LED Colours?

[emilyandmiles]

Am i right in thinking you can only get certain coloured LED's?

Any one got a link to some purple or pink ones?

Or a list of available colours?

[tiges_ tendres]

you can get violet ones, but they are semi-ultra violet.

http://www.smallbearelec.com  has a great range of leds in all sizes and colours

[niftydog]

I've gotten pink LEDs from Oatley Electronics in australia, no idea about their UV content, but I use 'em anyway!

[stm]

Today you can fin LEDs in basically any imaginable color. You just need to search for the proper manufacturer/distributor and you can get:

1) Pure Colors (wavelength range is approximate):

Infrared (over 700 nm)
Red (700-640 nm)
Orange-Red (630-620 nm)
Orange (610-600 nm)
Amber (590 nm)
Yellow (580-570 nm)
Yellow-Green (560-550 nm)  *this the traditional "puke green" LED
Green (540-520 nm)
Cyan (510-490 nm)
Blue (460-480 nm)
Royal Blue (440-450 nm)
Violet-Blue (420-430 nm)
Violet (390-410 nm) *this LED emits some UV-A radiation as well
UV (below 390 nm)

2) Combined colors:

Purple (not violet!) *made with a blue LED that excites a RED fluorescent element (blue+little red = purple)
Pink *made with a blue LED that excites a bright RED fluorescent element (blue+intense red= pink/magenta)

3) White LEDs: *these are made by a blue LED that excites a fluorescent that emits red/orange/yellow light. The proportion of the main blue emission to the scondary fluorescene determines the degree of "whiteness", AKA as color temperature.

Warm White (2700-3200ºK) *looks yellowish to orangish, matches the light of incandescent lamps
Daylight White (5500-6500ºK) *similar but not equivalent to the sun's light
Cool White (over 6500ºK, up to 10000ºK) *looks slightly bluish, like some fluorescents and the bright lit sky
Gold (!?) *apparently a variation on the warm white LED (I've seen this on one manufacturer's site only)


As you can see, LEDs are subject I've studied and applied with passion.

Check www.ledmuseum.org for additional information.

[Nasse]

Then there is RGB (red, green, blue) combination led, that has three or more colors in same pack, but it has more legs and needs some control circuit if you want to change color. But it can do almost any color

[MartyB]

Here's an example:

http://aronnelson.com/gallery/MartyBs/rgbrnge?full=1

[ninoman123]

I have some 12mm pink LEDs that I used for a recent amp build. They look amazing but after keeping them on for about 30 minutes the color dulls into a blue purple-ish. After I turn the amp off and leave it off for like 30 minutes or so the color will be pink again when turned on. Its odd and even when run under the recommended current it still does it. Oh well I still like it.

[Paul Perry (Frostwave)]

That site stm mentioned has the full dope on pink leds: http://ledmuseum.candlepower.us/ledpink.htm
I imagine the leds that are temporary (but return to pink the next day) have an organic dye which bleaches out (changes into something else) and then reforms overnight. There are plenty of light sensitive chemicals that do that.
Also as you can see from the link above, there are "single use" pink leds also.

[stm]

Well, your LEDs have an integrated hour-meter.

Consider yourself lucky that pink LEDs retain their color when power is cycled. On the Led Museum link I posted above the guy tested some pink LEDs that vanished their pink colorfor good after less than an hour of usage, turning light-bluish (or "baby blue", as described).  These were like the first attempts at pink LEDs, and the weakness lies in the use of organic pigments for the red part of the spectrum which degrade quickly under the strong flux emitted by the LED.

I think a much more "robust" pink tint can be achieved by multicolor LEDs. Mixing red and blue in the right proportions will let you change from Purple to Magenta. If you also add a slight amount of green you will turn into Pink.  % mm LEDs are readily available to do this.  I think I've seen some 10mm multicolor LEDs as well.  One last thing is that these LEDs mix colors better if the surface is slightly dulled with a 3M scrub.

[stm]

Nice theory this of the regeneration. It is one possibility.

The other possibility I see may have to do with temperature. Even if an LED is just warm on the outside it may reach well above 70ºC (160ºF) on the inside when running at nominal current (20mA?) for a sustained period.  If you like you could experiment by putting a "vanished" LED in the fridge for some minutes to see if the recovery is faster. Also you could pre-heat it in torrid water before using it to see if it starts vanished. At least this is what I would do due to my curious nature! Of course it won't solve your problem, but at least provide some insight.