teach me a bit about adding envelope control rather than a pot, please?

[pinkjimiphoton]

hi fam,
i built another mouthmeister from tiny dazzler.
https://tinydazzler.blogspot.com/2010/09/pedal-power.html



really cool little "talking filter pedal" using fredrik's excellent vero layout

https://www.parasitstudio.se/stripboard-layouts/mouthmeistor-talking-filter-fuzz



<the original one got sold to someone, so built another>

anyways, i was thinking, this pedal would be really cool under envelope control, instead of a treadle.
but thats kinda above my braindead fuzzmonger paygrade.

is there a "cookbook" approach to grafting in an envelope control to this circuit to replace the wah pot?
i haven't boxed it yet so i can definitely add a daughterboard. since the mouthmeistor kinda needs a fuzz in front of it anyways, i was wondering if maybe i could adapt the envelope section of an uglyface?



i really need something self contained and bone-headed simple. thanks guys

[Mark Hammer]

The easiest way to add envelope control is to use a simple half-wave rectifier - just like what you see on the Doctor Quack - to drive an LED, instead of a diode.  The LED makes an LDR go low-resistance.  Of course, that assumes that whatever you intend to control depends on a simple variable resistance, and not a voltage divider.

[pinkjimiphoton]

thanks mark,
still above my paygrade

i DID find this... hemmo's cmos follower



sorry mark, i'm really much dumber than folks realize, lol... so with the dr quack, i'd need to recreate the buffer input and the "bottom half" of the schematic in its entirety?




or just use the buffer part to drive an led, and replace the 20k pot in the mouthmeistor with an ldr?

sorry for the dumb questions... having a hard time wrapping my head around this kind of circuit. i downloaded and printed your "technology of" article and will be studying it tonite. i'm not getting where to tap the signal off to the envelope detector, and unsure of which part of the dr quack is necessary to make it work. i always loved the mouthmeistor circuit, but having it myowl via pick attack will be the bee's knees

i just need something stupid simple. thanks for the advice

[Rob Strand]

i built another mouthmeister from tiny dazzler.

One simple idea is to replace one or both of those 10k output mixer resistors with opto resistors.
If you replace only one you might need to tinker with the resistance range or the fixed resistor to get it to sound right.   If you replace both 10k's you need to have one opto going on when the other is going off, which is a bit of a pain.  It will take some stuffing about getting either one spot on.

i DID find this... hemmo's cmos follower

Hemmo was a good tinkerer.  He come up with quite a few things which were off the normal path.

[pinkjimiphoton]

brother rob!

i think i get what you mean, but i need to control the "sweep" pot... its grounded in the middle, which made me think i could use a vtl5c4, i've got a couple where the ldr is center tapped.  i figured i'll tie that to ground and use pins 1 and 3 to where it says "sweep pot".

looking at the dr quack, thinking maybe i can graft in part of the circuit from the sensitivity pot to the range pot <well, the red led before the range pot, i am assuming that second led, i'd wire up the led part of the vactrol, right?

i gotta question... would i  be better off taking the tap for the envelope from above the stock gain pot, an emitter follower <i think> or better to tap it off the collector like the rest of the circuit? will an EF have enough voltage/current to drive the envelope follower? i'm thinking maybe better to have too hot a signal and pad it down than not enough signal...

appreciate the advice. the more i look at it, the more a bit of understanding MAY  be dawning.

sorry for the dumb questions.

i gotta billion of them

[Rob Strand]

i think i get what you mean, but i need to control the "sweep" pot... its grounded in the middle, which made me think i could use a vtl5c4, i've got a couple where the ldr is center tapped.  i figured i'll tie that to ground and use pins 1 and 3 to where it says "sweep pot".

OK got it.    For those dual vtl5c4/2's  the resistances both come one at once.  What you want is,

Pot              R12                   R23
position       Rpot 1 to 2         Rpot 2 to 3
7:00            0                       10k
9:30            2.5k                   7.5k
12:00          5k                      5k
14:30          7.5k                   2.5k
17:00          10k                    0

Basically they always add-up to 10k, which is the resistance between 1 and 3 of the pot.

You need two vactrols.   One to emulate the pot 1 to 2 terminals and the other to emulate the 2 to 3 terminals.
To emulate the pot terminals one vactrol needs to be low impedance when the other is high-impedance for one end of the pot setting,  then the opposite of that for the other end of the pot setting.  So one  going on when the other is going off and visa-versa.

With no signal you want,
- R23 to sit at 10k.  For a VCL5C4/2 you need about 150uA LED current. 
- R21 to sit at 0 ohm.  You can get zero but something like 200 ohms might be a good start which equates
  to about 10mA LED current.  5mA might also be OK.

For the mid part of the envelope you want,
- both reistances to pass through 5k point at the same time to emulate the middle of the pot.
   So about 200uA each.

For full envelope you want the resistance R23 to go short (say 200ohm) and R21 to open to 10k.
So the R23 needs the 5 or 10mA and the R21 needs the 150uA.

So you will need two LED drives circuits one fed with the output of the envelope circuit and one fed with the inverse of that.

Off-hand I can't see an easy way to do it.  Maybe a web search will give some better ways to do it.

looking at the dr quack, thinking maybe i can graft in part of the circuit from the sensitivity pot to the range pot <well, the red led before the range pot, i am assuming that second led, i'd wire up the led part of the vactrol, right?

The part you have circuit is a good start.   You would need to add the LED drive circuit for one vactrol then a second circuit which did the "inverted" LED drive for the other vactrol.

A similar sort of circuit is Fancy Lime's Quackmire.  This one drive the LED directly.  Not sure if that's easier or not to get the inverting part working.

https://www.diystompboxes.com/smfforum/index.php?topic=119101.20

[As a side issue I think the quoted resistances of those dual vactrols is the resistance across the outer terminals of the vactrol resistor.]

