Somebody may ask now "But you can also buy a yellow LED. Can we make the color white just by using the yellow and the blue LED?" Yes, we can. Prepare the yellow LED as described before and fix it into the ball together with the blue LED. After some fine adjustment of the currents you should get the white color as in the previous experiment.

At this point, signs of confusion may be noticed on some students' faces. What is the difference between the "proper" yellow light and the red-green yellow, which is really the mixture of red and green light? Are there two yellows, or only one? How can our eyes perceive both yellows as the same color shade?

As we know, the cones in the retina are responsible for color sensation in human eyes. In the simplified explanation, there are three types of cones that are most sensitive to red, green, and blue light respectively. The sensitivity curves are broad and their tails overlap. There are no "yellow cones." The sensation of yellow appears when the red and green cones are about equally stimulated. This can be done in two ways — either by the mixture of red and green light, or by a light with a wavelength that falls between the red and green where the corresponding sensitivity of red and green cones is about the same. This happens for the (proper) yellow light of the wavelength around 570 nm.

The explanation calls for more experiments that can be used either as an illustration of the phenomena, or the evidence that supports the explanation.

Following the instructions from the first paragraph, make a light ball using three yellow LEDs. Connect the yellow LEDs in a series to a 9-V battery following the instructions from the first paragraph. For my choice of LEDs,³ I used a100- variable resistor in series with 120- (0.25 W) fixed resistor. Switch on the yellow light ball and the color mixer from the previous experiment with only the red and green LEDs on. Adjust the currents in the setups so that the yellow lights emitted by both balls look as equal as possible [Fig. 4(a)].

Figure 4.

Take different acetate color filters and observe the balls through them. Try red, green, magenta, and other filters. You will notice that for most color filters the "proper" yellow ball appears to glow dimmer, or may even look like it's been switched off, while the red-green yellow ball appears to glow in different colors [Fig. 4(b-e)]. For these filters the transmittance of the yellow light is low, but it increases for longer or for shorter wavelengths making the red-green yellow ball look redder or greener, respectively.

Ask students at this point what kind of a color filter would cause the red-green yellow ball to glow dimmer than the proper yellow ball. Obviously, this would be the filter that blocks the light from the red-green ball and passes the "proper" yellow light through (a bandpass filter). But, since acetate filters are very broad, one realizes after some experimentation that there is no single filter that would meet this requirement. However, combining two filters, one that cuts wavelengths below the yellow (orange filter) and one that cuts above the yellow (green filter), may give the desired result [Fig. 4(f)]. Of course, the price we pay is low light intensity, because the transmittance of acetate filters is slowly changing with the wavelength and that causes the overlapping of the tails. That is why the exposure time for the last photo in Fig. 4 needed to be so much longer.

The described experiments and related concepts become even clearer if students can observe the emission spectra of the LED color mixer. Spectra of white, red-green yellow and "proper" yellow light from the experiments are shown in Fig. 5. The spectra have been obtained with a pocket prism spectrometer.⁴

Figure 5.