Drat! Killed another LED....

[PlayboyJoeShmoe]

NO, not an LS (pretty sure I couldn't take that!).

You might recall that I found some Linrose Sunburst Series 10 Candella green LEDS at Frys.

You may also recall I killed one with 2x123 in short order.

So I have had luck running them on 3N. I had one with a nice closer to cyan than green cast to it. I have a 3AA Brinkmann Blackmax that I was trying to make something usefull out of.

The LED doesn't really care for 3AA. Now I know I probably should have used nimh for less voltage. But that don't change nothing, it's still dead. /ubbthreads/images/graemlins/twak.gif

I guess what this really says is I don't know anything about resistors and diodes and am stumbling along blind. /ubbthreads/images/graemlins/icon15.gif

Good thing stuff like Madmax, Badboy, and Opalec exist! :thumsup: /ubbthreads/images/graemlins/bowdown.gif /ubbthreads/images/graemlins/grin.gif

But I have all these crappy little M*g copies that really need help! /ubbthreads/images/graemlins/help.gif

Oh well. /ubbthreads/images/graemlins/icon23.gif

 

[James S]

I've had some surprises hooking up LED's to various battery combinations too /ubbthreads/images/graemlins/smile.gif Actually, running on NiMH, even though the voltage is lower may actually put more current through the LED and cause it to burn out faster as there is less resistance in the cells.

Use one of the various online LED resister value calculators on the web to figure out a safe value for the resister and then tweak it until you get the output you want, at least you'll blow fewer LED's that way. My favorite one is at:

http://linear1.org/ckts/led.php

But there are many. You just enter the led forward voltage, battery voltage and the current you want to pass through the LED and it will tell you a good value for the resister.

 

[UK Owl]

build a constant current circuit instead of playing with resistors, that way you will maintain a steady brightness throughout the battery life.

The simplest way is using an LM317 in constant current mode, but your battery voltage will need to be about 3.7 (2.5 volt dropout on an LM317 & 1.2 volts dropped across the current sense resistor) volts higher than the foreward dc voltage of the led. (this solution will be ok if you wish to run a single LED from a 9 volt pp3 battery.

Alternatively you can build a constant current source from two diodes, two resistors and a NPN transistor that only requires 1.2 volts more than the foreward dc voltage of the led.

Sorry I can not post a picture /ubbthreads/images/graemlins/icon23.gif, but the circuit can be built as follows:
(It will help you to understand this circuit if you draw it on paper as you read the description)

1) connect resistor 1 in series with the two diodes between the + and - on the battery (with diodes connected so a current will flow through them) with one lead of the resistor connected to the + on the battery.

Choose a value for resistor 1 using the following formula :

Resistor 1 value = ( battery voltage - 1.2 ) / 0.002

2) connect the junction of the resistor and one diode to the base of an NPN transistor.

note: use a high gain transistor i.e. hfe > 100

this will cause the base of the transistor to be biased at (typically) 1.2 volts (voltage will vary with diode type) providing up to 2 milli-amp base current.

3) connect your LED between the + on the battery and the collector of the transistor.

4) connect resistor 2 between the emitter of the transistor and the - terminal on the battery.

this will cause the current to flow through the transistor until the voltage dropped across resistor 2 reaches 0.6 volts.

for 20 milliamp resistor_2 value needs to be = 0.6 / 0.02 ohms = 30 ohms

actual current flowing through the LED will vary slightly with different transistors or diodes and as battery voltage drops, however the regulation is far better than using a resistor.

I hope this is of help to you, and I have not confused you too much. /ubbthreads/images/graemlins/icon15.gif