Warpspeed
Solar Wizard
This should not be too difficult.So I guess the question is, how can I build a more ideal diode here? I've heard that you can use PFETs (or perhaps NFETs) to provide a better solution here, but I am clueless...My area of expertise is batteries not FETs. Any guidance in the right direction is very much appreciated!
First requirement is a suitably fast current sensor to detect the direction of current flow. We don't actually need to measure the actual current, only detect its direction, so a fairly sensitive Hall device that has a low current measuring range might be best.
As long as the hole through the sensor is large enough to fit your really big fat dc wire, feeding 100+ amps through a five amp rated Hall sensor is not going to be a problem. The output just saturates harmlessly. As long as the wire passing through it can carry the required current, it will work fine with high sensitivity and especially stability around zero current.
Output from the hall sensor then goes into a voltage comparator to determine the direction of current flow. Even milliamps of forward/reverse current will produce a rail to rail output from the voltage comparator, and it will be fast.
An N channel mosfet acts as a diode between source and drain with the gate grounded. With a positive voltage on the gate it turns on hard and has a very low resistance in BOTH directions. Resistances of a couple of milliohms per mosfet (or better) are possible.
If you hook that up to your voltage comparator you get a dc switch that has a low "on" resistance, and can turn off in a few tens of microseconds if the current reverses.
A reasonable expectation might be each individual mosfet perhaps 2 milliohms, run at 50 amps = 5 watts dissipation per mosfet.
Four or five devices run in parallel, and they have a positive temperature coefficient, so they will current share quite well.