I saw a circuit the other day that was interesting. It was a standard buck converter with a high side driver except there was a second identical FET in series connected backwards to act as a switch instead of a diode. This was also driven by the high driver through a diode. I thought it was a novel idea because it used the gate capacitance to keep that FET solidly on while the other FET pulsed. Gate drain resistance was a very high 470K. A neat trick to keep in the back of your head.

What really interested me is the reverse current flow of several amps through the FET. It must work but I wonder if the ON resistance the same in the reverse direction. Just arrived at camp after being away for 7 months and I got bigger problems right now. I'll be trying it soon though. Just wondered how many circuits you guys have seen with a reverse connected FET. I am a little suspect of the circuit though. Even if it works I'm thinking there will be enough leakage through the high side driver to turn on the FET when the power rail is at zero and there is a battery supplying 12V reverse current. Here is the circuit http://www.timnolan.com/uploads/Arduino%20Solar/Ar...

Yep, the channel resistance from S -> D is the same as from D -> S in a MOSFET.

So Q1 is being used to protect against reverse polarity connection of the source.

Is indeed a neat trick that I will keep in mind. Prevents discharge from battery to panel without having to cop a diode drop in normal charge mode.

Certainly found the backwards FET unusual, but on consideration, isn't that exactly what Q3 is doing? And Q3, being the synchronous buck converter low side switch, is a very common arrangement. Now that I think about it, I believe Ron is the same in either direction, and that either direction is perfectly acceptable for a FET.

I share your concern about the battery being able to turn on the gate when the solar is at zero (this is what happens in my simulation). I think it might only work if you signal the shutdown pin to disconnect both gates.