Dynamically Controlled LFO

[thedefog]

Hey guys,

I'm wondering if anyone has tinkered around with this idea here. I was thinking it would be cool to have a tremolo thats speed/depth/whatever was dynamically controlled by your playing. This could be useful for any LFO based circuit. I'd imagine it would be easy to do. My current line of thinking is to just replace a speed pot with a Photocell (or make it switchable) or maybe even just run it inbetween the pot itself. How about just adding a dual opamp into the currently passive section of the Tiny Tremolo and having one opamp be a buffer for an LED going to the photocell and a pot to control intensity? The other opamp would just be a buffer for the guitar signal with a minor volume boost to compensate for the perceived volume loss that occurs with tremolo effects. The only problem I could see would be ticking from the 555 timer bleeding into the guitar signal.

[Mark Hammer]

Use of a photocell or optoisolator to vary LFO speed dynamically has come up frequently here (though I'm thinking about the last 10 years  ).  Simple enough to do with the classic 2-op-amp LFO.

The chief challenges are to:
1) make the dynamic changes feel as "organic" and musical as possible, rather than jarring or disruptive.  Part of that is dynamic sensitivity (i.e., how much speed change with picking change) and part is the onset and offset (timing) of the speed change.
2) provide some means to tailor the dynamic adjustment of speed to suit the needs of the tune.

So, you'd want some way to set a minimum speed, and a maximum "envelope influence", and you'd also want some way to tailor how quickly it speeds up in response to playing, and settles back down.

[CynicalMan]

Pigtronix has done it:
http://www.pigtronix.com/products08/tremvelope.html

Their pedal changes the depth of the tremolo along with the speed. That's probably something you'll want to look in to, since faster tremolos often sound more "effected" even when the depth of the tremolo is the same as a similar slower tremolo.

[thedefog]

Use of a photocell or optoisolator to vary LFO speed dynamically has come up frequently here (though I'm thinking about the last 10 years   ).  Simple enough to do with the classic 2-op-amp LFO.

The chief challenges are to:
1) make the dynamic changes feel as "organic" and musical as possible, rather than jarring or disruptive.  Part of that is dynamic sensitivity (i.e., how much speed change with picking change) and part is the onset and offset (timing) of the speed change.
2) provide some means to tailor the dynamic adjustment of speed to suit the needs of the tune.

So, you'd want some way to set a minimum speed, and a maximum "envelope influence", and you'd also want some way to tailor how quickly it speeds up in response to playing, and settles back down.

So should I look into Envelope generators and VCAs then and just skip the whole Vactrol route? Or is there a way to mess with the speed in which the LED lights and dims just using some caps and pots? Either way, I'm going to breadboard some circuits up tonight and play around with it a bit more.

[Mark Hammer]

Oh heck, it's a pretty simple thing to tap the signal off the input buffer stage in any effect and use a single op-amp to drive an optoisolator that is placed in series or parallel with a regular speed adjustment pot.

All my blathering is not about how complicated the circuit needs to be, but rather about how much trial and error you'll need to engage in to get the values and the feel just right.

Alex's point about the tradeoff between sweep rate and width is a cogent one.  This is also a reason why you need to think through how much you want it to respond to picking strength.  If the change in speed is substantial enough, you likely will want to incorporate some compensation in width.  If the change in speed you produce by picking/strumming harder is not that great, you may be able to easily live with not compensating for sweep width.

It's all about the tradeoffs, and thinking through whcih ones you want to make.

[jimbeaux]

Craig Anderton had a circuit called a "Pluck Follower" for the guitar synth type modules - featured in Device Magazine (back in the day).

It takes the trigger pulses from an envelope extractor - averages them & outputs a control voltage - using it to control an LFO is the idea I think you're after.

Instead of a filter sweep for each note you hit - it would track how slow or fast you are picking > feed a voltage to an LFO. Ultimately you would control a voltage-controlled Filter (VCF) or Amplifier (VCA) (tremolo effect) - or both simultaneously.
(there's also a simple inverter in one of these issues - which would "invert" the voltage - slower picking / faster LFO & vice versa.

Pluck Follower - page 7 in this pdf > http://hammer.ampage.org/files/Device1-11.PDF

[R.G.]

See Vibra-Magic at GEO for one way the old Thomas Organs did something like this.
oops. Forgot link. http://geofex.com/FX_images/vibmatic.pdf

[thedefog]

Wow! Nice information. Thanks guys. I'm very interested in both that pluck follower and that EAN Tremolo daughterboard. They're both very cool looking.
I guess the advantage to the pluck follower is that it will work in all of the CV gear I have as well. I'd love to plug that into my Soundlab Ultimate. I'm gonna breadboard it and mess with that tonight, and probably just build that EAN tremolo on perf. You can never have enough tremolos.

