3
$\begingroup$

I would like be able to measure LED brightness using a cost effective instrument.

What are the available instruments to measure LED brightness and respective cost effective techniques?

Background: I'm interested in converting what the customer thinks is acceptable brightness to actual LED brightness specification?.


Reference:

$\endgroup$
6
$\begingroup$

Customer specified "acceptable brightness" is going to be a subjective affair unless you have something to calibrate their responses against.

Loosely, you'll need to measure the available light within a particular area and then correlate that measurement with customer surveys. With a large enough sample of responses, you should be able to determine what constitutes an acceptable level of lighting for your customers.

I would expect that customer responses will fall along a curve of some sort and you'll eventually be able to determine what the standard deviations are for "acceptable."

One factor to be concerned about is that differing rooms will have differing levels of reflection. So even though two rooms may be of similar size and furnishings, the more darkly painted room will reflect less light and will be viewed as "more dim" on a subjective basis. To workaround that, you might create a dedicated "test room" and have potential customers come through and provide their feedback. While this is a more expensive approach, it's a very good way to isolate out variables that you can't otherwise control against.


To get you started with measuring available light, you'll likely need a light meter of some sort. This and this are just two examples of what I'm sure you'll be able to readily find with some searching.

Once you have a lux measurement for the area, you can convert that to lumens using a standard calculator. Searching will turn up a number of sites, such as this one. I didn't put the equations here as there are a number of steps involved, including determining the area of the room.

$\endgroup$
  • $\begingroup$ Thanks, your explanation is what we are planning on doing. But we need define a system for test, validation, as well as manufacturing. We have one of the exact instruments you have recommended $\endgroup$ – Mahendra Gunawardena Jan 21 '15 at 3:01
7
$\begingroup$

You can easily measure the brightness of any light source in the visible with a silicon (Si) or indium gallium arsenide (InGaAs) photodiode. Both can be picked up at relatively low cost especially if buying in bulk. A photodiode works in the reverse manner to an LED. With an LED running a current through the device causes it to emit light; with a photodiode light incident on the device causes it to become a current source. The responsivity of a Silicon detector is shown below.

enter image description here

Once you have your photodiode you need to build a simple circuit around it which converts current to voltage, and the voltage readout gives you a measure of the light intensity incident upon the device. This type of circuit is typically called a transimpedance amplifier. A high SNR transimpedance amplifier can be built with a single op-amp so the cost again is relatively low.

This setup will make relative brightness measurements very precisely as long as you don't saturate the photodiode or circuitry with too much light. Converting the measured voltage to an absolute intensity of the light is also possible though any errors in the parameters will turn up as errors in the inferred intensity. To do so you use the circuit parameters to calculate the amount of current generated by the photodiode. The photodiode responsivity is then used to convert to incident intensity.

$\endgroup$
  • 1
    $\begingroup$ This is the approach I'd take. People's perception of color and brightness are relative anyways. Also note that perception of certain colors (greens, blues) is more sensitive in most humans. So if designing an electronic sign using RGB LED's, you'll probably want to profile the hue component against human eye sensitivity, otherwise some colors will seem "painfully bright". Additionally, ambient brightness should be taken into consideration. There are many outdoor billboard LED signs in my area. During the day, they look fine. But at night, some are repulsively bright. $\endgroup$ – rdtsc May 23 '15 at 14:02
5
$\begingroup$

If you have a DMM with a uA current range, then a photodiode is the simplest detector. I get about 15 uA of current in my lab (PD area is 44 mm^2) An LED held up close can give more than 1 mA of current.

$\endgroup$
3
$\begingroup$

The brightness of a light emitting diode (or anything else) is measured by luminous flux, which is the impact of the light on an eye, adjusted for different wavelengths.

Luminous flux is measured by lumens, which corresponds to candelas of light, or luminous intensity, emitted over a solid angle.

$\endgroup$
3
$\begingroup$

There are two aspects that you need to consider here: One is physical measurement of luminance, luminous flux, or illuminance (I'll come on to the differences in a moment), and the other is the much less clear-cut issue of human perception.

Let's deal with the physics first. There are three concepts to explain, which are all related:

Luminous flux is the total amount of light emitted by a light source, in all directions. Its SI unit is lumens. This is difficult to measure without expensive equipment (an integrating sphere). Luminous flux is relevant because it gives a measure of the total amount of light that a source is putting into a room, and will typically be given in the specifications of a light source.

Illuminance is the amount of light that falls on a surface of unit area. Its SI unit is lux, where 1 lux = 1 lm/m2. It is easy to measure with commonly available "light meters" and photodiodes. Illuminance is the most important value when asking, for example, "will enough light fall on this desk for somebody to work by". Building codes or standards often specify the illuminance required on floors, walls, or work surfaces, for specific tasks. This is a function of the overall design of lighting in the space rather than just of the light source - type of fixture, quantity and spacing, colour of the walls, etc., are all important - but marketing materials for some LED light sources in particular will claim "equivalent to xxx wattage incandescent" by comparing the illuminance on a small area of work surface from an omnidirectional incandescent lamp vs a narrow-beam LED, and imply that they are equivalent - despite them having radically different effects everywhere else in the room.

Luminance is a measure of the amount of light being given off by a surface in a specific direction, per unit of area. If that surface is the surface of a light source - e.g. if you look at a light bulb - then this is related to how bright the source appears. This measurement is relevant because if a customer looks directly at a range of light sources to compare "how bright they are", it is differences in luminance that they will see. Luminance meters are much more expensive than illuminance meters.

So in summary,

  • Luminous flux is the total amount of light given off by a source
  • Illuminance is how brightly a surface is lit
  • Luminance is (part of) how bright the light source appears to the eye

Which one(s) you need to measure will depend on exactly what you are trying to achieve.

Human perception is the tricky part here. There is more to a feeling of "brightness" than either of the measures above. For example, people tend to perceive light of a higher colour temperature as being brighter than light of a lower colour temperature - a common issue when comparing LEDs to other sources. People are also strongly influenced by contrast; we mostly see relative rather than absolute brightness, so a patch of light on a small area of workspace in an otherwise dark room may seem brighter than the same illuminance throughout the space, especially once the viewer's eyes have adapted to the dark. So be very careful if giving practical demonstrations, or you may find that the customer's perception does not match your calculations. (it's also worth noting, of course, that for some purposes it's the perception that matters...)

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.