Hi Mark,
It is difficult to truly simulate the dynamic resistance of a high vacuum rectifier because the voltage drop versus current isn't linear and as you approach the full normal current draw the drop (and power dissipated in the resistor) become very high. The current draw is probably in the neighborhood of 500 mils in AM on voice peaks which would result in 135 volts drop and around 70 watts dissipation in a 270 ohm dropping resistor.
The Apache uses a choke input HV supply with a fairly stiff bleeder which does help regulation and the voltage rating for the caps gives you a little leeway for increased voltage.
In general I usually leave high vacuum rectifiers in place, it is the MV units I don't trust
Exceptions for me are my Hallicrafters HT-30, HT-32, HT-32B, and HT-37 transmitters which have the known weak insulation for the rectifier filament winding. For those I use plenty of series resistance and inrush limiting.
On edit: MV tubes are very easy to properly substitute using silicon rectifiers. Voltage drop across a MV rectifier is fairly constant ranging between 5 to 15 volts for a healthy tube within rated current flow. I just add a small series resistor (in addition to inrush limiting) to reduce the initial heavy capacitor charging current. You could almost perfectly simulate the voltage drop of MV by using a large number of lower PIV rectifiers (each with a ~.7 drop) instead of a few higher PIV units. A combination of this plus a series resistor could more closely simulate the operation of a high vacuum rectifier but the number of junctions required would be impractical and is rarely necessary.
