After the first part of the entry (read here), the local car mechanics had something to tell. Thanks to the men for the ideas for this entry!
First counterargument: it does not sound plausible that, with the standard amount of oil (6.5 liters), oil could pool in the area of the 6th cylinder. Say, it would need some 20 litres of oil! The car we are talking about reaches 100 km/h in around 4 seconds. It means: first gear (50 km/h) in around 1.5 seconds. I named these values to serve as a reference for further experimentation.
For the experiment, we will need only one accessory, available to everyone. What is that?
Exactly – the simplest cup!
Prepare the cup, fill it with water.
Prepare for the test.
What do we have to do?
Count to 3 and run as fast as possible, holding the cup “vertically”!
I think no one has any doubt – the acceleration of a car equipped with a B58 is significantly more powerful than the acceleration of the average person while running. Within just one second, the car will reach a speed faster than the world’s fastest runners (world record holders run a 100m distance at a speed of just over 30 km/h).
What happened? Did the water spill? Hmm. Who would have expected this (pardon the irony)! You will say – but there is oil in the engine, not water? The oil is more viscous!
Ok, buy a litre of oil. It will be enough for several experiments. Did the oil spill, too? How so? It’s physics. Inertia.
Want an even more beautiful experiment? Then you will need this:
Fill the canister with about half a litre of oil. Why half a litre? Such an amount/proportion would be similar to the situation in the engine. Place the canister on the car seat. Film, how the oil “behaves” at the moment of the acceleration. If you perform this experiment, I assume that the topic “No, it can’t be that the oil spills at the end of the engine” will be closed.
Do the BMW engineers know this phenomenon/problem? Of course they do! No idiots are working in BMW AG. BMW engineers have performed a massive amount of oiling system simulations and tests. The exploitation of actual engines shows that if the engine is used reasonably, the oiling system works perfectly! If the car’s owner “tortures” a completely cold engine, it is an attempt to destroy the engine. Even the most powerful oiling system will not help against such behaviour!
Second topic, which the car mechanics raised: no, the piston is damaged due to knocking. Such damage has been observed previously in older engines!
I want to argue that low-speed pre-ignition is nothing new, so the argument “seen before” is not really relevant.
Could it be the impact of the knocking? Let’s see the image:
I marked the most important with red. As we see, the piston “above” the compression ring is completely intact! Not a single sign of the damage!
At this moment, it’s worth reminding what is knocking. Knocking is abnormal (not even, but very aggressive – in a form of microblasts), a burning process that happens if the temperature in the combustion chamber is too high, too high pressure, the atomization quality is poor, and other problems. Knocking creates a large thermal overload at local points, resulting in a high mechanical overload across the entire piston group.
Let’s look at the piston again. Its upper part is intact! How is that possible – knocking has damaged the side part of the piston, “missing” the upper part of it? No, the logic says – it is not possible! No, in no way! If your mechanic keeps talking about the knocking in this situation, my suggestion is to change mechanics!
One more nuance which confirms the burning (LSPI) directly by the wall of the block – the undamaged part of the piston (above the compression ring) is covered with burn, and it has changed colour. The petrol burns relatively clean. What does burn so dirty, with residues? Yes – the oil!
One more argument of the highly qualified mechanics – I was wrong that the piston started to melt. It is clearly visible that the piston has “broken pieces”.
For the moment, let’s assume that the piece of the piston is broken. Obviously, the piston pieces are stuck on “something”! There is no other explanation why a significant part of the piston can break off. The only thing that it can stick against is the engine block. But there is one problem. The engine block has NO damage! NOT THE SLIGHTEST! For the exact engine, the piston was replaced, and that’s it; the engine was returned to the customer. How is that possible? The only explanation is that the piston in this sector became “soft”. How is that possible? The only answer I can imagine is that the piston was “one millimeter” from melting.
