Synth guys, Audio to CV?

[NewYorkDave]

Feeling totally discouraged yet, Dave? :wink:

 

[NewYorkDave]

Hey, I just poked around and found some info on the Korg box. It's called the MS-03 and it looks to be just what you need. Here's a block diagram:

Specs and online copy of owner's manual
http://www.korganalogue.net/korgms/specs/ppsp.html#ms03
And what it looks like

It has Hz/V and 8ve/V outputs, which is nice 'cause you can use it with a variety of synths. Since they're separate outputs, it stands to reason that you could use it with a Moog and a Japanese synth at the same time :wink:

None of these appear to be for sale on evilBay at the moment, unfortunately. I couldn't find a schematic online. I do have a collection of old Korg schematics in storage; next time I'm at that secret underground hideout, I'll take a look.

This schematic shows the "external signal processor" portion of the MS-20 synth, which probably gives some clues to the basis of the MS-03
http://www.korganalogue.net/korgms/images/service/ms20/circ2.gif

In theory, you could add a log amp to the CV output of this circuit and have something that should do the job.

 

[PRR]

> The library terminal I am using has right click disabled so I can't copy & paste the link.

Highlight text, Ctrl-C (copy), put cursor where you want it, Ctrl-V (paste).

Also Copy and Paste are under the top-bar "Edit" menu.

Similar but different on Mac. Same C and V letters, but I think you use the funny-key with it (sorry Mac-fans, everybody calls the cloverleaf/Apple/bug key something different).

I have seen web-site javascript that "broke" most copy/paste functions. Reboot, and don't go to those sites.

Disabling right-click seems like a pointless way to "protect" library terminals. There is always another way to do right-click evil, and many innocent things (like copy/paste) go best with right-click. But I know many different librarians, AND their bosses, and their geeks, so this does not surprise me.

 

[NewYorkDave]

...or the filtering/shaping circuitry before the F/V has to be highly tuned to musical demands, to present a more unambiguous signal to the converter. :wink:

 

[alk509]

[quote author="al_p"]after a quick search this might do the trick. http://www.synthtopia.com/items/Roland/ROLGK2B.htm[/quote]

Now that I think about it, using a GK2 (or any other similar pickup) may be a good idea, since it would let you filter each string individually, making it easier for the f-to-V circuit to know what the hell it's supposed to be tracking. Just a thought.

Another idea I remember some people talking about is to overdrive the living shit out of a pre-amp with your bass' signal and then have Bissel's PV-1 circuit track the square-wave-looking output, which - at least in theory - should be easier for it (Hey, it tracks a square wave VCO fine...)

Peace,
Al.

 

[NewYorkDave]

That reminds me, there used to be some crude gadgets that were supposed to make your guitar sound synth-like, using a 555 timer. Basically, your guitar drove the timer and the output was a square wave. A more refined version of something like that might be usable as an input conditioner.

 

[PRR]

> Is there an easy solution to get an audio signal to generate CV?

No.

And many complex solutions don't work well.

It is harder than it looks. First consider how you would pitch-follow by ear. It takes many milliseconds to guess a pitch that just started. Even if you had infinitely fast fingers on the tuning knob, you would not know what to tune to for quite a long time; long if the goal is to "play with" the instrument being followed. A 40 milliSecond lag starts to sound like an echo, rather than an accompaniment. And the ear is amazingly good at pitch-guessing. Especially since many musical instrument note-starts are full of noise, or off-pitch from the sustain (guitar and piano start sharp because the string is plucked/struck and stretched at the start). You know from experience that a piano that starts A-440+5cents will settle to A-440 in the first few cycles; a box has no experience. And note that 40mS is less than 20 cycles of most mid-range tones. And we have not begun to look at inharmonic overtones. And struck/plucked strings have two very high and generally unrelated pitches while the pick/hammer is still in contact and the string works as two shorter strings.

I'm frankly surprised that any musically useful boxes exist. As an effect, sure: loosely following behind. As a cheap back-up player: seems to me the errors would make it stand-out like a minor league player in the world's series. But just putting a box "F-V converter" on the plan isn't the answer: just-any F-V is not going to work, it has to be highly tuned to musical demands.

(Portamento covers many sins, if your music allows that.)

 

[PRR]

Dave, I think you posted to something I retracted for editing, but maybe it will come together...

> "external signal processor" portion of the MS-20 synth, which probably gives some clues to the basis of the MS-03

Yes, the lower-left corner "External Signal Processor" IS a music-aim F-V converter.

It starts with some gain to normalize levels. (So much maximum gain, that layout will be VERY critical.) Then they cut the lows 12dB/8ve with a user-knob, and cut the highs 24dB/8ve with another user-knob. At this point you have practically told it what the pitch is! Then D2 D3 TR2 are a tachometer. C2 fills up through R16, but is discharged by TR2 on both zero-crossings of (what is left of) the input signal. The faster the input (higher pitch), the lower C2 is held. There seems to be lots of ripple here (it may not be a simple tachometer), because IC4 seems to square that up, meaning it is PWM at that point. IC3 filters that back to DC (with a settling time set by the user-knob for bass-cut) and outputs a voltage. This must be linear; you need another stage if you need V/8ve control voltage.

(The function of IC2 10-11-3 escapes me. It seems to have one input at each supply rail. Tying pin 10 to V+ will blow the chip. This is either a mistake or deliberate deception.)

 

[StephenGiles]

This is an extract from the Gentle Electric Pitch Extractor. Can you see the little box - "Fundamental Extractor Module"? That's quite important for pitch extraction, which is why the Electroharmonix Guitar Synth had a good one, which is why it worked very well, despite its poor sales, and those lacking a decent fundamental extractor are largely a waste of time. I had the first one in England demonstrated to me at EH UK, and had the circuit explained to me - didn't understand a word of course, but a good story to tell!
Stephen

 

[StephenGiles]

You may care to look at my drawing of the input section of the EH Guitar Synthesiser:

Something to build on a rainy day.
Stephen

 

[thermionic]

Has the tracking got to be in real time? Forgive me if I'm stating the obvious, but if you've got a studio project you just want to get finished, you could try this (bear in mind my studio doesn't have a standalone pc, so I'm probably the wrong guy for this advice):

Record the bass to a hard-disc track. Certain DAW packages (does Logic do it?) have audio-midi conversion. Derive a midi track from the recording, and then use a midi-cv converter to drive the Moog?

If you can get a fairly accurate representation of a pitch progression via the insidious AutoTune, I would've thought the midi map to be fairly accurate providing the playing is clean and relatively monophonic.

Just an idea.

Justin

 

[soundguy]

thanks for all the input everyone. I guess sometimes things which appear simple are really difficult. If I ever see one of those korg boxes, I would check it out. Im basically trying to do something in a mix now, and its really not a dire thing, nor sometihng I would use alot, I didnt want to waste a track recording the synth, which is just what Im going to do, however, it woudl have been nice to have just been able to trigger it during the mix. Justin, thanks for the tip, but I dont use a computer for recording at all, just email and video games.

dave

 

[StephenGiles]

Personally I found the Korg boxes to be useless, very bad tracking indeed.
Stephen