Light frequencies and the human eye

[j0h4n]

If I have a blue and red LED (or theoretical light sources with pure blue and red light frequencies), and illuminate a white surface on the same spot, I seed pinkish purple light reflected. Is the reflected light of the same frequency as light from a light source with pure pink light frequency, or is the light a mixture of blue and red light. I believe it is a mixture, but I don't understand why a mixture of two lights could appear (to a human) the exactly same color as a completely different light frequency, in this case blue+red compared to pink.

 

A related question; I have powerful LED light sources with blue and red light, which light up the entire room, and to my surprise some green surfaces appear eerily fluorescent and bright. How is this possible if no green light is emitted from the light sources (which should be the case as I'm pretty sure the LEDs have monochromatic light).

Edited June 4, 2009 by j0h4n
edited title to take away reference to biology

[Klaynos]

As long as the surface does not floresce, then the light reflected can ONLY be the two frequencies that are incident on it. They eye and brain then interpret the signal and give you colour information.

 

Humans only have 3 colour receptor types, they each have different peak detection frequencies, and over lap the others significantly. The different intensities each of these detects is used to work out the colour. It's not a physics effect, but a biological one...

 

Without knowing the nature of the things giving off the green light, or the spectra from the LED's I can't really comment. I suspect it's florescence as opposed to reflection.

[Mokele]

Humans only have 3 colour receptor types, they each have different peak detection frequencies, and over lap the others significantly. The different intensities each of these detects is used to work out the colour. It's not a physics effect, but a biological one...

 

Precisely. It's also why you need at a minimum of two different photoreceptors to see color - with only one, you can't tell if reduced stimulus is due to different wavelength or just reduced light levels.

[McCrunchy]

Hi,

 

Interesting question. Complementing on Klaynos's answer, taking a try at color theory:

 

- http://www.normankoren.com/Human_spectral_sensitivity_small.jpg

 

shows a graph of the spectral sensitivity of the three (blue, green and red) cones in your retina. One sees that a red and blue mixture triggers all three detectors (blue the beta, red the gamma and rho). On the other hand, the sensation is equivalent to pure 400 nm purple, which would only trigger the beta detector. Putting that in equation gives :

 

c1 beta + c2 gamma + c3 rho = c4 beta

 

Where c1, c2 and c3 are the output coefficients of your three cones for the red+ blue mixture and c4 the one for the pure purple.

If one considers that c4 >c1, one gets

 

c2 gamma + c3 rho = (c4-c1) beta

 

with all coefficients positive. So a mixture of signals from gamma and rho can be equivalent to that of the third one (beta) ... this doesn't make sense, so we should have c4 < c1.

 

Then the coefficient up there on the right (c4-c1) is negative, and I don't know what this means.

 

I'll check back once I figure that out ...

 

In any case, you'll maybe want to check out the name of Ewald Hering, who's the founder of modern colour theory.

 

- http://mysite.verizon.net/vzeoacw1/coloradd.html

 

has a nice applet that enables you to see the color resulting from 3 different primary colours.

 

Best,

 

McCrunchy

[j0h4n]

As long as the surface does not floresce, then the light reflected can ONLY be the two frequencies that are incident on it. They eye and brain then interpret the signal and give you colour information.

 

Thanks, this verified what I thought too.

 

Without knowing the nature of the things giving off the green light, or the spectra from the LED's I can't really comment. I suspect it's florescence as opposed to reflection.

 

I looked up fluorescence, and basically it's light being reflected with a longer wavelength. So it makes sense that it's green fluorescent light coming off my computer that I've painter with green fluorescent paint, as green is near blue in spectrum and has longer wavelength. Also some of my collection of beer cans light up in orange fluorescent light, and that must be from the blue led too, as orange has a longer wavelength than red and fluorescence isn't supposed to be light reflected with shorter wavelength.

---

Merged post follows:

Consecutive posts merged

I understand color theory in practical applications (thanks McCrunchy for interesting links), but this color mixing and interpretation is still a mystery to me There must be some memory effect in the eye, as an example when red and blue light hits the eye in rapid succession (e.g. a computer screen flashing red and blue alternately with 60Hz), then it looks purple. Basically that's what happens with the LED's too, the photons from the differently colored LEDs don't enter the eye simultaneously, but with a few attoseconds (?) between. The eye and brain don't seem to care about this several orders of magnitude difference in time. And apparently it is possible for light of different with different wavelength hitting the corona on the same spot, to cause the same effect as one "pure" light beam. Fair enough, maybe understanding this is still beyond human's scientific understanding. But as this seems to get into biology, now that the mixing question was solved, maybe this isn't the proper forum for it.

 

Another interesting topic could be how digital cameras can make this same distinction (or lack of distinction) between a pure pink beam and one blue and red beam hitting the same spot on the sensor..

Edited June 6, 2009 by j0h4n
Consecutive posts merged.

[Bone Daddeo]

LEDs are not really monochromatic. Pick up a pair of diffraction grating glasses - cheap paper ones with plastic diffraction grating lens are cheap, work great, tons of fun and rather addictive and you will embarrass your wife if you wear them into a restaurant - and see for yourself. While preparing a unit on light, I looked at a whole bunch of different LEDs I had kicking around the old Laboratory for the Study of the Reanimation of Life here in Confusion and kept myself amused for longer than I would care to admit.

 

It's not unusual to have more than 1 color LED in a single package - especially in the water clear packages - whose light output is mixed to give a particular color of light. High quality white LEDs are made of a red, blue and green leds in a single, 2 lead package. You can even buy 4 lead devices (common ground) that let you mix the different LEDs by varying the amount of current driving each color.

[Ndi]

But as this seems to get into biology

 

It doesn't have to, the color is still purple when captured by a camera sensor, since it still has R, G and B filtered light ADCs. Advantage being that with a recorded image one can objectively measure resultant color, and latency, persistence, etc are fully known and understood. We have the specs.

 

Bone Daddeo:

Some LEDs are - Laser diodes (Laser LEDs) are monochromatic and coherent - thus the beautiful patter.... ooooohhhh...

[Bone Daddeo]

There are indeed laser diodes - injection diodes - and while they are a class of LED, they are truly a horse of a different color. A similar comparison would be a gas discharge lamp - fluorescent light tube - and a gas discharge laser.