2012 Volvo XC60 T6+LPG - rough idle and misfires

I have a problem with 2012 XC60 T6+LPG. Engine is idling roughly, especially when cold.
After leaving the car to idle for a longer period of time, ECU registers OBD errors for random misfires with misfires in each cylinder.
No noticeable problems (hesitation to accelerate or lack of power) or misfires when accelerating or cruising at constant speed.

Spark plugs have been replaced. Coil packs seem to be Ok. There is no noticeable vacuum leak in the intake - we have performed a smoke test on intake manifold. Compression in the cylinders is fine.

Once engine gets warm, there are no misfires on LPG, and on petrol there are misfires in single cylinder #5. As the car is driven mostly on LPG, I assume there may be problems with 5th injector.

I have performed live data on the car. What I found strange is very hectic ignition angle when engine is idling, despite RPMs being rather steady and only intermittent misfires. I am wondering if car is trying to adjust ignition angle based on some invalid data and thus introducing misfires itself? Is it possible/likely?


Any ideas how could I proceed?

* Is such variation of ignition angle possible when car is idling roughly because of vacuum leak or any other problem?
* MAF signal is not that constant, but I assume this variation is within tolerances?
* I have read that engine load is one of the major factors affecting ignition angle when idling. I haven't checked OBD readings of engine load yet.
* I was also considering maybe the car gets invalid readings from knock sensor, thus actively correcting ignition angle.

jonasltu wrote: I have a problem with 2012 XC60 T6+LPG. Engine is idling roughly, especially when cold.
After leaving the car to idle for a longer period of time, ECU registers OBD errors for random misfires with misfires in each cylinder.
No noticeable problems (hesitation to accelerate or lack of power) or misfires when accelerating or cruising at constant speed.

Spark plugs have been replaced. Coil packs seem to be Ok. There is no noticeable vacuum leak in the intake - we have performed a smoke test on intake manifold. Compression in the cylinders is fine.

Once engine gets warm, there are no misfires on LPG, and on petrol there are misfires in single cylinder #5. As the car is driven mostly on LPG, I assume there may be problems with 5th injector.

I have performed live data on the car. What I found strange is very hectic ignition angle when engine is idling, despite RPMs being rather steady and only intermittent misfires. I am wondering if car is trying to adjust ignition angle based on some invalid data and thus introducing misfires itself? Is it possible/likely?

Any ideas how could I proceed?

* Is such variation of ignition angle possible when car is idling roughly because of vacuum leak or any other problem?
* MAF signal is not that constant, but I assume this variation is within tolerances?
* I have read that engine load is one of the major factors affecting ignition angle when idling. I haven't checked OBD readings of engine load yet.
* I was also considering maybe the car gets invalid readings from knock sensor, thus actively correcting ignition angle.

Is the LPG system an aftermarket conversion or production line ? the reason I ask is the two types of system are quite different aftermarket systems are vapour injection and factory fit systems are usually liquid injection.
Assuming the car has an aftermarket conversion.

When swtitched to LPG modern aftermarket conversions work by intercepting the control signal to the petrol injectors, diverting it to the LPG computer which alters the signal giving it an extended open time to suit the much slower speed the LPG injectorstake to open and close.
Especially because it is an aftermarket fittment the LPG wiring has to be checked very carefully for breaks and chaffing
Because the wiring to each petrol injector has to be intercepted either by plug in connectors or cutting and soldered connection a bad connection can cause a misfire either on petrol or LPG or both.
Because of the petrol engine computer is still in control of the fuel trims on both petrol and LPG.
If an LPG system is setup correctly the total fuel trimson the Petrol Engine Computer should change by only about 2 or 3% when switching between fuels.
LPG system are very dependant on the vapour temperture sensor which will usually be located between the vapour filter and the injector block, to read this sensor you need to use the LPG system diagnostic software and a low cost interface.

LPG systems require additional l and more frequent servicing . When running on LPG the ignition system is subject to much higher voltage requirements than on petrol so the spark plugs plugs should be be set to a narrower gap ( I used 0.8mm ) than on petrol only engines.
Compared with petrol injectors the moving parts inside an LPG vapour injector are relatively big and heavy and so are subject to wear and change their characterstics over time. This will affect the fuel trims of the petrol engine computer.

An LPG s service entails new wet and vapour fillter a connection to LPG software general check of the LPG wiring and if required overhaul of the injector block.

Yes. The LPG system is an after-market one.

The spark plugs have been replaced with matching NGK spark plugs recommended for LPG operating engines.

The car tends to drive rich on LPG (STFT -10s), whereas on Petrol it tries to make the mixture richer (STFT >0). I assume it may also contribute to the problem, if the engine drives rich on LPG. And once it is started cold in open loop operation it may still use maps from previous drive and try to make the mixture less rich.

What is strange that, once engine is warm, idling on LPG seems to be better than on Petrol, so I don't think there is any problem with LPG injection. The idling is worst when the engine is started cold on Petrol. Somebody has recommended to check engine coolant temperature when the engine is cold - maybe it's misreading and falsely putting car in closed loop operation. So will do that tomorrow morning

Point noted regarding injector soldering points - no doubts in this installation the wires were connected by cutting and soldering them together. I have LPG diagnostic interface cable, I can read the commands sent to petrol injectors (the ones that LPG ECU intercepts), but, as you mentioned, I may not be certain if contact is not lost with petrol injectors. But I should probably be safe to assume that if the fauly wiring is the cause, I should have issues when driving too (if one of petrol injectors is intermittently opening for a shorter period of time)