LED/LDR based dynamic overdrive?

[obblitt]

HA! Uhhh... So I was thinking about the use of the LED/LDR set up in vibe and envelope filters etc., and was also randomly thinking about the Boss Dyna-drive.

Can you see where I might be going with this? Thinking: is it possible to somehow rig the LED/LDR thing to respond to how much signal is going into the circuit, and from that how much drive is added to the signal? Or am I way off on how the LDR's actually work?

I'm just getting started in the world of DIY and am thoroughly fascinated by it. As you can see, I'm already dreaming big! (Or at least I think I am, has this been done?)

I can't imagine this is the way the Dyna-drive works, because it's Boss, and most (I think all) of their stuff these days is digital....

[Cliff Schecht]

The Boss deal is a digital unit (I believe).

I'm not a big fan of LDR's, they're slow and I don't think they're all that "warm" (as I've seen them affectionately described). They also vary widely in parameters from device to device and don't make the greatest of tremolo's, at least if you want to run your trem past 30 Hz.

For a dynamic type of overdrive, I would probably play around with jfet channel length modulation first, although this isn't simple to do if you're just getting started.

[petemoore]

Can you see where I might be going with this? Thinking: is it possible to somehow rig the LED/LDR thing to respond to how much signal is going into the circuit, and from that how much drive is added to the signal? Or am I way off on how the LDR's actually work?
 That's pretty much the nutshell of how compressors work.
 LDR works by having low resistance in light, high resistance in dark, there are some LDR based comps floating around.
 LED...to get one of these to light use the envelop detector circuit from Nurse Quacky.

[obblitt]

I guess it just sounded super cool when I thought of it. There's always the good ol' volume knob on the banger! 

[flo]

What are the dynamics of the overdrive that you after?
Something like: Less overdrive when playing louder?

[Mark Hammer]

"Dynamic overdrive" is not so much a question of simply throwing more gain at a diode clipping stage.  Rather, it would ideally be something that changes the nature of the clipping in ways that are not simply more/less.

Cliff is spot on when he rules out LDRs.  One needs to consider that the speed requirements vary with type of control sought.  In the case of phasers, LDRs are great for providing smooth transitions spanning multiple notes.  They are fine for producing transitions that span a few notes when tremolos are sweeping at very modest sub-audio rates (e.g., 6hz and lower), but once you need more speed out of them, they start to falter in terms of being able to go from full on to full off as fast as you want them to.  And of course, if you're looking at limiting of very short signal peaks in a "transparent" way (i.e., sub-threshold peak B should not be affected simply because it was preceded by supra-threshold peak A), it's gonna take one heckuva LDR to do the job.

This turns our attention to either FETs or OTAs as control elements.  For instance, maybe a FET in parallel with a fixed resistor, but in series with a diode pair to ground, could provide a way of automatically adjusting the extent to which the diodes clip harder or softer.  That FET would, in turn, be envelope controlled, but with a very quick set of time constants.  So, you're not changing the gain of the stage preceding the diodes, but changing the behaviour of the diodes.  I suspect this is closer to what people are thinking of.

[obblitt]

Wow, thanks for that info. The vision I had was the softer I played, the less overdrive would come into the signal.

But after posting I thought it best to stick with the volume control on the guitar .

[Cliff Schecht]

"Dynamic overdrive" is not so much a question of simply throwing more gain at a diode clipping stage.  Rather, it would ideally be something that changes the nature of the clipping in ways that are not simply more/less.

Cliff is spot on when he rules out LDRs.  One needs to consider that the speed requirements vary with type of control sought.  In the case of phasers, LDRs are great for providing smooth transitions spanning multiple notes.  They are fine for producing transitions that span a few notes when tremolos are sweeping at very modest sub-audio rates (e.g., 6hz and lower), but once you need more speed out of them, they start to falter in terms of being able to go from full on to full off as fast as you want them to.  And of course, if you're looking at limiting of very short signal peaks in a "transparent" way (i.e., sub-threshold peak B should not be affected simply because it was preceded by supra-threshold peak A), it's gonna take one heckuva LDR to do the job.

This turns our attention to either FETs or OTAs as control elements.  For instance, maybe a FET in parallel with a fixed resistor, but in series with a diode pair to ground, could provide a way of automatically adjusting the extent to which the diodes clip harder or softer.  That FET would, in turn, be envelope controlled, but with a very quick set of time constants.  So, you're not changing the gain of the stage preceding the diodes, but changing the behaviour of the diodes.  I suspect this is closer to what people are thinking of.

Mark, I'm wondering why the phototransistor has been left out of pedals for so long.. I've designed heterodyning mixers (two and four quadrant) with FET's, transistors, OTA's, the classic MC1496 and even LDR's, but I have never seen or tried to implement a phototransistor based unit. Recently, with my "transparent" effect submission, I used a phototransistor to control the emitters of a diff-amp (with a bit of extra circuit trickery of course) and had some great results, at least in the lab. I haven't gotten to test the unit on a guitar yet, but I did noticed some very useful side-effects from working with the phototransistor. Since it has an on-off time of about 5 us, the device adds the perfect amount of slewing to a square wave tremolo which prevents audible popping. It is fast enough, however, to avoid a "trapezoidal" sounding square wave trem (where the transition times can be heard in some situations). The speed in which this device acts is about perfect for music applications and may even be characteristically quite musical (although this I haven't really said yes or no to yet). The part I was using was the Fairchild QSC112, for reference .

[R.G.]

(a) the output is a bipolar transistor, with all that implies for switching
(b) variations

Have you tried the H11F1/2/3 family?

[Cliff Schecht]

(a) the output is a bipolar transistor, with all that implies for switching
(b) variations

Have you tried the H11F1/2/3 family?

Why not bias it in the linear region? I had mine configured in a common emitter configuration with a 10 meg collector resistor, giving about 1.5 V (or more) of change in a linear fashion. I was controlling the phototranny with a sinusoidal source and heard a nice sin-based tremolo effects, so I wouldn't limit the phototransistor to just switching.

I have no idea what you mean by (b) variations, unless you are talking about wide device parameters. In this case, I'm sure compensation could be developed without too much hassle..

If I was to use a H11FX family optocoupler, then my effect wouldn't be transparent and wouldn't count towards the theme of the month .

[snap]

Expandora?