High oil consumption and alusil cylinder liners

[viperbluecerb]

FordPrefect56 said:

viperbluecerb said:

Everyone is here to learn and share and that is, after all, the entire point of this thread.

A modern closed loop engine can't run at anything other than stoichiometric without throwing up fault codes, or producing black smoke, failing emissions tests and generally misbehaving. If the injectors leak a bit the fuel trims will just compensate. Your engine can't have been running rich at all on average, never mind rich enough to cause some sort of obscure bore damage.

And I wouldn't argue otherwise when the engine is running. But when it is not, the fact it is a DI system means residual pressure and leaky injectors causes fuel to leak directly into the cylinders. This, at a mininum, was causing fuel fouling of plugs which I observed and so pooling in the cylinder, and specifically around the piston bore gap on top of the top compression ring is not inconceivable. Consider the corrosive nature of ethanol in some fuels, the fact that it could be sitting there for some time between engine run cycles and it is not an unreasonable conclusion to come to.

Edited by viperbluecerb on Sunday 15th May 13:22

[viperbluecerb]

One other point I would like to note. Whether I choose to fix the issue at this stage is somewhat irrelevant. Sure, I appreciate going through that process would have yielded some useful experience and information. However, the main point of this thread was to root cause the issue in the first place, and not just in terms of what the problem is, but more importantly how it occurred. The credible suggestion of unfinished bores at manufacture is one possibility. But let's consider it isn't the issue. It makes sense, IMO of course, to perhaps explore some other possibilities, however unlikely they may seem. Sure, it may not be possible to prove one way or another, but nevertheless it may be useful for others in the future.

[viperbluecerb]

FordPrefect56 said:

A modern closed loop engine can't run at anything other than stoichiometric without throwing up fault codes, or producing black smoke, failing emissions tests and generally misbehaving.

In the interest of accuracy, this isn't entirely true for the N53 engine. It can also operate in stratified lean mode under light load, for which the combustion dynamics are different and the standard step-ramp control strategy applied by the closed loop fueling algorithm to feed the catalyst(s) by keeping the engine switching either side of stoichiometric conditions is not in force. A lean NOx trap and NOx sensor are employed for emission control during this operating mode due to the larger levels of NOx produced, reverting back to homogeneous mode of operation to purge the NOx trap when required and when off of light load (e.g. vehicle accelerating).

[viperbluecerb]

stevieturbo said:

Modern engines can and do operate outside stioch quite often without throwing fault codes of any kind

Technically that isnt a fault just it running incorrectly, and they only run stioch at lighter loads.

Sometimes the random st codes do appear for are crazy...yet when some major problems exist, it throws no codes at all ! Modern cars are a nightmare at times.

I have also seen vehicles over the years that you would have though would have thrown codes but didn't. All depends on how good the OBD strategies have been designed and calibrated, and some are a lot worse than others. In my experience of implementing a variety of different engine management subsystems, closed loop is only an operating mode that is run during steady state. As soon as a large enough transient condition is entered, e.g. heavier acceleration, the system goes open loop. In a torque based or MAF based engine management strategy (torque based for all modern vehicles running electronic throttle control), the requested torque will be achieved using various calibrated maps to look up ignition timings and injector pulse times for a given manifold air pressure or mass air flow (with many other inputs such as coolant temperature, ambient air temperature, learnt trims, etc.), and even if you wanted to, achieving a stoichiometric mix is simply not possible for larger torque demands. Whether or not adaptive trims during open loop mode of operation are sufficient to compensate for things such as leaky injectors I guess depends on a variety of factors, but a rich condition when operating under open loop may not be picked up by the OBD system.

Completely accept that an overly rich condition continuously will kill the catalyst eventually. I remember one issue during vehicle development where we had several catalysts failing and I discovered that during an ECU running reset (which occurred due to a bug we had at the time), the ECU would recalculate fuel quantity using an uninitialised coolant temperature value of -40 degC, which caused a very rich mixture for only a short period of time until the coolant temperature sensor was read correctly, but it didn't take many of these events to cause damage. Issue went away when this was fixed.

[viperbluecerb]

viperbluecerb said:

After 1K miles, it is unfortunate but not unsurprising that I must report that the oil consumption continues. It was always a strong possibility that the issue was ring/bore related and after all the work that I have undertaken, this has proven to be the case. Slightly overfilling led to a longer that expected time for the oil level gauge to report a drop in oil level and then it continued at the previous rate. Generally running very well with a stable idle and no visible smoke under any observed conditions.

Actually, I jumped the gun a bit with this conclusion, reporting after just a single drop on the electronic dipstick and over estimating how much I may have overfilled. I have done a few more miles now and a better estimation of oil consumption is around 1 litre per ~1000-1500 miles, which is actually quite an improvement over the original 1 litre per 350-400 miles. Previously after my 15 mile drive to work there was a light grey smoke from both tailpipes. This has now also completely gone.

BMW spec up to 0.7 litre per 1000KM (~620 mile) consumption for this engine, which while still horrendous, the car would actually now be in spec. I will do some more miles to see how things continue.