Cathode Bias vs Fixed Bias
For those who wonder whether fixed bias of small signal stages makes a difference in sound quality, here is a comparison between a Sovtek 12AX7LPS stage using cathode bias and bypassed vs the same tube with a battery substituted for the cathode bias resistor and capacitor.
The system was set up for 260V High Voltage feeding a 100k load resistor to the plate. There was a 100k resistor in series with the grid to the signal source, the source having a low resistance DC path to ground. In one case the cathode bias resistor was 1.8k. This provided a bias point of 1.52VDC on the cathode and 172 volts on the plate. In the other case, the resistor and capacitor were replaced by a D Cell. The D cell had an open circuit voltage of about 1.518 volts. When connected in the circuit, the AC voltage gain was the same as with the resistor and bypassing capacitor, and the DC quiescent point was also 172 volts on the plate.
Here's the difference in distortion performance:
Notice that there is a small but noticeable difference in performance. The battery biasing seems to have slightly lower higher order distortion components.