What is the best way to balance cells in a 12.8V LiFePo4 battery with a passive BMS?
A friendly hello to the forum community,
Almost a year ago, I switched my small off-grid PV system from VRLA/Gel to a LiFePO4 battery with a passive BMS. Since there's not much solar harvest in the winter, I'd like to manually balance the cells using a small lab power supply.
During the "brighter" season, the LFP on my PV charge controller normally runs continuously with a charging voltage limited to approximately 13.9 to 14.0 volts.
However, after extensive research, I still don't quite know how a passive BMS performs cell balancing, even if this control voltage has been applied to the battery for hours with a barely measurable charging current of < 100 mA. (Without loads)
Do I then have to increase the charging voltage with an external power supply to compensate, or does cell balancing with a passive BMS already take place very slowly over, for example, 48 to 72 hours at approx. 14.0 volts constant voltage and < 100 mA charging current and < 100% SOC level?
I would be very grateful for recommendations from users with solid basic knowledge of this subject.
A laboratory power supply with adjustable voltage and/or current limiting as well as several multimeters for current and voltage monitoring between the battery terminals and the power supply would already be available.
Kind regards and thanks in advance for helpful information.
(1 votes)
Loading...
The resistors allow the current to flow past the respective cell.
Here: https://de.wikipedia.org/wiki/Balancer
The battery is charged, you can see that the cell number 3 is already full, so you let the current flow past the cell #3 with the help of the resistor, so that it is no longer charged. However, the other cells are still charged.
However, the system with the resistors often does not work really well because the charging current and the value of the resistor often do not agree, the charging current is often too high and the resistance value is often too large, so that there too low a current flows past the respective cell.
Better are capacitive, active balancers, which can compensate the cells exactly up to a few millivolts.
Unfortunately, the battery cannot be opened, but an adjustable laboratory power supply is already available. At present, the battery is connected to this power supply by the CC method for a charging current of ~ c/100. The solar charger is currently not connected to the battery electromechanically.
At the moment, the whole of the 0.07A fixed is about 13.9 volts fed into the laboratory power supply. I also measure voltage directly to the battery terminals via a separate multimeter. Maximum voltage at the laboratory power supply is currently limited to 14.3 volts, charging current to 0.07 ampere max.
I continuously measure charging voltage from the battery terminals using a separate multimeter.
First of all:
Can you retrieve cell voltages from the BMS via Bluettooth or UART-USB adapter?
There are very simple BMS that cost 2.50€ and there is not really much on it. A small chip, but no output of the data.
Then there are the better BMS, which have a higher range of functions, because then you can also look at the measured values (current, voltage of each individual cell) and enter new limit values.
Normally, the balancer becomes active when the battery is just charged and at least one cell has exceeded a cell voltage of 3.45V. But that doesn’t have to be the case with everyone, so everyone cooks his own sweetheart.
Right, the absolute charge-fit maximum is between 3.6 to 3.7V, depending on the cell architecture. From this, my passive BMS also turned off perfectly at 15.0 volts measured battery terminal voltage during the charging process. This point I could only find exactly with the laboratory power supply since my (lower) PV charger works too inaccurately.
My BMS performs the cell compensation only in the area of the top level balancing just before reaching the charging voltage. With me, this balancing starts at about 14.2 to 14.3 volts in the ongoing charging process. I could then measure this very well at a slight heating at some points of the battery cover. (only about 1-2 degrees more)
The passive BMS is contained at work within my LFP battery, but I cannot open the battery myself. Now I know how and when the balancing works on my battery.
At the moment I measure a quiescent voltage of approx. 13.38 to 13.39 volts with my small multimeter. The charger pulls out of the battery in the standby about 50 to 60 mA, but since yesterday this controller has been without input. But would probably still fall a little further to about 13.32 to 13.33 volts if I would leave the controller further without input in the standby.
Island is now full, battery balanced, and so everything goes to PV output on the Ongrid inverter. In this state, the island makes Winter-Bubu.
For the PV recharging of my island, I have set 13.9 volts on my controller, which can meander between about 13.8 and 14.0V in case of alternating weather at the battery terminals. (is just a cheaper ReVolt MPPT 10A)
I set cutoff to 12.6 volts.
15V clamping voltage? That’s 3.75V per cell. 3,65V is the absolute maximum.
I built my own system here to measure the tension very accurately. I had bought BMS at that time, but it’s actually so easy to build it yourself and also better.
I have a simple ESP32 board with which I present the website and list the readings there. But I might put that on the RaspberryPi, there is more possible. I will be able to determine the measured values via an STM32 module and this will also turn everything into the energy saving mode (DeepSleep) if there is a problem.
The tensions look like this in my case:
Voltage LiFePO4 battery with 16 cells in series:
2,5V*16=40V (remained empty)
3,0V*16=48V (empty)
3,2V*16=51,2V (ideal empty state)
3,3V*16=52,8V (Nominal voltage)
3,45V*16=55,2 (ideal full condition)
3,65V*16=58,4V (full – should not be reached too long)
My battery has only one passive BMS. This functionally only switches the cell(s) with the respectively highest charging level during charging via a recurrent circuit in the vicinity of the charging-circuit voltage, so that this cell is then removed in the compensating process max a few dozen to a few 100 mA current for conversion into thermal energy. These cells are thus braked so that the cells with even lower charging levels can pick up accordingly (slowly).
(So far I have been able to find out more)
Over night, I had my battery charge at 120 mA for set 14.0 volts at the laboratory power supply. The terminal voltage in the morning was about 13.85 to 13.9V per multimeter at the battery terminals.
After that, I switched to CV mode and in 1/10V – steps still a few hours slowly falling between about 200 mA initial current, up to about 100 mA to 14.4 volts. In the last of these 100 mV steps, the BMS then switched off at approximately 15.0 volt terminal voltage.
Thank you so much for your helpful support. 🙂
You will get the star as soon as this rating is unlocked. 😉
LG