Solar charge controller, full design.

[Codebird]

I've been working on this for a long time, and the point has come where I need feedback. I have a design put together, and a prototype built, for a solar charge controller. It's PWM, not MPPT, but it's a pretty impressive PWM one.

 - 20A handling easily, both charge and load, and able to handle as much current as you want. No P-channel power MOSFETs here.
 - Not a single obscure or SMD-only part, good for enthusiasts.
 - For 12V lead-acid batteries.
 - Logging of charge and load current and battery voltage to serial output for analysis and monitoring of battery health.
 - True two-stage charging process, controlled in software so you can enter the optimal values for your battery.
 - Automatic load disconnect on flat battery, and reconnect when charged.
 - Automatic disconnect of load on over-current condition.

https://ipfs.io/ipfs/QmU4GJQvMNpc2W7WYVf65LdH52N9CLcbLYwDCLG6KUcXQQ/

I'd like some evaluation. I've got one built to this design in my garage right now, and it's performing a lot better than any of the cheap PWM controllers I got off Amazon. I'm just working on refining the voltage regulation now - need another 100W panel up so I can test it at full charge current.

[fourtytwo42]

I am sorry to be 3 months late! I dont know much about commercial charge controllers but I do know about power conversion. Looking at your circuit I see no inductors, so no energy storage and no voltage conversion.
Am I right in assuming the PV voltage is forced to be the same as the battery ?
If so the panel will only operate near its MPPT for a small range of insolations corresponding to the equally small range of battery voltages.
I can see the merit in this circuit preventing overcharge and that in itself is very important but it's PV performance without any MPPT will be woeful, but then I would not be surprised if that is the case with cheap commercial products also.
Perhaps in the last 3 months you have moved on or this is adequate for your needs, I move on all the time

[Codebird]

You are correct - like all low-cost controllers, this does not do MPPT. I actually used an external, separate unit for that. It's really an economic thing - this charger is designed for 200W. An outbuilding, a cabin or a mobile home, not a house. At that scale it's more beneficial to spend money on just getting an extra panel than on MPPT.

The design could be adapted for MPPT, but it would greatly increase the cost.

[fourtytwo42]

Thank you for explaining You are right about cost, something I always keep in mind otherwise it's easy to end up in a situation where payback wont occur in your lifetime There are some silicon solutions for MPPT from TI and others, I saw someone here was promoting an open source hardware project but it seems to have been abandoned.