Measuring lights wavelength!

[zorthgo]

Hi guys, I was wondering if there is a sensor out there that would allow me to measure the wavelength of a beam of light hitting it. I have these multicolored LEDs that I am putting together to get different color lights. I would like to be able to measure the wavelength of the resulting light. If anybody could point me in the right direction, or maybe give me a wake up call!

Thanks!

[ejeffrey]

Mixing colors doesn't change the wavelength, it gives you a mixture of all the wavelengths.  To measure that you need a spectrometer.  Assuming you got sticker shock looking at that, you can measure the color (i.e., how humans perceive) by using a set of 3 photosensors with colored plastic filters in front of them.

No matter how you measure, the result will be completely linear.  The response on your detector with several LEDs lit will be exactly the sum of the response of each LED individually, at least assuming that your detectors themselves aren't saturating.  Red and blue making purple is a perceptual effect -- it is how your brain handles the red and blue photo-receptors in your eye going off at the same time.  It has no basis in optics or electromagnetism at all.

[Smokey]

It depends on how accurate you need to be.  Check out this link/video for a cheap solution.
http://thesignalpath.com/blogs/2012/06/04/tutorial-on-waveform-shaping-and-signal-constellation-in-an-optical-link/

He reviews a real wavelength meter here:
http://thesignalpath.com/blogs/2012/07/16/teardown-and-operation-of-an-advantest-tq8325-wavelength-meter/

[Psi]

Mixing colors doesn't change the wavelength, it gives you a mixture of all the wavelengths.

It's easy to understand this when ya consider that there's no wavelength for white

[Mechatrommer]

Mixing colors doesn't change the wavelength, it gives you a mixture of all the wavelengths.

It's easy to understand this when ya consider that there's no wavelength for white

its a square wave.

[poptones]

square waves have a precise mathematical relationship. white noise is called white noise for a reason.

[XynxNet]

For spectroskopie there is always the low-tech way:
Take an aperture and a prism or a diffraction grating and measure the peaks on some sort of projection screen with a ruler and calculate the frequency.

If you want to build a spektrometer i suggest you follow the same principle and use a difraction grating and a line-ccd from a scanner. This setup has the advantage that it can also detect the intensity.

[zorthgo]

Thanks guys for the replies!

ejeffrey, You got me a little confused now. In physics I, we learn about the electromagnetic spectrum and that the colors in the visible portion of the spectrum change with the change in wave length. But now you are telling me that the different colors are actually a bunch of different wavelengths bundled together. This would be consistent with what we learned in Physics II when studying optics (prisms and stuff). Can you please elaborate a little on the difference between the colors when the waves are of different length and when there is a bunch of waves superposed on each other? When do we see one, as opposed to the other?Thanks!

[caroper]

This may interest you http://publiclaboratory.org/wiki/video-spectrometer-construction


Cheers
Chris

[rolycat]

Thanks guys for the replies!

ejeffrey, You got me a little confused now. In physics I, we learn about the electromagnetic spectrum and that the colors in the visible portion of the spectrum change with the change in wave length. But now you are telling me that the different colors are actually a bunch of different wavelengths bundled together. This would be consistent with what we learned in Physics II when studying optics (prisms and stuff). Can you please elaborate a little on the difference between the colors when the waves are of different length and when there is a bunch of waves superposed on each other? When do we see one, as opposed to the other?Thanks!

Zorthgo, ejeffrey isn't saying that at all. You need to make a distinction between physics and physiology.

All colours are perceptual effects determined by the different types of cone cells in the eye. It so happens that stimulating them with certain combinations of wavelengths causes our eyes to 'see' certain colours, although a single wavelength may result in the same perceived colour. 

For example, 'yellow' can result from a single wavelength, or a mixture of 'red' and 'green' wavelengths.

Similarly, 'white' can be a combination of all visible wavelengths, or of just three.

[zorthgo]

Thanks rolycat for the explanation! So, let me see if I got this right. In physics, you get a specific light color when light has a specific wavelength. But because the cones in our eyes can only see RGB, when you mix those wavelengths together, you are able to simulate different colors. Kind of like the trick that televisions employ by using RGB to display pictures on the screen. These simulated color, don't actually change the wavelength of the original individual lights, it only superimposed them to create the illusion of a different color. You guys rock! I love this forum!

[rolycat]

Thanks rolycat for the explanation! So, let me see if I got this right. In physics, you get a specific light color when light has a specific wavelength. But because the cones in our eyes can only see RGB, when you mix those wavelengths together, you are able to simulate different colors. Kind of like the trick that televisions employ by using RGB to display pictures on the screen. These simulated color, don't actually change the wavelength of the original individual lights, it only superimposed them to create the illusion of a different color. You guys rock! I love this forum!

That's not a bad approximation, although rather than having red, green and blue cones we have cones which respond to ranges of long, medium or short wavelengths. There's quite a lot of overlap, and the peak sensitivity of long wavelength cones is more in the yellow than the red part of the spectrum. Our brains do a lot of post-processing of these signals to produce the colours we see.

Also, it's worth noting that we can see colours which aren't part of the spectrum, and only exist as mixtures of spectral colours - pink and purple, for example. As was recently noted elsewhere in the forum, purple and violet are not the same - violet is a spectral colour and purple isn't.

[PA0PBZ]

I always wondered:

As a child you learn that grass is green, the sky is blue and blood is red, so when I see something that I learned is green I call it green.
Someone else looking at the same object also says it is green because that is what he/she learned.
But if what he/she sees is in fact how I see red? I don't think there is any way to find out, because how would you describe a colour other than by example?

_{Hmm, this could be a lengthy discussion, or completely ignored...}

[rolycat]

I always wondered:

As a child you learn that grass is green, the sky is blue and blood is red, so when I see something that I learned is green I call it green.
Someone else looking at the same object also says it is green because that is what he/she learned.
But if what he/she sees is in fact how I see red? I don't think there is any way to find out, because how would you describe a colour other than by example?

_{Hmm, this could be a lengthy discussion, or completely ignored...}

There was an excellent episode of the BBC program 'Horizon' devoted to this question - it was called 'Do you see what I see?'.

Apparently there is a tribe in Namibia called the Himba who do seem to see colours differently to other peoples, and there is evidence that our perception of colours is shaped by the language we use to describe them. There's an article about this here:

http://www.apa.org/monitor/feb05/hues.aspx

[zorthgo]

Thanks guys for the help! I really appreciate it.
I had a very interesting conversation with my buddies from school about this topic!