Issues Charging LiFePO4

[David_CA]

About a year ago, I put together a battery setup for camping, mostly based on reading posts on these forums. For this setup, I use a couple 100 watt panels in series, an Ampere Time 100Ah LiFePO4 battery, an EPever Tririon2210N controller with MT50, and a 24V 15A power supply feeding the charge controller's solar input when charging at home. This setup has worked perfectly for numerous camping trips. This past weekend, I didn't take panels, since we were pressed for space, and the battery never dropped below 45%.
When I got home, I noticed the battery appeared to charge much quicker than normal. Instead of actually charging, it had stopped at 45.4% with a battery voltage of 10.5. When I first connect AC to the power supply, the MT50 shows it charging at 14.6V 10A for a couple seconds with the solar input (power supply) at 24V. Then, the output voltage from the power supply still shows 24V but zero amps. Also, I noticed that the controller LED that's normally green for the battery is now red, likely due to the low voltage, I assume.
I've verified the input voltage to the controller and the battery's voltage with a separate meter. I used parameters I found here to set up the controller, and they are listed below. Basically, I need some help troubleshooting this and can certainly do any testing necessary to help with this. My first thoughts were that a setting is wrong, and the battery is at a lower state than normal, but now I wonder if the power supply is at fault. I don't have a spare power supply but can order one if necessary. Thanks much in advance!


Over Voltage Disconnect 14.8
Charging Limit Voltage 14.6
Over Voltage Rec. Voltage 14.6
Equalize Charging Voltage 14.4
Boost Charging Voltage 14.4
Float Charging Voltage 13.8
Boost Reconnect Voltage 14.4
Low Voltage Reconnect Voltage 11.8
Under Voltage Warning Rec. Volt 11.8
Under Voltage Warning Voltage 11.5
Low Voltage Disconnect Voltage 11.1
Discharging Limit Voltage 10.8
Equalize Duration 0
Boost Duration 180 mins

 

[timselectric]

I assume that the BMS is disconnecting.
Due to one of its alarms.
Possibly cell imbalance.
Can you access the BMS data?

 

[David_CA]

I assume that the BMS is disconnecting.
Due to one of its alarms.
Possibly cell imbalance.
Can you access the BMS data?

Unfortunately, it's sealed, so I can't get into it without destroying the case, which I might have to do.
One question about BMS being disconnected... would the cells output anything? They're at 10.5 volts presently.

 

[timselectric]

Discharge and charge are on two different sets of mosfets. So they are independently controlled.

 

[RV10flyer]

It did not get below 32F did it? If so, you may need to warm it back up. If BMS is bad, throw an Overkill Solar in there.

 

[mikefitz]

I assume the battery state reading of 45% is from the Epever controller. Since this is voltage based it is a poor guess and cannot be relied on.
At 10.5 volts the battery is depleted and its probable the BMS has shut down. If the battery is under charge the values of 14.6 volts and 10 amps for a few seconds, suggest something is wrong, your charge voltage , boost volts, is set to 14.4 volts, thus you should never see 14.6 volts at the battery.
Personally I do not approve of using a low cost power supply to power a MPPT controller, there is a high chance of damage to one or both units.
Each component part of the system needs to be tested independently, the battery, the Epever controller, and the power supply. Since you have the panels available that gives you an option to test the controller, perhaps with an alternative battery (car 12v battery)

To wake up the battery, if its 'gone to sleep', a power source of around 12 volts with a few amps capability, connected should do the trick, a car battery or simple car battery charger. Many solar controllers cannot do this 'wake up' .

For charging the battery from AC a quality battery charger , idealy a unit with user settings, is recomended.

Mike

 

[David_CA]

I assume the battery state reading of 45% is from the Epever controller. Since this is voltage based it is a poor guess and cannot be relied on.
At 10.5 volts the battery is depleted and its probable the BMS has shut down. If the battery is under charge the values of 14.6 volts and 10 amps for a few seconds, suggest something is wrong, your charge voltage , boost volts, is set to 14.4 volts, thus you should never see 14.6 volts at the battery.
Personally I do not approve of using a low cost power supply to power a MPPT controller, there is a high chance of damage to one or both units.
Each component part of the system needs to be tested independently, the battery, the Epever controller, and the power supply. Since you have the panels available that gives you an option to test the controller, perhaps with an alternative battery (car 12v battery)

To wake up the battery, if its 'gone to sleep', a power source of around 12 volts with a few amps capability, connected should do the trick, a car battery or simple car battery charger. Many solar controllers cannot do this 'wake up' .

For charging the battery from AC a quality battery charger , idealy a unit with user settings, is recomended.

Mike

I'm curious as to how feeding 24 volts into a controller is different than 24 volts coming in from a couple panels, or is the issue the 'low cost' part? I did set the boost to 14.6 from its normal 14.4 in an attempt to see if it'd make any difference.
Are you saying to connect a 12V battery to the battery in question's terminals and then put it back on the charger? I've had to do something similar to a DeWalt tool battery that had been run down and left for an extended period.
As to temps, the lows were 59, so no issue there.

 

[mikefitz]

Your panels are a current source, current limited to a maximum, the panel I sc, a low cost power supply may not have effective current limiting.
Re 'waking up' a BMS. The power source is removed after a brief connection to the problem battery.
The cells in the battery need to be in perfect balance to charge up to 14.6 volts. If BMS is shutting down due to cell overvolt protection, lowering the charger voltage, Epever 'boost voltage', may be sucessful. Any voltage 13.8 or higher will charge.

 

[David_CA]

I assume the battery state reading of 45% is from the Epever controller. Since this is voltage based it is a poor guess and cannot be relied on.

I forgot to address this. The 45% is from an ammeter / volt meter that measures all current into and out of the battery. When I initially set it up, I charged the battery completely and set the meter at 100% per instructions. I don't recall if the controller showed the same percentage or not, as I don't use its display for anything other than panel voltage and current and battery charge voltage and current.
What's odd is that the 12V refrigerator's display never showed displayed under 11.8V, as I recall. The voltage displayed on it is maybe .2 or .3 volts higher than what's my multimeter shows, based on my checking to see how accurate it is previously.
I suppose that once this issue is resolved, I need a method to disconnect the battery once the voltage drops below a preset threshold. Most of what I use is 12VDC, so there's not an inverter in the mix to alert and disconnect on low voltage. Other than using the load outputs on the Epever controller to trigger a relay, is there such a device that will do this?

 

[David_CA]

Update: The issue is resolved, and I'm almost embarrassed as to what caused it. Apparently, the fuse holder I used was not up to the task. Thinking back, I seem to recall using it but was planning to replace it with a better one, which never happened. This fuse goes between the charge controller and the battery. As one can tell from the pic, there was far too much resistance between the contacts and the fuse, and things got hot. The battery and controller were fine all along. I replaced the fuse holder with a nice, heavy duty Maxi Fuse holder and fuse, and all is well now.

Back to draining the battery... I use mostly 12 volt devices. Is there a good option for a low voltage disconnect that will prevent me from running the battery too low? An option is a relay connected to the charge controller's load output, since that has a threshold that can be programmed. I'm not sure if this is the best option, as the relay will obviously draw current itself.

Thanks to those who responded. Even though the issue wasn't the battery being too low, it was a learning experience, which is always a good thing!