Determine emission spectrum of an LED

[Lotic7]

I recently purchased some 660nm LEDs. They look kind of orange not deep red. What is the easiest and cheapest way to determine the emission spectrum. Eventually I wanted to try to use the 660nm LEDs to grow some plants.

 

[berkeman]

I recently purchased some 660nm LEDs. They look kind of orange not deep red. What is the easiest and cheapest way to determine the emission spectrum. Eventually I wanted to try to use the 660nm LEDs to grow some plants.

Check the datasheet for the LEDs -- it may have that information. Or else, maybe use a prism?

 

[Lotic7]

I purchased the 660nm LEDs off amazon. There are no no manufacturer markings on them, but they work. I guess you get what you pay for. How well would a digital camera work in analyzing the spectrum? If a picture was taken of the lit LED or after the light from the LED was put through a prism and refracted. The RGB values of the individual pixels could be observed. It would be interesting to see how accurate this would be. Would the image compression affect the RGB values (TIFF vs JPG)?

 

[OmCheeto]

Eventually I wanted to try to use the 660nm LEDs to grow some plants

Me too! I've been thinking about it for over a year now. [ref: PF]

I purchased the 660nm LEDs off amazon. There are no no manufacturer markings on them, but they work. I guess you get what you pay for. How well would a digital camera work in analyzing the spectrum? If a picture was taken of the lit LED or after the light from the LED was put through a prism and refracted. The RGB values of the individual pixels could be observed.

I just did this experiment. The RGB values are of no use, IMHO.

A couple of weeks ago, my fish tank cast a rainbow on my La-Z-Boy, so I took a picture. I also took pictures of the light coming out of the fish tank.

I just now took a snippet of the red section from the La-Z-Boy image, and the only thing it told me, was that everything was red.

It would be interesting to see how accurate this would be. Would the image compression affect the RGB values (TIFF vs JPG)?

My camera is quite, um, "inexpensive", and only takes JPG images. But, I think it's a moot point.

Anyways, here's an interesting article that I just found that may answer a couple of your questions:

Build a high resolution spectrograph in 15 minutes


ps. Science!

 

[Andy Resnick]

How well would a digital camera work in analyzing the spectrum?

Very well: I took these by simply placing a diffraction grating in the optical path

That's the spectral output of an incandescent light, and on my camera, the full size the spectrum covers 950 pixels. There are some subtle aspects (the shape of the light is convolved with the rainbow), but if you have a standard spectrum, you can indeed make quantitative measurements. A good standard is the output of a fluorescent bulb:

Each emission wavelength has a clearly identifiable location- overlaying an image of the LED will give a pretty good estimate of the output spectrum. Even so, I have to mention that I'm not sure what the measurement limitations are- can I distinguish between 660 and 650 nm? Unclear, but that's why I have a spectrometer in the lab.

Edit: OmCheeto's link describes essentially what I did.

 

[OmCheeto]

...
Edit: OmCheeto's link describes essentially what I did.

I was going to drop your name, as a known authority on the topic, but I thought it might be seen as a bit presumptuous.

 

[Andy Resnick]

I was going to drop your name, as a known authority on the topic, but I thought it might be seen as a bit presumptuous.

Thx- feel free to drop away. I don't know about 'known authority'... maybe 'known blabbermouth' :) Cheers!

 

[Lotic7]

Thanks for the help. I will have to get some diffraction grating, and try this out.

 

[Tom.G]

Save your money for a diffraction grating. A CD or DVD works fine.

 

[Andy Resnick]

Save your money for a diffraction grating. A CD or DVD works fine.

CD/DVD are not transmissive gratings and have curved 'slits', complicating everything. Transmission gratings are so cheap as to be nearly free:

http://www.hometrainingtools.com/di...e=2&fep=2082&gclid=CNTn4f-TissCFQEGaQodrpUIIw

We buy then by the case.

 

[Tom.G]

Thanks Andy, that's a good site. They have spectroscope for under $10.

 

[OmCheeto]

CD/DVD are not transmissive gratings and have curved 'slits', complicating everything. Transmission gratings are so cheap as to be nearly free:

http://www.hometrainingtools.com/di...e=2&fep=2082&gclid=CNTn4f-TissCFQEGaQodrpUIIw

We buy then by the case.

Beings how today yesterday was Sunday, and the store that sells diffraction grating was closed, I decided to try the CD approach.
I now can appreciate why there are "optical benches", as my data is somewhat boogered.
I'll try again in the morning.

