This page is a collection of my observations with the Project STAR Spectrometer. I started this in Dec. 2001, and have slowly added to it. (last updated Nov 4, 2007)

What is the Project STAR Spectrometer?
The spectrometer is a very simple yet useful device made from cardboard or plastic (I have the plastic one) containing a diffraction grating, a strip of transparency film, and a lens. You are looking at spectrographic images taken using the spectrometer, the scales are on the transparency film, and the rainbow is caused by the diffraction grating. For more, click here.

Three very different examples of fluorescent lights. All of them show strong emission due to mercury. The first is a compact fluorescent (coiled type):

A more ordinary long tube fluorescent light. Appears to have the same spectrum as the compact version:

But a seemingly identical long tube fluorescent light has a very different spectrum! I think the first two are "warm white" lights, because they have a slightly pink/orange tint compared to normal lights, like this one:

Ever hear of "full-spectrum" lighting? I bought a couple of these and my wife and I immediately noticed that the color was weird. Not something we wanted in our house.
Naturally, I had to find out why:
(jan 2, 2007)
Hmmm... seems like the secret is revealed. What happened to the 570nm - 590nm band? People in the lighting industry who know way more than I do claim that by notching out these wavelengths a better "quality of light" is produced. Maybe. But it ain't full spectrum!

The color TV above uses an old-fashioned CRT. We recently replaced it with an LCD high-def. The LCD spectrum is identical to the compact fluorescent lamp, which tells us which light source back-illuminates the LCD!

The nice little sodium line is very strong visually, but unfortunately lies right in the gap between the camera's red and green color filters!
A better exposure might show it.

Sometimes you have to travel to get what you want:

..And that day has finally arrived. The main difference is a new digital camera, a Canon Powershot A540! Watch as I slooowly update all my shots on this page using this camera.
(Dec 2006)

(Nov 2, 2007: Note improvement in resolution! Thanks Registax!)

Monochrome webcam image showing spectral lines in near IR, to the left of 700nm. Apologies for the low resolution. B&W camera has no IR filter and no color problems! Image taken through double-paned window glass, which I suspect causes the large absorption line near 750nm.

More experiments: Color image turned black&white, then sliced and plotted using Iris. The plot says more about the camera sensitivity than it does about the Sun, but you can see the absorption lines and get a sense of their relative strengths.

LEDs

(jan 2, 2007)
Warm White LED Light Bulb (array of phosphor-coated blue LEDs and plain red LEDs. Note blue spike at 460nm from blue LED, yellow phosphor (red+green bands), and red LED at 630 nm). The bulb has a color temperature of 2700K to 3000K.

(fall 2006)
LED flashlight (Nichia). Note the blue spike again, and the lack of a spike in the red. This flashlight looks bluish white, typical of phosphor-coated LEDs. This bulb probably has a color temperature of around 6500K.

(Jan 7, 2007)
This light bulb contains an array of about 64 LEDs with roughly an equal mix of red, green, and blue chips. It looks good, but has the same weirdness as the GE Reveal due to the very pronounced gap between red and green. It made our fresh-cut butternut squash turn from yellow to bright orange!

Spectrum of exposed ion tube. The fluorescent line at 633nm is the one that lases, but some other lines can be made to lase too.

As seen in Sky and Telescope, Feb 2000, Page 77.

Real Science: Spectral comparisons of 5 light pollution filters!