More on Fav Designs

[gstopp at fibermux.com]

The slumbering list wakes up in fine form!

On 4/17/97 Paul Schreiber <synth1 at airmail.net> wrote:

>
>First, no VCO has a "sound", sterile or otherwise. They may have THD, = 
>but a pulse is a square is a saw. But what they DO have is drift and = 
>have linearity errors.
>

You may catch some flack for that one! I tend to agree, but there are some 
little reservations in the back of my mind, like about the spectrum of the 
vertical edge in real-world waveforms and jitter and things like that. What 
you said looks good on paper, but remember I said:

>
>yet somehow sterile-sounding, so the common wisdom goes.
>

Very subjective, this topic.

To align this response back towards the origin of the thread from Steve 
Jones, I'd like to contribute a list of modules that I'd consider to be a 
nice selection for the builder:

VCOs:

Here's a good one - the EN tri/square VCO using a 3080 current switch into 
a 3140 integrator into a 3080 comparator, with an MAT PNP expo source. Very 
nice tracking and stability. Fairly simple with easy-to-get parts. The 
sawtooth resulting from the invert/don't invert FET switch gets ugly around 
15K, but still sounds fine up there (where only the fundamental really 
counts, I suppose). Triangle shape is absolutely flawless even into the 50K 
range, which just gives me warm fuzzies about this circuit. Low end can go 
into periods measured in minutes, once again with nice waveshape. A real 
workhorse (so says Bernie, and I agree). Sync is a little tricky but I got 
it to work pretty good.

Another good one - the EN sawtooth discharge VCO using a 3140 integrator 
with a JFET hard-discharge on it, into an LM311 comparator with a cap 
hysteresis trick to determine cap reset duration. This is fed by a MAT NPN 
expo converter. Really simple, about as simple as they get. Very long 
periods at the low end, frequencies into the 120K region at the top, with 
excellent tracking all the way and a very nice clean sharp sawtooth to the 
very end. What with the pico-amp input bias on the 3140, plus the clean 
short of the JFET, plus the high speed of the 311, this VCO is perfect for 
sub-audio and audio. Also makes a good 1v/oct hi-freq clock for EPROM 
waveform scanners, so I hear. Triangle conversion tactics differ, but I've 
tried the single-transistor FWR which gets ugly in the 15-20K region but 
again not a real problem. Hard sync is dead simple here, and you can 
"soften" it up with a simple attenuator on the sync input.

Notice that both of these use the 3140 which has a sukky output structure 
meaning that you probably don't want its output to make it to a panel jack. 
Perhaps somebody can suggest a more modern beefy-output pico-amp (or even 
femto-amp?) input bias practical substitution....

The Chroma VCO works well too, although the sawtooth shape gets a flat spot 
towards the 5K region as I seem to remember, but tracking is maintained. 
It's the simplest yet, with an EXAR 4151 as the combined cap 
reset/comparator doohickey. I made two VCO's together used a single 3046 
for both. Cheesey maybe, but it is quite musically useful for hardly any 
parts.

VCFs:

Moog ladder: Copy the Minimoog circuit. Use 2N3904s. Have fun.

SSM 2040: good luck finding the chips, boy do these sound sweet. Very 
sproingey resonance, almost like a Minimoog!

EN State Variable: easy and fun. Watch out for high-amplitude breakaway 
oscillation if Q is turned up. Fry your tweeters.

EN cascaded 3080 4-pole: Surprisingly smooth-sounding. Nice resonance, 
similar to the 2040.

VCAs:

Despite my grumblings about 3080-hecklers, I really *would* like to find a 
low-feedthrough low-thump low-noise 2-quadrant design, that I don't have to 
call up a chip company to get. My discrete adventures lately really haven't 
caused me to give up on the 3080! Back-burner project at this time, 
however, but if something really elegant pops up I'd love to see it.

ADSRs:

Wow! 1200? Whatcha want fer those again? I do have some fine discrete 
designs but if 3310's are available then it seems foolhardy not to use 
them.

Regarding gate reset of the attack - in my discrete ADSR designs I have to 
*add* this in! The attack/decay flip-flop should be set by the rising edge 
of the gate and reset by attack peak reached. This will give you the "keep 
going even after you let up the key" behaviour. If you then add the 
gate=low reset to the clear input of the A/D FF, then you get the cancelled 
attack when the gate goes away function. A switch would be nice. Any way to 
get the 3310 to do this?

Regarding adjustment of sustain whilst holding a note - the EN ADSR which 
combines AD and AR with ganged attack pots has this, with no lag. In most 
other ADSRs, you must suffer with the lag of the decay pot as you move the 
S knob. Bernie went into quite a bit of detail here in later issues of EN.

