Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[demolite]

Hi!

I've been tackling a problem with cold starts with my 2000 2.5 V6 (X25XE) facelift for some time now. Cold start here being any start at engine temperature of 30C or less. For the first 20-30 seconds the engine is overfueling and running rough and there's a strong smell of raw petrol in the cabin. This has happened on every cold start for atleast two months.

There are no fault codes, and the car runs fine for the rest of the day once the morning cold start is done with. There are no fuel leaks from the pipes, hoses, filter, pump assembly, injector rail, injectors, pressure test port or fuel pressure regulator. I ran the fuel pump for 30 minutes independent of engine with no fuel smell from anywhere (intake manifold was open to see injector leakage).

Typical start at ambient temperature 20C:
Upon starting, MID shows a fuel consumption of roughly 10 L/ hour which starts dropping slowly. Rpm is held at about 1300. About 3 seconds after start, engine starts running rough, fuel consumption is at 7-8 L/h. About 5 seconds after start strong fuel smell enters cabin, if car is creeping forward or stationary (smell coming from engine bay). It runs rough for about 20 seconds after this until the fuel consumption has dropped to about 4 L/h then suddenly engine smooths out. Fuel smell starts to fade and disappears in a matter of seconds. All is well for the rest of the day.

BUT, the engine smooths out instantly if the accellerator is pressed in any amount during rough running. Fuel consumption goes down when revs go up. Again if I let off the gas, the rough running returns. If I keep the revs at 1500 as soon as it starts for that 30 seconds there is no fuel smell or rough running at all.

Logs dont tell me much, but look at the spark advance when pressing and releasing the throttle .



Here is log and a video of a typical cold start.

Log:
https://www.dropbox.com/s/n0u55lr5wmsds29/cold_start_overfueling.csv?dl=1
Video: (sorry for bad audio, the blower vent was pointed to the microphone, and can't really re-shoot today)
https://youtu.be/FLPs0B-wbcw

[Doctor Gollum]

Ecu coolant temp sensor

[demolite]

Ecu coolant temp sensor

But it is reporting a plausible temperature in the diagnostic log, though might be a couple degrees off? (engine temp 17C, air temp 20C)

[Doctor Gollum]

Ecu coolant temp sensor

But it is reporting a plausible temperature in the diagnostic log, though might be a couple degrees off? (engine temp 17C, air temp 20C)

Ok

Being cabled throttle, how does the idle control valve behave?

[demolite]

Ecu coolant temp sensor

But it is reporting a plausible temperature in the diagnostic log, though might be a couple degrees off? (engine temp 17C, air temp 20C)

Ok

Being cabled throttle, how does the idle control valve behave?

The idle control valve was my first suspect but as far as I know it works fine. I've cleaned it multiple times to be sure and it works smoothly when tested with a variable power supply. Looking at the logs, the spark advance is very low and sometimes negative, but at the slightest touch of gas it goes to 10+ degrees advance. Spark retardation is usually a sign that the ECU tries to lower the rpm, but why?

[biggriffin]

I would change the coolant temp sensor. Then go for a decent drive to get every thing nice n warm,
New sensor about€20.

A faulty thermostat can cause rich running.

[Diamond Black Geezer]

Recently similar issues, party a stinky interior for the first few mins on startup. Did a massive service, and can't say for certain this was the sole cause, but me old girl has certainly run better since. 

[demolite]

Just to let others know, the problem was a faulty MAF.

Tried another from the scrap yard and problems went away, cold starts now at about 3-4 l/h. It also hugely improved low end throttle response.

As to the negative spark timing, I found out from a Saab engineering manual that Bosch Motronic 2.8.1 uses multi-spark when the engine is cold (multiple ignition events per cylinder up to 12 degrees after TDC). The spark timing in diagnostic simply reports the last ignition event timing, which due to multiple ignitions can be negative (as in after TDC).

[Bojan]

Glad you've sorted.
Did it by any chance had a sporty after-market filter (oily type)?

[Diamond Black Geezer]

PFL MAFs are known to be 'bulletproof' this is most odd if this was the cause, and worth adding to our collective knowledge if this is the case.

[demolite]

Everything is stock engine wise.

PFL MAFs are known to be 'bulletproof' this is most odd if this was the cause, and worth adding to our collective knowledge if this is the case.

Yeah, thats what i have read. I didn't even suspect it until it was one of the last possibilities. It wasn't fully broken but seems to give out of spec values when cold and/or at low airflow. No fault codes. Old and new are Bosch 0280217519. The surface of the measuring plate looks a bit oxidated.

I tried cleaning the MAF before replacement though.

[Nick W]

PFL MAFs are known to be 'bulletproof' this is most odd if this was the cause, and worth adding to our collective knowledge if this is the case.

They're a lot older now, and much more likely to be 'used up.' MAFs are not the only part that will suffer like this.

[Diamond Black Geezer]

Picked up a spare MFL MAF along with the intake system t'other week, it's just sat in a box of 'useless spares/for selling/giving away', I shall keep my hands on it, then 

[demolite]

Just an update in case anyone has the same problems. The fault was fixed with a new MAF sensor, but the old one could be resurrected too.

Just for fun I disassembled the old sensor to see whats inside.
A spider had made its home in the old one and some of the crud on the plate was too stubborn to get off with "maf cleaner" alone. So I mechanically cleaned it the best I could with isopropanol alcohol and q-tips and rags and it started working fine again! I tried it for a few weeks without any problems. So now I have a spare 

Heres what the insides looked like:


I think the crud and dirt caused some issues at low airflow in the cold.

[Diamond Black Geezer]

VERY good to know!


Up until recently is 'the word' that the PFL ones were bulletproof, but the FL ones were a bit more likely to fail. However time marches on, and we're hearing more and more of PFLs/MFLs breaking, too.