Photo diode for clipping?

[casey73]

I was looking through a surplus catalog and saw some photo diodes. It occurred to me that in all the clipping type circuits I've seen over the years, and ones I've built using various diodes and even FETs for clipping, I've never seen a photo diode used in a pedal application. I suspect there is a reason for this, like maybe they don't work well in typical clipping circuit. Anyone have insight into this?

[mikeboone]

Try Monte Alllums at www.monteallums.com He uses LEDs for clippers in some of his kits.

Also look at www.cmatmods.com Chad Mathews is a very cool guy and will likely have an answer or opinion on the subject.

[Mark Hammer]

LEDs are used all the time for clipping purposes, and often to pleasing effect. Please note the following:

1) LEDs have a much higher clipping threshold; generally at least 3x higher than "normal" silicon diodes, and some 6x higher (or more) than germanium diodes. This means that fewer of the peaks in the signal will actually have enough amplitude to send the LED into conduction...UNLESS the input signal is boosted, or whatever gain structure exists in the circuit itself is adjusted to bring those peaks up to the requisite level. Now, it is often the case that the signal IS strong enough to produce clipping without altering the circuit or requiring external boosting. What you will find is that the clipping appears to be less pronounced (because proportionately fewer of the peaks in the signal will be robust enough), and that may well be the goal of using LEDs, so count that as a success.

2) Depending on the circuit involved, whatever diodes are used for clipping may set the ultimate ceiling on maximum output level (which is then attenuated by the volume control). Using LEDs for clipping will raise that ceiling substantially, such that the pedal will be "louder". That may be at the cost of clipping intensity, but since the output is hot enough to send subsequent stages (including the amp itself) into overdrive, you can sort of consider use of LEDs as a sort of "redivision of labour" in that it redistributes the source of clipping across several devices instead of providing it all on its own. Depending on your amp, that can be a very desirable thing.

3) Not all LEDs are created equal. In two ways. First, red LEDs are generally preferred because they have a lower forward voltage than other colours, which still permits them to clip without radical circuit overhaul. So, while you can certainly feel free to tinker with green, yellow, blue or whatever the flavour of the month is, don't expect improvements via colour experimentation. Stop at red, so to speak. Second, even within colours, there are variations in luminance for the same current. So, where a "garden variety" 100mcd red LED will not glow when clipping, a 6000mcd "superbright" LED may well glow perceptibly when clipping. The glow is, as far as anyone knows, entirely separate from its function as a clipping diode, and should not be construed as signifying anything. So, for instance, one hears about "seeing eye" mods to pedals. Such mods assume the use of LEDs with sufficient luminance that they will light up on clip. If they don't it may simply mean you miss the light show, and not that you miss the clip. Given that your eyes should be on your fretboard or audience, rather than on your pedal, that's no great sacrifice.

[Enzo]

Um...

He did say Photodiodes, not LEDs.

TO me that means diodes sensitive to light - the kind and LED shines ON in a light beam interrrupter circuit. So the amount of light falling on them would affect their action in a circuit. That is both a problem, in that you'd have to protect them from ambuent light, especially when the pedal was open for service, and it is a blessing in that it opens up certain opportunities for control.

[R.G.]

All diodes are photodiodes.

If the packaging is arranged so that light can fall on the junction, light of the right color will pump a charge carrier across the depletion zone. The difference between any old junction producing current flow and highly efficient solar cells is just fifty years of semiconductor technology getting the junctions well exposed to light and sensitive to a wide range of light energies so most of it isn't wasted.

When impinged by light, junctions produce a current proportional to the light which pumps charge carrriers across the junction. In normal circumstances this looks like excess leakage when the diode is reverse biased and extra current flow when the diode is forward biased.

The question is probably something like "what has optimizing the junction of a photodiode for light reception done to its forward conduction characteristic when it's kept dark.

In this, as in most effects, the thing we're looking for as effects makers are pure side effects to the main purpose of the devices.

[casey73]

RE: all diodes are photo diodes

R.G. - I think you are getting at my question. I had in mind an arrangement with a photo diode and standard diode (white maybe?) in a light tight capsual, like the common LED/Cds arrangement. Varying the intensity of the light that the LED gives off would vary the photo diode's conductivity. So, would this be able to provide something new/better in a clipping circuit, above the traditional
clipping arrangments?

For instance, having the LED in such an arragement run directly off the power supply in series with a pot to brighten and dim the output. My assumption is, if this were a practical appllication it would have already been done.

[Koreth]

Please forgive the necromancy.

So, more or less light across a diode junction causes said junction to conduct current. Since the relationship between current and voltage in semiconductors is not linear like it is in normal conductors, I have to ask: does the current across the diode due to light exposure have any effect on the threshhold forward voltage? If so, is this effect on said threshhold forward voltage large enough to be of useful or detrimental to a clipping circuit?