According to Fender's schematic, the Champ 5E1 single-ended power amp consists of a 6V6GT operating at a screen voltage of 305 volts.
The parts values are
CG = 0.02uF
RG = 220k
RK = 470
CK = 25uF
The Fender's measured voltage across the 470-ohm cathode resistor is 19 volts, so by Ohm's Law the current through it is 40 milliamps. This is the current through the cathode, which is the sum of the plate and screen currents. Let's see how that corroborates with published transfer characteristics. Here is the screen current as a function of control grid voltage.
The screen current is approximately 3 milliamps, so subtracting that from the cathode current we get 37 milliamps of plate current. This is very close to the 36 milliamps we can estimate from the transfer characteristics shown here:
This is closer than most bench test results, so we'll accept the 19 volt grid bias at face value. If we imagine a curve on the plate current plot corresponding to a screen voltage of 305 volts, we can envision the lower left end crossing into cutoff somewhere in the vicinity of -36 grid volts, so the tube swings from cutoff at
VG = -38
to saturation at
VG = 0
centered at the DC grid bias voltage of
VG = -19
We conclude that the DC operating point is defined by
VGQ = -19
VPQ = 305
IPQ = 37mA
representing the quiescent grid voltage, plate voltage, and plate current.
The transconductance of the 6V6 is around 3.75 milliamps per volt. The cathode impedance is the reciprocal of this value in parallel with the cathode resistor RK, which is 170 ohms. The -3dB bass cutoff frequency for the cathode capacitor CK is thus 37 Hertz, so the cathode is fully bypassed for maximum voltage gain and the power amp reaches full power when driven by a signal amplitude of 19 volts.
At a plate voltage of 250 volts, a screen voltage of 305 volts, and a grid voltage of zero the plate current should be close to 140 milliamps, as shown by the green circle here:
This corresponds to the green circle on the plate characteristic curves shown here:
So we can imagine that the plate characteristic curve for a screen voltage of 305 volts and a grid voltage of zero volts roughly follows the green lines. (We ignore all the black curves because they are for lower screen voltages.) The knee of the curve is marked by a red circle. If Fender had designed the amp so that positive grid voltage swings at maximum power strike the knee of the curve (which is typical for push-pull amplifiers) then the AC load line would be the red line connecting the knee to the DC operating point (the blue circle).
A load line through the knee raises some significant concerns. From its idle value of minus 19 volts a full-power sinewave input drives the grid positive to zero volts and negative to minus 38 volts. This represents a symmetrical swing of plus and minus 19 volts centered on the idle value of minus 19 volts. From the load line the output is substantially distorted, because the plate voltage swings from an idle value of 305 volts to a minimum of 25 volts to a maximum of 425 volts. This represents a 280 swing in the negative direction and only a 120 volt swing in the positive direction. A positive grid voltage swing of 19 volts produces a negative plate voltage swing of 280 volts, an amplification factor of
280 / 19 = 15
A negative grid voltage swing of 19 volts produces a positive plate voltage swing of
120 / 19 = 6.3
This produces considerable 2nd harmonic distortion, not necessarily undesirable in a modern guitar amp, but almost unthinkable in the 1950s. At full power the total plate voltage swing is 425-25 = 400 volts. The total plate current swing is 120 - 0 = 120 milliamps. The output transformer primary impedance reflects the ratio of voltage swing to current swing:
RP = 400 / 0.12 = 3.3k
As an alternative, let's look at what we would need for a symmetrical output at full power. Since the idle plate current is 37 milliamps, at full power the plate current needs to rise to double this value: 74 milliamps. So at full power the plate current swings between 74 milliamps and zero milliamps, centered on the idle value of 37 milliamps. This is shown by purple load line.
The symmetry greatly reduces 2nd harmonic distortion, but sacrifices power. From the load line we see that the the plate voltage swings from 305 volts down to 20 volts, a negative swing of 285 volts. The positive swing, although beyond the right edge of the graph, is therefore 285 volts higher than 305 volts: 590 volts. Total voltage swing is 590 - 20 = 570 volts and total current swing is 74 - 0 = 74 milliamps. The output transformer primary impedance is
RP = 590 / 0.074 = 8k
This represents the upper extreme of what most designers use. The previously calculated 3.3k represents the lower extreme. The Radiotron Designer's Handbook suggests that a good general rule for single-ended designs is a primary impedance of
RP = (0.9)VPQ / IPQ = 7.4k
Fender chose a value close to this number: 7k, which is represented by the blue line here:
This provides almost 6 watts of output power measured at the primary.
Based on our Grid Bias Excursion Calculator the -3dB bass cutoff frequency is 31 Hertz, well below the lowest note on a guitar with standard tuning (82 Hertz). It's still high enough to help prevent blocking distortion and motorboating, however, as indicated by the relatively short bias recovery time.
1Richard Kuehnel, Vacuum-Tube Circuit Design: Guitar Amplifier Power Amps, (Seattle: Pentode Press, 2008).
Copyright © 2005-2016 Amp Books LLC All Rights Reserved