Flooded engine!

[essayer]

GF's Lupo started, ran really rough for 10 seconds then stalled and just wouldn't restart.

So call the RAC, guy cranks it for about 2 minutes, started 'catching' and suddenly started up fine. (in a massive cloud of white smoke)

He reckoned it had got flooded somehow.. something about no compression and bore wash. Starts up fine now. Done so quick the RAC guy didn't even bother with the paperwork

I didn't think in the days of fuel injection engines could get "flooded". I know it has had an intermittently faulty coolant sensor (sometimes reads cold when it's not) so perhaps - combined with quite a few short trips lately - it has had a long period of overfuelling? Seem possible?

[S7Paul]

Don't know about the cause, but I've certainly had experience of fuel injected cars failing to start due to flooding. In every case but one I've managed to get them started by holding the accelerator flat to the floor & cranking the engine until it fired.

[cmsapms]

Both 'er indoors Elise S2 and Mother's 2008 Fiesta have been known to flood. I may be wrong, but I think the act of flooring it whilst cranking tells the ECU to inject zero fuel to "dry" the engine out.

[SuperchargedVR6]

Is it a single point injection Lupo, or the later 1.4/1.6 16V multipoint?

Single points can sometimes do that as they're essentially just a posh carburettor and have pretty basic control.

Certainly in multipoint injected VW engines, I've only ever seen proper bore wash occur when the injectors stick fully open. Due to either an electrical fault, or failed injector driver/s inside the ECU.

A failed coolant temp sender usually defaults the reading to a failsafe number. Normally 70 deg C in Motronic managed VWs, which should prevent over-fuelling.

[essayer]

It's a 2003 plate 1.4 16v.

Will have to keep an eye on it.. and keep those jump leads handy in case another 2 min cranking is required!

[SuperchargedVR6]

Did this suddenly happen out of the blue? Did Mr RAC man interrogate the ECU for fault codes?

[martycossie]

if i was u ill be putting new teemp sener in as that could be faulty

[eliot]

cmsapms said:

I may be wrong, but I think the act of flooring it whilst cranking tells the ECU to inject zero fuel to "dry" the engine out.

Indeed it is.

[DEANO87]

martycossie said:

if i was u ill be putting new teemp sener in as that could be faulty

This is normally the case.

In the eyes of the rac/aa they see it as problem fixed because it started but abit of investigation might have shown a faulty coolant temp sensor that is under reading and causing over fueling.

[essayer]

Cheers all, I have ordered a new coolant temp sensor from VW and will fit it at the weekend.

There were two faults logged on the memory but I didn't get a chance to memorise them, I think one was about a lambda sensor, the other was about the coolant temperature being over or under voltage.

Lambda I'm not sure on, could this have been triggered by the rough start/flooding?

I will swap the coolant sensor, as this definitely reads zero from time to time, and if we get the check engine light again perhaps the lambda sensor is faulty?

[Nick1point9]

essayer said:

Cheers all, I have ordered a new coolant temp sensor from VW and will fit it at the weekend.

There were two faults logged on the memory but I didn't get a chance to memorise them, I think one was about a lambda sensor, the other was about the coolant temperature being over or under voltage.

Lambda I'm not sure on, could this have been triggered by the rough start/flooding?

I will swap the coolant sensor, as this definitely reads zero from time to time, and if we get the check engine light again perhaps the lambda sensor is faulty?

Most ECUs ignore the lambda sensor until they are up to temperature, so it's probably related to the fact the dodgy temp sensor and the fact that when you crank without the engine firing you're effectively pumping fresh air down the exhaust, so if the ECU did listen to the lambda for some reason it would likely give "an impossible reading".

[Mr2Mike]

essayer said:

I know it has had an intermittently faulty coolant sensor (sometimes reads cold when it's not) so perhaps - combined with quite a few short trips lately - it has had a long period of overfuelling? Seem possible?

The ECU and the dashboard gauge normally use separate sensors, so the gauge reading low intermittently does not mean the ECU is seeing the same thing.

[Pumaracing]

Lots of clues in the first post which as usual no one has spotted. A cloud of white smoke would indicate coolant in the combustion chambers burning off which would obviously prevent proper starting as the spark plugs would be wet. Low coolant level, and/or an intermittent air lock in the coolant could cause the temp sensor to be in an air pocket or above the coolant level in which case it will read cold because it isn't sat in hot water to give a proper reading.

So it could have a blown head gasket leaking coolant into one or more cylinders. Check coolant level and do a compression test to verify head gasket sealing.

[essayer]

Ouch, hope not. Makes sense, although maybe the white smoke was water vapour, it was late evening when Mr RAC showed up.

I'll check the coolant tonight, as far as I remember it's never used a drop. But it is a bit mystifying how it worked fine since forever and one day randomly just wouldn't start..