Hi-Intensity LED light meter?

Madness

Several of my projects involve Hi-Intensity LEDs.··· I have been using the PNA-4601M for IR Remote control, and ULN-2003 & 2803s for drivers.· All works well.· I also gang together several ULN outputs with different low-ohm resistors, to·make variable intensity drivers for strings of LEDs.· This too, works.

Over the years, I have accumulated more than few unlabled Hi-Intensity LEDs, and would like to be able to measure their outputs to some·crude, but useful, level of specification.· The most common LEDs I use are Whites, Reds and Blues.· I tried to configure the VT900 Photo-Resistor in various configurations supplied in the book and have not been successful in getting something that will reasonably, repeatably, reliably·measure a wide range of Hi-intensities, Colors or Angles.· I would relish being able to measure mcd ranges from 1k, to 100k, in 1k steps ...·viewing angles from 15d to 75d,·to 120d, and colors.

I looked around a·bit and did not see anything addressing this, might anyone have some suggestions?

metron9

I used a small tube with a photocell mounted at one end. the tube is just 1/2 inch long. I simply place it over the LED and apply power. If using resistors and a bench supply you can bring the power up until the photocell stops or slows down to changes as the voltage increases. At that point you know the light output has reached it's maximum. It also works well to monitor an LED on a heat sink. Full power is given to the LED and the output of the photo cell is monitored. If the light output starts to go down it means the led may be getting too hot and the current should be lowered or the heat sink should be increased.

Gives you a rough idea without having to take in the different angles at various distance at least when it comes to identifying maximum current to an unknown LED.

For example I have 3 watt and 1 watt luxeon LED's in a drawer from various projects. Can't really tell them apart as they look the same however the 1 watt is much lower in output as you go past 350ma and the 3 watt keeps increasing over 350ma.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Think Inside the box first and if that doesn't work..
Re-arrange what's inside the box then...
Think outside the BOX!

Madness

Many thanks for the reply. I assume when you say "photocell" you are referring to a Photo-resistor similar to the VT900?

Typically, I use a meter and fixed resisitors to approach the nominal specified current rating (and Vf) for the LED. But again, that only tells you when you have reached a maximum current level, or exceeded it. I am trying to quantify the Mcd output of a wide variety of Hi-power LEDs, which may have similar max currents (typical 20-30 ma), but different Mcd output levels (1500 Mcd to 100,000 Mcd).

I understand that viewing angles and color only complicate the matter, but one problem at a time. [noparse]:)[/noparse]

Kevin Wood

I was roaming the web one day looking for info on the TAOS light-to-frequency sensor that Parallax sells, and came across an article that may be of interest to you... roamingdrone.wordpress.com/2008/11/13/arduino-and-the-taos-tsl230r-light-sensor-getting-started/

It's written based on the Arduino, but the concepts should translate to PBasic ok. One note... if you look at the datasheet that is eventually linked, you'll notice that there are part numbers listed for versions of the sensor with more accurate tolerances. If you want one of those parts for you're purposes, you can find them at Digikey. Otherwise, you can use the version Parallax sells.

Madness

Cool, that's it, I think. A Light-to-Frequency IC ... I hadn't thought of that. I'll have to read more, but it explains the difference between Luminance (Radiometric) and Color (Photometric) measurement, with a section for HI/LO ranges. There is also a ton of info in the comments, and links.

Thanks.

erco

That's the sensor on sale at Radio Shack, BTW. Clearance item while they last.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
·"If you build it, they will come."

Phil Pilgrim (PhiPi)

When measuring the output from the LED, be sure to take two samples: one with the LED on; the other with it off. Then subtract the latter from the former to cancel any effects of ambient light.

-Phil