envelope filter nightmares...arghhhhhhhhhhh...

[deadastronaut]

right, ive been messing around with envelope/detectors/filters/followers etc etc.. a hell of a lot lately...probably too much..

anyway its been a breadboard marathon..

1:sound to light, with morley wah.....the 386 makes it noisey

2:sound to light with earthscums lowaht....the 386 makes it noisey

3:nurse quacky....too distorted. and no B+E string response...

4:nurse quacky envelope detector part..

5: and now breaded this @#$%er twice, but no quack  (led/ldr)......will this schematic actually work?....i have signal in/out ok, just no quack!!..
http://www.beavisaudio.com/schematics/DOD-440-Envelope-Filter-Schematic.htm

i'm trying to get a 'sensitive' EF so notes will quack with just hammering/tapping notes...without having to whack the ass out of my guitar..

so

A: will the above schemo work?...anyone tried it?

B: anyone got a 'good' envelope filter with 'subtlety'...but with common parts too.?

C: arggggggggggggggggggggggggggggggghhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhh........


i have yet to breadboard the

DR Q
phunvgnosis..

cheers guys.

[cpm]

i'm trying to get a 'sensitive' EF so notes will quack with just hammering/tapping notes...without having to whack the ass out of my guitar..

thats a pain, any envelope detection usually have a treshold range that is fixed for a very specific range on input levels.
I have protoyped that a compander, using common ic like the NE/SA571, can be tapped for a sort of rectified/averaged output signal, and then it could be processed for an appropiate ADSR control for other stages (filter, etc). The nice thing is that this signal is proportional to the rate of the variation in input, but not so much on the input level of that variation. Simply put, would get a similar "bump" by playing a note hard or a gentle tap.

I suggest you to lead research on that..

[digi2t]

The only envelope detector circuit I've played with is the Foxx Guitar Synth pedal, which is a glorified Mutron III. I've got two different veros and schematics over on this thread;

http://www.diystompboxes.com/smfforum/index.php?topic=95016.0

Might help you.

[Pyr0]

Maybe get some ideas from the Mad Professor Snow White Auto Wah, seems to work pretty well for me on all strings.

[jaapie]

You could try using a comparator to create a gate from your envelope, and then follow that with an attack/release generator. I've had pretty decent results with that method, and you get an even quack across the entire fretboard (depending on how you have the filter set, of course). I've been using a simplified version of the A/R generator in the MFOS subcommander and it works great.

[midwayfair]

Rob, someone either Mark Hammer or RG Keen gave cap values to get a better b/e string response out of the Nurse Quacky. The sensitivity knob also has a fairly small range where it will work properly with any given guitar ... mine seems to hit the b and e strings fine with the stock values, but it's a pretty bright guitar.

I haven't built the Beavis version of the DOD, but it's not radically different from BYOC's, which I *have* built and it sounds crazy good. It's not subtle at all, though.

Also, Madbean's got a Mutron III derivative coming out in a few weeks, might be worth a look. Not a cheap project, though (it uses two vactrols).

Have you tried just running a boost into the Nurse Quacky? A treble booster especially would probably solve the string definition problems you're having as well. You could combine them in one circuit and adjust the decay to keep it from getting out of hand with the boost running.

[WhenBoredomPeaks]

You could try using a comparator to create a gate from your envelope, and then follow that with an attack/release generator. I've had pretty decent results with that method, and you get an even quack across the entire fretboard (depending on how you have the filter set, of course). I've been using a simplified version of the A/R generator in the MFOS subcommander and it works great.

that is almost exactly what i am doing (mfos gate extractor or whatever it is called into yusynth adsr into various synth filters)

my only problem is that if i don't leave a little bit of silence between notes then it won't send a new gate signal so i can't really filter fast licks with it where an envelope follower would work perfect (but then an envelope follower won't give the flexibility of an ADSR)

[Devius]

When I first build the nurse quacky i had a b and e response problem then figured out one of my caps was the wrong value (10pf instead of 10nf if i remember correctly) one I changed it the b string was responsive. I still have a problem with the e string. I tried tweaking the values of the 4n7s to 4n4s but didn't really do much.