[Rob Strand]

It occurred to me the switched resistance method used by the MXR Envelope Filter works might be a good way to emulate a pot.   The "on"-switch would go in the position of R23 and the off-switch would go in place of R21.   These types of circuits aren't so easy to understand (and can have problems).

[pinkjimiphoton]

thanks rob!
i will re-read that all tomorrow... too late right now

but i don't see why i can't just use the one vactrol to replace the sweep pot? i know i'm dumb lol but doesn't it end up working the same as the pot would?




or am i misunderstanding? oh,, i think i get it... you need one to go , lets say, pos while the other swings neg to sweep the filter? with just the one vactrol, both filters would sweep on at the same time, right? and to get it vowel-y , they need to sweep against each other?

i'm gonna def have to re-read in the morning, and break out the breadboard.  i'm guessing that's why you suggested the 10k mix resistors being replaced.. use the output of the filters to drive the envelopes? i thought it needed to trigger off the input for the env, not after being shaped by the filters.

so confused. n brain dead

thanks bro
nite!

[Rob Strand]

or am i misunderstanding? oh,, i think i get it... you need one to go , lets say, pos while the other swings neg to sweep the filter? with just the one vactrol, both filters would sweep on at the same time, right? and to get it vowel-y , they need to sweep against each other?

Yes both sides of the vactrol come on at the same time whereas with a pot one side closes when the other side opens.

[nocentelli]

Would this work? Using envelope DC rather than LFO varying DC?

[anotherjim]

Build this envelope follower. Fit x2 instead of only x1 100k resistors with one each feeding the sweep pot lug 1 or 3. At those lugs, fit the x3 leds to +9v to one and the x1 led to 0v to the other. For extra, fit a DPDT in between those 2 LED arrangements to each lug to select which filter sweeps up or down first.

[pinkjimiphoton]

well, dicked around with this all day yesterday, no dice. tried all the ideas presented pretty much, tried a few "insertion" points and just couldn't get it to fly.

but with a pot, it sounds great.  anything from b2k to b50k will work fine for the sweep pot... the difference is how much of the pot travel is static before it yowels. 2 k is a very wah-ish response, 20-25k is about perfect. 50 k ya get a bit more yoy, but ya gotta work it harder, as it's spread out at the extremes of the pot travel.

i MAY try working something up with another 4069 using hemmo's cmos follower... i could get two of them going with one chip.  could probably do it all with one 4069, but ... i'm reaching well above my paygrade already

[BluffChill]

My favourite env follower is a combination of the phuncgnosis, dr. quack and various others:





It only needs one half of a TL072 or a 741 works just as well. You can play with the values of C3 and R5 - larger gives a slower release - and take the envelope either from behind the LED in the picture or after it for a more conventional application (I like a bit of ripple). You can then hook this up to a transistor for control in lieu of a pot, or replace the LED with a vactrol or LDR/LED combo for smoother control.

[pinkjimiphoton]

thanks bluff, i already gave up on this one, pretty much.
it would be easy, other than how the sweep pedal is implemented.

its grounded to the wiper, so you should be able to treat it as two 10k pots wired opposite,  but THATS what makes it so bloody weird and difficult.

i see ya gotta new project, gonna have to check it out. peace!

[pinkjimiphoton]

as it turns out, a roland ev5 will work with this circuit with a bit of rewiring. use an insulated plastic socket, like a wah style socket.. but it has to be one where there's NO connection directly to ground. on the circuit side,
wire ground to the RING of the jack, and pins 1 and pin three to the tip and sleeve.
now, this has a problem, obviously, in that signal is going down the shielding, stock in the ev5. this obviously won't work...without a quick rewiring inside the exp pedal.
swap the sleeve and ring connections inside the ev5 <i just happen to have a really beat pos one i don't mind messing up> so the shielding connects to ground, and the ev5 has a great sweep for this circuit.  the trim pot for minimum on it has to be full up or full down, one of them 3 am things so i forget... but it DOES work.
at least as a temp solution til i find a cheap plastic wah or vol pedal to gut. i have crybaby shells, but i hate lugging them around.

[Eb7+9]

Pink, just a rough idea ...



of course, you'll want to set the bias for Q1 higher and Q2 lower

I'm curious if this does anything useful
~jcm

[Rob Strand]

... um (deleted)

I didn't check it but I thought you had it.  I was thinking the same thing the other day.
You need an inverter then that feeds two identical circuits, say NPN, pretty much like you had on one of the arms.

The trick is to tweak the base current so they are both 5k for some control voltage.

[Eb7+9]

Thanks for checking Rob,

after a handful of blunders ... maybe this will be ok
previous version only went from common mid-point to extremes

I've never mucked with bipolar devices wired as VCRs
but I do remember buying a Dr Q sometime around '77 and digging it

~jcm

[pinkjimiphoton]

JC
love it!! will try it tomorrow... got an outdoor gig all day today. the shrooms will be flowing lol so no condition to work on stuff when i get home, but will give it a shot tomorrow...
one question, better to use rail to rail i assume, but can i just use one quad oa, or is it important to use three separate ones?

the ca3140's i bet will be perfect for this if they need to be discrete. thanks brother!!

you always have the cool solutions to impossibility. rob tried to lead me there, but i'm too dumb

rock on!

[Eb7+9]

Jim, I updated to idea and there's only one op-amp now ...

you could use one half TL072 for this side-chain
which is hovering around mid-rail, so no need for rail-rail
any op-amp should do

and perhaps use the other half of a dual op-amp as an output buffer
to the signal path

or not ...


enjoy the New England rain dude
~jcm