I've damaged LED's in the past by passing too much current through them, such 
that they go dim and don't recover.  But they don't die altogether either.  
Since have many LED's on the board, try swapping a dim one with a bright one 
to verify it's the LED (and not the circuitry), then test the circuit to see 
if there's too much current through it... it's possible that the LED specs 
are incorrect (I've seen that before as well).

-Neil.



On Monday 09 April 2007 18:06, alan smith wrote:
> Have a board with several LEDs on it....ok....about 100.  Getting some
> strange failures, in that some of them will dim out...but not die.  Its the
> standard configuration of the current limit resistor hooked to +V, and a
> NPN switching ground.  I had them short the collector to ground to see if
> that fixed it (ie...transistor not turning on hard enough) no effect.  I
> think...need to verify, that there was no change in the voltage drop across
> the current limit resistor.  They put a higher voltage across the LEDs and
> no change (that I saw on an earlier failure).  These have been running for
> better part of a week now, and all of a sudden they do this.  Granted its
> one or two out of 100....and they are chinese manufactured (I did say to
> them....the quality can be questioned...but usually its brightness between
> batches) and they have used them in the past without issues on a different
> product.
>
>   Just wondering....what might make them....sorta fail dim like this?
>
>
> ---------------------------------
> Expecting? Get great news right away with email Auto-Check.
> Try the Yahoo! Mail Beta.
-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist