As little as one resistor per channel is all it might take to give this trick a whirl. If the polarizing voltage is sourced from the line stage amplifier, then a high ohmage resistor is shunted across the coupling capacitor at the output. The value of the resistor will depend on the amount of desired polarizing voltage and the value of the resistor that references the coupling capacitor to ground.  Basically, what happens is that the addition of the resistor creates a voltage divider that leaks a small percentage of the plate voltage into the output.  For example, given a 10 volt polarizing voltage and a plate voltage of 210 volts and 500k terminating resistance at the output of the line stage, then a 10M resistor will be needed to shunt across the coupling capacitor.
 What happens when the modified preamp or power amplifier is connected to a DC coupled output stage of a solid-state preamplifier or a DC coupled input stage of a solid-state amplifier? Maybe the loss of your speaker's woofers and output stage of your solid-state amplifier. Solid-state amplifiers are frail in this regard. Ten volts of DC at the input of a DC amplifier with a gain of 30 is catastrophic. So catastrophic, in fact that I will not make the recommendation that a switch be used to switch the extra resistor in and out of the circuit, as it might get switched by accident and toast your solid-state amplifier. The only mitigating grace is the fact that most solid-state amplifiers have a relatively low impedance on the order of 10k to 47k. This low value will reduce the potential 10 volts down to a low of 0.1 volts and a high of 0.5 volts, which is obscenely high for an input offset voltage, not at least it is not as horrendous as the full 10 volts. Remember, some solid-state amplifiers have a much higher input impedance, as they use FETs or MOSFETs at the input of the amplifier. These voltage driven devices allow for high impedance inputs, as they do not require the input current that the transistor does. If you blow up your solid-state amplifier, you cannot say that you were not warned.   What happens if the power amplifier receives the modification and it is hooked up to a solid-state preamp or line stage? Maybe the loss of the output stage of the unit, but probably not, as the low output impedance of the unit, 10 to 300 ohms will voltage divide the 10 volts down to millivolts. Furthermore, the output may sport a coupling capacitor or a DC servo loop that will pull down the 10 volts to 0 volts (the amount of current needed to pull the 10 volts down 0 volts is a trivial 0.21 mA). In other words, this is the safer modification.   If anyone tries this trick, please keep us posted with your findings                                   //JRB
 On the other hand, if the polarizing voltage is sourced from the power amplifier, then the high ohmage resistor is shunted across the input  capacitor at the input of the amplifier.
 pg. 14