The car runs poorly at idle and often stalls when idling. Idle is unstable and the engine “grunts” like a pig.
At higher RPM the engine runs reasonably well and the car can be driven, as long as it’s not allowed to return to idle.The following actions and tests have been performed, but the fault has not been found:
When the issue first appeared, it repeatedly gave fault code P0170.
Both lambda sensors cleaned → No effect.
Both lambda sensors produce a signal, and the waveform appears normally on the Delphi DS150 tester.
Catalytic converter inspected with a camera → Looks clean and intact.
A new MAP sensor was tested in place of the original → No effect.
New spark plugs installed → No effect.
New camshaft position sensor installed → No effect.
A new crankshaft position sensor tested → No effect.
Throttle body opened and visually inspected → Looks clean and intact, moves according to the pedal without issues. Throttle body was also replaced, but no effect.
Injectors removed, cleaned, and tested with brake cleaner spray → Do not leak without control signal, and with control they produce a clean spray pattern.
New injector seals installed → No effect.
Used accelerator pedal assembly installed (from a scrapyard) and pedal adaptation performed → No effect.
Spraying brake cleaner into the intake manifold at idle causes the engine to stall immediately.
Front lambda sensor heater resistance measured → 8.9 ohms.
Intake leaks tested with a smoke machine → No leaks.
Compression tested on all cylinders → All between 14.5–15 bar.
Cylinders inspected with a borescope → No scratches or abnormalities.
Engine oil and filter replaced.
Tank filled with 98 octane petrol and Liqui Moly catalytic converter cleaner added.
New ignition coil installed about six months before the issue appeared (Valeo). Coil tested with a spark tester.
Fuel pump removed from tank, visually inspected, cleaned, and tested with an external power supply → No abnormalities, appears to function normally.
VVT solenoid opened and cleaned → No effect.
PCV-valve checked and found to be intact/working okay. I have not found an EGR-valve to check in this car.
Secondary air pump removed and tested with external power → Appears to work normally.
Oil separator (“PCV box”) removed and cleaned, hoses blown clean → No effect.
ECU removed and connectors visually inspected → No corrosion or faults found.
All fuses checked → All OK.
BSI / fuse box opened, visually inspected, and cleaned with electronics cleaner → No effect.
Engine bay fuse box opened and cleaned → No effect.
Electrical connectors around the engine and battery cleaned and greased → No effect on idle, but other electrical issues were resolved.
Front lambda sensor signal measured with an oscilloscope → The signal correlates with the “grunting” at idle.
Front lambda sensor signal oscillates normally at 1500 rpm, and the engine runs reasonably well if RPM is held steady with the throttle.
Throttle pedal adaptation performed as follows:
Turn ignition on (do not start) for 30 sec
Turn ignition off for 30 sec
Turn ignition on (do not start) for 30 sec
Turn ignition off for 30 sec
Turn ignition on, then slowly press the throttle pedal to the floor and hold for a few seconds
Turn ignition off
Start the car
For some reason, the engine runs normally when the crankcase ventilation hose (going into the intake manifold after the throttle body) is disconnected — i.e., when a vacuum leak is introduced. So what could cause the engine to run too rich at idle?
Still not replaced: brand‑new injectors and lambda sensors. Timing chain/belt has also not been opened. What would you suggest? Attached is a picture of the front lambda graph at idle. This has been a hobby since I have stopped counting how much time I have used to this, but I still don't agree to give up as this has become some kind of challenge to me even the car should have scrapped the day this problem occurred (+260k kilometers and 20 years behind).