[ QUOTE ]
evan9162 said:
What really needs to be done is to charactarize a Pila cell at various current levels to see what its terminal voltage is at those current levels. Then you can definitively see what kind of danger you will be in with DD.
Lets say the Pila maintains 3.7V at 900mA (mid-life), and your Luxeon's initial Vf is 3.7V at 900mA. Then for the majority of battery life, the current will be around 900mA, excepting the first few minutes (which is really the most critical, since it's when damage can be done).
Anyways, if the Vf drops by 0.3-0.4V (which I've observed with K Vf luxeons running at 1A), then the Vf of the Luxeon will be around 3.4V-3.3V at 900mA, and 3.7V at about 1.5A. Now, you won't get 1.5A, because the extra current will pull the cell voltage down, but, you may get around 1.2-1.3A, which may not sound like a whole lot more, but consider that your design just shifted by 40%, and power dissipation in the emitter went from 3.33W to 4.5W (+35%). Even with perfect heat sinking, junction temperature jumped from about 70C to about 85C (when the body is at room temperature). Initial current has changed in a similar manner, which means with a fresh charge, you could be slamming the luxeon with nearly 2A, for about 8W of dissipation, for an initial junction temp of 130C (5C below absolute maximum)
So you really should characterize your power source, and calculate the conditions of the worst case scenario. So that means measuring cell voltage at various current levels, both freshly charged, and about half discharged. Is this something that SilverFox has done with his battery testing?
Just some food for thought.
[/ QUOTE ]
Yes, but re-do the numbers for an aged emitter, and with heating of the flashlight. Also consider a K bin has what, a 0.3 V range to start with....Try it with emitters on the low end of the bin, dropping the Vf for age and for temperature...