Light Meter for LEDs

[4runner freak]

I have a request that I thought that I ought to come to the pros for.

I have never used a light meter before and I am wanting to buy one to be able to label some lights for their brightness. I am going to be testing standard incandescent and halogen lights as well as LED lights.
I know that LEDs are very directional lights but can anyone here suggest a good light meter that I can use for them?
I will be using this along with my new Canon 40D(Just upgraded from a Pentax SF-X) as well.
I want one that works on AA(or 9V) batteries, so I'm not having to source the 123 6v batteries(IIRC)

I'm sorry that I'm a little out of my league on this, just hoping for some help.
Thanks in advance

 

[oftheherd]

I don't know how many light meters give lux or candle power readings. The Gossen Luna Pro and Luna Pro SBC do. Also the Sekonic L28c2, or at least readings that you can convert. Am I correct that is what you are looking to do?

If you are just looking at how it lights up a scene, any light meter will do that for you, reflective or incident, including the meter in your camera.

 

[sevo]

You need a flat diffuser head, as lux is a measure of surface illumination. These come with (or are at least available as a accessory for) most professional incident meters, with an EV to lux conversion table in the accompanying leaflet (or even programmed into the device software). Reflective only meters or meters with a permanently installed dome diffuser are no good for the purpose, for the reason above.

 

[oftheherd]

You need a flat diffuser head, as lux is a measure of surface illumination. These come with (or are at least available as a accessory for) most professional incident meters, with an EV to lux conversion table in the accompanying leaflet (or even programmed into the device software). Reflective only meters or meters with a permanently installed dome diffuser are no good for the purpose, for the reason above.

The Sekonic L28c2 has a flat head. Also used for contrast measurements.

 

[gshybrid]

Maybe this Sekonic i-346

 

[4runner freak]

I don't know how many light meters give lux or candle power readings. The Gossen Luna Pro and Luna Pro SBC do. Also the Sekonic L28c2, or at least readings that you can convert. Am I correct that is what you are looking to do?

If you are just looking at how it lights up a scene, any light meter will do that for you, reflective or incident, including the meter in your camera.

I am trying to measure lumens, but lux and candle power does just as well. I'm wanting to compare the brightness of lights for use as attention getters vs. how well they light up a room.

Thanks for the help so far.
I have a friend that has been a pro photographer for a long time and she is going to lend me her meter(I don't know what make/model yet) to try out before I buy one.

 

[geotrupede]

If you are looking for a comparison why not just request to the manufacturers the ies (or eulumdat if ies is not available) files, which contain the intensity distribution of the sources?
With software like dialux (free) you can then test different options without having to buy a luxmeter...

Note that to measure the flux distribution of a light source you may need a goniophotometer...

If you are looking into measurements of illuminance with a degree of latitude (=some error) and the room has walls, floor and ceiling which are not extremely specular (I'd say if they look enough Lambertian) then you can use HDR imaging and Radiance ray tracing suite to work back the illuminance at any point of the room
You will effectively be measuring the luminance (HDRShop , Photosphere, etc) by combining bracketed exposures with a calibrated camera, then you will use a sample to estimate the reflectance of the surfaces and convert to illuminance by E = L x PI / r ...

If you are interested to know more feel free to send me a message.
Question is, why are you looking into this?

Good luck
G.

 

[divewizard]

measuring illuminance with a exposure meter

measuring illuminance with a exposure meter

Using an indecent light meter that measures EV you can get the intensity of the light hitting the measurement dome. If you set the ISO (ASA) to 100 the (exposure value @ ISO 100) EV = LV (light value).

Each increment of LV is a factor of 2.

0 LV = 0.25 candles per square meter = 0.36 foot-lamberts = 2.5 lux = 0.232 foot-candels = 0.232 lumens per square foot

Lets use lux as an example.

to get the illuminance value (light intensity) for a reading with your light meter you use one of the following formulas

illuminance = 2.5 · 2 ^ LV (lux)

illuminance = 0.232 · 2 ^ LV (foot-candels)

Example:
set the meter to ISO 100, you read 12.33 EV = 12.33 LV, then using lux

2.5 · 2 ^ 12.33 = 12872 lux = 12.9 Klux

To get relative brightness you compare the illuminance for the two light sources.

 

[divewizard]

comparing relative brightness

comparing relative brightness

If you don't really care about the absolute brightness, but only the relative value then you you don't need to worry about ISO settings on the meter. You just need to make sure the ISO setting is the same for both readings.

Example:

light A reads 11.66 EV

light B reads 13 EV

the relative difference in brightness is 2 ^ (13 - 11.67) = 2 ^ 2.33 = 5.03

Therefore light B is 5 times brighter than light A

 

[4runner freak]

Wow. I am amazed at the help.

So, to explain a little better, I will be comparing 12v lights (1156 bases) similar to these. Lights

Most if not all are made in Asia and don't have any documentation.
I am comparing light designs that we are going to sell to see how they compare.
There is a point where no matter how many LEDs you have(or close together), the eye cannot tell if it is any brighter then a smaller quantity of LEDs.
I am needing to have a numerical basis to label these lights.

I will try it out with my friends meter and use the calculations divewizard posted to compare them.

Thanks again

 

[geotrupede]

To be fair I do not think you are going to be able to compare sources just by measuring illuminance and especially with a device that measures exposures (factor of 2 as the "unit"). It is not going to be accurate enough.

I think you mentioned luminous flux, well that's it. You got the idea, but need also to translate in into practise.

Luminous flus is about the distribution of your light. For example where the peak intensity is, is it a bat-wing photometry, asymmetric, flood...

If you measure illuminance (or exposure as you have been advised) you are just getting point values, which you need to revert to the angle between source and reading and estimate with distance. A lot of computation to go backward to intensities and then again to flux.

If a light source has a very narrow distribution, than you will get a very high value, but in a small area, this will lead to big errors unless you oversample the area where the spot lies.... obviously for a wide distribution will be different and perhaps easier...

But how to compare the two? Or how to compare two samples before knowing the solution?

The simpler way would be to build a test box, place the light source in it and take an HRD image of the inner side of the box. Take the same picture with the second light source.
Then divide the two images by each other and average. The average of the ratio should be the ratio of the fluxes... there may be some correction factor to be applied to account for the perspective view (i.e. there are more pixels to sample some areas than others) but this is something that can be easily implemented.

In order to start with this you just need a digital camera, a tripod and the test box.

Links here for how to:
http://luminance.londonmet.ac.uk/webhdr/
and
http://anyhere.com/gward/pickup/hdrgen_linux.tar.gz
and
http://anyhere.com/gward/pickup/photosphere_uni.tar.gz
and
http://radsite.lbl.gov/radiance/framew.html

I can only say "try", it does not cost anything and it is fun.
It is also science so...

G

 

[geotrupede]

and BTW.

light intensity is one thing
luminous flux is another
flux density even another
and luminance again another...

Brightness is not an objective term for anything.
Sometimes luminance is defined as the objective measure of the subjective sensation of brightness....

The intensity of a light is not really a correct expression as well.

I'd suggest to see here and check the names:
http://en.wikipedia.org/wiki/Photometry_(optics)

G