identificationledpolarityterminal
In the figure above, LED A is a regular 5mm LED and the visual identification of its terminals polarity is well known.
As you can see, on LED B the anode is identified by the longer leg, but the other rules for visual identification are inverted.
I've confirmed the situation above with a multimeter.
I have no datasheet or reference for LED B.
I never saw an LED with these issues before.
Is LED B the result of a manufacturing error?
Thanks in advance.
Micro Power Systems: http://www.datasheetarchive.com/pdf-datasheets/Datasheets-111/DSAP0025577.html
The package is called by different trade names by companies,
Lumileds Corp calls them "Barracuda",
while
Osram Opto. calls them "Dragon".
The important feature of the package is a heat-conducting metal center, the "heatsink slug", that can be soldered onto a larger heatsink below:
For details and source, see the chapter on LED packaging of the somewhat-retired LED-related website Light-Emitting-Diodes-dot-org - (note different domain).
See the bottom-left image: Chapter 11: Packaging - Fig 11.5.
A possible reason for not having a clear type code for this package could be that it was not actually a LED package, but just a more generic heatsink-with-extra-plastic-lens.
Best Answer
Super-Red (AlGaAs) LED dice are mounted the opposite way of most others, so they appear backwards from the silhouette of the leadframe.
The leadframe pin length should tell you the difference but there's been a case of one identified in another question here that's reversed even for that (presumably so they didn't have to buy two leadframe types or trim them differently- it's described in the datasheet so it's not an error).
So, the bottom line is "no" you can't trust anything that's not specified in the datasheet (and even then, stuff happens, I've seen a 1N400x diode with reversed markings, presumably dropped and replaced on the line- post test, pre-marking by a worker in Taipei where the factory was at the time).