|
Page 13 |
|
2003 |
|
Copyright © 2003 John Broskie All Rights Reserved |
|
Well actually, I did once read in my horoscope how this UFO had landed on the UC Berkeley campus and how it had forced a few grad students to rewrite the code used by the SPICE engine so that… |
|
Returning to earth, at least six different amplifier topologies can be had from two large power supply caps, two tubes, and a signal reference. We have seen four so far. The remaining two are variations on the circlotron circuit. In the first as shown on the top of the next page, we shift the reference to the left output terminal and, in the second, to the right terminal. Once again, after the driver circuitry has been updated to work with the new reference points, the amplifiers perform identically with the other variations, offering the same gain, distortion, and bandwidth. But not output impedance. Both the totem-pole configuration and the circlotron configuration whose signal reference lies midway across the output have higher output impedances than the four other variations. Why? The reference point is used by the output tubes countering any grid-to-cathode voltage perturbations. So when we placed the signal reference at the midpoint of the output, we effectively halved the transconductance that the output tubes use to oppose perturbations at the output, as each grid only sees half of the perturbation voltage. So is the output impedance exactly half that of the other four variations? It is, if the triodes used have an infinitely high mu. While the transconductance has been halved, the each triode’s rp remains constant and the two rp’s are effectively in parallel with the output load impedance. Put mathematically, Zo = rp/(mu +2), |
|
Circlotron output tubes conduction over two cycles. Single ended? Push pull? Class-A? You decide. |
