In the previous entry, I explained how (this time – differently) the ignition coil of the 6th cylinder crashed.
Accordingly, now I had several ignition coils, with two different defects:
a. several ignition coils, which (by overheating) caused sporadic misfires in the cylinder (at the same time, Combustion time stays relatively correct – DME does not record error messages regarding it);
b. ignition coil, which (not necessarily overheating – at least one crash happened when driving with normal speed) sporadically “loses” energy and DME identifies Combustion time problems (more exactly: 0 ms Combustion time).
Briefly about what I checked in the damaged ignition coils (and compared the parameters with new ELDOR/Bosch ignition coils:
a. active resistance of primary winding/connection [R];
b. inductance of the primary winding in four frequency ranges (100Hz; 1KHz; 10KHz; 100KHz);
c. quality factor [Q] in the primary winding in four ranges (100Hz; 1KHz; 10KHz; 100KHz)
d. basic parameters of diode (which is connected in series with the secondary winding): leakage current Ir when feeding 15V voltage of reverse polarity; Uf and If feeding 15V voltage;
e. active resistance of the secondary winding [R];
f. leakage current between primary and secondary winding;
g. visual inspection of the coils’ high-voltage part.
Here are several images from the measuring process:
In the image: DC measurements of the ignition coils
AC/Z measurements of the ignition coils
What are the results of my measurements?
Honestly, I was disappointed. All parameters were identical (of course, taking into account technological tolerances). The only difference was that for the used ignition coils, the spring, located in the pipe, was oxidized. It should be noted here that the oxide layer was very thin and, most importantly, this peculiarity does not affect the functionality of the ignition coil in any way.
Accordingly, the first conclusion: unfortunately, even complicated measurements of the electrical parameters of the ignition coil (at room temperature) do not allow for identifying the defects of the coil.
Seeing such a situation, I cancelled the ignition coil performance tests with my custom-made ignition coil tester (description here). Obviously, there is a significant impact (for the defect to manifest) of increased temperature/load, which I can not (more precisely – do not want, because such a test is very laborious) repeat.
A short digression. Several observations:
a. Already for the N53 series engine, I noticed that the density of the misfires reduces if the decorative cover of the engine is taken off. I think that there are not many options – the reason is the lower temperature of the ignition coils! I don’t have another explanation.
b. For N53 series engines (observed also in part of ZB N52 engines), if the car was left in idle, after a short moment, the engine switched to Mapped mode (with engine temperature of 80 °C) and forcibly (close to the maximal power) cooled the engine/motor bay with an electric fan. Again, the temperature of the ignition coil was maximally reduced.
c. For most powerful versions of N53 (and forced versions, intended for a warm climate) in the engine bay, you can see a specially constructed aluminum air duct which “collects” the air from the front side of the car and drains it to the exhaust manifold, reducing the flow of the hot air in the motor bay.
Not taking into account that the B58 series engine is significantly more powerful, with a worse efficiency (accordingly, the amount of heat released from B58 is considerably larger than from N53), we don’t see solutions described in points b or c.
Returning to damaged ignition coils. What is the reason for these ignition coil defects?
a. For the storage of the energy, the iron core is used. The ability of the core to store energy significantly lowers when the temperature rises.
b. dielectric abilities of each insulator (body of the ignition coil, pipe) significantly worsen when the temperature increases;
c. Reverse-connected diode parameters deteriorate significantly with increasing temperature.
I’ll tell you a little bit more about point c: reverse-connected diode.
This diode is intended to prevent a spark at the “wrong moment” – the moment when the power transistor opens. In the moment, when the transistor opens, to the diode Ur=U1*k is supplied (where U1 = onboard voltage, including losses in the Uce of the transistor; k = transformation ratio of the ignition coil). Tipycal Ur = 1000 .. 1500 V.
Let’s see the parameters of a typical diode (which is used in this application), for example, Ir vs Ur (leakage current vs temperature):
The graph indicates how swiftly Ir increases when the temperature rises. It’s a leakage current. For a perfect diode, it is 0. The larger this leakage current, the worse the performance of the diode. As we see, this leakage current increases from 3uA (by 25 °C) to 700uA (by 175 °C), which means more than 200 times! In addition, as an example, I took a relatively expensive diode. For a cheaper diode, this parameter increases even more destructively, but the max work temperature is lower: 150 °C!
This leakage current, in addition, heats the diode itself by 1000V and 700uA current, thermal power of 700mW flows via the diode. It is a lot! This additional heating can create a “chain reaction”: the diode heats up – the leakage current increases – the diode heats even more – the leakage current increases even more – the diode overheats – thermal runaway is created.
Such a thermal runway can both damage the diode and create short-term damages (which disappear when the temperature lowers). More specific defects are also possible – the diode’s ability to handle load/overload after such extreme events is decreasing. It means the ignition coil will work, but will start to “refuse” at lower temperatures than earlier, before the first failure event.
Exactly the same – degrading – is the manifestation of another defect of the ignition coils. If the insulator (body of the ignition coil or pipe) is damaged for one time (the spark “shoots” via it), then the “scar” is formed in the place of damage – the material becomes less sturdy. Next damage will happen faster and with a lower load.
Judging by the symptoms, ELDOR (which is also Bosch) ignition coils have several typical defects:
a. high-voltage “piercing” inside the coil. As a result, the “useful” energy of the spark is reduced, and misfires appear (it is typical that the Combustion time is not below the minimum 0.5 ms threshold, no error messages recorded about this). DME poorly recognizes this defect (because they can not use the indications of the flywheel sensor – consequences of the Siemens patent war), typically – the exhaust is flooded with fuel; white smoke, total crash of the adaptations is guaranteed;
b. possibly – also the thermal runway of the diode, due to which the energy is not stored in the ignition coil (it is used before storage in the iron core), and stable misfires are created. In this case, DME detects 0 ms Combustion time and correctly performs the cylinder shutdown frequency.
