What Don is saying is that it doesn't make sense to have a 500mA light built to run it on a li-ion that will cause the light to run at a current level in excess of 700mA. If you intended to run at 700mA with the li-ion, you may as well have built the light with a bias of 700mA, or more.
The "math" that I'm referring to is this, and it all assumes a worse case scenario. For arguments sake, let's say that you have a J Vf luxeon that falls at the bottom of the Vf range and its actual Vf is 3.27v. The NexGen will tolerate an over head of .7v. A battery will experience a voltage drop of about .5v per amp of draw. Let's also assume that a fresh li-ion has a voltage of 4.2. As I said, we're assuming the worst.
With a NexGen400, the converter will draw about 600mA from the battery, so assume a voltage drop of about .3v. Of course it will draw more than that from a li-ion cell. That puts us at 3.9v. Then consider the Vf of the luxeon as being 3.27. So, 3.27 + .7 is 3.97 which makes the overhead within tolerable limits, but it begins to get a bit muddied here.
This is because we are within safe tolerable limits as defined by Wayne, by .07v, which isn't much. The danger comes with the fact that li-ion cells do not drop voltage under a load like primary cells do.
OK, so I have to admit (and have always done so) that I am not a math nor electronics wiz. If fact I'm an electronics idiot.
As I said in my previous post, as a practical matter, I've never fried a NexGen by running one on a li-ion cell, although using a li-ion does get more current to the LED than intended by the bias of the converter.
With out making any warranties or claims, I'll say this. Running a boost converter with a li-ion cell is nothing new. I (and I'm sure that many other have also) began running boost converters on li-ion cells almost two years ago. The way that I stumbled on it was when I found that a DownBoy (buck converter) would not run in regulation (in most cases) on a li-ion cell. I just figured that if it wouldn't run in regulation on a downboy, a badboy must be in order. The BadBoy NexGen is exactly that. The next generation of badboy converters, so I've continued to run the new converters on li-ion cells.
I'm just not willing to warrant that it will work.
As Don said, it just doesn't make sense to build a light to run at 500mA and then force it to run at a higher bias by using a li-ion cell. If you want it to be brighter, then bias it that way. Then you'll get a predetermined output no matter what you run it with.
If you want to take full advantage of a li-ion cell and want to run a light at a more conservative current level, then go for a Wiz2 board.
As for the NG500 being run on a li-ion, I've also had several reports from others that it's working out for them. I've even had one person tell me that they've (definitely not recommended) been running their NexGen LE on 2X123. The NexGen may be a tougher board than it's gotten credit for. I know that DDing a lux III on 2X123 kills it pretty quick, so I don't know what's going on there, but the point is that the board didn't fry.
The bottom line is, "so what do I know?" We all live and learn through trial and error and experience.