 

[davenn]

Beings how today yesterday was Sunday, and the store that sells diffraction grating was closed, I decided to try the CD approach.
I now can appreciate why there are "optical benches", as my data is somewhat boogered.
I'll try again in the morning.

View attachment 96274

Brilliant

also good to see you have the obligatory roll of duct tape in there, Om Dave

 

[OmCheeto]

While I'm collecting new data, I thought I'd share some notes, images, comments, retractions, excuses, and theories as to what the heck is going wrong.

Notes:

1. My camera is old and cheap. The only adjustment available is "focus", which consists of markings: "Flower"(1 ft), "Mountain"(∞), and a "4 feet" line, which I drew on with a sharpie pen one day.
2. When overloaded, my camera turns colors white.
​

Images:

1. In the course of the experiment, I rounded up 7 unique light sources from around the house, of which, I collected spectral images:

Comments: Looking at the first raw images, I didn't think this experiment would work.

But, I would discover that the fact that my camera had 2560 x-axis pixel resolution, would yield ≈1 nm information, based on:

white LED bulb range

x-axis____ limits of visible range
1418______22 pt red
1702______64 pt blue
284_______difference​

----
google:

A typical human eye will respond to wavelengths from about 390 to 700 nm.
700-390= 310 nm​

Retractions: I may suffer from multiple personality syndrome, as, well...

Om; "...everything was red"

Everything was NOT "exclusively" red.​

Om; "The RGB values are of no use, IMHO"

Thank god I included the "IMHO"...​

RGB values are somewhat useful, to certain points.

Wavelength to Colour Relationship
A simple tool to convert a wavelength in nm to an RGB or hexadecimal colour.​

Fortunately, Lotic7's problem fell outside of the usefulness of this tool.
The tool freezes the RGB settings at 255,0,0 from 650 nm to 700 nm.
And the 700 to 780 nm measurements are done with diminishing "red" values, so this is a useless tool for studying monochromatic light

Excuses:

My younger brother showed up on Saturday, and was not impressed with my cereal box and cd, that I was going to build a spectrometer from. Things went downhill from there.
My sister is flying in from out of town tomorrow. She will be here for 7 days. Do not expect much, after today.
​

Theories:

1. My red LED really warmed up. Could there be some black body radiation effect, affecting the emitted wavelengths?
2. When I looked at one of my LEDs(they are all clear lensed), I wondered if human eyes can be oversaturated, kind of like my camera, and give an off color interpretation?
​

ps. RGB values courtesy of @lpetrich 's most awesome "Image Measurer" software, Version 1.0 (1) [ref]

 

[OmCheeto]

I hate science...

I swear to god, it laughs at me...

But I did capture a missing mercury vapor yellow band there.

Off to recalibrate my devices.

ps. I added a lens.

 

[davenn]

While I'm collecting new data, I thought I'd share some notes, images, comments, retractions, excuses, and theories as to what the heck is going wrong.

Notes:

1. My camera is old and cheap. The only adjustment available is "focus", which consists of markings: "Flower"(1 ft), "Mountain"(∞), and a "4 feet" line, which I drew on with a sharpie pen one day.
2. When overloaded, my camera turns colors white.
​

..... big snip ...

outstanding effort considering the setup !
am impressedDave

 

[OmCheeto]

Neither science, nor my setup, are treating me kindly.

@Lotic7 , invest in the diffraction grating...

argh!

Off to take my 3rd set of photos...

ps. did I mention, that science, is stupid?

 

[Andy Resnick]

While I'm collecting new data, I thought I'd share some notes, images, comments, retractions, excuses, and theories as to what the heck is going wrong.
<snip>

Nice! Here's one I put together of the planets (JPG):

I did this a long time ago, before I had a tracking mount- I can now leave the shutter open longer, getting a better signal-to-noise ratio. In any case, plotting the (B, G, R) values shows the color differences. Here's Venus (the top planet):

Note the pixel position of the red and green peak. For comparison, here's Mars (the next one down), which is redder in hue:

It's noiser, the peaks have shifted to larger pixel values and the red channel is generally brighter, reflecting the redder hue. Here's Jupiter and Saturn:

And the bottom one is a star, Porrima, bright and conveniently located:

The two dim ones are moons of Jupiter, which I can definitely capture next time.

 

[1oldman2]

Beings how today yesterday was Sunday, and the store that sells diffraction grating was closed, I decided to try the CD approach.
I now can appreciate why there are "optical benches", as my data is somewhat boogered.
I'll try again in the morning.

