I have a 2004 Mazda 3. It got a new battery just over 2 years ago. About 8 months ago, the alternator died and I had that replaced and everything seemed fine in the months since then, including as recently as last weekend.

A few days ago, the car wouldn't start. When turning the key, absolutely nothing happened. Battery didn't even have enough energy to unlock the doors via key fob. Called AAA to take a look. Technician gave the car a jumpstart, but lights were flickering and the car died again shortly after jump kit was disconnected from battery. He thought the alternator was to blame.

This morning, we had the car towed to a repair shop. To get the car out of my garage and onto the truck, tow driver had to jump the car and leave the kit connected while driving the car. Even doing this, the car would die within seconds of ignition each time he started it, so it took a few attempts.

Repair shop ran a battery diagnostic and discovered that one of the battery cells was totally dead, and the cause was likely the added strain from last year when my alternator failed. They replaced the battery and reported that the alternator voltage was fine. The car drove fine all the way home (~30 minutes in slow Chicago traffic).

However, I am suspicious that my problem has only been fixed in the short term, and the car will start dying again soon. I thought that, after jumpstarting a car, it would run with a totally dead (or even disconnected) battery because everything would be powered by the alternator. That did not seem to be the case when the tow driver was getting the car out of my garage, or when the AAA guy was trying to diagnose the issue.

Am I paranoid or ignorant? Could a battery with a dead cell really have made the car die even after a successful jumpstart? Or should I be expecting more issues to crop up in the near future?

  • Welcome to Motor Vehicle Maintenance & Repair! Commented Mar 31, 2018 at 22:53

2 Answers 2


The issue of the car not running on alternator only is normal - yes I know some people say you can test the alternator by disconnecting the battery - DON’T - an alternator has lots of sensitive electronics and they can easily be damaged.

This is a common misconception, most likely due to the behaviour of dynamos. Where it was possible to start a vehicle by the starting handle and with a dynamo, it would charge from dead flat eventually. Also, one could start one tractor, remove the battery and then start another...

The behaviour of the alternator is also very dependant on engine speed - they don’t do much charging below about 3000rpm and, for that reason, they have a smaller pulley than the crankshaft - which means that the engine needs to be above 1500 rpm to get much of a charge. Also why some people with tuned engines with increased rpm find alternators fail (bearings, rotor contact), the solution there is to fit a slightly larger pulley to the alternator.


No no no no no

I really have to laugh at some of these posts

The alternator WILL charge the battery at low revs extremely successfully

First of all disconnect your current battery and jump start the car, when you remove the jump leads the alternator completely takes over the electrical demands on your vehicle which will have a voltage reading around 14 volts (14.4 is perfect). If the engine struggles or dies then the alternator is goosed

As for sensitive electronics, it’s just diodes and a voltage regulator, no damage will be caused by connecting/disconnecting

The voltage regulator keeps the voltage to where it needs to be whereas a dynamo had no voltage regulation as this is more complex for direct current, an alternator is alternating current and is very easy to produce and regulate

Alternators work perfect at idle speed and due to the voltage regulator they have an almost exact output regardless of engine speed

Also did you know alternators are all 3 phase circuits Well they are 😁

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .