I have a domestic battery bank of 3 Trojan 31 XHS batteries purchased February 2016. They are charged by a Victron multiplus at the recommended charging voltages for Trojan batteries with Trojans recommended temperature compensation. Fortuitously I monitor the battery state remotely using a raspberry PI running Victorns Venus software which connects to a Victron BMV 700, the solar charge controller (MPPT 75/15). Last week I noted the current rising and the battery temperature increasing. I went to the boat and found that one of the batteries had failed. Once disconnected from the charger it read under 11V. Fortunately the other pair seem OK reading about 12.8V fully charged with no load after a days rest.
In the remaining batteries the plates are clearly distorted
The boat is not in daily used and has suffered very few deep discharges. It is typically used about 8 weeks per year so the cause of failure is not deep cycling the batteries. The battery bank is used to drive the inverter in the multiplus. When it does this the current is typically up to 100A. Could this be the cause of failure or is something else in my use causing a problem.
The multiplus on on and it follows Trojans recommended prescription for i.e.
Max charge current 17A per battery
Temp comp -30mV/deg C
Repeated Absorption time 1hr
Repeat Absorption Interval 7 days
Storage mode set which gives 13.2V after 1 day at float.
However the solar panels are not so constrained so that it goes on bulk mode each day.
There is only about 200W solar on deck in partial shade which in Sept for example raised voltage to about 14.5V so did not get to absorption on the solar controller. The net current was typically 3A compared to the storage current of 0.2 A for the bank.
I suspect that in this regime I have overcharged the battery.
When in the Marina, I leave the mains on since it is used to run the dehumidifier. I wonder whether the best course of action is to turn the bulk voltage down to 13.2 on the solar system when connected to the grid.