The battery that didn’t bark

This was one of 4 batteries, in parallel, providing house power on a boat. They had been installed by the boat builder, 3 years ago, when the boat was built (in Europe), and you might forgive the manufacturer for penny-pinching and installing wet batteries under each of the stern bunks. After all, why worry about sore eyes and sore throats for people sleeping directly above the batteries when you can save a bit of money on the fit-out of an, otherwise, very well-appointed yacht.

Anyway, enough of my cynicism. The yacht was returning to the marina, under motor one day, when the couple on board heard a huge bang. They thought that they'd hit a log or similar, but didn't find any damage. It was only when the battery power seemed a bit weak, next time out, that they investigated and found this under one bunk - the battery had exploded, the box that contained it had shattered, the tie-down strap had been ripped out of the timber below it, and the acid had soaked into the timber and left the boat via the bilge pump. 

So, what had happened here? Well, the battery had built up a huge pressure of gas and either the breather valves had failed to release the gas, or an errant spark had caused the gas to ignite. Either way, the battery case had failed dramatically. 

And that's what happened, but the interesting thing is why it happened. 

So we removed and replaced all 4 batteries (with agm batteries - safe to use under bunks and what should have been there in the first place), and we tested this battery, and the other 3 that it was paralleled with, and we found that 2 of the 3 intact batteries were in pretty good shape for their age, but one had "dropped a cell" which, in this case, was caused by a shorting out between 2 cells due to debris in the battery. The result of the cell failure was that the one battery dropped voltage by 2 volts, from 12 volts to 10 volts (and a maximum of well under 12 volts when under full charge), causing an approximate drop of half a volt across the bank of 4. The charging source (alternator) took this voltage drop as a signal to deliver as much power as possible to the batteries to try and restore them to full voltage (something over 14 volts) and, having no luck getting things back in order, kept on pushing too much power into the whole bank. The duff (10v) battery wasn't absorbing much of this, so the 3 batteries that had all cells intact took the brunt of the extra inrush of power, and the one with the weakest case or worst breathers failed dramatically. 

So, interestingly, one battery failed by dropping a cell, and the outcome was that another battery in the chain exploded. Not at all the conclusion that one might jump to after a look at the picture.

A couple of key take-outs from this:

• A good battery monitor is essential, especially when more than one battery is involved, and keeping an eye on the battery monitor might give you an early warning that a single battery is in trouble. A half volt drop on a battery bank should get you scrabbling to switch off the batteries, cut the charge to them, and check them out. If one is warm, or making some odd noises, or if you smell sulphur, let the batteries settle, stay well clear, and then get a professional with suitable protective clothes (especially eye protection) to take the suspect battery out of circuit and test them all. 
• If there is any indication of a significant drop in the performance of a battery bank, it's a great indication that a battery test is called for. In this case, with 4 batteries, there was enough power available that the first battery failure (the dropped cell, not the explosion) wasn't obvious.
• Use the right batteries. This has to be a classic case of nearly spoiling the boat for a ha'p'orth of tar.