This time for the experiment, we will use another subject, available to everybody:
Unwrap the chocolate and move closer to your car’s air vent. Turn the heating on to the hottest possible setting. Let’s melt the chocolate. Observe how the chocolate “behaves” in the part which is in “between” the melted and still “hard” part. Shortly before melting, the chocolate becomes softer and more plastic. If this pliable part of the chocolate is “fractured”, the “fracture” site looks as if the chocolate is really broken, but the power applied to do that was very tiny. The piston material (despite the fact that the piston is made of aluminium alloy, not chocolate) behaves very similarly. A short moment before melting, it loses hardness – the material becomes loose/pliable and deforms very easily.
Returning to the piston, take a closer look at the edges of the fracture site. There are no sharp edges or fracture lines. All fracture lines are “rounded”. How so? Very simple – they are melted!
One more obstacle, which promotes low-speed pre-ignition in low temperatures, in addition to the thick oil (which is a difficult task for oil rings to rinse from the block walls), is that the gap between the piston and the block walls has been increased. The reason for this is that the piston expansion ratio is higher than that of the engine block. At room temperature, this gap is around 2 times wider (as for the engine in the work temperature), but in temperatures below 0 °C, it is even 3 times wider! Wider gap – more oil can “squeeze” between the piston and the block shell. Thick oil, wide gap, oil spill during the swift acceleration, high RPM – if all conditions agree, low-speed pre-ignition is almost guaranteed.
At this point, I want to talk a little about knocking. For a correctly working engine (by which I mean the DME software is not damaged by performing some “upgrade”), BMW uses a range of advanced tools for knocking management.
First of all, the compression and power cycles of each engine are controlled – DME measures knocking (how aggressively the fuel burns).
For this purpose, knock sensors are used.
For BMW engines, two knocking sensors are used.
Data from the knocking sensors are used to:
a. detect the quality of the fuel and its octane number;
b. manage the ignition maps;
c. control/reduce the maximum allowed torque if later ignition does not give the desired result;
d. identify super-knocking and, in case of it, immediately switch off the cylinder;
e. DME performs the control of the knock sensors (self-diagnostics); modifies the ignition to maximum late if hardware damages are detected.
In N and B series engines, the ignition maps are managed very advanced. More about this can be found here. As we can read, DME adaptively changes both the common for all cylinders and the individual ignition map for each cylinder. DME analyzes how changes in ignition timing affect each cylinder’s knock data and predicts each cylinder’s behavior. DME performs both long- and short-term, entire-engine/group and individual ignition adaptations and corrections using multidimensional maps.
Suppose the DME software is not damaged (unfortunately, it is very often damaged by “upgrading” the engine’s performance, because the partial or complete switching off of the knock control allows for increased torque/power). In that case, prolonged and powerful knocking is not possible. I have data on thousands of N and B-series petrol engines, but I cannot remember any case where an engine with stock software has damaged a cylinder due to knocking.
Several comments regarding super-knocking (also called LSPI). How does DME distinguish a “regular” knocking from super-knocking? Very simple – super-knocking starts BEFORE the fuel is ignited. If DME notices this problem, the appropriate cylinder is immediately switched off (for some time). After some time, DME tries to restore the engine performance. If the super-knocking is noticed again, the cylinder is immediately switched off again. Super-knocking is not a common problem. Of the many diagnostic data and user-posted “code” lists on the Internet, I have only seen errors about this a few times. Despite the problem being identified very rarely, I do not have a reason to think that DME “does not see” this problem, because, as I already mentioned, I cannot confirm the case when I would see the piston damaged by knocking.
My experience says – knocking is not the thing that, in the case of BMW, would endanger the piston group the most. More dangerous would be catastrophically poor injector atomization. If the injector “pours” the fuel into a single sector of the piston (not sprays it as a fine mist), local overheating can damage it. I have seen several N43/N53 engines with a melted piston due to incorrect injector operation. Here I have to add that, in all cases, the owners ignored for a long time a strong vibration, the EML (Check Engine) symbol in KOMBI, and all possible error messages regarding the fuel mixture.