I don't just "like" this I love it

 

[Jeff Rosenbury]

Some fairly bright ideas. (Bad pun, bad. ) But perhaps you are overthinking it?

Go down to Home Depot to the paint department and use their spectrograph. Tell them you want to match some paint to this light color. (Not a bad idea for the grow room, BTW.)

Of course the machine is designed to match paint, not LEDs, so it may not work. But if it doesn't you are only out some gas.

All hail OmCheeto. His setup shows a deep understanding of the problem. With a little more bubble gum and some bailing wire I now expect him to build a 747.

 

[OmCheeto]

outstanding effort considering the setup !
am impressedDave

I decided that the flexibility of the cardboard box was one major problem.
I placed the lamp inside the box each time, and I'm sure it shifted the slit in the end, thus throwing off measurements.
On my 4th data collection run, my red LED finally burned out.
So for my 5th data collection run, I decided simply to take some photos of the light reflected by my LEDs, and see if I couldn't extract useful information from the RGB levels.
The data there was also screwy. I used my red laser pointer as a reference, and my camera thinks there is blue light coming out of it.

R    G    B     Notes
203  0    46    red laser
196  0    51    red laser

183  0    25    red led
165  0    9     red led

I believe it was at that point, that I decided that I should start over from scratch.
And perhaps get a new camera, as this one is obviously just making things up.

I will probably do future optics experiments on the kitchen counter, as that little table is a bit wobbly.

 

[Andy Resnick]

The data there was also screwy. I used my red laser pointer as a reference, and my camera thinks there is blue light coming out of it.

I see this issue as well, when I illuminate with a strong monochromatic source. I think it's a combination of Bayer filter artifact and intensity, but I don't know for sure. Here's an example, a diffraction pattern from a 532 laser:

And the color line trace from the central peak outwards:

The red and blue channels get excited when the intensity is high.

 

[OmCheeto]

I see this issue as well, when I illuminate with a strong monochromatic source. I think it's a combination of Bayer filter artifact and intensity, but I don't know for sure. Here's an example, a diffraction pattern from a 532 laser:
And the color line trace from the central peak outwards:

The red and blue channels get excited when the intensity is high.

I think my Bayer filter is broken.

Here's a thumbnail of my red laser reflected off of a white surface.
I knew it would oversaturate my camera, so I duct taped my burned out led to the end to scatter the light.
Even the dimmest regions show blue.

hmmmm... There's a total eclipse next year, and I should probably get a new camera for that. I think I'll go shopping in the morning.

 

[Tom.G]

Even the dimmest regions show blue.

If the camera has "automatic white balance", turn it off!

 

[Daz]

The red and blue channels get excited when the intensity is high.

I too have seen this occur. It happens with CCD detectors where each pixel is essentially a potential well with a finite density of states. If the incident light intensity is high, the individual colour pixels saturate and charge starts spilling over into adjacent pixels. It’s called white-out and reducing the intensity should resolve it.

Even in the absence of saturation, a monochromatic input will still lead to some signal in all three colour channels because the red, green and blue pixel filters aren’t perfect. Here’s a plot of quantum-efficiency taken from the data sheet of a common CCD image sensor. Note that none of the filters completely block wavelengths outside of its pass-band.

I guess this might explain OmCheeto’s observation. Also, the signal processing in consumer cameras usually includes white-balance correction which adjusts the R-G-B signals to try to compensate for different lighting conditions. So interpreting those values to determine wavelength should be done with great caution.

Coming back to the original post about LED colour, I’ve noticed (through error, rather than design) that if I overdrive an LED the emission blue-shifts and broadens. I once accidentally connected an LED display directly to the supply without a current-limiting resistor. The normally red display lit up brightly with an almost whitish-orange hue. Then started to smoke! Maybe that’s why the OP observed orange emission from a red LED?

 

[Jeff Rosenbury]

Coming back to the original post about LED colour, I’ve noticed (through error, rather than design) that if I overdrive an LED the emission blue-shifts and broadens. I once accidentally connected an LED display directly to the supply without a current-limiting resistor. The normally red display lit up brightly with an almost whitish-orange hue. Then started to smoke! Maybe that’s why the OP observed orange emission from a red LED?

Important application note: Don't let the magic smoke escape.

(I know from experience.)

 

[jerromyjon]

Important application note: Don't let the magic smoke escape.

(I know from experience.)

OMG I've done this, a looonnng time ago, what bad could happen?

