DImyself measurement of actual lumens (of a Fandyfire a10b) using a simple lux-meter

Having read quite some threads about measuring luminoux flux of flashlights, I was not satisfied with a)the costs and trouble of making an intergrating sphere, b)several methods that just measure the relative light output compared to other flashlights. So I started out with just a flashlight and a luxmeter and did an attempt to measure absolute output of the flashlight at hand: a Fandyfire A10b (with XML-emitter), with a freshly loaded ultrafire XSL 18650 2600mAh protected battery. The current at the tailcap was 2.9A. I believe this is a primitive version of what was called in an earlier thread the method of photometric goniometry (always good to give it a catchy name). The method and results are shown in the picture below:



fandyfireA10B_luxmeasurements by djozz, on Flickr


Some thoughts on what I did:
-It took me an hour to do the measurements, half an hour to do the basic calculations.
-I measured the average lux in discrete segments, thereby saving some integrating maths and also saving the need to account for the size of the lux-sensor (which was 60mm wide)
-of course this leads to some measurement error, especially at the edge of the central spot, where you could do more precise measurements, but I can think of quite some more experimental errors (like heating up the emitter during the measurements, or how well the lux-meter is calibrated) and I still have an idea this is a practical way to measure luminoux flux within let's say 50 lumens or so.

Any comments on this?..
djozz

Nice job.

Using a raw picture from a camera, in combination with a lux measurement could probably make this a lot easier (I you can make a program to sum all the raw values).

HKJ said:

Nice job.

Using a raw picture from a camera, in combination with a lux measurement could probably make this a lot easier (I you can make a program to sum all the raw values).

Click to expand...

What I quite like about the way I did it is that it requires a minimum of equipment and only little knowledge/use of mathematics and computers, just a little spare time (that we all have anyway since we wander around on the internet and make a fuss about the output of flashlights ). I wonder even if the use of camera plus computer would save you time (perhaps if you are planning to measure a whole load of flashlights).

Furthermore, every extra step away from direct measuring potentially introduces extra measuring errors. I am not an expert on camera's (there must be a camera-forum out there that has all the answers), but required to a certain minimum level must be a)a lambertian reflecting wall to point the flashlight at (an ordinary plastered wall will do the job quite well), b)a camera that evenly illuminates the ccd-chip (most modern camera's are ok in that), c)the spectrum that is picked up by the camera must be more or less the same as the spectrum used for lux-measuring (which is not the case, camera's pick up at least some of the infrared that is added to the visible light in the image, lightsources that also emit a lot of infrared will be overrated in lumen-output).

It may very well be that the errors introduced by using a camera-image to add up the emitted light are minor compared to the errors made by measuring a limited amount of lux-values, than this can be useful alternative.
djozz