How to measure the Intensity of Light?

HungryChemist

While I and my friends were talking about some topics in optics, we come to realize that we do not know how to measure the the intensity of light. Does anybody have a good idea? I don't want to hear 'Oh, You use blah blah to measure the Intensity of light' but rather what's being directly measured. Since the Intensity of light relates to Power and the Radius. One should measured the radius. How should one measure the power given off by the light source?

ZapperZ

Unfortunately, you will have to hear "blah blah" and more. I don't now about you, but I measure light intensity using a photodiode. It is a PN junction that is reverse biased [I could have sworn we had a thread on this a few weeks ago]. A photon (within the working range of the diode) hitting the diode will cause a promotion of a charge carrier into the conduction band and thereby increasing the current. So the larger the current in the circuit, the higher the number of of photons hitting the diode.

This will only make sense in terms of the connection to light intensity if you understand that light intensity is really a measure of the number of photons hitting an area per unit time. Such a concept may or may not have been covered in your classical optics class.

Zz.

Claude Bile

ZapperZ, I've found that PIN photodiodes are good for measuring powers, but not so good for measuring Intensities, unless the intensity profile is relatively well known.

An array of CCDs (Or Charge Coupled Devices) such as those found in digital cameras work very well, they can measure intensity profiles to a good resolution (about 30-40 microns). Even better is SNOM (Scanning Near-field Optical Microscopy), which can acheive sub-micron resolutions, although you do make a number of significant trade-offs such as a reduction in detection area and signal to noise ratio.

HungryChemist, Intensity (or more correctly, Irradiance) has units Watts per metre squared. That is, it is a power divided by an area. To measure intensity, you need to be able to measure power and an area. Measuring Intensity is easy if the intensity is relatively constant over the area you are measuring over. PIN photodiodes that ZapperZ mentioned will measure Power, while it is easy enough to measure area.

IF the intensity varies which is usually the case, then you can do either one of two things;

- Measure the total power over the total area to get an average intensity.
- Use an array of detectors to measure the power at a number of points, then reconstruct those points into a 1D or 2D intensity 'map' (similar to a contour map). PIN photodiodes are generally a bit large for this purpose, so an array of CCDs would be more suitable (or SNOM if you are looking for hyperfine resolution).

Claude.

rbj

here's my curiousity (being an electrical engineer): how do you calibrate those detectors so you know what indication corresponds to some standard intensity of some given watts/m^2? how do you measure absolute intensity of light? (i can think of using some large black tank with water and measuring the rate of rise in temperature.)

r b-j

HungryChemist

Thank you for your reply but I still have further question. What you said about measuring the light intensity using a photodiode sounds alot like 'photoelectric effect' where you increase the light intensity you measure the increase in photoelectric current. Imagine, that I happened to be the person who's task was to find out the relationship b/t the photoelectric current and the light intensity. So naturally, I will have to know when I vary the intensity of the light, how much of intensity of light I am changing. How would I do that? I guess, what I am ultimately asking is 'how did the people back in times of Newton measured the intensity of candle light?' Is my question making any sense???? -_-;

ZapperZ

It's similar, but not identical. In a typical photoelectric effect experiment, you liberate the electron into vacuum. Here, you don't. It stays in the semiconductor and becomes a current. Furthermore, depending on the doping type and value, and the gradient of the reverse bias, you can change the field gradient between the PN junction - you can't do much in changing the work function of a metal in a photoelectric effect.

Not sure I understand this. Can't you just read off how much current has changed from your photodiode?

I don't know how they did it back then. From classical wave theory, the "intensity" is a measure of the amplitude of the E-field oscillation. So I suppose one has a pick-up antenna and measure the signal strength from it, something we still do today.

