Conceptually, this the only way to go, as four triodes configured in a fully balanced layout and run in a strict Class A mode will yield the greatest voltage swing with the lowest distortion. In this arrangement, each stator will see the same impedance and voltage swing. The output impedance would equal the rp of the triode type used.

    Still, if the right transformers can be found and the wallet can take the abuse, this arrangement might prove to be the optimal solution, in terms of elegance of design and purity of sonics. The transformer, however, can be replaced by circuitry.
   The top triodes require a drive voltage but the easy circuit that readily comes to mind, the SRPP, shouldn't. While the SRPP does provide for simultaneous driving of both top and bottom triodes, it works best with a strictly resistive load, not a purely reactive one. The capacitive load has an impedance that decreases with rising frequency. In fact, from 20 Hz to 20 kHz the impedance will fall by one thousand fold. Shunting the capacitance with a fixed resistor would help the circuit work better, but at the cost of less output voltage swing. Of course, experimentation is welcome.

   In practice, however, the problem of equally driving all the triodes must be solved. The bottom triodes' grids are easy to drive as their cathodes are remain at a fixed voltage. The top triodes, on the other hand, have cathodes that swing with the signal. Keeping all these triodes running with equal grid-to-cathode voltages is tricky, but not impossible.
   If high quality coupling transformers can be found this might make for the simplest and best solution. In fact, if a high step up ratio, say 10 to 1, were used, no other tubes may be needed, as a 1 volt input signal would become 10 volts of peak drive signal for the output triodes. A capacitorless amplifier! Optimally, four transformers would be used per channel, one per triode. This would help keep parasitic capacitances between windings to a minimum, which would better allow each top triode's transformer to float with the large cathode voltage swings. The problem with this rosy scenario, is that high quality transformers are anything but common or cheap.

< PREVIOUS

pg. 4

NEXT >