LED's as Light Detectors

Tanya M. Sansosti
Stony Brook University
Laser Teaching Center
Optics Rotation Project 2, Spring 2002

Advisor: Dr. John Noé

Why This Project?


Given my strong interest in optics and atmospheric sciences, for my second Optics Rotation project Dr. Noé and I were originally considering doing a project which focused on how a laser beam might behave if you shine it through various fluid media that scatter light. However, we came across a website that involved using an LED as a light detector rather than a light source to estimate the amount of haze present in the atmosphere. Naturally, I was immediately intrigued to have the opportunity to work on an atmospheric optics project.

Haze-Span website

The Haze-Span website, intended for amateur scientists and students, is run by Dr. Forrest M. Mims III. It offers detailed instructions on how to build a home-made sun photometer and use it for understanding haze in the atmosphere. For my purposes, understanding what an LED is, how it can be used to detect light rather than emit it, and which wavelengths are best detected, were more important than anything else.

What is an LED?


"A light-emitting diode is a semiconductor device which emits light when an electrical current flows through it. Semiconductor diodes can also be used as light detectors. Light-emitting diodes are good detectors of light of specific colors. Generally, the wavelength of light detected by an LED is shorter than the wavelength of light emitted by the same LED. For example, certain red LEDs are relatively good detectors of orange light." [1]

Light Emitting Diode

Using an LED to Detect Sunlight


LEDs are designed to make, not detect, light. But it happens that they detect light quite well. Even better, unlike detectors designed specifically to detect light, LEDs detect a relatively narrow band of wavelengths. For example, an LED that emits greenish-yellow light at a peak wavelength of about 555 nanometers (nm) detects green light at a peak wavelength of about 525 nm over a spectral width of about 50 nm. [1]

For the haze detector: light from the Sun causes the LED detector to generate a tiny electrical current much like a solar cell does. This current goes to the operational amplifier, which is connected so that the LED current is transformed to a voltage and boosted by the amount of the feedback resistance in ohms.

haze detector

At the Haze-Span website, Dr. Mims uses the LED to build a sun photometer. This is a device which is designed to measure the intensity direct sunlight penetrating the atmosphere. It can be used to measure haze since this blocks some direct sunlight and the photometer is not designed to measure diffuse or scattered light in the atmosphere. So, basically, this device designed by Forrest Mims is an inexpensive, efficient tool, practical for the amateur scientist, that can detect how much haze exists in the atmosphere.

Using an LED to Detect Laser Light


Rather than building a copy of the sun photometer, I decided to start by studying the wavelength response of colored LED's using a monochromator as a source of colored light. The first step was to understand more clearly how the instrument worked and to calibrate it. Here is some of what I saw:

The above images show the monochromator and it's interior workings. There is a rotating diffraction grating inside whose motion is controlled by the dial and the angle is determined by the reading on the dial. According to a plate on the front of the instrument, the grating has 675 lines/mm. There is also a mirror, which can be seen to the right in the third picture. The mirror is facing downward to reflect the light source inside which can be better viewed in the following three images:

Top View of Light Source Light Source with
Light On Light Source with Light Off

Once I had a better understanding of what I'd be working with, my next task was to calibrate the instrument. Here is a list of the steps I took:

1) I used a protractor to determine the angle of the diffraction grating for several different dial settings. For one end of the dial (R=0.7) the angle was estimated to be 10 degrees, and for the other end of the dial (R=1.6), the angle was about 30 degrees. The mechanism inside is set up in such a way that the angle is a linear function of the dial reading.

2) Next, I set up a spreadsheet that calculated the wavelength of light passing through the instrument as a function of the dial reading, using the grating equation: n * lambda = d * sin (theta), where n is the order of diffraction, lambda is the wavelength, and d is the grating interval. This gave a prediction for lambda versus dial reading R that was almost a straight line.

3) The two calibration constants that fix the predicted curve were adjusted slightly to make yellow light in first and second order agree with observations. Yellow was picked because yellow is a specific color to match, and wavelengths shorter than yellow were off the dial in first order.

4) Next I measured the dial reading for four specific colors in second order. For red, we compared the color with a helium/neon laser (633 nm). For yellow, we compared it with a sodium lamp (589.5 nm), and for green, we used a green laser (532 nm) in the lab. For blue, unfortunately, we had no other tool to use but our unaided eye, and so visually determined the wavelength to be 474 nm, which is the average wavelength for the blue range (455 nm - 492 nm).

5) The four data points were found to be very close to the predicted calibration curve. A further very small adjustment in the calibration constants brought data and curve into exact agreement as shown in the plot. The plot shows first, second and third order curves. The red line is an exact calculation of the grating equation while the blue lines are a linear approximation, which can be seen to be quite accurate.

Calibration Curve

The LED and My Results


Under the recommendation of Dr. Mims, The LED that I used to make measurements of laser and sunlight was shaved, or flattened, to a point where there was no longer a rounded edge. He suggested doing this with a piece of coarse sand paper.

Next, I had to create a set up for the LED and monochromator where a movable lens could be placed in between the light source and the LED. For me, proved to be a bit of a challenge itself. I have displayed: a movable monochromator facing the LED edge on with an adjustable lens in the middle. Off to the side is an amplifier that is reading the current off of the LED.

Flattened LED Ampllifier

View from LED Close-Up View from

Using a Radio Shack 276-022A green LED, I was able to produce the following curve:

I was also able to take some reading with the LED outside, under sunlight. For these measurements, I used different color LED's to compare what different readings I might get. Many of the LED's were not flattened on the end which made it difficult to get a steady reading.

Outside, using different color LED's with unflattened edges, we got the following readings on a voltmeter with *magnified* sunlight. The sun was quite bright even though it was late afternoon (about 5 PM) because the sky was very clear that day.

red: ~.1 microamps
yellow: ~ 330 microamps
green: ~ 365 microamps

And for using different color LED's with just indirect sunlight (no magnifier):


red: ~.1 microamps
yellow: ~ 4 microamps
green: ~ 6.6 microamps

I also took some measurements with the flattened, polished green LED outside under sunlight without a magnifying glass and got readings that ranged from 1.4 microamps to 1.9 microamps, depending on the angle the sunlight was hitting the LED. These readings, of course, were much more steady since the LED had a flattened tip.



I would sincerely like to thank Dr. John Noé for his extreme patience and kindness in working with me on this project. He has been of great support. My deepest gratitude to him.

I would also like to thank Professor Bob DeZafra for his input and support on this project.

And finally, I would like to thank Professor Harold Metcalf for giving me the motivation and courage to sign on for a class that I had such little background in. Truly, thank you.



[1] Haze-Span

[2] Britney's Lasers

[3] What is an LED?

[4] When Hazy Skies Are Rising

[5] Mims III, Forrest M. "Sun Photometer with Light-emitting diodes as spectrally selective detectors" Applied Optics. Vol. 31, No. 33. November 20, 1992.

[6] Mims III, Forrest M. "The Amateur Scientist: How to Monitor Ultra-Violet Radiation form the sun" Scientific American. August 1990.

[7] Morys, Marian; Mims III, Forrest M.; Anderson, Stanley E. "Design, Calibration and Performance of Microtops II Hand-Held Ozonometer" Presented at the 12th International Congress on Photobiology, Vienna, September 1996.


Created by:
Tanya M. Sansosti