5F4 Tweed Super‑style · 1957–1960 · 26 W

The tweed Super shares the Bassman's front half — a 12AY7 two-channel front end, a 12AX7 gain stage feeding a direct-coupled cathode follower, and a fixed-bias pair of 6L6-family output tubes — but its back half is its own circuit: a split treble/bass tone network unlike the Bassman ladder, and a driver + split-load (cathodyne) phase inverter rather than a long-tailed pair. Produced across the late-tweed years (1957–1960), it shares the 5F6-A's power iron almost part-for-part (PT 8087, choke 14684, a 5881/6L6G pair biased near −40 V) while running a slightly lower rail set. Its direct ancestor is the narrow-panel 5E4-A, which it follows stage for stage — the same two-channel 12AY7 front end, the same cathode-follower treble/bass network, the same driver + split-load cathodyne, the same selenium bias supply — changing the output pair back to 6L6G and lifting the rails.

Circuit walkthrough (short form)

Two channels, two jacks each (each jack through its own 68k stopper; the 1M leak at jack 1's tip) → V1 12AY7 (100k plates, shared 820 Ω cathode with 25 µF bypass) → 0.02 µF couplers → 1M volume pots → 270k mixers → V2A 12AX7 (100k plate, 1.5k/25 µF cathode) → V2B cathode follower, DC-coupled (100k cathode load) → the split tone network (below) → straight into the grid of V3A, a cathode-biased 12AX7 driver stage (100k plate, 1.5k cathode) → 0.02 µF → V3B split-load cathodyne: 56k plate, 1.5k

• 56k under the cathode, 1M grid leak returned to the 1.5k/56k junction → 0.1 µF couplers from the cathodyne's plate and cathode → 6L6G pair, fixed-biased at −40 V through 220k leaks (1.5k grid stoppers), screens tied straight to the +410 V node → Fender 45216 output transformer, its centre tap on that same +410 V node, into the two 10-inch speakers. A 56k negative-feedback resistor returns from the speaker to V3A's cathode, where the 5k presence pot bleeds it to ground through 0.1 µF.

Power: 5U4G rectifier → +415 V reservoir → choke (14684) → +410 V screens and output-transformer centre tap → 10k → +332 V phase-inverter supply → 10k → +280 V preamp. A selenium rectifier with a 6.8k/56k network supplies the −40 V bias, fed from a dedicated tap on the 8087's high-tension winding — between one end and the centre tap, the lead the layout letters red-blue — and not from a rectifier plate; the schematic here draws it from that tap.

The output tubes are lettered 6L6G on the C-EG sheet; Fender equipped the tweed Super with 5881s, the ruggedized 6L6 the Bassman and Twin also used, and the two are interchangeable. Simulation here uses the 6L6-family 5881 model.

The tone network the sheet draws

The 5F4's two knobs are not a two-knob cut of the Bassman stack — the C-EG sheet draws a different network, in which treble and bass ride two separate branches off the cathode follower and recombine at the phase inverter's grid:

• Treble: the follower's cathode feeds a 250 pF capacitor into one end of the 1M treble pot; the pot's other end reaches ground through 0.01 µF, and the wiper is the output.
• Bass: the cathode also feeds a 0.1 µF coupler into a node carrying a 220k leak to ground (the amp's 4.7M feedback resistor returns here too), then 100k in series into the 1M bass pot's wiper — the branch injects at the wiper, not at an end lug. One end lug reaches ground through 0.005 µF, the other is grounded outright. From the wiper node, 220k carries the branch to the output.
• The recombined output node is the driver's grid: no coupling capacitor and no grid-leak resistor follow the network — its own 220k-and-track path to ground sets the grid's DC.

There is no slope resistor and no shared cap ladder, so the two controls interact far less than the Bassman family's do — and the treble and bass curves cross over with no scoop carved between them.

The phase inverter the sheet draws

The C-EG drawing prints no long-tailed pair. V3A is a plain cathode-biased gain stage — 100k plate at the printed +213 V over a 1.5k cathode at +1.7 V — and V3B is a split-load cathodyne: 56k in the plate (printed +270 V), 1.5k

• 56k under the cathode (printed +55 V and +53.3 V), with the 1M grid leak returned to the 1.5k/56k junction so the stage biases itself. The two 0.1 µF output couplers leave from the cathodyne's plate and cathode — one inverted output, one not, which is the whole job. The 56k feedback resistor and the 5k presence pot land on V3A's cathode: cathode-injected feedback, with the presence control varying how much of it the 0.1 µF bleeds to ground.

The heater circuit, as the sheet draws it

The 6.3 V supply is single-ended. At the transformer one green lead meets the red-yellow high-tension centre tap at a solder point and goes to a ground hatch; the other green runs to the pilot light — which the sheet gives a single terminal, the feed passing through it — and on to the heaters. Each socket then returns to chassis on its own. At all three nine-pin bottles a short bow outside the socket ties pins 4 and 5 together and pin 9 carries a lead up to a ground hatch: the 6.3 V parallel arrangement, in which the two heater ends are one leg and the centre tap is the other. At each 6L6G a bow ties heater pin 2 to the base sleeve at pin 1 and a second bow carries pin 1 on to the cathode at pin 8 — this circuit takes fixed bias off its selenium rectifier, so that cathode is grounded — and pin 2 takes the lead out to a ground hatch, leaving heater pin 7 the fed leg.

One detail there is worth naming, because it looks alarming and is not: at the noval sockets the grounded centre tap doubles as the tie point for the cathode at pin 8. A wire between a heater pin and a cathode is exactly the kind of thing that reads as a fault, and on a floating supply it would be one; here the centre tap is grounded at the socket, so it is simply the nearest ground lug.

Verification — and what the chart gets wrong

The rails, the 12AY7 front end, all five phase-inverter values, and the 6L6 fixed-bias supply verify against the printed chart (worst gated node 13.1 %, against the chart's own ±20 % convention). The five phase-inverter figures — +213/+1.7 on the driver, +270/+55/+53.3 on the cathodyne — are mutually consistent once the stage is read as the driver + cathodyne the sheet draws; an earlier revision of this page modelled a long-tailed pair here and had to dispute three of them as contradictory, which they never were. Three printed values remain excluded as disputed, with the arithmetic shown in the voltage table: the chart's V2A pair (140 V plate with a 2.2 V cathode) is physically impossible for a 12AX7 — that cathode voltage implies a current the tube can only pass near 280 V — and the cathode follower inherits the same printed value through its direct coupling. Measured period data put the real operating point near 190 V, and simulation lands near 178 V. The full analysis is in the archive's 12AX7 calibration study.