The lead-acid battery: what it can tell you, and what testing it properly means

A car battery does one demanding job — deliver several hundred amps for a few seconds — and one undemanding one, holding the memory circuits alive while the car sleeps. It is very good at neither for very long, and it is the component most often replaced when something else was at fault.

What is inside

A lead-acid cell, invented by Gaston Planté in 1859 and essentially unchanged in principle. Six cells in series at about 2.1 V each gives the nominal 12 V of a car battery — 24 V on heavy vehicles, which is six more cells rather than a different chemistry.

Each cell has positive plates of lead dioxide, negative plates of sponge lead, and dilute sulphuric acid between them. Discharging converts both plate materials towards lead sulphate and dilutes the acid; charging drives it back.

The original article calls the positive plates "the anode". That is wrong in the case that matters, and the reason is worth knowing because the confusion is universal. Anode and cathode are defined by the reaction, not by the sign — the anode is where oxidation happens. In a battery that is discharging, the negative plate is oxidised, so the negative plate is the anode and the positive plate is the cathode. On charge the roles swap. Since the labels reverse depending on what the battery is doing and the plate signs never do, say positive and negative and the ambiguity disappears.

"Atomic" batteries, and a warning that is backwards

Sealed maintenance-free batteries are sold in Iran as baatri-ye atomi — "atomic" batteries. There is nothing atomic about them. The name is marketing that stuck.

What they actually are is lead-acid with calcium alloyed into the plate grids instead of antimony, usually calcium on both plates or calcium and silver. Calcium grids gas far less on charge, so the cell loses very little water — which is what allows a sealed case with no filler caps and a service life often quoted at up to seven years.

The original then says these batteries suit cold regions and should not be used in hot ones. That is the wrong way round. Lead-calcium grids were adopted specifically because they resist corrosion at high temperature and lose far less water than lead-antimony, and heat-driven water loss is a leading cause of battery failure in hot, dry climates. If anything the calcium battery is the better choice in heat, precisely because you cannot top it up.

There is a real effect underneath the folk advice, and it is about charging rather than climate. Calcium batteries want a slightly higher charging voltage and tolerate deep discharge poorly. On an older car whose regulator sets a low charging voltage, one can sit chronically undercharged and sulphate — which will show up sooner in a hot climate, where everything ages faster. The fix is to check the charging voltage, not to buy a different battery.

Reading its state

Two different questions, and conflating them is the usual mistake.

State of charge — how full is it? Measure open-circuit voltage after the battery has rested, ideally a few hours off charge, because a surface charge left by the alternator reads high and disappears under the first load.

State of health — can it still do the job? Voltage cannot answer this. A battery with a dead cell or heavily sulphated plates can read 12.6 V at rest and collapse the instant the starter turns.

That is what a load test or a conductance tester is for, and it is the only measurement that separates a flat battery from a finished one. A conductance tester gives a figure in CCA that you compare against the rating on the case; a traditional load test pulls roughly half the CCA for fifteen seconds and watches whether the voltage holds above about 9.6 V.

The cheapest version of the same test needs only a multimeter: watch battery voltage while somebody cranks. It will dip — that is normal — but a healthy battery should stay above about 9.6 V through the crank. Below that, with the starter and cables known good, the battery has failed under load regardless of what it read at rest.

Why it died — because it usually is not just age

A battery that fails well before its time is reporting something else, in the same way an early catalytic converter failure is:

• Undercharging. A slipping belt, a weak alternator, or a bad connection between alternator and battery. See the alternator and charging system.
• A parasitic drain, which flattens it repeatedly, and repeated deep discharge is what actually kills lead-acid. See finding a parasitic drain.
• Short journeys. Starting costs a great deal of charge and a ten-minute drive does not replace it. Batteries on cars used only for short trips die young and there is nothing wrong with the car.
• Corroded terminals or a poor earth strap, which look like a weak battery and are not — a voltage drop test across each connection finds them.
• Heat, which is the underrated one. High under-bonnet temperature accelerates grid corrosion and evaporation, and it is why batteries in hot climates typically live shorter lives than the same part elsewhere.
• The wrong type of battery in a stop-start car, which is the only item on this list you can buy by mistake. It has its own section below.

Stop-start needs a different battery, and the wrong one fails quietly

A conventional starter might see thirty thousand engine starts in its whole life. A stop-start car can reach that in a year — every set of traffic lights is a start, taken out of the battery while the alternator is not turning. Nothing above changes; the arithmetic does.

So those cars are built around one of two types. An AGM battery holds its electrolyte in a glass-fibre mat, and an EFB — enhanced flooded battery — is a conventional design reinforced for repeated shallow discharge. Both tolerate being cycled in a way an ordinary starter battery does not, and both cost more, which is exactly why an ordinary one gets fitted in their place.

The failure is early and it is quiet. What you notice is not a warning light but stop-start simply no longer happening: the module measures cranking speed and switches the function off rather than risk a slow restart at a junction. A car that has stopped doing something it used to do is reporting its battery.

Two rules follow. Replace like with like — an AGM car gets AGM, an EFB car gets EFB or better, never worse. And check whether the car needs the new battery registered with a scan tool, because the charging strategy is calculated from the battery's type and age; fit one without telling the car and it is charged on the old battery's profile for the rest of its life. The starter on these cars is reinforced too, and the starter motor covers that half.

Practical notes

• Disconnect negative first, reconnect negative last. With the negative off, a spanner touching bodywork while on the positive terminal is harmless. The other order puts a short across the battery through the spanner in your hand.
• Charge a flat battery properly before judging it. Testing a discharged battery tells you it is discharged, which you knew.
• Never jump-start onto a frozen battery, and never charge one. Electrolyte only freezes when heavily discharged, which is itself the diagnosis.
• Expect the car to need a moment afterwards. Disconnecting power loses learned idle and fuel adaptations, so a rough idle for the first few minutes after a battery change is normal, not a new fault.

Related codes

Charging and supply faults surface as P0562 system voltage low and P0563 system voltage high. A battery too weak to hold the system up during cranking can also produce a scattering of unrelated codes across several modules — clear them and retest after the battery is sorted rather than chasing each one, because low voltage makes modules report faults they do not have.

Video guides

Video, for the parts of this that are easier watched than read. The English-language ones come first.