Lpc2000

newmanrf@... wrote:

>Hummel,
>
>    These guys are pulling your leg. That means they are teasing you for
>not knowing any better.
>
>    It is not eloquent to connect a led to a i/o pin without a resistor if
>your pin can source or sink more current than your device can
>withstand and survive. Yes it will work for a while but then either
>your led or your pin driver will fail so no you should not do it.
>
>  
>
Something that I did not realize from the LED datasheets, until I did a 
dot matrix display bar, was that LEDs have a max current and a max 
voltage rating (breakdown voltage).  That breakdown voltage is 
surprisingly small!  IIRC, the VF parameter is the maximum voltage 
across the LED junction, exceeding that value will cause the junction to 
be punctured.

If you think of the LED as a current device and use series limiting 
resistors, you are ok.  This is because the resistor soaks up some of 
the voltage in the loop (kirchoff's law).  When you "burn out" an LED 
due to too small a resistor, it becomes a question of did you apply too 
much current and overheated the junction -- OR -- did you overvoltage 
the junction?

I typically calculate the VF (voltage across the junction) to be within 
80% of the rating, then subtract that voltage from the applied voltage 
of the circuit.  Once I have that "excess voltage", I then calculate the 
resistance needed to "soak" up that excess votage.

TomW

-- 
Tom Walsh - WN3L - Embedded Systems Consultant
http://openhardware.net, http://cyberiansoftware.com
"Windows? No thanks, I have work to do..."
----------------------------------------------------