48v DC power LED (on indicator)

"Gene" schreef in bericht news:8xTCf.14784$sA3.13130@fed1read02...

Well, err... Grab a textbook on electrics basics and learn something about Ohms law, Kirchhoffs laws, power and energy.

Assuming your LED will light up enough when using about 10mA you will need a

4k7/0.5W resistor, which will become pretty hot. A 1W resistor will be better. When you want to use 20mA (more light, more heat) you'll need 2k2/1W for a minimum. This one will also become hot and live short. Use a 2k7/2W resistor to stay on the safe side.

petrus bitbyter

You're asking the resistor to dissipate 1 watt - no wonder it let the smoke out!

I find that LEDs are bright enough for me at less than 10 mA, so you could use a 4K7 0.5 watt, which would have to dissipate 0.45 watts, or you could use two or three resistors in series to spread the power dissipation around. If you do want the 23 mA LED current your 2K provided, you could use three 680 ohm 0.5 watt in series - each would have to dissipate about 1/3 watt.

--
Peter Bennett, VE7CEI  
peterbb4 (at) interchange.ubc.ca  
new newsgroup users info : http://vancouver-webpages.com/nnq
GPS and NMEA info: http://vancouver-webpages.com/peter
Vancouver Power Squadron: http://vancouver.powersquadron.ca

Or you could put several smaller wattage resistors in series to spread the dissipation between them. For example, a string of four 500 Ohm,

1/4 Watt resistors in series has the same equivalent resistance and power dissipation rating as a single 2 KOhm 1 Watt resistor but takes up a bit more space.

--
James T. White

Use a low-current LED - these produce decent brightness at 1-2mA which will keep resistor dissipation a lot lower, but it will still get warm.

Of course it cooked up real nice. Obviously, you didn't do the math!!! Ohm's law still rules. Your 2K resistor is dissipating a bit over a full watt.. what did you expect it to do? Increase the resistor value to 4700 ohms @ 1 watt; then it will dissipate less that 1/2 watt, and should still provide adequate intensity from the LED.

--
Dave M
MasonDG44 at comcast dot net  (Just substitute the appropriate characters in 
the address)

Never take a laxative and a sleeping pill at the same time!!

In , P. bitbyter said: (edited by me for space)

It's easy enough to get LEDs that get bright enough for most indicator lamp applications at a lot less than 10 milliamps.

Plenty of the various "superbright" and "ultrabright" red LEDs should be bright enough and possibly more at 2 to 4 milliamps, even considering the way they lose efficiency at low currents. Just make sure the "viewing angle" is adequate. Most diffuse wider viewing angle 3 and 6 mm LEDs have a rated viewing angle of 60 degrees. If you can only find ones rated 45 degrees that you like, I expect chances are they will be adequate - especially if you have a dispersion lens or diffuser or a light pipe over the LED. Look for GaAlAsP, Agilent's similar "AlGaAs" and InGaAlP chemistries as clues for the LEDs being "superbright"/"ultrabright" or the like.

As for green - GOOD GRIEF!!! Look for InGaN chemistry or for wavelengths in the 520's or 530's of nm. One nice thing about those - at least in my experience, these have efficiency maximized by currents of only a few milliamps. (Exception of ones big enough to normally require heatsinking, and having "characterization current" 150 milliamps or more.) Heck, many of these could be "bright enough" at 1 milliamp!

And if you want a yellowish shade of green like that of "old tech" green LEDs - InGaN green ones mostly meet you halfway when you give them low current. These are among the LED chemistry/color types that have color varying somewhat with current. Most InGaN green LEDs at rated current produce a "very slightly whitish pure green". But at low current, most of these are more of a "lime green" to "yellowish side of lime green", although not as yellowish as "old tech" green LEDs.

Most blue and white LEDs do well at only a few milliamps, often as little as 2 milliamps. Nearly all white ones use InGaN chips, known for efficiency that is high and which I know to generally have efficiency maximized at fairly low currents (a few milliamps when maximum current is

30 mA and "characterization current" is 20 mA). Most blue LEDs are InGaN also, with GaN being second-place and still likely to be "bright enough" at 5 milliamps or maybe less. (GaN LEDs still in production mostly have peak wavelength about 428 or 430 nm and "dominant wavelength" about 465 nm, while most InGaN blue LEDs of "indicator lamp size" usually or at least often have wavelength specification(s) of/around 470 nm.)

For really good white, green and blue LEDs go to the website of Nichia, who makes better ones. They sell direct - minimum order 100 pieces, last time I checked close to $80 plus shipping, still about $90.

For lower cost production size orders, there are competitors with good ones - as an example, ETG - a manufacturer of LEDs that have chips by Cree.

But if you want to spend far less than $90, then get a Digi-Key catalog.

If you only need one or two or 3 or 4 LEDs, go to Radio Shack. Although none of their LEDs are efficiency champions, they have some that are efficient enough to get plenty bright at just a few milliamps. Many of their more efficient ones have narrower beams ("viewing angles") and if you need a wider beam on a budget that low, then get some fine sandpaper and grind off about a millimeter (.04 inch) from the tip of a narrowbeam LED and then re-round the tip. Radio Shack LEDs that I have noted as good at 2-5 milliamps -

*conservatively*, as in being no less bright at such lower currents than "regular brightness" LEDs are at 20 mA, and that is after doing the above "sandpaper treatment": 276-307 (red - narrow beam, and very efficient for "pure red") 276-066 (maybe discontinued) - similar slightly dimmer but still "ultrabright"

In my experience, these GaAlAsP red ones are good at 2 milliamps.

The ones below *could* outshine these red ones, but I have yet to test them for maintaining such high efficiency all the way down to 2 milliamps, although they may.

276-309 - a wider beam GaAlAsP red of similar efficiency - should be good at about 2 milliamps. 276-306 - an InGaAlP reddish orange one good at about 3-4 milliamps. 276-301 and 275-351 - InGaAlP yellow or orangish yellow, beam may be narrowish, should be good at 5, maybe 3-4 milliamps. 276-316 - InGaN blue and in my experience well short of "good InGaN blue" in brightness and efficiency, but still plenty bright at 2.5-4 mA, maybe somewhat less. This is a narrower beam one. 276-311 - Has a Cree GaN chip, slightly whitish very slightly violetish blue, narrower beam one, but still good at 5 milliamps. 276-320 - white with InGaN chip, good at 3-4 milliamps or so.

- Don Klipstein ( snipped-for-privacy@misty.com)

Very good information Don though I doubt the OP can do much with it due to lack of even basic knowledge of electronics. Nevertheless I appreciate your efforts.

petrus bitbyter