Stalling/nearly stalling at idle

[Jono8]

Hi

My partners car, a 2004 1.6 Mazda 3 saloon has developed an odd issue. When idling, every now and then the revs will die down and the engine light and battery/stall light flash on briefly before the car recovers and idles normally again. Sometimes however it will stall completely and doesn't manage to save itself. It will fire up perfectly again after this has happened.

It has been in the garage and they said the inlet manifold gaskets needed doing as there was a vacuum leak detected there. This was done, but did not solve the problem.

When it seems to nearly die it sounds and looks like the engine is suddenly starved of something (it does kind of look vacuum related).

There are no issues when moving or accelerating. This only happens after the car has been idling for a bit (anywhere between 30 secs and a few minutes).

Any ideas? The garage it is at is now at a loss as to what it is. The only fault code it has thrown up is that the engine was running too lean at one point. However the engine light for this isn't always in and doesn't come back for a while after getting rid of it.

 

[Funky-Melon]

Have you tried cleaning out the idle control valve/throttle body? This can cause the symptoms you're having.

Have you also tried disconnecting the battery and leaving it for a bit before reconnecting? On some cars that'll force the ECU to re-learn the idle point, if for some reason it's got confused.

 

[Jono8]

Have you tried cleaning out the idle control valve/throttle body? This can cause the symptoms you're having.

Have you also tried disconnecting the battery and leaving it for a bit before reconnecting? On some cars that'll force the ECU to re-learn the idle point.

Yes, I should have mentioned that the garage has cleaned the throttle body too. Not sure about the idle control valve or the battery.

Will suggest that.

 

[bJN]

I'd pull the MAF sensor and give that a good clean, as well as the idle control valve - usually the obvious offenders for these symptoms!

 

[Phemo]

Had exactly this on one of my cars. A Ford so different engine so of course this may not apply but a Ford garage had it twice and said there was no fault found. Yet over time it got so bad that it would actually stall when the engine was idle, so on every approach to traffic lights, roundabouts etc.

I took a gamble and bought a used throttle body and it’s still fine even many years later. I had cleaned the old one but it made no difference. Obviously something was knackered with it in such a way that it didn’t throw up any error codes.

 

[grudas]

see if there's a procedure to "relearn" the idle.

for e.g. in my s2k if the battery has been disconnected the ecu needs to relearn the idle and the procedure is along the lines of -;

start car - make sure all electric draw is out so lights, radio etc.
hold at 3k until fan kicks in/temp is at operating
leave car idling for 5-6 minutes

ecu will learn the stable idle point.

thinking back about it, I had to clean my TB on my old mazda 3 2.0 and had to follow this procedure to "teach" it after! mine used to stall too, after this it was fine.

https://www.autoserviceprofessional.com/article/91698/how-to-fix-a-mazda-that-stalls-at-stop

 

[Clarkey]

So it will drop to idle and then idle stable for a while and then start stalling out? Pretty odd, does sound like MAF, vac leak, and/or o2 sensor issue. A bad idle air valve would normally cause it to stall out as it drops down to idle, not while it's already idling stable.

 

[EvilDreams]

I had the same idle issue in a 1999 Mazda 323 with an auto.

There was a vacuum leak and I replaced the intake manifold myself, and after I did that, it fixed the idle but instead of struggling to idle it would struggle and cut out when coasting up to junctions/roundabouts and braking, which was worse and I had to coast up in neutral to stop it happening. I thought it might be the brake booster but since putting it in neutral stopped the cutting out I was a bit at a loss at that point.

I gave up and sold the car.

Hope that helped

 

[Jono8]

Thanks for everyone's input.

For those interested, it is sorted now - basically, they completely reset everything/disconnected the battery for a few hours etc. They then left it to re-learn the idle for 20 minutes or so and then all was good.

Apparently the ECU and throttle body were still thinking there was the vacuum/manifold gasket leak so were compensating for that, causing the stalling still.

Been a week now since they did it and idles perfectly and no EML so far so seems to be all good.