I was just testing the LED indicators on a pedal I'm workin' on, and one of them seems a little...sick.  It's a yellow LED and it barely lights up even with a full 9v going through it.  I don't recall taking very long to solder it up -but is it possible that I may have fried it?  :shock:

It's certainly possible.  Too much heating can be a problem with LED - especially if the leads are short.

Try a 680 to 1k resistor with the LED on a 9V battery without any additional circuitry - if it's still dull then the it's sounds like the LED is kaput (assuming the battery is OK).

Too common scenario....

"Hmm, I clipped these leads.  I wonder which lead is the anode and which the cathode.......I'll just yank out a 9v battery, nestle the leads against the battery connectors and flip the LED around to see.......oh......I guess that *would* have been the anode....if it hadn't glowed so brightly for a moment there......hmm, wonder if there are any more working LEDs in here"

I have personally never experienced a *partially* functioning LED, but I've seen stranger things so...why not?

Always good to keep around a 9v battery you consider "the walking dead" for LED-testing purposes.  I usually have something in the 3-5v range handy for just that purpose.  If you can just barely tell it's not dead, using your tongue (see other threads) that's ideal.  The LED will light up and live to do so another day without any current-limiting resistor having been needed.

...no one has actually come out to say this in so many words, but LEDs for the most part, can't handle 9v.

I've always known the majority of them to like 3v to maybe 5v. If you're testing them with 9v, you're punishing the LED. I will second the 680K resistor comment.

Roger

Is this already in the circuit? Wouldn't an LED just barely glowing indicate to much resistance? OR to weak of a battery?

Just a thought.

I think you have an under-done LED there. You know, you can under-cook meat, why not LEDs? :?

i have also never experianced a "half dead" LED.  Sure its not a 12ver?  Never seen those in yellow, only red, but the exist.

Over the years I have killed and/or damaged many an LED.

The voltage isn't important.... it's the current. The forward voltage is constant and dependent on the chemistry involved.

If you force feed an LED more current than it wants (usually 20mA max), you overheat it, causing the semiconductor junction to diffuse and fail. If this is done quickly, the LED may actually explode with a very loud pop as it shoots plastic chips around (usually two or three pieces ). If the overheat is slower, you can watch the brightness dim as the damage happens. If you do this with green LEDs, you will see the chemistry change by watching the wavelength of the output. It will usually shift quite a bit in the death process from green to yellow or even slightly orange. Red LEDs sure do get less bright, but I wonder how much of that is the spectra shifting towards infrared?

So.... yes.... you can partially fry an LED very easily and it will put out less and less light. If you watch it on a meter, you can see the current consumption rise proportional to the failure of the junction.

Take care,
-Peter

See, my under-cooked LED theory was right after all

In one of my EE classes, once in a while the professor liked to show what happens when components are given voltages beyond what they can take-blowing up a polarized cap, burning out Christmas lights, etc.

An LED will spark very bright and burn out if you connect it to 9V without a resistor. Somehow he got it to spark and burst, but I'm unsure if the votage was applied forward or reverse across the diode.

Ben, you're not the only one who blows stuff up playing with wallwarts. :lol:

It's a phase.... you'll get over it and graduate to blowing up expensive chips with regulated power supplies. (5 volt signal processors don't like 12 volt supplies. :oops:)

For anybody who wants a good amateur video "equipment death" special effect.... try running 120V into a 10 ohm 1/4 resistor (I assume a 22 ohm resistor would be correct for Europe).

Take care,
-Peter

Quote from: Peter Snowberg
If this is done quickly, the LED may actually explode with a very loud pop as it shoots plastic chips around (usually two or three pieces ). If the overheat is slower, you can watch the brightness dim as the damage happens. If you do this with green LEDs, you will see the chemistry change by watching the wavelength of the output. It will usually shift quite a bit in the death process from green to yellow or even slightly orange.

A long time ago (over 10 years) one of my friend connected a LED he found on my table to a 12v powersuply (also on my table) and turned it on before I had time to say anything. It let out was a loudish snap and the front part flew across the room and made a hole to one of my posters.

I have actually seen green LED to turn quite red before burning out. Some of them actually seemed to recovered from it if I disconnected it early enough.

Quote from: Peter Snowberg
For anybody who wants a good amateur video "equipment death" special effect.... try running 120V into a 10 ohm 1/4 resistor (I assume a 22 ohm resistor would be correct for Europe).

I tried to connect smallish (maybe 100ohms or something, maybe 10) 1/4W resistor to speaker output of my small 15W combo amp and playing my guitar through it... I could "play" "smoke signals" with it.  

The LED is not partially but totally fried: what glows is a very thin gold wire that feeds the top of the semiconductor wafer. You now have a fragile and very inefficient incandescent bulb

I guess the heat from my iron damaged it, and the 9v put it out of it's misery.  :oops:

The pedal has 3 indicators and I applied 9v to all of them to make sure they were wired up properly, and even though the other 2 still work, I guess I'll replace them too just in case.  The Red LED lit up yellow!

I've learned my lesson.

Playing smoke signals! COOL! Time for my amp to be re-cased.

I'm thinking of taking the amp head out of my practice amp's enclosure thing and just having the speaker by itself. That way I can use both my SS amp and my future FireFly rev. II tube amp without having to buy another expensive speaker... :? It would also mean that I can use 100 ohm 1/4W resistors ghehehe :twisted:

i had an led explode on me the other day.  i was trying to push it in to a hole that was too small.  I dont think it works anymore.

yes... if green leds get too much current they glow red.

OUCH MY EYES.

Well, I bought some Blue LEDs to replace the damaged ones (and blue matches the box colour), and I put 'em in with a 680k resistor on the battery as suggested.  Now, I don't know if these are just super bright LEDs -but looking directly into them leaves little blue LEDs on your retinas.  The picture on the right is the beam of light on the ceiling with the pedal at floor level!  :shock:

But like I said, it's got the 680k on them so at least they're not gonna fry like the last ones!

These are the specs for the LEDs:
LED 5mm Blue 1500mcd

Emmitted colour: blue
Lens colour: waterclear
Wave length: 470
nm: 26
Pd W: 130nW
If mA: 30
If mA (peak): 100
Min: 3.2
Vf (V) Typ: 3.5
Max: 4.0
IV Min: 1000
MCD Typ: 1500
Viewing Angle: 20

So uh...which value resistor would bring these down to eye-safe levels?  Or are these just really bright no matter what?

Yeah, those blue LED's get pretty bright. I take advantage of that fact by sticking a 10K resistor in series, which leaves the LED plenty bright enough to indicate on or off, but drinks less current.

Is it me or did bobbletrox say he has a 680k current limiting resistor?

I'm sure that's a 680 ohm resistor .

You can just keep adding resistance to dim it down to a place you like. You only need to worry about having too little resistance. Try adding a 10K or 25K pot in series with the 680 ohm resistor and turn the dial until you get the brightness you want. Once you find that, remove the power and use a meter to find the resistance you selected with the pot. Replace the pot with a resistor close to the measured pot value and there you go.

Take care,
-Peter