Been awhile I have'nt posted but you could use some guidance here.
The alternator, it is not that efficient cause it lose power thru the rotor coils. Still, this is acceptable as it is rugged and potent and in your case you may be able to waste a few watts here. Many hydro has been built with those.
The charge controler,tough, is not efficient and does'nt charge appropriately for a homebrew system. This controler will just reduce the load on the alternator when it's not charging. This would mean impeller overspeeding and explosion from centrifuge forces as the water speeds it up and up and up if there is no load.
Thus, you need a suitable controler so that the less you charge, the more you dump in a load. This is not the same as just switching the load on and off to charge and no charge as in windmills. This usually involve Pulse Width Modulation. This mean you switch the lenght of the Charge Time and DumpLoad times in a cyclic fashion quite high (usually around 10,000 times per sec).
You will need to find or hack a charger that sense battery voltage and vary charge time in consequence. You don't want to dump 50A in an almost full battery. The Off time must be diverted to the load in the circuit. Should be pretty easy to come up by some electronic wiz. I might design one in the near future as I will build a micro-hydro at my creek.
The other ways is to make a centrifugal vane so that water deliver get reduced as the speed get higher. So that you'll limit the speed of the impeller. This essentially is one of the biggest catches, but really worth it as hydro is full-time ON and quite powerfull. For mechanically inclinded, centrifuge van is not that hard to do. You just have to keep an eye on it.
Btw, marine deep discharge batteries are the winner as for availability and fit pretty well the need. More is better than less, especially if you have 50 amps comming your way all the time... Since you would have the capacity to keep them pretty well charged with a good hydro, you can take the peaks without hurting them if you have a few.
Think well about speed or charge/dump control before spending 200$ on an impeller. Prepackaged solutions are one venue, but homemade is so rewarding...
Well, good luck with the system.
Para