[sdiy] Discrete Op Amps

[Ben Bradley]

There's too much here I want to respond to, but I'll try to pick a few things.

Yes, you can make a good discrete op-amp, but it takes more than three
or five transistors, as the schematics of the 741 and later chip
op-amps demonstrate. There are "classics" such as the Jensen 990. But
yes, modern chips are quite good, and the difference between sound is
quite subtle, at least too subtle for me to hear the difference. As
has already been suggested, if you want to design/make/use discrete
op-amps, do it for fun rather than with the idea that it's going to
sound better than a modern chip. Going discrete lets you have higher
supply voltage rails and output voltages, but I can't think of where
you would need more than +/-12V or 24Vp-p that you get from a +/-15V
supply and most any chip op-amp.

There are many discrete high-power op-amps around, they're usually
called "audio power amplifiers" and include hard-wired gain resistors
for a non-inverting gain of 10 to 100 or so. There's also "chip"
high-power op-amps - I remember seeing some expensive ones decades
ago, but now there's the popular LM3886 made especially for audio that
can put out about 50 watts with proper power supply and heat sink.

For op-amps in guitar distortion effects, I can see where a difference
can be made depending on the peak output voltage swing, where the
clipping might be done by the op-amp's max output voltage (resulting
in hard clipping) rather than the external limiting circuit. I tend to
worry that the 9V battery is not enough voltage for op-amps, and
indeed the common TL072 series is listed as having a minimum power
voltage of +/-5V, or 10V total. Oddly, germanium transistors are in
demand to recreate "classic" distortion pedals of the 1960s, when
germaniums were about all that was available, and the whole box could
operate on one or two 1.5V batteries.

I've got a Peavey Rage "Transtube" 158 amp - the preamp circuit is all
transistors, with four Darlington pairs apparently attempting to
emulate four 12AX7 triodes. There's a couple of diodes and capacitors
in each stage that appear to change the bias dynamically when the
stage goes into clipping, much like a real tube stage does. I'm not a
"golden ear" guitarist and can't say how it compares to the distortion
in a real tube amp, but it's an interesting circuit, and different
from the usual symmetrical back-to-back signal diode distortion
circuit.



On Tue, Apr 11, 2017 at 3:24 PM, ASSI <Stromeko at nexgo.de> wrote:
> On Tuesday, April 11, 2017 9:14:21 AM CEST Roman Sowa wrote:
>> With all due respect I had so far to Phil Cirocco, that website is
>> overfilled with attitude like "I'm the best in synth design and everyone
>> else are clueless morons who know no shit".
>
> Or he had an epiphany of what to do to live off his particular set of skills
> (which I still highly regard).
>
> Reminds me of Nathan Marciniaks hilarious "Elemental Voice" pages, only that
> those were meant as a prank.  He's taken them down somewhere in 2013, I'd like
> to suspect because people couldn't tell the difference and actually wanted to
> buy those things.  :-)  Wayback Machine to the rescue:
>
> http://web.archive.org/web/20130926143924/http://www.nathanmarciniak.com/
> elemental/
>
>
> Regards,
> Achim.
> --
> +<[Q+ Matrix-12 WAVE#46+305 Neuron microQkb Andromeda XTk Blofeld]>+
>
> SD adaptation for Waldorf microQ V2.22R2:
> http://Synth.Stromeko.net/Downloads.html#WaldorfSDada
>
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at synth-diy.org
> http://synth-diy.org/mailman/listinfo/synth-diy