FXBus 1 Module 1 - a Buffer/Gain/Booster Module

Copyright 2000 R.G. Keen. All rights reserved.

Module 1 is a little deceptive. First, it looks needlessly complex for either a buffer, gain block, or booster. Second, for those of you who are into understanding what circuits do by looking at schematics, it looks like it couldn't possibly do anything useful. What are all those parts, and why are they there?

It's about options. By selectively leaving parts out and using jumpers for others, I counted 75 different circuits/functions that can be built on that schematic before I got tired of counting. That does not even include variations that only have slightly different values of parts. In fact, there are doggone few single transistor gain or buffer stages that *can't* be built on Module 1's PCB.

Q1 is shown as a circle with legs, no designations on the legs because it can be a bipolar junction transistor, JFET, or MOSFET, and either NPN, PNP, P channel or N channel. Proper choice of the other parts will make any of those transistors work in there. For NPN and N-channel FETs, the drain is pin 1, the base/gate is pin 2, and the emitter/source is pin3. For PNP and P-channel FETS, pin 1 is source, pin 3 is drain.






Here's how it's laid out:



Here are a few examples of what you can make this thing do:
(If a square is blank, that part is not used; a "0" resistor is a jumper wire.)

Circuit Q1 C1 C2 C3 C4 C5 C6 R1 R2 R3 R4 R5 R6 R7 Pot
Emitter Follower 2N5088 0.1uF 1uF      1uF   10K 100K 0 10K   0 10K  
JFET Source Follower 2N5457 0.1uF        1uF     0 0 1M   0 10K  
"Improved" JFET buffer  wider input range 2N5457 0.1uF 10uF      1uF   75K 1M 0 22K   0 10K  
JFET Follower with LED bias per Orman 2N5457 0.1uF 10uF      1uF   1.5K 1M 0 LED   0 10K  
MOSFET Src. Foll. BS170 0.1uF 0.1uF      1uF   1M 1M 0 1M   0 10K  
Bipolar gain of 10 2N5088 0.1uF 1uF     1uF   150K 100K 0 15K   10K 1K  
MOSFET gain of 10 BS170 0.1uF 1uF     1uF   150K 1M 0 15K   10K 1K  
Bipolar high gain 2N5088 0.1uF 1uF     1uF 22uF 150K 100K 0 15K   10K 1K  
MOSFET high gain BS170 0.1uF 1uF     1uF 22uF 150K 1M 0 15K   10K 1K  
JFET gain with "coloration" per Orman 2N5457 0.022uF     1uF   22uF   0 0 1M   4.7K LED  
Big Muff Input Stage 2N5088 1uF     1uF       0 39K 100K 470K
15K 100 Use pot output; 100K-A series with 1K to gnd.
Big Muff Output Stage 2N5088 0.1     1uF     390K 0 0 100K   10K 2.2K Use pot output; 100K-A
Dallas Rangemaster 2N527A 0.0068   22uF       470K 0 0 68K   4.7K   Use pot output; 10K-A; hack wire pot to R7 pad


















It's easiest to understand by separating it into input options, biasing options, and output options.

Input options.

The input circuit consists of C1 and R3, and it interacts with the value of R2 and either R1 in parallel with R4 or R5 in parallel with R4. C1 sets the input low frequency rolloff point in conjunction with the rest of the resistances. Making C1 big allows the full audio spectrum to pass into the input, making it smaller rolls off bass. You could make a treble booster for instance, by making C1 cause a 500 Hz to 1000 Hz rolloff of bass. Where R5 is present (that is, a voltage feedback biasing arrangement), R3 determines the stage gain by the opamp gain equation; gain will be close to R5/R3.

Biasing Options

R1, R2, R4, and R5 bias Q1 to get the right voltages across R6 and/or R7.  We'll consider NPN and N-Channel FET's first; the PNP and P-channel FET's are similar, just work from the positive voltage rail instead of ground.

The following diagrams show the main types of biasing variation. The nominal bias voltage set on pin 2 of Q1 (which will be the base or gate) is determined either by the voltage divider R1/R4 (as in A and B) or by R5/R4 (C and D).  In general, either R1 or R5 but not both will be in the circuit. Sketch A shows the standard "stabilized bias" circuit. Sketch B is the so-called "noiseless" bias scheme, which is very similar to the usual trick of setting up a bias voltage for opamps in effects circuits. Where a resistor in the full-blown circuit is replaced by a wire, you simply put a cut-off resistor lead in its position on the circuit board. Where a capacitor disappears, it's just not populated on the board.

Sketches C and D show the two main types of voltage feedback biasing. These have the hallmark that there is a feedback resistor from terminal 1 (collector or drain) of the transistor. These are familiar stages from circuit sections like the Big Muff Pi distortion.

Output options

This is where we get to pick whether we get gain out of the circuit and how much. First, we pick whether we have  buffer with a gain of +1 or a gain stage with a gain of roughly R6/R7 by choosing whether to take the output from pin 1 through C4 (gives inverting gain) or from pin 3 through C5 (gives a follower, or buffer). We make this selection by leaving out C4 or C5 - the pin where there's a capacitor does the output. The "outboard" sides of C4 and C5 are tied together, and this point is also connected to to the "hot" lug of the off-board pot. The pot allows us to have a variable output if needed. Soldering in a wire in the four-pad block to connect the full output (from C4/C5) or a variable output (from the wiper of the pot) selects which it will be.

 For an NPN/N-channel device, the ratio of R6 to R7 deterimines the mid band gain. This can be modified to change the AC frequency response by selecting values for C3, C4, C5 and C6.

A final option is to make the module into a phase splitter as shown in "G". We do this by making R6 and R7 equal in value, connecting C5 as the output (gives a signal gain of 1 to the output pin) and cutting the connection between C4 and C5. We then use a jumper wire to connect C4 to the SCout pin, and so an inverted version of the signal is passed to the next module on the Sidechain input. We'll later use this one to make an octave up module set.