Yes, absolutely, a fully on diode or not without a resistor is not safe for a microcontroller pin. And with the digital multimeter once the current settles in it's too late to have seen that spike. Is that kind of what you're saying ? Also below forward bias current I believe the LED produces extra heat if I'm not mistaken.With unfiltered PWM, you still need a limiting resistor (or better a constant current source), otherwise the current will exceed the normal range when ON, LED response being very quick (ns). It will produce more heat than light. The eye perception could be misleading.
Yes it seems better to use an NPN transistor to turn on the LED. Voltage drop is certainly more predictable. Nice document thanks for sharing!
When I tested these with the 9 volt Supply I did use an indirect method to look at the light intensity like you were saying before. Over 20 mA lthe brightness was certainly brighter then at 30 and 50 Etc but not really anything that I would consider useful. I suppose that could be different with a diffused red LED but again over Vf you're certainly using quite a bit of power.
A constant current source?, I am not familiar with these.
Statistics: Posted by breaker — Sun Dec 29, 2024 6:50 pm