the bias network. The value of the resistor to the bias pot wiper is an obvious one to reduce - so long as you are not then overloading the phase splitter.
3) Reduce the output impedance of the phase splitter.
Split the longtail pair anode load resitors into to two resistors each of equal (half) value and feed the output tubes from the mid point. This halves the drive signal voltage and halves the drive impedance. Both will help.
I'd start with 1) and ONLY try 2) and 3) if you absolutely have to.
Cheers,
Ian
- if not ignore this.
What you are seeing IS "blocking" - just not enough to cause full blocking distortion.
When you drive the amp hard enough such that the signal peaks go above 0V then the 6V6 start conducting grid current and the coupling caps from the phase splitter charge up, thus shifting the bias negatively.
You can "play" with this by:
1) Increasing the 6V6 grid stop resistors so that grid current is reduced and the coupling caps don't charge as much. Try values up to 47K. Go as high as you can without sacrificing brightness.
2) Decreasing the grid 1 to 0V impedance through the bias network. The value of the resistor to the bias pot wiper is an obvious one to reduce - so long as you are not then overloading the phase splitter.
3) Reduce the output impedance of the phase splitter.
Split the longtail pair anode load resitors into to two resistors each of equal (half) value and feed the output tubes from the mid point. This halves the drive signal voltage and halves the drive impedance. Both will help.
I'd start with 1) and ONLY try 2) and 3) if you absolutely have to.
Cheers,
Ian
Thanks Ian, I measured the voltages on the phase splitter side of the coupling caps. When this begins the plate voltages on the phase splitter dives by 20V. I made the measurement at the junction of the power tube grid resistors. Same thing is happening at the power tube grids. I'll try the fixes listed and let you know how it goes.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.