Playing with light (RGB modded DX MTE's)

Time ago I purchased a bunch of cheap seoul MTEs at DX for giving away and experimenting, this post is about the later

Replaced the seouls with luxeon III red green & blue and kept the drivers




You can see how colors blend into yellow, magenta, cyan and white!

One thing I noticed about this flashlight is that the orings aren't the proper size, the right ones are a little bigger, the same size as the E-series tailcap (parker 2-015)

My kid will be happy


Pablo

:kewlpics: Nice photo's Pablo, and an excellent practical demonstration of colour theory.:twothumbs

It's worth doing the mod for that alone.



Colin.

Nice pics. Would be interesting what kind of beam this LED would have.

light_emitting_dude said:

Nice pics. Would be interesting what kind of beam this LED would have.

Click to expand...

I would like to know that too. Unfortunately the datasheet doesn't specify which bins they use, but I've been toying with various combos of the highest bins offered for red, red/orange, amber and green rebels. One thing I learned is that if the three rebels are driven in series, the green rebel will dominate the output as it appears to be about 4-5 times as bright as the reds to the naked eye in the 100-500mA range. Red and red/orange rebels appear to be about the same brightness and appear 2-3 times as bright as the amber ones. Understand that this is all very unscientific and I am just going by the seat of the pants perception of relative brightnesses here, so please don't crucify me, but I just find it amazing how much brighter and more efficient the green rebels, and green LEDs in general, are compared to the rest of the pack.

Here I was trying to make a composite yellow made of green and red emitters. I have two clusters of 4 rebels each. Everything is driven in series at 350mA. The right cluster is made up of 4 amber rebels (LXML-PL01-0030 30lm) and the one on the left is made up of 3 reds (LXML-PD01-0040 40lm) and one green (LXML-PM01-0080 80lm). It may not be clear in this picture, but the one green rebel (third emitter from the left in the left cluster) outshines the reds and ambers combined. And it does this while it has a lower Vf (about 2.4V) than both red and amber ones (2.5-3.2V). I found through trial and error that combining one green and five red rebels produces a nice warm yellow tint, about the same tint as an incandescent 1W or so night light, maybe a liitle more yellowish and much, much brighter of course. So if anyone's interested in making a really, really bright and efficient yellow flashlight with absolutely no focus, well, here you go.

Relative brightnesses:


Reflection off a white piece of paper:

For simplicity sake, why not use amber leds?

What would be nice to try is increasing the perceived color temperature (measured as white light) of amber leds using other color leds.


Pablo

PEU said:

For simplicity sake, why not use amber leds?

Click to expand...

Amber leds in general, not just the rebels, are not very efficient and they get very hot. If you look at the first picture, the amber cluster uses slightly more power than the red/green composite but outputs a fraction (like 1/3) of the light.

PEU said:

What would be nice to try is increasing the perceived color temperature (measured as white light) of amber leds using other color leds.

Click to expand...

Yes, interesting idea. Actually I have something like that on the drawing board.

I mean decreasing temperature to make it colder, more white/blueish.


Pablo