This Project, with a newly upset circuit-expansion to the "very heart of the Univox Superfuzz Concept", is addressed to all Guitarists and Elektor´s Community Members, who may also have their interests in designing and working with pure analog circuits, but also to those who are interested in some innovative function-settings, on a formerly good known and much popular guita

This Project, with a newly upset circuit-expansion to the "very heart of the Univox Superfuzz Concept", is addressed to all Guitarists and Elektor´s Community Members, who may also have their interests in designing and working with pure analog circuits, but also to those who are interested in some innovative function-settings, on a formerly good known and much popular guitar-effect. The circuit is not only experimental, but fully working in simulation and also partly on breadboard. A final version gets meanwhile built as an vero-pcb, identical to the presented schematic and I hope to give a little later a youtube-demo with this unit, in maybe a few days.

Also intended for all active builders of todays older univox-superfuzz-clones, this project may come as a nice challenge and invitation for to do some really effective modification again, where the project as extension on existing units may pass completely over one weekend and keeping always a little experimental thing in it.
With this additional aspect for experimenting, I present the circuit with as much trimpots as possible, where some of those of course may get in future replaced with fixed value parts, when getting used with the unit and it then sounds mostly ok. Here I promise in the foreground to present something really innovative and again with a very individual sound ability, done on the basic principle of the superfuzz´es octaver-effect.
As a main background for the project - meant as a suggestion for almost not getting too much to modify, when starting with existing units - your intention could get mostly on the fact, that you may not believe in the possibilty of reaching with this formerly outstanding univox concept (and this extension), again something really unique and once more superfuzzy.
Another goal of this project is not to get into something completely new, but on expanding some older circuits with new ideas and to change this concept only a little bit, at a quite relevant point and in a very uselful manner, based on the great advantage of a current-source and to combine this for a possible adjustment onto the used diode-types, but hereby without affecting the octaver´s sound abilities.
The result of importance is a circuit-part, which gets like this "adjustable as audible" on its sounding, done with ease for various types of Ge-Diodes (basic truth that each pair comes really individual), at a formerly very critical point on the output of the octaver-section, which was for general not possible with existing superfuzz-units and also with respect to get the whole thing working more stable over wide temperature variations (also possible by modifying the octaver´s inputs and by changing one npn-transistor - T3 - into a germanium device with OC139) and with different decoupling this signal splitting expander-transistor, finally on the basis of an innovative biasing technique.
The zero-setting of the octaver is here also easy to adjust with a trimpot on the sound of the audio-product and by picking single strings on lower volumes, with not too much fuzz-level, so obviously not strictly needed to get adjusted by the additional use of an oscilloscope.

Again there should be noticed - and that´s a really quite important fact at all - the background to the awesome sounding univox superfuzz units is based upon the individuality of the used germanium´s, in conjunction to an optimized coupling onto the octaver´s output impedance and if this is choosen for best optimization, then you will get those possible "eighthundret dollar devices", where each unit again gets unique, due to these circumstances. With this circuit you will get this possibility to optimize once more - and as I hope so - also an invitation to try several diode-pairs for different soundings and each time easily adjust the circuit onto those variations!
As the octaver of the original "Shin-Ei Superfuzz-Unit" is in its core function a special kind of a matched differential amplifier (and a much temperature sensitive device too), I changed this section into something most similar, with also a matched dual-transistor, where this is now looking a little bit like the OTA-Cell of a "SSM-Synthesizer-IC", with a current-mirror set on top of the arrangement (2 x BCY78-IV/BC557B, or 1 x BCV62B/SMD) and another one coming as a "wilson-style-current-mirror" placed at the bottom of the device and replacing the formerly used 1.8 kOhms resistor into a dynamic, signal-driven and temperature compensated biasing system, where some possible mismatches of the used parts may get easily adjusted with the trimpots.

Some more information about current mirrors you may find here:



(The whole npn-transistor-section of the octaver is best prepared for the usage of matched transistors with +/-1mV ube-accuracy and on close hFE´s, or using a LM3046-Transistor-Array. These sometimes obsolete ICs are again still available at valuable prices and have got relaunched by Texas Instruments after "swallowing" National Semiconductor Corp.)

The enhanced octaver-arrangement is best descibed as it works like a "self-biasing ota-cell", which gets its bias-current generated from the two resistors/active collectors of the differential-amp in conjunction with the trimpots P2 & P3, controlling this "bias-loop" over the npn-mirror applied to the emitter-node at the bottom.
The whole circuit-block works with an average bias current at around 500-600 µA, where the mirrored peak-level for driving the diodes with large signals then may increase up to 1-2 mA, for preset recommended values, but of course this is alterable. Another alternative to use, instead of the transistor array, is to change that into some standard npn-transistors, but of course you will get hereby a greater sound difference of the unadjusted octaver-effect, due to unavoidable mismatches of the differential amp. In the end, it is not necessarary to spent much money for expensive dual-transistors as also single units are there a quite good solution, and then the LM3046 on not too much cost is coming here for almost perfect.
Also only meant as a recommendation is the changed tone section, which is almost the same circuit as in the "ota-overdrive-project".
Additionally, I have also implemented a potentiometer control for soft & hard clipping, where the usage of a stereo-device adresses here a better setting onto the diode´s leakage currents, so to speak to take therefore more part of the produced sound.

Important and useful directives for experimenting with that circuit:

The one and really important directive for a proper handling of the circuit with its extentions and needed to say for not to do this with any current-mirrors, is to modify the inputs of the devices with a directly short-circuit-condition, or some too low impedances (forcing more than 2-4 mA respectively at higher supply voltages is much and an increase may damage the transistor´s basis paths!) towards ground potential, or to the opposite polarities, where respectively this node connected to the collectors of the diffamp-transistors should be no problem and on the other side of the current-mirror, its output (is an open collector) could get easily shorted to ground, or any other location, by patching over R26/P4 without any damage, but then of course the audio resulting more in silence...
(If you are formerly not much used to working with current-mirrors, then read this upper section once again, before you will start with some trials on these circuits !!)

Have a nice day and much fun with this circuit !

More information on principles, varying transistors and support you may get by contacting: jo2030 at gmx dot net !
Several variations for pcb´s in Eagle V7.3 are in progress and will be uploaded here sometimes later...

Please also pardon my bad english...