I need to build a bipolar power supply for my CA ring mod

[Chris Goodson]

I've never been happy with it because the carrier is so loud.  I tried a better trimpot to no avail.  But, I recently read that it helps to have equal voltages.  So I'm looking for something I can put inside the box.  I've got piles of transformers around, and theres enough room in there for one and another small board.  So where's a good design to use?

[Paul Perry (Frostwave)]

http://sound.westhost.com/project05.htm

this is pretty much how I do it, you just tie one transformer output to earth & the other side goes to two half wave rectifiers.
I don't know exactly how fussy you are, the carrier never COMPLETELY nulls out with a 3080. You can do pretty well, though.

Oh, and here is the Word: http://www.geofex.com/Article_Folders/Power-supplies/powersup.htm

[Mark Hammer]

The CA/EPFM ringmod is a decent sounding device.  What would benefit it immensely is to have an integrated gate for the effect signal, in the same way the A/DA flanger and Boss CE-1 chorus do.  

Both of those have constant clock noise to contend with, an oscillator of a different sort (in fact TWO different sorts).  When you play, it isn't noticeable, but when you stop playing, it is.  The gate is applied to the effect signal only, rather than the entire wet+dry mix, so that you never miss the onset or tail of a note because you still hear the unaffected version.

In this case, a simple, "Dr. Q" style half-wave rectifier driving a FET would do the trick.  Now that I think of it, the potentially slow attack time (and by "slow", I'm talking about maybe 10-20milliseconds) of such a circuit could work to the advantage of the player, since it could lead to the modulated sound being faded in AFTER the initial note transient.  The initial attack transient is much more harmonically complex than the decay phase of the note, and more likely to lead to harsh-sounding sums and differences, largely because there are so many of them given all that harmonic content.  The "classic" RM sound was based on oscillators providing fairly pure waveforms, and the part of the note after the initial pick attack approximates that a little more closely, without having to engage in complicated filtering.

Luckily for us, the output of the 565 does not have all that much dynamic range, and may even be a fixed level if I remember correctly.  What that means is that we can have a fairly fast decay time without having to suffer envelope-ripple based clicking that would normally come from little sub/suprathreshold fluctuations as notes decay.

So, to sum, although not to ignore your power supply needs entirely, a side-chain/noise-gate, tapped off the input stage, and applied to the 565 signal reaching the input of the mixing op-amp, could prove very useful in making the RM not only quieter but more musical in tone.

A daughter board sandwiched with the main one could work just fine.  One would need a lead from the output of the first op-amp as the input, a lead from the gating point in the actual RM going to the drain pin of a FET on the daughter board, and the +/- and gnd leads for power.  Attack/decay parameters and follower sensitivity can be purely a matter of either trimpots or fixed-value components.  I only wish I was clever enough to work out the particulars.

[moosapotamus]

What kind of power supply are you currently using for your CA ring mod?

Before you go stuffing transformers into your box, you might want to just try a pair of 9V batteries to see if it gives you better carrier rejection. It might / might not. But if it doesn't, you will have saved yourself some time, effort and disappointment.

btw - I've always gotten great results with the MAX1044.

~ Charlie

[Chris Goodson]

I've been using two nine volt batteries.  I've been concidering getting a few of the MAX1044's to play around with.  I wish I'd included them in my last Small Bear order.

[Mark Hammer]

Stick with a pair of batteries or a regulated supply.  Look at the specs for the current needs, and you'll see that this thing will suck your bettery dry in no time if you intend to use a single 9v and 1044 to bipolarize it.  Indeed, I suspect that one of the primary reasons why so many of Craig's projects used a pair of 9v's was not so much because he was bullish about headroom, as much as because there was no way a single battery could power these things for long with a floating ground.

[moosapotamus]

If you don't want to have to worry about batteries, the MAX1044 seems to work well with a regular 9VDC wall wart power supply, too. I can't say if it would be any better or worse than a transformer/rectifier based design, tho.

~ Charlie