Using red + green + blue Christmas LEDs to form white light

For a few years now I've been lighting our bedroom mainly with white Christmas lights strung across the ceiling. They've gotten pretty dim and burned out by now, and need replacing. But they seem to heat the room up quite a bit in summer, plus I'm getting somewhat bored with them, so I'd like to try something a little different.

Red, green, and blue light is supposed to stimulate the red, green, and blue cones in the eye to give the sensation of white in combination, so I thought it might be nice to use strings of those colors, rather than white strings. I think it'd be interesting to see colored lights when you look up at the ceiling, but essentially white down below, where the light is better mixed. But it seems like a bad idea to use conventional incandescent lighting for that, since each bulb produces a full spectrum of light which is then selectively filtered by the bulb color coating, resulting in dimmer light, warmer room, more electricity, sadder polar bears, etc.

It's my understanding, however, that LED Christmas lights are very efficient, more so since the colored ones only give out narrow-spectrum light. So I think I'd like to use them. But I have no idea about the relative intensities of the R, G, and B that LEDs give of, and thus how well they'll "fool" the cones. (And I see people complaining that "white" LEDs often seem more blue than white.) Yeah, I know, I could just buy three strings and check it out for myself, but before doing that I thought I'd ask whether anyone else has tried this before, or has any ideas on the subject. In particular, is there any way to get output specs for the strings, and use that somehow?

Also, does anyone know how many LED lights one can safely attach to an outlet?

And on a semi-related concept, after reading about how LED computer monitors work, the question occurs to me: would it be better for screen longevity and power consumption to use a desktop background that's all white, or all black? Or doesn't it make any difference?

Reply to
Weltanscha
Loading thread data ...

Have you considered using a compact fluorescent bulb?

;-)

Reply to
mrdarrett

If by "safely", you mean "won't burn the house down", then with a standard (US) 120V, 15A outlet, you could burn 1,800 watts of power.

Say your LEDs have an average Vf of 3V and an average current of

20mA - that's 3 * .02 = .06 watts per LED, so 30,000, if you want to sit there and solder them all up. ;-)

Nah, cancel that - you're going to lose anywhere from 10-50% of your power in the current limiting resistors, albeit series strings would help there.

Have Fun! Rich

Reply to
Rich Grise

I was keeping the story short, but I use both the Christmas lights and track lights with the compact fluorescents, both on one switch. The resulting atmosphere is homey/romantic meets nasty interrogation scene. Don't like it that way, meant to change it to two switches, never got around to it. Mostly I point the track lights so they're as unobtrusive as possible. I'll change it, someday. On the other hand, if I wait a little longer I'll be creeping up on a decade of procrastination, which is surely something to be proud of.

Reply to
Weltanscha

On Fri, 29 Dec 2006 15:41:23 -0800, Weltanscha wrote: ...

A decade??? ONE decade???????

***AMATEUR!!!!!*** ;-)

I've been meaning to join the Procrastinators' Club for going on 40 years now! ;-)

Cheers! Rich

Reply to
Rich Grise

A 35 LED set wired in series with a current limiting resistor was labeled as taking about 4.8 watts per set max so 35 LED off 120v would be about 0.04A max which is a bit higher than typical LED forward current. A 70 LED string would probably be the same but without the current limiting resistor since the total max voltage would be 140v DC.

So theorically you could have more than 300 LED light strings off a single typical 15A household outlet and not even get close to warm. Time to stock up on assorted multi-outlet adapters.

Power consuption wise: all black = all LED off. As for the advertised life cycle, the companies tended to be on the conservative side. When I bought an LCD monitor, it was claimed to about 10,000 hours for the backlight. My monitor has been on for almost 3 years 24/7 now and that is over 30,000 hours.

LED can last a lot longer than CCFL so you could probably get 7-8 years of 24/7 without a problem. By then, somehting better would come around to replace the LED monitor.

Reply to
Impmon

You have a club? No one bothered to form any locally because no one could be bothered to show up and organize a club charter.

Reply to
Impmon

Impmon wrote in news: snipped-for-privacy@4ax.com:

Was it formed Jan 2? You know, the day you try to keep your resolutions.

Puckdropper

--
Wise is the man who attempts to answer his question before asking it.

