"The life of a cell is maximized by operating as nearly as possible to a fully charged state. When one...paralleling...and so operates closer to fully charged."
A perfect example of fallacious logic! You are confusing the two issues, of discharge depth and parallelling. By using the correct cells, you can accomplish the same low-discharge-cycling WITHOUT paralleling. If you want to make a proper comparison, the comparison would be using a series of cells (a simple battery) versus a parallel array, of the same total capacity. In which case you'd find the cycling effects from the parallel setup to have no advantage at all.
Remember, if you have two batteries of "x" amp hours each...You can still cycle each one separately to 20-30-50% depth and still get the same total number of amp hours in and out of them--regardless of whether they are paralleled, or used sequentially. The depth-of-discharge argument holds no electrolyte here, if I may coin a pun.
Regarding boost charges and such..."Other special charge cycles are possible but not recommended unless you can monitor temperature, outgassing, current draw, etc." All of which can't be accomplished properly with parallel batteries, since there is normally one charge voltage sensor, one battery temperature sensor, one &cetera and now, the question is which battery or bank do you hook it up to? And of course, that's all stuff that the smaller systems using automotive alternators just aren't going to have.
"This is from both theoretical and empirical data. and the only time I did not get these results some other factor was present" And what other factors did you get? Factors related to the parallel construction, or factors that might, again, affect it differently?
Let's look at another issue that has been ignored. Cell failure. Whatever the odds of cell failure are (1:1000? 1:100,000?) for any individual cell, once you start using parallel batteries, you multiply those odds. If you have six cells, each with a 1:10,000 chance of failing, you now have a 6:10,000 chance of having a battery failure from a bad cell. But, if you parallel two such batteries...you've doubled the odds of having a failure, to 12:10,000. And one cell failure will still take down both batteries as it starts those current loops and failure modes that you say can be ignored--except in the case of a bad cell.
Keep the batteries separate, and use them sequentially, and your risk of failure is now effectively halved, since you've got redundant and separate banks. Use them in parallel...and all your eggs were in the one basket. I can't call that a good idea. Or do you dismiss that?
Last edited by hellosailor; 09-05-2006 at 04:52 PM.