[Mark Hammer]

I think the pluck follower is a great idea....except that you need to derive a trigger pulse, which will, itself, require you to have an envelope follower.  The pluck follower is a kind of "meta-envelope follower" in that it takes events produced via an envelope follower, and collates them over time.

Great idea, just potentially more complex than what you were intending.

[thedefog]

I think the pluck follower is a great idea....except that you need to derive a trigger pulse, which will, itself, require you to have an envelope follower.  The pluck follower is a kind of "meta-envelope follower" in that it takes events produced via an envelope follower, and collates them over time.

Great idea, just potentially more complex than what you were intending.

The complexity part doesn't necessarily scare me away if the result is a more musical sounding effect. And there is more than enough information on here to get me started. I'm going to play with a few different designs tonight and see what I come up with. I think I've finally (after 6 years of tinkering) got a decent enough understanding of basic circuits whereas I can really start experimenting without wasting a lot of time guessing. Although I have to admit that I've had a few fun moments in the past, especially the time when that tantalum cap exploded. 

[deadastronaut]

http://www.diystompboxes.com/smfforum/index.php?topic=91246.0

[StephenGiles]

One of my long lost Tech Tips articles published in ETI back in the 1980s was an abomination which was a tremolo if played softly and a fuzz if played hard. I think it used a CMOS device for the tricks.

I remember that my Mum typed the text for me - and would you believe my ex wife tried to muscle in on the mean £20 per article I was paid!!

[petemoore]

  Throw an LDR across 'the' pot.
   Use stop and jumper resistors, white-out the R values around the control pot [ie if there's a stop resistor on the pot] and the pot on the effect schematic, redraw R value/sweeps that will satisfy the circuit and afford control parameters desired [or not, there are limits to LDR min. etc.
  The Env detector circuit isn't my specialty, but dark to medium-bright LDR range seemed enough to get good 'what' sweep [in this case LDR thrown across SS phaser speed 500kL pot]. Put a capacitor across the LED so the ramp down from fast wobble was a bit slower.
   Reverse LED env. effect is possible also...
  Main limitation that was outside the parameters I wanted [also did some LDR cotnrollin' on other effect controls] is the LDR min. resistance.
   Often enough the 'shaped' R ramps [ie stop and/or limiting resistors] had ample minimum resistance with and without existing circuit modifications.
  Sometimes throws right in and gets 'it', other times a pot finds a static value [very difficult to read a peak min resistance of Env controlled LED driving LDR] as starting point for say a stop or limiting resistor.
   Some controls are larger value than the circuit would care about anyway..the pot R-value 'position' can be reduced [pot can be removed entirely and replaced by fixed/swept resistance].
 
 

[thedefog]

 Throw an LDR across 'the' pot.
  Use stop and jumper resistors, white-out the R values around the control pot [ie if there's a stop resistor on the pot] and the pot on the effect schematic, redraw R value/sweeps that will satisfy the circuit and afford control parameters desired [or not, there are limits to LDR min. etc.
 The Env detector circuit isn't my specialty, but dark to medium-bright LDR range seemed enough to get good 'what' sweep [in this case LDR thrown across SS phaser speed 500kL pot]. Put a capacitor across the LED so the ramp down from fast wobble was a bit slower.
  Reverse LED env. effect is possible also...
 Main limitation that was outside the parameters I wanted [also did some LDR cotnrollin' on other effect controls] is the LDR min. resistance.
  Often enough the 'shaped' R ramps [ie stop and/or limiting resistors] had ample minimum resistance with and without existing circuit modifications.
 Sometimes throws right in and gets 'it', other times a pot finds a static value [very difficult to read a peak min resistance of Env controlled LED driving LDR] as starting point for say a stop or limiting resistor.
  Some controls are larger value than the circuit would care about anyway..the pot R-value 'position' can be reduced [pot can be removed entirely and replaced by fixed/swept resistance].
 
 

Nice notes. Thanks. I've been messing around with it a little more today, and it definitely is tricky controlling the way it feels. When I finally have something that sounds good to my ears, I'll share my results. I still haven't settled on what tremolo circuit to mess with yet either. I have 1/2 of the EA Tremolo built, and the Tiny Tremolo on my breadboard currently which is what I've been playing with with the speed and depth LDRs.