Resistance???

Robocop · May 23, 2004

If I understand correct,a resistor in a circuit for powering an LED,will restrict the amount of current the LED receives.If you remove a resistor it opens the circuit up and allows more power as in direct drive.
Now I have recently read that in modding for more power some say they added resistance and increased power.How can adding resistance increase power.Would it not be better to have no resistors than to have multiple?...sorry if the answer is simple and I am just overlooking it.I tried a search however found nothing...

shiftd · May 23, 2004

i think what you meant is on regulated circuit. Usually, the current flowing to the lux is determined by the voltage drop across the sense resistor. So there is always a resistor on the regulator. Now, adding resistance in *paralel* with the original resistance will decrease overall resistance on the sense resistor, and thus would allow more current to pass to the lux.
hope this explains

mattheww50 · May 23, 2004

The answer is that there is more than meets the eye. LED's (and most other semiconductor devices) are inherently non-linear devices, and as a result can be subject to thermal runaway. So for example increasing the applied voltage from 3 volts to 4 volts may not increase the current 33%, it may triple it! Generally the device switches to a near short as you go much past Vf. This is what allows devices like Zener diodes to work.

In the short term , the resistor does limit the output, however without it, the LED may get well past the design limits, and most semiconductor devices suffer from serious performance loss at truly elevated temperatures, often resulting in absolute failure. This can happen in literally seconds.

In effect at higher temperatures that result from higher power, you don't get more light output, you get less, as more energy goes into I^2R heating of the junction. By limiting the current, you can restrict the I^2R heating, and keep the device within design limits. The results is that while initially there may be some loss in brightness relative to the resistor equipped version, if you drive beyond the design limits, as the junction heats up, you will in fact get less output rather than more as a result of overheating. This can happen with truly surprising speed.

If you can afford it, the best solution for optimal output is a constant current source, so it stays within ratings pretty much independent of the actual Vf, which wanders more than just a little.

gwbaltzell · May 23, 2004

I think shiftd's answer is the main one you're after, however it applies to unregulated circuits as well. The formula for resistors in parallel is 1/(1/R1 + 1/R2 + 1/R3 + ...). This simplifies to just the resistance of one divided by the number of resistors in parallel if they are all the same resistance.

Relying on the internal resistance of the battery (direct drive) is a very iffy process. I wish I had a Luxeon for every one destroyed while attempting direct drive.

idleprocess · May 24, 2004

[ QUOTE ]
mattheww50 said:
In the short term , the resistor does limit the output, however without it, the LED may get well past the design limits, and most semiconductor devices suffer from serious performance loss at truly elevated temperatures, often resulting in absolute failure. This can happen in literally seconds.

[/ QUOTE ]
In my experience, a severely overdriven (or overheated LED) will die in milliseconds.