|
Add to these advantages the lack of an output shunting capacitor, the lower voltage parts, the reduced shock hazard, the cooler operation, and the current sourced version looks even better still.
If current source are so good, why are they not used more often in tube circuits?
Part of the answer lies in the unfamiliarity of the topologies used. A PNP transistor with a breakdown voltage of 300 volts did not exist in the 1950's and consequently none of the classic tube audio gear from that era used solid-state current source loading. (Besides, for many, a Cathode Follower is the very limit for exotic circuits.)
Beyond the newness of the current source topologies, there is a mistaken belief that voltage regulators are not in the amplifier's signal path, in contrast to the current sources, which certainly are in the signal path. In reality, both are very much in the signal's current path and consequently, very much part of the amplifier's signal path. Whether we start at the beginning or middle or end, as long as the AC signal flows out of or into the device, it is in the signal path. Thus, another advantage of the current source loading of a triode is that we are less likely to forget that it is actively part of the electrical circuit.
Because it is part of the circuit, it should receive as much care in design as any other portion of the circuit. Current sources cannot be bypassed by high quality film capacitors, like the three-pin voltage regulators used to set the cathode bias. Therefore, if we are concerned about the quality of the parts and the use of feedback in the design of the tube portion of the circuit, we should be equally concerned about the parts quality and use of feedback in the design of the current source. My recommendation is to stick with elegant, i.e. simple topologies, whose small part count will allow easy part substitution, testing and optimization.
|
|
