On 12 jun, 23:42, Owen Duffy wrote:
walt wrote :

...

What you have said above is the key to the concern over the output
resistance of a Clsss C amplifier being non-dissipative. What seems to
be universally misunderstood is that there are really two separate
resistances in the operation of these amps; one, the cathode-to-plate
resistance, which is the dissipative resistance Rpd that accounts for
all the heat, due to the electrons striking the plate; and two, the

It is my understanding that the average power (heat) generated at the anode *
of a triode can be found by averaging the product of the instantaneous
anode current and anode-cathode voltage over time. In a Class C amplifier,
the voltage and current are not linearly related to each other, ie there is
no constant of proportionality, no constant or fixed resistance.

I don't understand why then, that people try to explain the anode
dissipation in terms of some value of resistance.

Owen


Hello Owen,

I fully agree with you. When I am doing power electronics, I show
myself the instantaneous current*voltage plot, and the averaged
integral. The last one shows the dissipated power, the first one
tells me where I have my losses (and also what I have to change to
reduce the overall loss or reduce component stress).

Regarding the class C output impedance issue, I updated my simulations
with a 3.6 MHz output stage with a 6146 tube and real class C
operation.

You can almost get every impedance you want, also the conjugated match
condition, but you need to do many simulations to find that point.
With hours of time, I couldn’t get closer to an output VSWR of 1.58
with 79% efficiency. I know you can get closer, but at his moment I
cannot provide you a spice file as I don’t have it anymore and I also
dropped orcad long time ago. .

The smallest change in output loading or drive level results in
significant change of output impedance. So take an arbitrary class C
amplifier, measure its output impedance and it will very likely be way
off the intended load (in my example 3500 Ohm).

If you want to use load change method with some different resistive
loads, you should make phase measurements to get valid results, as you
don't know whether you amplifier has real or complex output
impedance.

For the class C amplifier I changed the drive by +0.36 dB, resulting
in an increase of plate efficiency from 79% to 80%, but the output
VSWR changed from 1.58 to 3.6. Also reduction of input drive gives
significant change in VSWR (magnitude and phase).

I hope that people will reproduce some of the simulation themselves to
develop a solid opinion on their own.

Best regards,


Wim
PA3DJS
www.tetech.nl
without abc, PM will reach me.