[Mark Hammer]

The sensitivity of the pedal is determined by the gain of U1A , which is set by the value of VR1.  As the total resistance on the ground leg (VR1+R2) gets smaller, gain is increased, to a maximum of 221x.

However, the gain applied is pretty similar for anything 160hz and up.  What you'd like to do is create greater gain for higher strings and notes.  You can do that by utilizing a 2nd parallel gain-setting path.

If you stick a 6k8 resistor and .033uf cap in series, and place them in parallel with R2/C1, at max sensitivity, the gain of that stage will be 324x for content roughly 710hz and higher.  If you drop the cap down to .022uf, that shifts to roughly 1060hz.  Max gain for content below that point will still be 221x, as before (actually a little higher for stuff just under the rolloff set by the 2nd pair of components.

To increase sensitivity for everything, regardless of string, consider increasing the value of the feedback resistor R1, from 2.2M to something a little higher, like 2.4M or 2.7M.  Another thing you can do is swap diode D2 (shown as a 1N4148) for  something with a lower forward voltage, like a germanium or Schottky type.

[Devius]

Mark, would that apply to the nurse quacky as well?

[Mark Hammer]

Mark, would that apply to the nurse quacky as well?

Sort of.  The Nurse Quacky uses an inverting op-amp configuration for the gain stage before the rectifying diode.  The gain is modified in a different manner here.  For the Quacky, rather than varying gain, the gain is fixed at [3.3M]/[47k] = 70x, and the sensitivity control adjusts how big or small a signal you apply that gain to. 

The rolloff going into the gain stage is set by 47k and .01uf, which is approximately 340hz.  So it already tames the low end a bit, relative to the middle and higher strings.  But you can still produce the same differential gain thing, if you use a parallel RC pair, alongside the 47k/.01uf pair.  Try out 27k and .0056uf (5600pf), which will get you a gain of around 122x (instead of 70x), starting around 1050hz.  If that seems a bit much, drop back to 33k and 4700pf, for a bit less gain (100x) startoing around roughly the same point.

[deadastronaut]

cheers for the suggestions/tips guys...that SWAW looks/sounds nice.

i have the beavis EF on breadboard at the mo...have signal but still no quack..ive checked all connections thoroughly
http://www.beavisaudio.com/schematics/DOD-440-Envelope-Filter-Schematic.htm

i am using an led/ldr in place of the vactrol...

e.g.  ldr across the 2x 220k's.....led lights up, responds etc.

if i get it going i'll try those mods mark suggested..cheers mark.


.........ldr.........
220k......220k..

going by this schematic its saying ''centre tapped ldr''?....but i'm using 1 x ldr 2 legs...hmmmm...i suspect this is my problem... cure?

http://www.google.co.uk/imgres?imgurl=http://experimentalistsanonymous.com/diy/Schematics/Filters%2520Wahs%2520and%2520VCFs/DOD%2520440%2520Envelope%2520Filter.gif&imgrefurl=http://experimentalistsanonymous.com/diy/Schematics/Filters%2520Wahs%2520and%2520VCFs/&h=481&w=486&sz=6&tbnid=aug6hcpJdUXuYM:&tbnh=90&tbnw=91&prev=/search%3Fq%3Ddod%2Benvelope%2Bfilter%2Bschematic%26tbm%3Disch%26tbo%3Du&zoom=1&q=dod+envelope+filter+schematic&usg=__Mqf_FPFys2cra3bB7Hq910Q0ZJs=&docid=THMqB97guVgAzM&hl=en&sa=X&ei=uxGFULzoOpPB0gWs_4DYBA&ved=0CB8Q9QEwAA&dur=3802


i guess what i'm saying is, can it be converted to use and led/ldr...i dont get the centre tapped thing?...the vactrols ive looked at have 4 legs, 2 for led, 2 for ldr...hmmmm...
here's the recommended one....wtf?

http://www.mammothelectronics.com/VTL5C9-Photoconductive-Cells-and-Analog-Optoisolat-p/400-1109.htm

[midwayfair]

going by this schematic its saying ''centre tapped ldr''?....but i'm using 1 x ldr 2 legs...hmmmm...i suspect this is my problem... cure?