On the subject of power supplies - I agree, switchers are bad! Stay away 
unless you know what you're doing. If you put a switcher in a synth, and 
then put on some magic glasses that allow you to "see" in the RF region, 
everything would look dark except for the switcher which would look like a 
lit Roman Candle, spattering sparks everywhere! Yuk.

OK I've about shot my wad... now all's somebody has to do is compile a 
bunch of these ideas into a kind of 90's preferred circuits collection! As 
always my angle is practicality, availability, simplicity, and 
functionality. I hope this helps. Keep up the great thread. I'd love to 
continue contributions but (again) I'm out of town till next week.

- Gene
gstopp at fibermux.com


______________________________ Reply Separator _________________________________
Subject: More on Fav Designs
Author:  Paul Schreiber <synth1 at airmail.net> at ccrelayout
Date:    4/17/97 5:02 PM


Hot Damn! Some intelligent activity in a group (there is hope).
     
VCOs
=3D=3D=3D=3D
     
First, no VCO has a "sound", sterile or otherwise. They may have THD, = 
but a pulse is a square is a saw. But what they DO have is drift and = 
have linearity errors.
     
Now even the CEM3340 will drift some. But not as much as most, because = 
there is a circuit on the die that generates the +3300ppm voltage. Note = 
that this must be MULTIPLIED with the input voltage. A resistor divider = 
is a multiplier (by a coefficient <=3D.9999). The on-chip multiplier (OK = 
there goes my trivia question: where did OnChip get their name) has = 
linearity errors. This is also called a "tracking error": instead of = 
1V/oct, get .993 in the first octave, .997 in the second, 1.002 in the = 
third, etc.
     
I think any synth "voice" needs 3 oscillators. Monosynths with 1 or 2 = 
just aren't very 'fat'. Polysynths using CEM3394s are 'thin' in omni = 
mode because a 3394 only has 1 VCO per voice.
     
Gene's point is for small # of VCOs, tracking can suffer some, and I = 
agree 100%. Now what remains is the nitty-gritty.
     
This is a QUICK overview.
     
a) #1 overlooked component in a VCO is the integration capacitor. You = 
just can't chunk any old cap in there. Why? The quick answer is that = 
real-world caps have ESR (effective Series Resistance), leakage, = 
temperature drift (just like a transistor!) and (what we worry about) = 
something called dielectric absorbtion. All of these contribute errors.
     
The best cap to use is a metallized polypropolyne.
     
b)#2 is the integrator used in the op-amp. A perfect opamp has zero = 
input bias current. But in the real world, the input bias current = 
"steals" current into the integration cap. So what? Well, the problem is = 
this input bias current DOUBLES every 10C. And, for low frequencies, the = 
input bias current on cheesy opamps can approach the integration = 
current.
     
c)#3 discharge transistor. Best advice is use a DMOS FET, like an = 
2n7000. They have a factor of 10 (to 1000!) lower ON resistance. This = 
eliminates errors at higher frequencies.
     
The 3080/LM13600
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
     
Hey, these are OK for poking around. Heck, I bought them in tubes of 100 = 
for years. But they suffer from:
     
a) control voltage feedthrough. This translates to audible 'thumps' when = 
the CV rapidly changes (ie fast attack)
b) they like current SOURCES to control them, which are harder to manage = 
(but doable)
c) their THD/noise is between AM and FM radios which is OK if you can = 
live with it
     
I like quiet VCAs with NO feedthrough.
     
The 3310 EG
=3D=3D=3D=3D=3D=3D=3D=3D=3D
     
All agree, this is the chip. So concern over the availability. Gene, I = 
GOT 1200!! Now, this is certainly enough for now. So don't be shy using = 
it. I will design an equivalent from discrete later this year.
     
Power Supplies
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
     
It's hard to beat a 723/D40xxxx pass transistor. This is a case where = 
'vintage' is better. I will ask Walt Jung for permission to publish his = 
"super high tech" power regulators. They have 1/1000th of the residual = 
noise of a LM7812. Amazing!! Can be built for about $8/side.
     
Running 500 Amps to heat a transistor 
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= 
=3D=3D=3D
     
Granted, this is the quick way to do it for 1-2 VCOs. If your building 2 = 
VCOs, be sure the both exponential gen transistors & heater are all on = 
the same die. In order for this to work, the DIE temperature needs to be = 
about 60C which will burn your little pinkie.
     
Doing this is a polysynth is, what is referred to in engineering = 
circles, a bad idea.
     
Keep the dialog rolling.
     
Paul Schreiber