Best of Deals Car Reviews Repair Shops Cars A-Z Radio Show

Starter tested ok, could it still be bad?

2001 Mitsubishi Eclipse GT 3.0…Car won’t start, Turns over very slow, then stops and clicks. After a few tries we get smoke and burnt smell coming from area where starter is found. I had the battery checked, it checked out bad, replaced the battery, still no start. Had the alternator tested, it tested ok, still no start. Took the starter out and had it tested, it tested ok, still no start. Checked battery cable connections, a little corrosion, not bad. Cleaned that up, same thing. No start. I feel that it’s an issue with the starter. Could a starter test ok, and still be bad?? Or does anyone have any other ideas on what to check or try??

Not very likely the starter is bad although the test isn’t usually under any load so there is a small possibility

More likely you have a bad connection somewhere made worse by the burning smell. Check all the grounds. The battery ground being the first thing you check. The ground from the engine to the chassis is next. Chase the battery cable until it reaches the starter and check for burns and corrosion. A little resistance cause by corrosion becomes a big drain when a couple hundred amps runs through it.

Good Luck!

Have you tried turning the engine over by hand at the crank bolt, to see if it rotates freely?



It depends on how they tested it. If they just put it on a bench and tested it by applying voltage with no load it could appear to be good. The fact that you saw smoke coming from it tells me it’s toast. But try testers suggestion first.

I will check grounds and look for breaks or corrosion. This is what a coworker told me also.

No, trying not to have to take the engine apart if possible. I don’t see the crank bolt as easily accessible.

Remove the front tire on the same side as the front of the engine.

There’s usually an access hole to get at the crank bolt.


Ok, I got to the crank shaft. It does turn. It’s not in a place that I can get much leverage, but with some strength, it does turn.

The battery ground may be corroded at the engine or body connection.

Burning odor and smoke indicates a high resistance connection in the starter motor circuit or starter motor itself. “High resistance” means on the order of 0.01 ohm, so the resistance doesn’t need to be very high in that circuit to become a problem, due to the 100+ amps the starter motor draws. There’s at least a half dozen places where the electrical connections could be bad. Rather than going through them one by one, and you’ve already verified the battery and battery connections are good, it is usually faster to use a volt meter to measure the voltage at the two starter motor terminals during attempted cranking. Measure from terminal to starter motor case. If both measure 10.5 volts or more, and it doesn’t crank robustly, replace the starter motor. If either or both measure less than 10.5 volts during attempted cranking, works backwards towards the battery to find out why.

1 Like

So I checked the volts to the starter, it’s just fine. Ground is fine. I tried to crank the engine to show my father what it was doing and the car started right up… WTF! The only thing that I had done since last attempt at start was turn the crank bolt. Could we have a dead spot in the starter? My father suggested we might have a problem with flex plate. I didn’t think that would be a problem because prior start attempts the starter was not turning. Thoughts?

just a suggestion if you go with a starter, might wanna check the price on the internet
I just replaced one in my truck and it was a hundred less then autozone
from a place called db electrical (online)
shipping was free and it came in like 3 days, perfect condition, new not rebuilt…

1 Like

Good tip! I ordered from db. It was cheaper than RockAuto, and less than half the price at the local parts store. For that kind of savings I will wait 3 days to get it.

it’s possibly (though unlikely,) that the starter was hitting the ring gear in just the right spot as to not engage, but instead get stuck. This would cause a no start, and high resistance on your starter. You turning the crank bolt would turn the engine a bit, allowing the starter to now be able to engage., I’ve seen it before, but not very often.

I would start with replacing the starter, checking all the wires for cleanliness and tightness, and hope for the best. If it happens again, you may have to dig deeper into the flywheel/ring gear area.

Yes. Dear Internet, sometimes starters go bad.

My problem was with my 2008 Civic. It started out as a weak start, sometimes starting with no problem, sometimes acting sluggish. It progressively got worse and more erratic until all it would do is click when I tried to start it. (About six months to a year from the first symptom)

I changed the battery, ground wire, nothing seemed to fix it. So I took out the starter (Do not attempt this on a late model Civic, just take it to the dealer. trust me.) and took it to the parts store to test on their fancy bench test computer. The test said the starter was fine, but the observant guy at the parts store (which will remain nameless, but it was a place like Advanced Auto) noticed that the bendix was not fully engaging, it would only come out about a quarter of an inch. He also noted that the tests are not always accurate because it does not test it under load. When he put the new starter on the machine you could see the bendix fully engaging.

Anyways, I took the new starter and installed it. The car now starts perfectly. Bad starter.

Well, the observant guy at the parts store is correct about bench tests not being accurate.

Um… didn’t you remove the starter for testing and reinstall it? Perhaps the installation was the problem? Was anything loose when you removed it?

So, the guy observed the Bendix not fully engaging, but your last words were it was engaging when testing? What was it?

Sorry for the confusion. The moral of the story is that a bad starter can test good.

The old/bad starter: I took it in to the parts store, he hooked it up to the computer machine and it said it was a good starter. The bendix looked like it was only engaging about 1/4 of an inch.

The new starter: Hooked up to the machine, also tested good (obviously) and the bendix was observed FULLY engaging.