Protocrash,
I read your reply with interest, and I like it. I think everything you say is very credible, and I really don't think I need to spend a lot of time doing my own research and bench-top testing. Thanks.
I'm not an expert by any means, but my engineering and video production careers did sometimes involve batteries of several different types and capacities. So, I do know, for instance, that capacity ratings are invariably variable due to various factors, two of the more important often being discharge rate and temperature. Thus, actual usable capacity is normally less at higher discharge rates due to some extent to chemistry, but I think mainly due to internal resistance losses (the good old P= I^2 x R ). Also, capacity is very much diminished normally by lower temperatures.
So, it doesn't surprise me when you talk about how difficult it is to control the discharge cut-off point with great precision, using such a simple voltage-only scheme. But I also suspect that your scheme should be good enough to hit the "correct range."
So, I tested my Angel using one, nearly new DJI battery with a voltmeter connected to the leads. I charged and discharged it using the Angel three times. Summarizing and averaging:
Fully charged at rest: 4 LED's, 12.5V.
Start Discharge: 0 minutes, 4 LED's, 12.3V
11 minutes, 3-1/2 LED's, 11.9V
21 minutes, 3 LED's, 11.6V
Auto Shutoff: 25 minutes, 3 LED's, 11.5V
Right after shutoff, rebounded at rest to 11.6V, and after about ten minutes 11.7V.
After shutoff, the LED indicators show 3 LED's remain lighted (no load).
If you look really closely at the DJI chart, you will see the red curve actually starts at about 12.3V. This corresponds favorably to curves for one cell at 1.0C discharge rate, times 3. The DJI flight battery I'm using is rated at 5200 mAh (57.7 W-h). The point is, give or take and allowing a little latitude in the numbers, the Angle with a 50W bulb discharges at about 1.0C rate. And if you go back and look at your chart and the 1.0C discharge curve, you see that my measurements show that the Angel cuts out at about 57%. (GIMME SOME LATITUDE HERE---my measurements are pretty good, but I'm rounding off some of the data and eye-balling the charts, etc.
So my conclusion is that you're hitting your intended 50% - 65% cutoff range pretty much in the middle. That's good. If it does anybody any good to know, that's with 3 full LED's shining. The 3-LED indication is pretty rough because 3-LED's shine across a pretty broad range, but it's better than not knowing anything.
So Protocrash, you get a good mark. I think your simple and very economical device is a great value and a everybody should consider getting one.
BTW, I LOVE Greece. We were there for a couple of weeks in 2009 as part of my daughter's college graduation family celebration. We rented cars and did our own touring and "our own thing." We spent five days on Santorini, and the rest of the time in and around (day-trip range) Athens. Love the Greek people. Y'all know how to live. Love the food. I just hope you can get your fiscal house in order.
Thanks for everything.