Hard Bargain (Discrete..)

[antonis]

Of course, we're talking about Electricduide's well known Hard Bargain..
https://electricdruid.net/designing-the-hard-bargain-distortion-pedal/

Below is an attempt to simulate overall circuit's behavior with discrete items (BJTs only – Fets strictly forbidden..) instead of ICs..
Tried to follow Tom's circuit architecture (partially 'cause I'm as lazy as Deadastronaut.. ) and, surprisingly, I'd to make only minor changes (e.g. R11/R18 stopper resistors different values due to port A hardly considered as "virtual ground"..)



Q1/Q2 stage is an input CFP buffer with some gain (x3,5).. Q3 circuitry is the main distortion stage (x1,6 to x116 gain) followed by D1/D2/D3 clipping stage (SW1 switches between symmetrical and asymmetrical mode).. Q4/Q5 stage is a buffer for TILT control "isolation" (particular configuration might be kind of overkill but let it be..)  
Q6/Q7 stage, as well as Q8/Q9 one are high gain (about x120) / low output (about 18R) CE amp drivers..
Q10/Q11 stage is an inverting buffer, just to set IN-OUT signal in phase..
(it can be omitted with no significant issues..) 

All 10μF caps (except output one) can be much smaller (but, as I said, I'm too lazy for particular filters corner frequency calculations..)   

Frequency response graphs will follow (in due course..)

[deadastronaut]

 oi!..

[Fancy Lime]

Antonis, this is so silly that I am not sure if you are making fun of me as well as Tom and astronaut. I like it! Taking a deliberately simple and efficient design and making it unnecessarily complicated by discreet implementation could have been my idea. Or a Boss circuit from the '90s...

[ElectricDruid]

I rather like it. Ok, it might be *slightly* crazy, but in a good way.

I mean, imagine the number of transistors there must be inside all the op-amps I used in my version. You could say that Antonis' version is more efficient!

I don't think I'd call the original Hard Bargain "simple and efficient" by any stretch. Six op-amps for a simple dirt pedal seems pretty excessive to me. I couldn't work out a good way to get it down to only four though, and there was no way I was going to do a design with five, so six it is.

[antonis]

That's the point, Andy..
(twice the parts count for an almost equivalent efficiency..)

To be honest, it started when I tried to implement NE5532 instead of TL072 and resulted into extra DC blocking capacitors, bias resistor bootstrapping, etc..
So I thought, why not discrete BJTs..??

P.S.
2 resistors and a capacitor saved due to Vbias abscence..

[antonis]

Finally, graphs..

Andeniably, op-amp version is much more effective..

[Steben]

I bet those discrete stages add slight even order harmonics which cannot be found with opamps.

[antonis]

Could you imagine the above circuit with push-pull class A amps instead of single-ended ones..??