 

[Jeff Rosenbury]

OMG I've done this, what bad could happen?

Worst case? Dogs and cats living together. Fire , flood (well, if it's a dam controller). Destruction.

Normally the component just needs to be replaced though.

 

[Jeff Rosenbury]

On a more serious note, LEDs are current driven, so need some sort of current limit. Often this is just a resistor.

 

[jerromyjon]

Worst case? Dogs and cats living together. Fire , flood (well, if it's a dam controller). Destruction.

Whew, as long as my dreams didn't go up in smoke from too many fried leds! The last one I fried was the best because the next one launched the rocket successfully. I used leds to make the switch that you had to push one direction on controller to arm it, then the other direction to launch, using a forward or turns in reverse wireless car chassis. I fired a D12-7 missile looking rocket but the fins didn't hold and those were the days.

 

[Andy Resnick]

I think my Bayer filter is broken.

I too have seen this occur. It happens with CCD detectors where each pixel is essentially a potential well with a finite density of states. If the incident light intensity is high, the individual colour pixels saturate and charge starts spilling over into adjacent pixels. It’s called white-out and reducing the intensity should resolve it.

Yeah, I was wondering about blooming. The way it appears and is controlled controlled in CMOS is different than CCDs: on CCDs, blooming leads to vertical or horizontal streaking which I never see with a CMOS sensor.

The more I think about Bayer filters, the more confused I get. First, they are thin-film reflective type filters as opposed to absorptive filters, because absorptive filters would degrade over time, leading to inconsistent and nonuniform bandpass changes. On the other hand, thin film filters only work over a restricted range of incident angle, so high-angle rays associated either with fast lenses or wide angle lenses would not be correctly filtered. I have never noticed such a thing, and have not heard anyone else noticing that.

But they have to be reflective- here's an image of light reflecting off the filter, then reflected again by the lens and captured by the sensor:

This is clearly a reflection, but... if this was light initially rejected by the filter, then the colors should be inverted: blue reflects yellow, green reflects magenta, and red reflects cyan. Although maybe the colors are inverted- I can't tell because I'm color blind :(

Imaging the filter directly is a challenge, this is the best one I've been able to make (so far):

This image was taken using brightfield reflection microscopy, but the light must be reflecting off the underlying pixels, not the filter, because the colors are correctly rendered. I don't understand the 'half pixel' appearance... I guess mine is broken as well :)

 

[collinsmark]

I don't understand the 'half pixel' appearance... I guess mine is broken as well :)

This is a pretty common Bayer pattern.

Notice that there are as many green pixels as there are red and blue pixels combined (or another way of putting it, twice as many green as red, twice as many green as blue). That's because the human eye perceives more detail in the green part of the spectrum. In the post-processed image luminance favors the green channel. By favoring green in pixel count, the silly human will perceive a more detailed image.

 

[collinsmark]

This image was taken using brightfield reflection microscopy, but the light must be reflecting off the underlying pixels, not the filter, because the colors are correctly rendered.

Maybe the filter material (of a given pixel) reflects and transmits the same color, thus absorbing the other colors (neither reflecting nor transmitting the other colors*)? 'Like stained glass.

*meaning the other colors are not reflected, but they don't make it to the pixel element either.

 

[Daz]

Yes, I’ve noticed that CMOS sensors don’t saturate in the same way as CCDs where you tend to get vertical white streaks in the image as the charge overspills along the read-out line. But I guess they do something similar, nonetheless.

On the other hand, thin film filters only work over a restricted range of incident angle, so high-angle rays associated either with fast lenses or wide angle lenses would not be correctly filtered. I have never noticed such a thing, and have not heard anyone else noticing that.

This is one of the reasons why mega-pixel cameras require the objective lens to be image-space telecentric. Virtually all modern lenses for nigh-resolution imaging are designed for telecentric output. With such a lens each pixel is illuminated more-or-less normally. Certainly, if you were to use an entocentric objective with a milti-megapixel image sensor you do get pixel shadowing and a shift in colour balance towards the corners of the image.

 

[OmCheeto]

If the camera has "automatic white balance", turn it off!

Even my new camera has no "Off" mode for "automatic white balance".

As far as I can tell anyways.
It comes with 131 pages of instructions.

I'm pretty sure my old camera had one page of instructions.

ps. It took me 3 hours of googling yesterday, just to figure out how to transfer the pictures from my new camera, to my laptop.

Things were a lot simpler, in the Canon A-1 days, of old.