[sdiy] Mixer module (was: Changing an LM13x00 with a LM13x00)

[harrybissell]

synth waveform mixer ??? either is fine

Audio mixer... use the CA3280 or other VCA (see THAT corp).

The LM13600/13700 are not that good, even with diode linearization.
CA3280 is much better.

Both need to be controlled from current sources... not hard to do at all.
If you want to pan, one CV needs to be inverted from the other. In fact to pan
properly... you need a curve.  With linear control, each signal will be down
50% and the overall volume will be too low.

In either case, you will attenuate the input signals first...to some level that
gives a good trade of noise and distortion.  If you use the diode linearization
(good for hi-fi) you may even need a buffer amp first.

Can't use caps for 'phase shift'... first it might be as much as a 90 degree
shift, but one that varies with frequency, and only works on AC.  Best to make
your VCA separate, then drive from appropriate source. If you just want back and
forth panning... triangle waves and inverters are all you need. If you want more
complex panning, probably you want a quadrature oscillator (that does have 90
degree phase shifted outputs).

Its very hard to make quadrature outputs from a signal that does not have them
to start with... (its easy if that signal is a sawtooth, or triangle)

H^) harry

Tom Adam wrote:

> Thanks Guys
>
> I'm thinking about making a mixer module with CV-pan possibilities with
> regular line outputs and 10Vp-p inputs.
> Since I got like 10 LM13x00 around I was thinking about using these.
> I've also got a bunch CA3280 around.
> Which one would you use for this kind of operation?
> What's the best way: attenuate to line levels and do the CV-pan option
> than or the other way around? (I guess the other way around would be
> better.)
> I also would like to include a way to inverse the CV (So one signal goes
> right while the other signal goes left) Do I need an inverse op-amp
> circuit or could I do this just using caps? (If I recall correctly a cap
> gives a 90° phase shift??? It's been a while...and I'm more or less a
> newbie;-)
> I'll post my final circuit for your comments ;-)
> Cheers,
> Tom
>
> harrybissell wrote:
>
> >uhhh no... Three...
> >
> >Three answers...
> >
> >OTA, Buffer, and 'who cares'
> >
> >Among the many answers...
> >
> >1) What Jim said
> >
> >2) The 13600 has a connection between the Gm bias input and the buffer...
> >so it is variable bias.  Better for full range, but a fast moving control signal
> >
> >might couple in through the buffer
> >
> >3) The 13700  has a fixed bias... you cannot change it. Its good for most apps.
> >
> >4)  TRY it.  In most cases they can be swapped with no noticible effect on
> >performance.
> >
> >H^) harry    (among our many weapons are fear, surprise, torture, and an almost
> >fanatical devotion... :^)
> >
> >James Patchell wrote:
> >
> >
> >
> >>There are two answers....
> >>
> >>         Using only the OTA section....interchange to your hearts content...
> >>
> >>         Also using the BUFFER section....No, they are not 100%
> >>compatible.  The buffer sections are biased differently in the two chips,
> >>and some applications, this makes a big difference.
> >>
> >>At 02:53 PM 12/16/2005 +0000, tom.adam at telenet.be wrote:
> >>
> >>
> >>>Guys,
> >>>
> >>>I was wondering if it's possible to interchange a LM13700 & LM13600
> >>>without  other changes?
> >>>I noticed on the datasheet that the buffer input current is entirely
> >>>different so I guess I need to change some resistor values?
> >>>
> >>>Cheers,
> >>>Tom
> >>>
> >>>
> >>         -Jim
> >>***************************************************************
> >>http://www.oldcrows.net/~patchell
> >>
> >>***************************************************************
> >>
> >>
> >
> >
> >
> >
> >
> >
>
> --
> Tom Adam
> The Big Ear
> http://users.pandora.be/the-big-ear/
> Check out the download section!