Zz.

mezarashi

Hi, I'm new here. Been surfing around SF.net, but just bumped into this forum, and I'll hope to be gaining alot from my time here. I'm majoring in Electrical and Electronic Engineering in college. Nice to meet you all ^_^

As far as I know, you can't measure an "absolute intensity" of light. There can be an absolute absence of light (i.e. not even 1 photon), but where would one set a limit for a max value? And when our units like Watts/area comes into play, there is possibly no other way but to calibrate it with a known reference. You can probably derive a way to do so using fundamental physics with say a 60W light bulb. Usually measurements are only important in a relative sense anyway.

mezarashi

Back in the times of Newton, there wasn't much you could do to study light intensity. Most of it was done subjectively, visually. I'm sure you know the term "lux". They would compare things to candles. The first objective devices came as our study of astronomy advanced, where they started finding a relationship between the apparent diameter of a star and its intensity. Later on when photography was invented, it came to analyzing photographs. Here's a nice link for you to read up on

http://www.astro.virginia.edu/~afs5z/photometry.html

The detection of light and use in electrical systems is a relatively new branch of physics/engineering known as optics and in engineering photonics. The photodetector and photodiodes are the engineer's most invaluable tools in this field.

rbj

how is it, then, that they tell us that there is about 1360 watts/m^2 of solar radiation at our distance of 93,000,000 miles from the sun (and from that they calculate that the output of the sun is about 3.8 x 10^26 watts)? how do they measure that? i think there is some way to calibrate photosensors, but i dunno what it is.

r b-j

HungryChemist

I don't think I can just read off how much current has changed from your photodiode. What you're trying to show(as a demonstrator of the photoelectric effect) is that as you increase the intensity of the light, you want to show that the photoelectric current also increase by some factor. But, don't you already have to know how much you have increased the intensity of the light in order to make fair judgement?

You can't just say 'oh since, our photocurrent doubled, we must have doubled the intensity of light and therefore as I doubled the intensity of light doubled the photo electriccurrent doubles. So, from now on we will just read photoelectric current to measure our intensity of light??

mezarashi

I did not mention that photodiodes cannot be calibrated. I was assuming you had a home made photo sensing device and wanted to calibrate it. Manufactured photodiodes are undoubtedly pre-calibrated. Calibration is when you set two points of known values.

Lights off - zero intensity: photodiode produces a amount of current - calibrated as 0
Lights on - 100W light bulb: photodiode produces b amount of current - calibrated as x (where x is the known reference intensity - you can use physics to figure out what x is)


For the most part, what you do is increase the power of the light source. Say originally you dissipated 30W of energy into the light source, now you dissipate 60W into it. Electric power can easily be measured. Now your intensity is known. Intensity is a function of power and distance right.

rbj

i didn't say you did. i am asking how they are calibrated.

but the efficiency of the light bulb is not easily measured (unless you already have an accurately calibrated photo sensor). how much of that 30W or 60W or 100W are dissapated as heat? (okay, first it's emitted as infared, but how much of that makes it to the photocell or photodiode?)

nope. that's what we're trying to measure. no circular logic is allowed.

yup.

r b-j

ZapperZ

You can't? Why not? If you're calibrated it to some level, why can't this be true?

Zz.

GENIERE

The practical way to calibrate is to obtain a source from a company that specializes in producing the a particular standard that is traceable to the international standards lab (CIC? I forget). You would obtain a source in the spectral range you are interested in. These sources produce a shower of photons, not the individual photon that I think HungaryChemist is interested in.

I think this thread would do better in the Stellar Astrophysics forum. I believe the astronomers have techniques to count individual photons, an absolute calibration.

ZapperZ

They do??!!!!

.. and what would this be?

Zz.

HungryChemist

Because, in my imaginary situation, the ultimate question I am asking (in the process of demonstrating photoelectric effect) is "If I doulbe the intensity of light how much of that photocurrent will change?" So, even before I look and measure how much the photocurrent change, I have to know precisely that I have indeed doubled the intensity of light.

GENIERE

http://www.pparc.ac.uk/frontiers/arc...&style=feature

“The first practical device
In practice these devices are not particularly simple to manufacture, and it took 3 years from our first ideas to our first laboratory measurements, and a further 3 years for the development of a small ‘camera’ consisting of an array of 6 x 6 of these STJs, each measuring 25 micrometres square. But now we can form an image with our tiny detector array, and we can count photons arriving at rates of up to about 1000 per second on each of the 36 junctions. We can record their arrival time with an accuracy of about 5 microseconds, and we can measure their wavelength with an accuracy of about 100 nanometres. This wavelength resolution is much lower than what can be achieved using filters or a spectrograph, but it is just the beginning. Superconductors with a lower critical temperature result in a larger number of broken Cooper pairs for a given photon energy, and we believe that larger arrays, which can count at much higher rates of incident photons, and which can measure each photon energy with an accuracy of a few nanometres, will be developed over the next few years. “