yes but how do you get the thermal protection TO a battery that is hanging out the side, that is the issue, not that the chargers or power adaption, has thermal protection but the battery is monitored, so if/when when the batteries overheat the computer can react. the thermal probes are very close to any battery that is in the original battery slot.
20amps, yes that is right the 9000 does have the capacity thing covered, so you don't have to keep recycling the battery back on.
if it was OHMS of resistance it wouldnt be a problem, heck some 1000feet of copper wire are only a single ohm or so, so i dont use that kinda math stuff, its the voltage drop with any of the connections and wires, and any looseness or burping or minor changes in those connections. A mere foot of say 22ga wire can change a power curcuit by a lot, its the reality not the math, to try and quantify the small ammount or determine a limitation would be a waste of time, the Rue Goldberging of the charger would just be done as well as you can do it, then it will work.
the way it comes batteries lock into the metal, solidly, the metal is connected to the board, with like 0 resistance, calibration and testing and programming junk was all done with the intended connection that doesnt get a inch away and is all in copper foils on the board. it would not TAKE much variations from that to have a variation to the Brain, not because your adding in large resistance, but because there is so little to begin with.
so you would try and keep all the wires short , of fair ga. size, of good copper quality, all conections very solid, nothing bumping things, or rattling them, ya know how many things can rattle a connection that is not perfect. within the capability of still accomplishing your goal, not nessisarily going overboard. so the brain still can do its testings and comparisons
a commonly sluffed off voltage drop of say .02volts would (for example) raise the 1.47 to 1.49v a huge difference in very carefully selected voltage termination. (assuming your model has this feature) that is why there should be as little resistance as possible.
a variation in the connections of mere milivolts used to determine the v-drop would effect any termination based on V-drop, which the computer samples and resamples and resamples, with PULSES of power occuring, pulses that can be changing any connection that is not very well connected. that is why no connection should "Change" in resistance at all.