Michael writes: <<<It seems there may be some discrepancies between theoretical extremes and real-world experience. Here's a link to one of the forums on which people are reporting on their results:>>>

I would caution that taking anecdotal "test" reports published on a blog as fact is a bit of a risk. Having a guy by the name of "Tostino" report he did a "few" discharge cycles with no observable imbalance tells us nothing. It certainly is not enough to contravene the battery manufacturer's published recommendations. How many cycles? What was the temperature? What was the rate of discharge and the depth of discharge? How was it measured? What was the rate of charge? What was the charge temperature? Did the discharge rate vary (like in an aircraft application)?

Hamid, on the other hand, has actually designed multi cell Li-ion and LiFePo power systems. One system was for Liquid Robotics http://liquidr.com and featured seven 6-cell packs that powered the electronics and payload of the Wave Glider (winner of the WSJ Innovation of the Year award). Photovoltaic panels were used to charge the batteries in sequence during sunlight hours. The prototypes we built for Liquid Robotics have logged thousands of charge - discharge cycles during several trans Pacific crossings and deployments from Baja to Alaska without any failures.

Reliability is, in part, the result of careful consideration of theoretical extremes. This frequently produces favorable real world experiences.

So, believe who you want, a guy who has 15 years experience designing scores of high reliability electromechanical systems including certified primary flight displays, Electronic Flight Bags and power systems for AUVs or a guy named Tostino you met online.

Oh, and here is another clue that implementing a Lithium battery power system is non-trivial; most cell manufacturers won't sell you the bare cells until they qualify your safety circuit.

Regards
Brent Regan