This now is the savings circuit, as it was used in earlier types of ignition coils. The primary- and the secondary windings are connected, at one end to each other, and with terminal-1.
Here the two circuits are completely separated from each other. Despite the galvanic separation, when working on the ignition one should never take it for granted that a high tension current can not also jump over to the primary system. In todays twin-spark twin-spark coils there is a spark plug on both 4a and 4b, whereas this, in the single-spark coil is only connected to 4, and 4a with earth.
Problems with the ignition coil should not only be solved by replacement, but detected through measuring. If one is certain about the power supply and one of the signals necessary for the ignition on terminal-1, the tendency is towards measuring both windings using an Ohmmeter and then possibly, comparing the readings with the manufacturers data.
This is not a very thorough or reliable method of checking the ignition, in this case, one can only find actual breaks in the coil. One can also determine the resistance of each one against the casing, indeed, this only helps if a direct contact is present.
These are not typical ignition coil errors. Since how much movement is possible, e.g., inside a molded system, without it also being visible from the outside? Temperature influence is more likely, particularly with single-spark coils, or also the development of hairline cracks through which a spark can jump to earth.
In addition, with older ignition coils, while in the stand-by circuit, measurement may still be possible without problems, later models however, may have a diode in the secondary circuit, which makes it advisable to measure the flow in both directions. The success of such a measure depends on the tension-passage of this diode, which, in the high tension part can be much higher than the well known 0,7 V.
Should the Ohmmeter find no through-flow, don't simply throw the coil out. It should be possible using 12 V and a (diode-) testing lamp, either that, or one has ascertained the defect successfully. Should plus be connected, one can generally, by connecting the individual one or the respective terminal-1 to earth, still manage to get a spark.