Considering B58’s problems with identifying misfires, I performed various experiments. Purpose of the experiments – to explore how B58 manages the detection of misfires of different causes. And to check if I am right (the symptoms raise suspicions) that Bosch still uses unevenness of crankshaft turning to detect misfires during the warm-up of the CO catalytic converter (during the first minute after the start of the engine).
Misfires can be caused by several reasons – incorrect fuel mixture and weak ignition. In case of ignition problems, several options are possible too – defects of the ignition coil, flooded spark plug, etc. Depending on the ignition defect, the time of the spark creation also changes (or does not change). DME can also use an incorrect spark creation time to identify the misfires. Is DME using these data, and how is it going – read further.
Attention – don’t try to repeat such experiments! You can get a fatal electric shock!
In my experiment, I used a brand new Magneti Marelli ignition coil. The experiment was performed with cylinder No.2. I connected the ignition coil correctly but did not fix it in place. I gradually dragged out the ignition coil from its defined position to provoke the misfires. In such a way, I increased the losses in the high-voltage circuit of the ignition coil – part of the energy is lost due to discharge in the air.
First, I dragged the ignition coil out for around 1 cm. How it looks:
Not even the slightest difference in the performance of the engine. Nor during the warm-up phase of the CO catalytic converter. The engine works perfectly.
I dragged the ignition coil out more. Around 2cm from the intended position:
Zero effect! The engine works perfectly!
I increased the air gap for some cm more. Zero reaction! I already had an idea – maybe this engine suddenly does not need the ignition at all? Anyway, first conclusion – the energy of these ignition coils is enormous! The energy reserve is huge! No, this engine is not unique – each industrially developed engine has a substantial spark energy. Accordingly – if the engine misfires, you don’t have to install more powerful ignition coils – that is the fight with some unsolved problems!
When I did not manage to cause misfires with simple methods, I acted in the following way:
I increased the air gap and used the pliers as an additional “load”. I put the pliers on the Rail – grounded them. Part of the energy should “bump” against the pliers.
Yes, such a solution gives the effect. Finally, the first misfires appear!
Directly after starting the engine (on the 133rd second), everything looks nice:
But, after a short moment, the first misfires can be noticed:
On the 165th second, a full-fledged misfire appears, and DME notices it. On the 171st second, a partial misfire – DME misses it. To be fair, it should be noted – the warming up of the CO catalytic converter is still in progress, and the performance of the engine is still less even than usual in idle. Such partial misfires can not be considered as a problem.
On the 182nd second, and INCREASED mechanical energy of the “damaged” cylinder – DME identifies a misfire (the last seconds of the CO catalytic converter warming phase are in progress). Why is the energy increased? Other cylinders will not be the ones to blame – in this mode, the vector type idle smoothing is not used. A premature ignition happened? DME identified incorrect Combustion time? At this moment, I have more questions than answers. But we still have to conclude – during this phase (warming of the CO catalytic converter), DME quite correctly recognizes all suspicious situations.
Here, a zoomed-in image:
As we see, a solid “spike” of the energy. A very exotic reason for this event could be – the fuel was ignited at the beginning, not the end of the dwell time! The reason is that these ignition coils have no diodes included (they were an integral part of the previous/pre-DI generation engines). This theory should be checked!
Here, a minute after starting the engine (remember – exactly at the end of the period of the heating of the CO catalytic converter, what is significant), the misfire counter of the 2nd cylinder is cleared.
On the 252nd second, a partial misfire – DME does not notice it. Since the heating phase of the CO catalytic converter is finished, DME suddenly does not do well with detecting the misfires!
Here, the next several tens of seconds zoomed in:
We see several partial misfires:
DME does not notice any of them!
It looks like during the phases of heating of the CO catalytic converters, DME becomes very “sighted”- my suspicions were justified!
Let’s see how DME manages to detect group misfires. The air gap was increased a bit more, and – yes, the engine now started to work on 5 cylinders!
On the 196th second, the engine is started. In the 2nd cylinder – continuous misfires. Starting with the 200th second (4 seconds after the start), DME began registering the misfires. Misfire’s increase speed is around 10 misfires/per second, which means – DME recognizes each of them! Excellent! In addition – no false misfires registered in any of the other cylinders!
Unfortunately, the method of continuous misfires does not work long-term (DME switches off the injector of the “damaged” cylinder), so I continued the experiments with relatively rare misfires. Here is the next test by provoking misfires time by time:
The engine is started on the 0 second. Till the 20th second, there are 4 events of incorrect efficiency of the 2nd cylinder, all noticed by the DME. Excellent!
On the 21st and 26th seconds, the incorrect performance of the 2nd cylinder – DME notices both events. On the 31st second, there is a controversial partial misfire – DME misses it.
On the 46th and 54th second two misfires – DME notices both. Excellent! But, what is even more interesting is that on the 56th second, the misfire counter of the 2nd cylinder is cleared. Again – exactly at the end of the CO catalytic converter warm-up phase!
In the next image – how DME manages to recognize the misfires after the CO catalytic converter warm-up phase:
We see group misfires on 194th to 196th seconds, but DME notices some 6 of them (in the graph, we see at least 15 .. 25 misfires). From 199th to 201st second – continuous group misfires (at least 8 .. 9 pcs), DME notices only 3. On the 204th to 206th, the second are some separate misfires – with these, DME manages better.
Here, I have to add – that all these misfires are caused by “ignition defects”, which also manifest as a significantly shorter combustion time. If the combustion time stays (at least partially) correct, DME (in idle, AFTER the phase of the CO catalytic converter warm-up) does not recognize such misfires at all. More exactly – compulsively pretending not to recognize for more than a minute (read more here).
In case such an “ignition problem” misfires, DME acts correctly – immediately on the background, turns off the injector of the damaged cylinder, and there is no smoke and pungent smell from the exhaust. True thou – after such misfires too, some surprises expected me, but more about that – in continuation.
At the end of this part – how DME manages the Combustion times measurements.
Till the 28th second, the ignition coil of the 2nd cylinder was in the intended position. On the 32nd second, I dragged it out for some 1 .. 1.5 cm. On the 41st second, the ignition coil is dragged out for some 2 cm.
On the 44th second, the ignition coil is dragged out for around 3 cm. As we see, DME identifies stable (and significantly reduced) combustion time.
On the 74th second, I started to move the ignition coil back to its correct position. On the 77th second, the ignition coil is back in the correct position. Everything is in order, and DME correctly measures Combustion time.
Read the continuation here.