To email me directly, send a message to puckdropper (at) fastmail.fm
Reply to
Puckdropper

In article , Weltanscha wrote in part:

I have a bit of past experience indicating that to get white light, you need:

1) Really good color mixing

2) Number of green LEDs roughly twice the number of blue ones, and number of red LEDs roughly 3 times the number of blue ones. This will vary with efficiency of your particular LEDs.

If you want a warmer shade of white, use fewer blue ones, replacing them with red more than green.

So far in my experience, equal number of red, green and blue LEDs gives a roughly sky blue color.

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

Reply to
Don Klipstein

I have only made one "New Year's Resolution" in my life, and that was that I'd never make another. I've kept it for over 40 years.

--
Service to my country? Been there, Done that, and I\'ve got my DD214 to
prove it.
 Click to see the full signature
Reply to
Michael A. Terrell

Hey! Stop stealing my lines! ;-)

Cheers! Rich

Reply to
Rich Grise

Oops! I meant to write "LCD," but was stuck in the "LED" groove.

The reason I'm asking is that, according to what I read, in LCD displays there's a white backlight that's always on, passing white light through a vertically-polarizing filter, then through the LC, then through a horizontally-polarizing filter in front. If there's a voltage applied across the LC, the vertically-polarized light passes through it with polarization unchanged, thus it's blocked by the filter in front. So it seems to me at the very least, there's no power cost to having a white-wallpapered screen versus a black one. So I'm wondering about burn-in (or rather the equivalent).

- Tom

Reply to
Weltanscha

Hey, thanks, Don, that's the kind of info I'm looking for.

Just for the hell of it, I set up the background on my three-monitor computer display to show a grid of R, G, and B squares, just to see how tiny I had to make the squares to get a good blend. Not too small at all, as it turns out. On the other hand, maybe it wasn't such a good test, because it wasn't very bright at all, and not LED of course.

BTW, Richard Dawkins asks what I thoght was an interesting question in one of his books: Pure green light stimulates all three kinds of color receptors, to a greater or lesser degree. What do you suppose you'd see if you could somehow stimulate the green, and *only* the green, receptors?

- Tom

Reply to
Weltanscha

Dunno, but you might like reading this:

formatting link

Reply to
mrdarrett

In an LCD monitor, the bulk of the power is in the backlight; the state of the LCD itself (pixels on or pixels off) makes little difference. Plus, there are both flavors of pixel technology - "white" when on, and "black" when on - in common use, so it's hard to make a general statement about whether all-white screens or all-black will be the most power hungry.

For LEDs, of course - organic LEDs, or OLEDs, being the only LED technology likely to make it to the desktop - you're right: "white" means "everything on," and is the highest power consumption.

Burn-in is not (today, at least) a significant concern with LCDs. There has been in the past some problems with what was called "image sticking" in various LCD types, but modern drive methods have in most cases eliminated this (or you may still see some sticking, but it is not caused by permanent damage to the panel - i.e., it can be reversed by applying the proper drive signals).

Bob M.

Reply to
Bob Myers

The problems I had with LED color mixing include the different color ones having different light distribution patterns, causing a smeary "pattern" of varying color. Some spots are a little more red, some less red, some more green, some less green, some more blue, and some less blue. Unless you solve that, the illuminated area could look messy.

A deeper shade of green. Try being in a magenta-illuminated room or cover your eyes with magenta glasses or a diffuser illuminated with magenta light to get your red and blue receptors less sensitive and your green ones more sensitive, then see what a pure green light looks like. (Such as a spot on a wall illuminated by a 532 nm laser or a green LED behind a few layers of green "plexiglas").

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

Reply to
Don Klipstein

In article , snipped-for-privacy@gmail.com wrote in part:

The curves in the page linked from there look "off" to me. For one thing, the blue one (often called "S" for shortest wavelength one) is shown at almost half its peak at 400 nm - depending on whose determination (see below), it's more like 4-10% of peak at 400 nm. Also, they all look a little broader and more symmetric than they should be.

For what I consider better "cone fundamental" functions, go to:

formatting link

Sorry, no curves - these are text files.

There are several versions and determinations.

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

Reply to
Don Klipstein

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.