Ah.

Yes, that's your problem. You need a 5-legged vactrol like the VTL5C4/2 for the DOD project. And they aren't cheap:

http://www.smallbearelec.com/servlet/Detail?no=350

I ... suspect ... you can just use two MATCHED photocells, point them at the same LED, and join two legs together.

[deadastronaut]

yeah i tried 2 ldrs....no joy..

@#$%ing envelope filters......arghhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhh

[deadastronaut]

ok i thought i'd post this little problem on here too..

heres what i have at the mo,

the sound to light circuit/EF is fine on its own.....

and the wah is lovely and clean on its own.....in fact its really nice.

but as soon as i connect the guitar input to both to get the led responding into the ldr it distorts...now i know you guys will spot  the problem easily, but i've tried a few things to no avail...

ive been ok with lf'os into this kind of thing, but have never tried the guitar input into both circuits.....any suggestions?

[midwayfair]

Well, that 4.7M resistor is a TON of gain. You're probably overdriving the envelope's op amp. Why that causes distortion in the audio circuit, I can't say, I just know it does. (I had this problem recently in multiple circuits.) Hopefully someone will jump in and 'splain it.

You should have a divider of some sort (you can't just short the guitar signal to the amplifier and add a resistor to the envelope, you'll never trigger the envelope) along with a series resistor going into the envelope section. You also probably need the hold cap in the circuit all the time -- maybe with the cap going to ground, not in series with the second LED. (I'd ditch the second LED here anyway and match it to the reference LED.)

[Gurner]

In that setup, the -ve input pin of IC1a is essentially shunting the guitar input signal to ground (the -ve input of an opamp presents a very low resistance to ground...which loads the guitar signal = much sorrow).....at the very least, you need a high value resistor between C2 & R2.....but even that isn't a particularly good solution.

As an aside, the input impedance of the other opamp is a tad too low (R10 could do with being higher ....same problem...i.e. you're using the -ve opamp input) ....even better use a non-inverting opamp configuration there. You're better following the path followed in the other thread (http://www.diystompboxes.com/smfforum/index.php?topic=90184.60), because the particular LED setup you've adopted above will *never* have a slow release tail (should you seek this)

[deadastronaut]

cheers guys, if only things were as simple as me...

ive changed the EF section anyway.. , i ended up pinching the nurse quacky led sectiion.......(i tried lowering the gain on the opamp, but then i lost the led..)..


@G: ok...looks like i'll have to do it properly then....

[deadastronaut]

ok, had a rummage around, this is from the classic EF mcmeat,
http://www.experimentalistsanonymous.com/diy/Schematics/Filters%20Wahs%20and%20VCFs/Lovetone%20Meatball%20McMeat.pdf

could i just lift the envelope follower from the ''sense''-----to the ''intense''....led. ?

it has attack, decay, etc...

[WhenBoredomPeaks]

ok, had a rummage around, this is from the classic EF mcmeat,
http://www.experimentalistsanonymous.com/diy/Schematics/Filters%20Wahs%20and%20VCFs/Lovetone%20Meatball%20McMeat.pdf

could i just lift the envelope follower from the ''sense''-----to the ''intense''....led. ?

it has attack, decay, etc...

yeah

btw i don't see how the envelope follower section could be this much of a problem, the way i see it it is just an AC-DC converter/rectifier with some gain and sometimes with some caps for attack/decay. i mean once i used a simple graetz bridge as an envelope follower (that was only on a breadboard) to control a synth filter. I had to boost the incoming AC signal (the guitar) though to get around 5-10VDC swing on the output.

btw after a while the guitar's envelope won't satisfy you i think (i mean the guitar's envelope is generally the same shape all the time, that is why they try to shake up things with the attack/decay cap) that is when you should try triggering an ADSR to control the filter instead of an envelope follower but then there are problems with reliable triggering imo

edit: also when you use an envelope controller your filter cutoff will depend on the input signal's loudness which is not always a good thing imo. When you wanna get huge filter sweeps your signal's loudness will swing quite huge too which a lot of the times not a good thing. (think about it from a mixing engineer's viewpoint)