LED currents?

how about PWM it?

Use a wide-viewing-angle diffused-lens LED, for an overall pleasant useful appearance.

Yeah LEDs are very efficient these days, almost blinded myself with a stupid SMT red LED on one of my projects. It was just supposed to be a power LED, not an optical warfare component...

For "blinky light" indicators, I use 1k per volt Vcc. So, 3.3k for

That's less than 1mA current (due to the Vf drop), but it's enough to easily tell that the LED is on without casting that warm glow throughout the room.

Example: for the average green LED (2.0V Vf) at 3.3v: (3.3 - 2.0) =

There is quite a range of "lumens" for LEDs, I like to buy the brighter ones because I can drive them directly from GPIO ports without the extra transistor.

If the green LED is one of those InGaAlP ones, half a milliamp is probably about right for "normal LED indicator lamp" brightness. At that current, voltage drop is about 2.9 volts, so the dropping resistor would be about 3.9K or 4.7K or so.

I have had some better ones work just fine at .3 milliamp.

InGaAlP green LEDs tend to have nominal wavelengths around 520-530 nm. Their color varies a little with current, and at half a milliamp expect a pretty lime green or greenish lime green color.

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

Does anyone know what those super-bright LEDs in the Harborfreight head lamp can stomach over the long haul? There is no part number on them.

Gosh, I don't think I've *ever* seen an LED (singles, not a special type) with a part number on it.

And those that do have numbers are hard (impossible?) to find references to, last I tried.

Tim

-- Deep Fryer: A very philosophical monk. Website @

Just test it. But I would wonder if they last longer than 100 hours.

- Henry

Most "regular size" white LEDs are characterized at 20 mA, and I hear a few life figures of 6,000 and 10,000 hours, though sometimes 50,000 hours for some.

I have seen some with a halflife appearing to me about 4,000 hours at 30 mA.

If you don't mind faster aging, my experience is that most don't mind 40 mA too much and you have a fair chance of getting away with 50, especially with some heatsinking on the negative leads (such as large copper areas in a circuit board).

Keep in mind that the usual white LEDs are nonlinear, with maximum efficiency when moderately underpowered (maybe 5 mA or so for "regular size" / "low power" ones).

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

On a sunny day (Wed, 26 Dec 2007 14:57:15 -0800 (PST)) it happened snipped-for-privacy@netzero.com wrote in :

LOL

On a sunny day (Wed, 26 Dec 2007 17:51:23 -0500) it happened Jamie wrote in :

Yes that would work, but I need timers for other things too... And that would add a user control. Not a bad idea... maybe add a photo cell and measure room light.... But it defeats simplicity :-)

On a sunny day (Wed, 26 Dec 2007 14:52:18 -0800 (PST)) it happened Winfield wrote in :

Thank you Winfield, I will have a go at some catalogs.

On a sunny day (26 Dec 2007 18:23:26 -0500) it happened DJ Delorie wrote in :

Yes, looks like I will have to go to lower currents too.

On a sunny day (Wed, 26 Dec 2007 23:39:59 +0000 (UTC)) it happened snipped-for-privacy@manx.misty.com (Don Klipstein) wrote in :

Thank you Don, that confirms also what DJ Delory suggests I use. Yes I think these LEDs are those InGaAlP, they are a different green then some older ones I have, I will play with the current a bit to see if the color changes.

Well, that's why I was asking since they might drive them a bit past their usual limits :-)

Thanks, Don, that's good information. So I'll keep them to 20mA then.

In the long run they will ALL go dead.

Best regards, Spehro Pefhany

Why is that? Sloppy processes or something like that?

As a kid we could only buy cheap non-diffused LEDs. So I roughed them up a bit with a sand paper sponge. Others dipped them into household chemicals to etch them but that was too dicey for me. Of course you guys also have a much better diffusor for the receiving end: Beerenburger ;-)