Every day more Fender Deluxe 5E3 circuits come to life for the first time than perhaps any other design. A simple internet search of "Fender Tweed Deluxe kit" turns up enough links to force even Fender itself to consider entering the hot DIY market. Let's take a closer look at the 5E3 power amp.
Here is the power amp and power supply. The voltages are calculated values, not measured values, as described below.
Let's assume a vintage screen supply of about 320 volts and a DC grid bias of about minus 20 volts and see if it makes sense according to 6V6 and 5Y3 data sheets. (This is at the low end of modern day measurements.) Subtracting 20 volts from cathode to ground, the screen voltage is 300 volts.
The current through the 250 ohm cathode resistor is 20/250 = 80mA, or 40mA per tube. The transfer curves for the screen current shows about 3mA for a 300 volt screen and a minus 20 volt grid.
This makes the plate current 37mA per tube.
We'll estimate 1.5mA per preamp triode for a total of 6ma DC in front of the power amp. That means the drop across the 5k power supply filter resistor is (3mA + 6mA)(5k) = 45 volts. So the plate supply is at 365 volts and the idle plate voltage is 345 volts (10 percent higher than the 6V6GT max rating of 315 volts). According to the 5Y3 data sheet, shown below, a load of 80 + 6mA = 86mA and a 330-0-330 transformer create a plate supply voltage of 355 volts, close enough to our estimate of 365. We'll call it 360.
Plate dissipation is the product of plate voltage and plate current: (240)(37mA) = 12.6 watts, slightly above the maximum rating of a 6V6. This is a hotly biased amp.
According to our LC Ripple Filter Calculator there is 35dB of ripple attenuation for 60-Hertz AC across the 5k resistor and 48dB across the 22k resistor. Class AB operation, a vacuum tube rectifier, and an RC front end instead of a choke create considerable screen supply voltage sag at full power, but none of the slowly oscillating transients that are created by LC filters like the one in the Fender Bassman.
To get an idea of how much voltage sag to expect at full power we can look at the GE data sheet, which states that for a screen voltage of 285 volts and a DC grid bias of minus 19 volts the total plate current rises from 70mA at idle to 92mA at full power. Screen current rises from 4mA to 13.5mA. The total increase is 32mA. Our screen voltage is higher at idle and lower after full sag. So the current increases will be proportionately somewhat smaller. We estimated that the power supply provides 86mA at idle. At full power an increase to 118mA, according to the 5Y3 curves, drops the output voltage from 355 volts to about 335. This means the plate voltage relative to the cathode is in the vicinity of 315 volts.
A 10mA increase in screen current creates 50 more volts of drop across the 5k resistor. Add this to 20 volts of power supply sag and we get a screen supply voltage at about 250 volts. Subtracting 20 volts from cathode to ground gives us a screen-to-cathode voltage of 230 volts. When seriously overdriven, where the input signal looks more like a square wave, we might even reach this level. We'll assume something in the neighborhood of 250 volts.
Here are 6V6 plate characteristic curves for a control grid voltage of zero. The top curve, highlighted in blue, is for a 250 volt screen. The effective impedance for pure Class A is half the plate-to-plate primary impedance, or 4k, which is depicted by the green load line. For Class B we divide by 4 to get 2k, which is represented by the red line.
The 5E3 sits somewhere in the middle between Class A and Class B, creating a load line closer to the knee of the curve. Taking the knee (plate voltage 30, plate current 92mA) as the point of maximum swing at full power, the plate voltage swings from 315 volts to 30 volts while the net plate current rises from zero to 92mA. Total output for a sinewave is (315-30)(92mA)/2 = 13 watts.
If 13 watts seems a bit low, consider the instant that a full power signal is applied, before the screen and plate voltages have had time to sag. We can imagine a 300-volt curve with a knee at about 40 volts, 120mA. This puts maximum power at (315-40)(120mA)/2 = 17 watts, closer to what we've come to expect from a push-pull 6V6 amp. Historically the original 5E3 tweed design doesn't push the 6V6 to its limits. The Fender Deluxe 6G3, for example, raises the screen supply to 365 volts, cuts the 5k filter resistor to 1k, and uses a higher capacity GZ34 rectifier instead of a 5Y3. Even a vintage 5E3 operates at higher power levels by the increased AC line voltages we experience today.
The input signal amplitude needed to drive the power amp to full power is equal to the DC grid bias, so the Fender Deluxe 5E3 power amp has an input sensitivity of about 20 volts. There is no negative feedback from the output transformer secondary to the phase inverter, so unlike the Marshall JCM800 Model we don't need to compute forward gain and closed loop gain. The input sensitivity for the phase inverter is simply 20 volts divided by the phase inverter gain. The 5E3 uses a Concertina with a gain of slightly less than unity, so its input sensitivity is about the same: 20 volts.
1 Richard Kuehnel, Circuit Analysis of a Legendary Tube Amplifier: The Fender Bassman 5F6-A, 3rd ed., (Seattle: Pentode Press, 2009).
2 Richard Kuehnel, Vacuum Tube Circuit Design: Guitar Amplifier Power Amps, (Seattle: Pentode Press, 2008).
Copyright © 2005-2015 Amp Books LLC All Rights Reserved