Will near max IO source current eventually damage PIC?

[Futterama]

Hello forum,

I have a PIC10F206 driving a red SMD LED. To get the desired brightness, I would need to drive it with 25mA. 25mA is the absolute MAX for the PIC source current. The voltage on the PIC pin will drop to about 2.5V (Vcc = 5V) when sourcing 25mA.

Will I damage the PIC over time when sourcing the MAX current?

Regards,
Futterama

 

[blueroomelectronics]

You still need a current limiting resistor. 110ohms should work.

 

[Futterama]

You still need a current limiting resistor. 110ohms should work.

Yes yes, I know, I'm using 120ohms right now, and the current is about 15mA and the PIC pin voltage is about 4V. But I was just wondering if the PIC could handle the current for extended periods.

 

[justDIY]

i don't see why it couldn't - as long as you're within the electrical specifications (on the datasheet), there's no reason to think it would "wear out", aside from manuf. defects or something

not that it is strongly recommended, but I've got leds running straight off pic pins, with no resistor, relying instead on the internal resistance of the pic to limit the current (the collective RdsON of the pic's gate fets). Source voltage for the pic is only 3.6v in this case however

 

[donniedj]

Yes yes, I know, I'm using 120ohms right now, and the current is about 15mA and the PIC pin voltage is about 4V. But I was just wondering if the PIC could handle the current for extended periods.

Yes it will handle the current for an extended period. Also, the output voltage will naturally drop to the LED bias voltage when the LED is attached raw as justDIY mentioned. You do knot that the PIC will not put out more the 25mA +- 4ma on that output right. Any Rlimiters is for the LED and not the PIC.

Will your LED drift from its wavelength sweet spot after an extended period at max current is the question you will find out in time that depends on the truthfulness of the LED vendor and tolerances stacked against it.

 

[Diver300]

I think you will find that you won't get much more current if you do reduce the resistor. The short circuit current is only about 20 mA and once the output voltage start to droop, there isn't much more left.

If you short out a PIC's output, it means that bitwise operations on that port (eg bcf GPIO, 1), will cause the output to turn off.

http://www.sxlist.com/techref/readmodwrite.htm explains the problem, although I try not to overload outputs and avoid adjacent port instructions instead of using a shadow register.