[sdiy] Two ways of mixing modulation routes

Ben Bradley ben.pi.bradley at gmail.com
Fri Jan 6 21:13:14 CET 2017

(I just messed up two ways: I accidentally sent my response Mattias
only thanks to the Gmail Reply function and I somehow lopped off the
first sentence or two. Trying to recreate it)

The three signals are all different waveshapes (sawtooth, pulse,
triangle) from the same oscillator being mixed together (they all have
the same frequency and phase). The first two go to the passive summing
resistors and the noninverting input of the opamp.  These two are
correlated (since they come from the same oscillator, they have the
same frequency and phase) and so they add together more than would
signals from two different sources (two oscillators at different
frequencies), so it makes sense that when turning both on, they should
both be attenuated, so the fundamental (or overall level) would be
approximately (or closer to) the same level whether one or both are
connected. But with series resistors of 100k and (if I read it right)
200k and a summing resistor of 33k, there doesn't look to be be much
change in level, so maybe that wasn't even the intention.

The triangle goes to the inverting input/node, which means it would
add linearly without any interaction between it and the others. I can
only imagine the resulting modulation sounds more "right" being
inverted relative to the other two waveforms, rather than added in as
non-inverted. Keeping in mind this is for modulation, polarity (of a
sawtooth and pulse) is important, so I think there was good reason
(though I can't say exactly what that reason is) for the circuit to be
the way it is.

I wonder what waveshapes these combinations give. I suppose I could
look at the 3340 datasheet and put this in LTspice.

On Fri, Jan 6, 2017 at 7:32 AM, Mattias Rickardsson <mr at analogue.org> wrote:
> Interesting. I agree it looks a bit strange.
> Non-inverting op-amps are often avoided in mixing stages because the
> different sources feel different impedances depending on each other,
> instead of just feeling a well-known constant impedance to ground.
> First I thought they might have used this fact in a clever way to
> attenuate the modulation signals a little bit if more than one of them
> are switched in to the same destination, giving the sum of modulations
> a more similar level regardless of how many modulation sources are
> used - but when I look in the schematic I think it seems to be the
> other way around: flicking the switch to add another modulation source
> seems to make the first modulation a bit stronger instead. Hmm?
> /mr
> On 6 January 2017 at 12:02, Tom Wiltshire <tom at electricdruid.net> wrote:
>> Hi all,
>> I've been looking at the Sequential Pro-One schematics, specifically the modulation routes:
>> http://machines.hyperreal.org/manufacturers/Sequential/Pro-One/schematics/pro1-schematics-a.gif
>> (Rotated image, but it's bottom right when it's the right way around).
>> I noticed that rather than mix the signals using an inverting op-amp mixer (as I might have expected), instead they do a passive mix, followed by a non-inverting op-amp.
>> Is there any benefit to this approach, aside from avoiding the inversion? Is that the reason it was done this way?
>> Thanks,
>> Tom
>> _______________________________________________
>> Synth-diy mailing list
>> Synth-diy at synth-diy.org
>> http://synth-diy.org/mailman/listinfo/synth-diy
> _______________________________________________
> Synth-diy mailing list
> Synth-diy at synth-diy.org
> http://synth-diy.org/mailman/listinfo/synth-diy

More information about the Synth-diy mailing list