Nuts & Volts LED photometer

Go To Last Post
8 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hi Guys,

There is an article in the new Nuts & Volts that describes using LEDs as photometers.  It is based on a Forrest Mims article that states that the LED will produce a current if the wavelength of light incident on it is in the narrow band that the LED emits at.  I made two photometers, one with a red LED and one with a green LED.  When I shine red light on the green LED I get no current, but when I shine green light on the red LED I do get current, inconsistent with Forrest Mims. 

 

I would expect that if the incident photons have an energy equal to or greater than the detector LED band gap I will get electrons getting kicked across and giving a current, consistent with my observation.  Doing a Google search, I find a lot of people post that the detector response is in a narrow band, and others, one from Analog Devices, saying the detector LED will respond to light at wavelengths equal to or less than the wavelength it emits at.

 

Do you guys have any comments on this.  I have to believe my observations and I trust Analog Devices to be a reputable source.  All I know about Forrest Mims is he seems to be quite the self promoter.

 

mark

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

MarkThomas wrote:
Do you guys have any comments on this.

I played around with this some years ago, as an alternative to using a "real" light detection device.

 

I remember indeed getting a green LED to "work".  But even a brightly-lit office gave a signal in the mud.  OK results right next to office fixture; good results in daylight.

 

Ended up using a "real" device in the app.

 

More fun to me are the [formerly] Taos detection chips.  Have used them in several apps with good results.  Pricier than a common LED of course.

http://www.taosinc.com/  (has links to the ams Web site)

 

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hi theusch,

I also use the Taos detectors.  The one I have been using has just been obsoleted.  It has a SYNC pin so you can integrate over the number of times the sync pin goes high, so I could integrate over a chosen number of PWM pulses, which is perfect for a dimmed LED.  The replacement devices have all done away with the SYNC pin, and the only option now is to integrate over some number of 2.3 mSec time intervals, which will not work so well for me, as I will get different results depending on where in the PWM the integrations starts.  Other than that they are nice devices.

 

mark 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I've done some experiments with leds as photodetectors by holding some random leds close to a fluorescent light and measuring the voltage on the led with a DMM.

Results varied a lot.

Some leds generated1800mV, others only a few mV.

Color of the led did not seem to have much significance.

Make & model of the led did seem to make a lot of difference.

Optimizing a led for light output is probably another task from optimizing for light input.

 

Fun story:

Read a story about a battery powered gadged with a few leds & a uC.

The led pins were switched to input. Maybe to save a few uAmps, maybe some multiplexing with buttons or whatever.

Sometimes the gadged had an unexpected high current consumption. It turned out that when a bright light was shone on the led's they generated enough voltage to turn the input of the uC in the linear region where both the N and P channel Fet's were conducting...

Doing magic with a USD 7 Logic Analyser: https://www.avrfreaks.net/comment/2421756#comment-2421756

Bunch of old projects with AVR's: http://www.hoevendesign.com

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

This is basically the circuit I used, only with 5V, and varying feed back resistances:

 

Image result for led as photometer circuit

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

It could be related to how monochromatic your 'green' light-source is.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

It could be related to how monochromatic your 'green' light-source is.

I am using high quality narrow spectral width LEDs, so it is not because the green LED has some red in it.

 

I tend to believe this guy, who was European applications manager for Analog Devices for many years:

 

http://www.analog.com/en/analog-dialogue/raqs/raq-issue-108.html

 

And this from Analog Devices:

 

https://wiki.analog.com/university/courses/electronics/electronics-lab-led-sensor

 

and not this from Forrest Mims, amateur scientist:

 

http://makezine.com/projects/make-36-boards/how-to-use-leds-to-detect-light/

 

You know how the web is.  Sometimes you will find the same wrong information propagated across many different posters, who all seem to get it from the same source.  My own observations agree with Analog Devices, who should be as reputable as they get.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

A reverse-biased LED holds charge I.e. is a capacitor. A leaky one, as it happens. The rate at which that charge leaks away is dependent upon the intensity of light falling upon the junction. Each photon kicks away some charge.

Charge the LED by reverse biasing for a short time, say a few ms (much shorter is likely to be fine, too), then measure how long it takes to fall to a specific voltage. The faster it falls, the higher the intensity of light.

I expect the same bandgap limit applies.

"Experience is what enables you to recognise a mistake the second time you make it."

"Good judgement comes from experience.  Experience comes from bad judgement."

"Wisdom is always wont to arrive late, and to be a little approximate on first possession."

"When you hear hoofbeats, think horses, not unicorns."

"Fast.  Cheap.  Good.  Pick two."

"We see a lot of arses on handlebars around here." - [J Ekdahl]