Preamp and power amp distortion (Acapulco Gold)

[pokus]

Hey there, I took a look at the Acapulco Gold schematic and noticed it's quite a simple design. It uses two JRC386D to create power amp distortion. The first one has a gain of 200 the second one 20, with a low pass filter in between them.
So I wonder if there's any difference when building it with a common IC (TL072 or else) and setting the gain of each stage the same? Or did the JRC386D provide more output?

What I also noticed is that there's a 10uF polarized cap right after the input, but it's reversed (in -, out +). Why? 

Thanks for answers!

[PRR]

The LM386 is NOT a generic opamp. It has internal biasing and gain-set. Subbing a generic opamp will need a complete re-think.

The LM386 is NOT expensive.

[PRR]

> 10uF polarized cap right after the input, but it's reversed (in -, out +).

Why do you think that is "reversed"?

[pokus]

Alright thanks. It's not about the price, I've already built one with the 386Ds. I'm trying to understand what's the difference between them with regard to the sound. Why is the sound of the distortion different when there's the exact same gain? Does a low voltage audio power amplifier distort in another way than a common opamp?

I called it reversed because in quite every schematic I had looked at it was in + , out - ,when in the signal passing path. But I guess it's just a matter of habits

[FiveseveN]

Does a low voltage audio power amplifier distort in another way than a common opamp?

Arguably. The way it's used here, the difference may be subtle. Build it both ways and decide for yourself.

I guess it's just a matter of habits

It's a matter of which side is more positive.

[PRR]

> I called it reversed because in quite every schematic I had looked at it was in + , out - ,when in the signal passing path. But I guess it's just a matter of habits

Not "habits". Get it wrong, an electrolytic cap will suck-out your DC bias.

I can't think any reason a -negative ground- pedal would have "+" to the input jack/switch.



The external source would normally be near zero Volts DC. (We don't like any DC on audio cables/jacks.) The place the cap goes is at half of the supply voltage, +plus+ 4.5V DC. This drawing is correct.

One popular confusion: there are several old-time +positive ground+ circuits.


Yes, the input cap is the other way because EVERYTHING is the other way: battery, transistor type.

Specific to a LM386 plan: the '386 is quite different from most other audio circuits because the designers "stood the input transistors on their heads" to get the input pin to essentially zero V DC. Why? Because (at least in radio/phono use) this way you do NOT "need" an input cap, which saves a few pennies, which is critical in mass-market stuff. Actually they wander up to about 0.050V DC. This means a volume-pot in front may "scratch" (which may be why '386 in pedals often have a cap). It incidentally means you do not HAVE to get a polarized input cap "right" because an e-cap will stand tenths of a Volt reverse.

[pokus]

Ok thanks a lot for making that clear!
I guess I just have to build the opamp version and compare the sound. Although I think it could sound pretty lame because there isn't much tone shaping in this circuit.

[pokus]

Hey, so I built another variation of it with a first high gain LM308 stage followed by the LM386-1 power amp stage with a gain of 20. I also put a tone stack behind. It already sounds quite fine, but there's one thing I can't get rid of.
Sometimes the tone breaks off at low input volume. It acts like a noise gate, which is quite cool, but not wanted.
Could it be an impedance (tone sucking) problem or isn't the LM308 able to drive the power amp at low volume? (But isn't that how an amp works? preamp+poweramp)  It's also doesn't matter what gain is set at IC1, softly picked strings, especially the low ones, sound weird.

[FiveseveN]

Did you properly bias the LM308? Show us a schematic of what you built, a "high gain LM308 stage" tells us nothing useful for debugging.

[pokus]

It's the RAT pedal without anything behind the opamp and less tone shaping. It's biased with 4.5V through an 1M resistor. The input impedance is about 500k. Is also tried to put an output buffer to the design, but the noise gate is still there.