Why are my red LED's REVERSE POLARITY ??

[MathWizard]

I bought some super bright LEDs, and the GRN and blue worked just fine, but the red's, not 1 of them turned on, of the handful I tried, not even with 2.5V on my PSU, so I stopped and planned to get a datasheet, and left 1 on the desk.

Then today, with a DMM, I go to answer a question about diode readings on reverse biased diodes, and lo and behold, THE RED LED WORKS !!

And the large terminal is the Anode/high side. I should take a picture, I assume it was a mistake a the factory? Or are some LED's opposite for some reason ?

[ebastler]

I assume you are talking about through-hole LEDs, and by "large terminal" you mean the long pin? That is generally the anode, i.e. the positive terminal, as you found with your red LEDs. Are you sure it was the other way round for the green and blue ones?

[MathWizard]

It looks backwards for sure, as in the high voltage is on the bigger terminal, and also, i have a normal LED right here too, I also doubt it's backwards, but I swear it really is.

I have to dig up my camera for something else, so I might as-well do both now. I'll put both back to back on a battery

[ebastler]

It would help if you clarify what you mean by "the large terminal" or "the bigger terminal". The longer leg? Or the larger piece of metal inside the transparent plastic body?

[mikerj]

It looks backwards for sure, as in the high voltage is on the bigger terminal, and also, i have a normal LED right here too, I also doubt it's backwards, but I swear it really is.

I have to dig up my camera for something else, so I might as-well do both now. I'll put both back to back on a battery

Does the flat on the body still indicate the cathode?  You can not rely on the lead lengths or the shape of the leads within the package.

[Gyro]

I remember coming across some early high brightness / ultrabright or whatever LEDs like that. I can't remember the colour but they were probably Red. They're probably lurking in a drawer somewhere but I can't remember the details now. It was either something to do with the process (reflecting out of the 'cup'?) or, more likely, two bond wires.

The flat and lead lengths should always be consistent [Edit: with polarity].

[janoc]

It looks backwards for sure, as in the high voltage is on the bigger terminal, and also, i have a normal LED right here too, I also doubt it's backwards, but I swear it really is.

I have to dig up my camera for something else, so I might as-well do both now. I'll put both back to back on a battery

Does the flat on the body still indicate the cathode?  You can not rely on the lead lengths or the shape of the leads within the package.

I have encountered LEDs where the side with the flat was anode and another where it was cathode. The same for the length of the leads. There is no standard for this, so some manufacturer somewhere is guaranteed to do it the "wrong" way around.

I always measure the LED before installation - saves me from putting it backwards and also the occasional dud.

[MathWizard]

This is worse than the time I tried taking a Bigfoot picture. My camera is not that good, and I basically have to build a TV studio and holding jig to position and light it correctly on camera.

I'll post what I have, I'm not James Cameron, so I'm not filming it.

[Vovk_Z]

I never look for led polarity in datasheet (when DIY). I always check it by DMM (so I never had such a problem as TS).
Of cause, usually longer one is "+" but not always.

[capt bullshot]

Yes, there's plenty of LEDs around that don't fit the rule of thumb "the bigger terminal (inside the dome, where the chip resides)" is cathode.
This was true in the past, all the LEDs sat on the cathode terminal, and a bond wire connected them to the anode terminal. Things changed with high efficiency red, blue, and a variety of modern stuff.

A real rare case, nothing to do with polarity, I've encountered many years ago was a red LED in a yellow housing in a bag full of yellow LEDs in the very same yellow housing

[ebastler]

The flat and lead lengths should always be consistent with polarity.

That's what I would also say: Electrode shape within the plastic body can vary, but there is a (quasi-) standard for the flattened body side and short lead indicating the cathode. Although others have already reported finding exceptions to these rules...

Anyway, I always measure LEDs before installing them, but that's mainly because I can't remember the conventions. 

[MathWizard]

I thought that maybe super bright LEDs needed higher voltages or something, that's what I wanted the datasheet for. They are Ligitek LED 3mm Red Water Clear Ultra Bright A-705, and indeed the inside terminals are opposite the norm. They look good

No data sheet shows up right away, and the seller lists them for everything usually

[eblc1388]

I gave up years ago looking at the led for polarity.

Instead I hot glued a coin cell CR2032, new or used will do, to my breadboard. I test every LED I use to make sure it works and the polarity is correct by placing it across the cell briefly, before plugging it into the board.
 

 

[Renate]

Somebody is probably screwing with you.
They went through your LEDs and snipped the long lead shorter.
 

[Renate]

(Sorry for the old post...)

I was building a project. I always check my LEDs with diode test on my DVM.
It's enough to get a good glow. Besides, I mix up all the colors.
I verified that the long lead was the anode.
After I soldered the whole thing together I looked in dismay at my handywork.
The "anvil" (the big supporting internal structure) on the two LEDs was pointing in different directions.
It turns out, I had wired everything correctly.
The long lead on anode was correct, the flat on the cathode side was correct.
It's just that the internal polarity of the two (different colors) was different.

Is there any rhyme or reason to this?
Is this based on color or technology?
Just wondering.
Pokes @Noopy

[MathWizard]

Well never provided an proof yet since don't have a camera that's any good for much of anything.

So far I never used any of the LEDs I bought back then, so this problem awaits me again, so better check any of these before i solder them in.

[musky4821]

These super bright red led's are in places like bicycle tail lights and optical computer mouse and yes the polarity is reverse bias

[BrokenYugo]

They're harder to find but there also exist "bipolar" LEDs that have two reverse parallel dies inside so they light up both ways on AC, or two colors set by polarity.

Personally I usually just check on DMM/component tester, then breadboard one from the bag in question to get the brightness/drive current figured out ahead of time. Not a fan of stuff where they clearly just put 20mA or some arbitrary current on the LEDs and finished development wearing sunglasses.

[CaptDon]

One thing I can say for certain, In the days of wholesalers offering 'grab bags' of 20 to 50 LED's almost all of them were QA/QC fallouts. Dies off center, bubbles in the epoxy encapsulation, low light output and of course lead polarity improperly indicated by the flat on the wrong side or the longer lead not being the positive input. And I am sure someone is still buying and marketing 2nd quality parts today. I worked for a couple of years in the QA/QC department of a 'board stuffing house' contract manufacturer and the one thing I ran into from dubious suppliers was capacitors with the marked shrink-wrap applied backwards!!! We had daily explosions of 2200uf@25vdc capacitors that were marked wrong. I finally created a 6vdc supply with a resistor in series and a mechanical (analog) voltmeter across the resistor for pre-testing and a normally closed switch across the capacitor. The switch would hold the meter at 6 volts, then with a capacitor inserted to the test clips (in parallel with the switch) you would press the switch and watch the meter quickly drop to near zero volts. If the cap was marked wrong the meter would only drop to about 1/2 scale. Releasing the switch would discharge the capacitor so it was safe to be placed on a final product.