Battery problems and LiFePO4

Author	Message
KeepIS Guru Joined: 13/10/2014 Location: Australia Posts: 2232	Posted: 11:23am 11 Jan 2024         Quote   How I see it with the Charge MOS off, the BMS is like a big Diode and the battery is still accessible to the loads when required. Hmm, I wonder if that makes any sense.     Also when the charge mos switches off it reads or triggers 100% SOC, probably some other settings going on, most of them are still set at default. That is how I see it, likely the body diodes. Edit: Of course it is, and utilizing FET body diodes as part of the Charge/Discharge switching design. The change to set 100% trigger is the pack-voltage setting, obviously they want you to set that trigger point to what might be a 100% charge, which is in reality a very rough guess. In my case, the cell voltage trip is set to 3.5v and the pack voltage is set to trip when each cell is 3.4v, so the battery pack trip is 54.4v. The cell voltage should normally never trip in mine, only the pack voltage. Solar charge voltage is set to 55.4v (1v above the Pack MOS trip) so there is no short term cycling, and any slight sag will not cycle the MOS switch. I've had it running like this all day, once batteries were 3.4v per cell, the charge MOS switched off and only came on twice for a few seconds the rest of the day. It smoothly switched on as the sun went down. That's two 48V batteries running in parallel and tracking each other perfectly with their respective BMS MOS switching. Basically, once the charge MOS switches off in each BMS the battery banks spend the next 7 hours disconnected from the charging. It requires Charge controllers that don't over-voltage with sudden load disconnects or load variations. An Inverter that had good CAPS and does not rely on the low impedance of a battery source for its transient response, and one that has a toriod wound for a wide DC voltage input range to hold good AC voltage regulation under heavy load and DC sag, without stress. I use four 60A Make Sky Blue chargers (lot of people rubbish them) I purchased these for $170 each 4 years ago and were a bargain IMHO, and have worked faultlessly. So now, onto the task of implementing it correctly for SOC tail current and charge cutoff. . Edited 2024-01-12 13:11 by KeepIS