Power from Green LEDs?

Oldbitcollector (Jeff)

For some reason I was hooking various LEDs to my meter late the other night,
(don't ask, it's a sleep deprivation activity for hardware geeks.)
and I noticed that I was getting current from LEDs, a rather nice amount from
green LEDs, when I held them under my shop light.

I'm starting to wonder if I were to collect a few more, maybe I could use them
in place of small solar cells? They certainly are smaller. I doubt I've made any
new discovery here, but does anyone understand why this is working this way,
or even more why/why not this is/isn't a practical way to generate a small amount
of electricity?

OBC

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
New to the Propeller?

Visit the: The Propeller Pages @ Warranty Void.

Mike Green

Any diode or transistor will produce some current if exposed to light. Solar cells are just diodes optimized for the production of power at low cost. The amount of current is proportional to the junction area and the intensity (and frequency) of the light.

SRLM

I bet if you were to take your LEDs outside and put them under a magnifying glass that the green LED would be the worst. The assumption is based on plant color: they reject green, and soak in everything else.

Peter Jakacki

I use green leds as daylight detectors either by reading the voltage directly into a high impedance input or else by feeding the led directly (no need for a current limit) into the base of an NPN. The minute current produced is enough to bias the NPN on and of course the voltage is clamped to Vbe but a common emitter configuration is a current amplifier after all. With a collector load resistor of around 47K the output is a nice logic level to indicate daylight or any white light such as fluros etc.

As to producing enough current to power anything I doubt that very much. If you connect a current meter you will be fortunate to read 10's of microamps but more than likely only 1 microamp or so even under very bright light. Of course in so doing your terminal voltage is now almost zero. Don't draw any current and you will read around 1.5V, but not both together. I think a lemon makes a better power source, and four lemons should be able to power a Propeller. But if the lemons aren't fresh then the brownout detector operates

@SRLM: If leds were plants you might be right.

*Peter*

Tracy Allen

You might find this light meter application from the Prop tutorials amusing and informative. The green LED measures the light level, and alternately serves as the means to indicate the light level.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Tracy Allen
www.emesystems.com

Craig Eid

OBC,

You won't solve the energy crisis with LED's but you can have some fun experimenting with different parameters using·a low input current op amp. You can compare the change in LED current as a function of·LED color (red, green, and blue), wavelength (red, green·and blue filters) and luminous intensity (standard and high output LED's).

Have fun!

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
························ Craig Eid


··········· Triad Research and Development
Electical Engineering Design and Consulting Services

··················· www.TriadRD.com

Tubular

Hey OBC,

If you get your LEDs right you can even roll your own optocoupler, one you can actually see working. (Perhaps something you can include in your book)

Put two high efficiency narrow angle 5mm leds directly facing each other. The voltage transfer ratio is surprisingly good but current ratio very poor (hence not much power). But good enough for driving CMOS inputs etc.

If you need more voltage you can use two pairs - two transmitters in parallel, and two receivers connected in series to give ~ 3-4 volts.

I believe there is also some kind of wavelength shift - so you might get best results emitting Yellow but Receiving on an Orange LED, or vice versa (cannot remember which way it shifts)

If you want to see a commercial implementation check out the PVI by IRF
www.irf.com/technical-info/appnotes/an-1017.pdf

tubular