I attempted to replace my brake/tail light bulbs with some SMD LED's. The Led's smoked and melted. I replaced them with incandescent bulbs and all is well. I'm guessing defective LED's (E-Bay) but is there something else I'm not seeing?
There was some corrosion in the lamp socket which I cleaned out. Any ideas?

 

Don't leds have pos and neg ?

 

I believe they are polarized (is that the correct term?). If they don't light up when first installed, they say to reverse them....

EDit - no, that's not the correct term. The polarity needs to match for the LED's to function.....

 

Your lucky the "long term lasting LED" didn't torch your ride, chalk it up to a lesson learned i.e. E-bay = shady.

 

Well stated and sadly, very true....

 

Common causes for LEDs to experience catastrophic failure:
- Manufacturing defect, which makes the "semiconductor" into a "conductor."
- Polarity reversed during assembly or installation
- Peak forward current exceeded
- Peak forward voltage exceeded.

1) You can't really do anything about.
2) Find out what got reversed - a(n) emitter chip(s) or the wiring
3) Several causes for this, not generally involved in installation of the final assembly.
4) Failure to install a voltage-limiting resistor is the fault here. Typical Vf for an LED emitter chip is down around +2VDC (they don't like AC, but you don't run into that outside of the alternator case, typically, unless the rectifier diodes have failed) and a resistor is installed in series to reduce the operating voltage. The assembler should have known this, based on the intended application of the final assembly.

Those are my first thoughts. Corrosion in the socket would have increased resistance, which would decrease both voltage and current, and therefore not have been a great problem.

(The low operating voltage/current of an LED emitter has much to do with why they've gained popularity in handheld lights, once high-flux emitters were developed. My SureFire 6P has an LED conversion head in it, the head has a voltage regulator inbuilt. Source voltage from the batteries is a total of +6VDC, the regulator hauls it down to +2.1VDC. As long as the battery voltage exceeds +2.2VDC, I have useful life in the LED - since current requirement is down around 50-60mA, as I recall. The current/time curve of the CR123A cells - particularly from SureFire, since they have a higher energy density than typical commercial CR123As - gives me SOLID life on the batteries - because voltage and current share a direct mathematical relationship. cf: Ohm's Law, Watt's Law.)