A special circuit is needed when charging a battery from a Solar Panel. When the solar panel is not providing any power the battery might start draining current into the panel. One common solution is to have a diode in series with the charging circuit to keep the current from going back to the solar panel. The problem with using a diode is the voltage drop across the diode reduces the available voltage for charging.
This circuit is great because the LTC4357 prevents the current back feeding into the panel without having the voltage drop limitation of the diode circuit
Via Electronics Lab.
There are some clever and low-part-count ways of doing this without any specialized controller. The LTC4357 looks really cool and is almost certainly ideal for the use, but you’re paying a lot for the operating voltage range compared to what you can get with a couple fets. The other thing is, you lose a lot by simple switching of a solar cell into a load like this. a switching controller, perhaps even maximum power point tracking, isn’t much more expensive for a huge performance increase.
Maybe I’ll reply in blog-post format when I get a chance, there’s a lot to discuss here!
Please do Alex! I have a few solar panels I’d like to put to good use.
I agree that an MPPT regulator will give the biggest improvement over a conventional regulator (PWM). I calculate 15% for a 12V system. Also, if you are using an array of panels in parallel, you should have series diodes in every panel to stop back-feed into shaded / dirty / broken panels. I have measured the back feed and it can be severe. 1 amp from a fully illuminated to a fully shaded 80 watt panel. Traditionally Schottky diodes are used, but these will drop 0.5V which in a 12V system is a wastage of 3%. Ideal diodes would waste almost zero, so by using a combination of MPPT and ideal diodes in a 12V system there is a saving of about 18%. I want to try the LTC4357 which looks like a very simple device to use.