Any good ghetto methods for comparing emitter lumens?

I want to compare LED emitters that I'm in the process of acquiring to determine quantitatively which throws off the most lumens (within, say, plus or minus 5%). I don't want to measure just center brightness (even though it would be easiest to measure).

So, my first thought was that I would need some kind of integrating sphere (like the kind used by spectrophotometers), with the LED emitter at one end and a measurement sensor on the opposite pole (so to speak) so as to funnel as best I can all the emitted light into the sensor. Is that how professional measurements are done? If so, what sort of measuring tool is that called?

Anyhow, I'm guessing a proper rig (with baryium sulfide coating the inside of the sphere) would probably cost $100-$200 or more if purchased, and my modest objective doesn't really justify that level of expense.

For less money and more measurement effort, I could setup like jtr1962 did in his awesome posting: http://www.candlepowerforums.com/vb/showthread.php?89607-White-LED-lumen-testing

However, is there an easier/faster ghetto method for comparing lumen outputs? Different model emitters will probably have different beam spreads and light distributions, so I'm afraid just "eyeballing" it won't be accurate enough. For instance, would using some kind of either colluminator or diffuser reformat the light distribution into something more standardized so that then maybe different emitters could be more easily compared (perhaps by then eyeballing them--maybe using sunglasses for safety?--side by side)? If there is such a thing and if it's something digikey might carry, I could order it at the same time as the emitter samples.

If all else fails, I'll trust the spec sheets. I'd rather not though, as I'm not entirely convinced it would yield a fair apples-to-apples comparison since some spec sheets might be more honest than others.

Here's how a pro does it: http://www.youtube.com/watch?v=mvyptpA-BmY

As you can see, it's not a trivial setup, and I bet the equipment cost is a lot higher than my initial estimate. The way he describes it, it sounds like you need a spectrometer to grab the data in order to convert it properly into lumens. I have an inexpensive lux meter (http://www.amazon.com/review/R2AZ8UESLBC4V/ref=cm_srch_res_rtr_alt_2) which I thought might figure into a ghetto solution, but now I wonder whether it even does the right type of measurement? The engineer in the video seems to imply that to calculate lumens you need to do some kind of weighted sum of the light intensity at different color wavelengths.

Whew! This looks like a great ghetto "lumen" meter: https://www.youtube.com/watch?v=3QO7-jZEfqk

Unless someone here can suggest something better, I'm going to roll with that design. It makes more sense to me than the ceiling bounce method discussed elsewhere on CPF as a poor man approach.

Apparently it's called a LumenToob, and it had its genesis on this very forum: http://www.candlepowerforums.com/vb...ate-Lux-amp-Lumens-with-Light-Meter-LumenToob

Unfortunately, the photos that were posted originally are gone.