Envelope Trigger

[liquids]

I've been working with envelope filters again lately, in conjunction with 'synthy' sounding effects (see the 'Synthbox' thread).  I definitely like being able to use the synthy sounds by themselves, minus any envelope filtering....but I'd also like to use them together with an envelope filter.

So I thought to build a dedicated filter in the same box as my (eventual) Synthbox final build,in order too use a clean, buffered signal to control the envelope of the filter rather, than run one into the other and have the gain/frequencies/dynamics affect the filter enveope.  Blah blah...nothing new there. 

However, as I mess around I realize, if I build other synthy boxes, I'd have to give each it's own dedicated filter...hmmm.

A couple EFs out there were build with effects loops or an additional triggered/envelope input jack.   I think this is a great concept.  I've never seen a dedicated buffer/trigger pedal  touted as such.  I suppose your typical a/b box that does true A+B could work in that regard so long as it's active/buffered, not just passive splitter.  So that's what I'm thinking of doing, with a few subtle tweaks....I'm in the brainstorming stage.

If run first in the chain (or after GE fuzzes, whatever), it would for one buffer my signal, and likewise split it, meaning the same clean signal would always trigger the filter, no matter what dirt pedals I ran before or after it.   That means no need to joint house it with anything.  It would ALSO solve a major issue I've had with leveling--with a gain pedal running into a EF, what to the ears seems like 'unity gain' compared to said clean sound is not so to the envelope filter.  This should solve that too!   

At first the idea was basic - build a buffer and tap off the output with multiple caps feeding the same number of jacks, and voila.  However, I'm wondering if it needs more complexity...as reading this made me ponder:
http://www.muzique.com/lab/splitter.htm

Hmm.  Now, I'm probably more likely to use op amps not jfets, bear in mind.

Questions: 

1) What are the advantages to having the input signal feeding two separate inputs of two op amp buffers, and then using the two outputs in this manner as a splitter?  Assume the worst - that the input is a passive, unbuffered guitar pickup signal at the input.

Likewise, I'm assuming that I'd be better also first to buffer, then to split the signal to at least two op amp buffers rather than expect the passive signal to feed both op amps....

2) If I'm using your average good op amp with good input impedance characteristics (TL072/Tl074 lets say, if we need to be specific) how much does it even matter that I buffer it first before I split it and re-buffer it? 

And then, a question more specific to the 'triggering' concept:  On the 'envelope' side of filters/compressors, etc, it's common to say 'use a smaller cap so that the bass frequencies don't trigger the envelope more than than the high frequencies.'  I get that, but since I also sort of understand sallen key filters, I imagine that using a sallen-key HP filter would be a good thing, after buffering, to feed the trigger out and make it more even in response along the fretboard:   http://sound.westhost.com/dwop-f18.gif

That in mind:

3)  Any tips on what frequency range to shoot for with such a HP filter for this use?

4) Since the part of the split signal that will be going to the filter's 'envelope input' will hence be getting some potential active filtering, how does this impact using one buffer vs two or three as questioned above?

The good news is the input signal will be very consistently seeing my guitar pickups, rather than an octave/fuzz, etc, so frequency content will be consistent outside of pickup variations. The bad news is, I have no idea where to start.  I realize guitar fundamentals are between 80-440hz +....From doing sound, my first guess would be that everything below 80hz is bad, possible ~100hz is a safe start.   But I also then think, maybe a 2-pole sallen key HP may be 'too much of a good thing' with such a steep roll off slope, and if not tweaked carefully, it could make my low strings overly unresponsive.   Is the more gentle sloped of a 1-pole active filter (series resistor to inverting input, cap across inverting input to output) potentially better?



Help...I think this could be a useful concept for quite a few different purposes if it hasn't been done before (And better)...well, at least, in my head it could be.      And I usually do things by trial and error, and I plan to do so in the coming days, but I thought other's might both appreciate the idea and run with it, and/or contribute.

Cheers!

[liquids]

Is the more gentle sloped of a 1-pole active filter (series resistor to inverting input, cap across inverting input to output) potentially better?

Pardon, I had low pass on the brain...and even still I probably would have it wrong...for HP, it would be probably wired like a buffer with a capacitor to non-inverting input, and a resistor across the inputs...even if not you get the idea.

I tried some preliminaries of this on the breadboard, and I liked it. Very useful, great improvement to always trigger from a buffered signal right at the input rather than variably what is at the 'input' of the filter wherever in the chain it may be.... Honing in on what frequencies, how many poles, what kind of high pass filtering is best for the 'trigger' is still not easy though.   Any tips would be appreciated (bump!). 

[StephenGiles]

Nothing like talking to yourself on these "ever so more complicated than fuzz box" threads Don't worry, I've been there and the teeshirt had already shrunk!!

I had every intention of looking at my angle on the "adaptive peak following" start/stop frequency style of envelope generator over Christmas, but I've had "timetables" thrust upon my by events and my dear wife! For instance - 4 visits to the Vet, and as soon as that red Qashqai of her's returns from the doctor's surgery, it's dog / health walk time.

However perhaps on Friday I may get a couple of hours "can you just do this" free!

On the subject of triggering, you might care to look at the conditioning filter at the front end of the "What"  compressor. This attempts to even out the level of 3 (I think) frequency bands before hitting the rectifier circuit.

[loss1234]

one thing i have found with triggering, (and the product i am releasing at namm has a trigger out for working with outboard fx or modulars) is that you want to figure out what you are using it for,  determine the level of what is going to be coming in, and then work a lot to fine tune the duration of the trigger. like if this is mainly for guitar than you know the frequency range and the basic level.



for my purposes, my box puts out a very SHORT gate which is useful for triggering an ADSR or a VCA.


the key for me was all in the peak detector values.

i spent a long time getting it right, and as you mentioned, it comes straight from a buffered input, NOT after an adjustable gain stage. This might bother some folks but i actually like it.

it gives you a pretty reliable pulse without messing around with settings. however getting to the point where i was happy with the pulse with a guitar input took months.

however, this is all very subjective. you might find that you like a longer pulse in your breadboard trials.

i didnt fully read the first post so i hope i didnt totally miss the point of this, as i am off to assemble some stuff, but thanks for starting an interesting thread that as Mr giles mentioned, doesnt seem to interest many people

thanks!

[StephenGiles]

however getting to the point where i was happy with the pulse with a guitar input took months.

Ah, that gives me a little breathing space then