How Can I Make Beam Optimization Experiments Easier?
Q. I'm interested in driving my IR LED to the maximum power without significantly reducing it's life span. I've heard that i can drive an LED above its rated power if i do it in pulses?
A. If you actually need 50% duty factor, you will find yourself limited to just double the rated continuous current for the LED, usually about 50mA. For best results, get an AlGaAs IR LED with as much directionality (narrow beam) as you can tolerate. At 100mA, it will drop about 1.5Volts, so choose limiting resistance accordingly or use a current source (Voltage level drives NPN base, emitter resistor sets current).
Contrary to what others have said, you can easily get 50% df from a CMOS 555. Just connect timing capacitor from 2,6 to ground, and connect a timing resistor from 3 to 2,6. Use pin 7 (open drain FET) to drive the load. The C555 makes a Schmitt oscillator this way. Don't forget to use bypass capacitor across pins 1 and 8.
If you can operate with less than 50% DF, you can jack the peak current up proportional to reduction in DF, but don't try to go past about 2 Amps, since the wirebond in most IR LEDs can't handle more. In fact, some will fail at significantly lower currents. The most fragile LEDs tend to be the ultra efficient visible units. A few IR units are rugged enough to handle about 10Amps peak.
If you are operating the LED in bursts with quiet periods in between, you can increase peak current somewhat. Exactly how much is beyond the scope of this reply. EMail me with further questions.