Bi-pulse convertor (ON-topic)

[KA4HJH]

Torbjörn Hörnfeldt wrote:
>Why don't you ask Ian Fritz? He just may have had something to do with it?
>:-) Actually, it is in the EN Preferred Circuits Collection and was first
>presented in EN#72. I think it has also been presented in Elektor(?)

That's the one I was looking for! It's basically what I envisioned (and
Magnus drew) with some extra knobs to play with. The schematic has the
input labelled "sine in" but obviously you could use any non-rectangular
waveform.


Colin Fraser wrote:
>I wonder what the effect of using separate modulators for the top and bottom
>width would be, with different rates.
>Should be thick...

Ian's circuit has a mixer panning between the two convertors, which I
hadn't thought of, as well as an offset control for the input. You could
get really fancy with this. The simple version I heard last night had a
really nice sweep to it that was distinctly different from plain vanilla
PWM. I'd say you'd be crazy not to have some wave-wrangling gadgets in your
modular (Tony's waveshaper looks pretty interesting, too).

Not that I have anything built yet myself...ordering parts is such a pain.


>You could do this with a couple of comparators driving a PNP and a NPN
>transistor - kind of like of a TTL tri-state output. If the top pulse
>comparator is on, pull the output high.
>If the low pulse comparator is on, pull the output low, otherwise it sits in
>the middle (with a potential divider to set the mid point).
>Alternatively, use an analogue switch to select variable voltages for the 3
>different wave levels, and modulate those too.
>
>It would take so few parts, it's got to be worth trying...

Ian's circuit uses six opamps (two for comparators). If you can make it
simpler...


Mike Granger wrote:
>Seems like you could use two identical saw-to-pulse converters (comparators)
>but feed them with pulse width control signals that  are inverted versions of
>each other. This should increase pulse width of one while it decreases pulse
>width of the other. You would need to have a diode connected to the output of
>each comparator so that one would contribute the positive phase of the output
>signal, and the other (diode connected opposite polarity to comparator output)
>would contribute the negative phase. The other ends of the diodes would go to
>the output jack.

This is what I was thinking but as it turns out you don't need any diodes!
Everything cancels just right. Unfortunately I can't just scan and post
this *particular* schematic and I don't have time to redraw it right at the
moment so everyone can look at it (see Magnus' simplified version for the
moment). Where's Ian?


>Very interesting idea. Where did you hear of this?

I got to play around with the software synth Reaktor last night. Looks like
simulators are good for getting new ideas.

I downloaded the demo version and the Bi-Pulse oscillator is in the module
list but I'm not sure how to make it work. I don't know if any of the demo
synths use it--haven't had time to check it all out yet.


-- 
Terry Bowman, KA4HJH
"The Mac Doctor"