Hello everyone! We used the S32K148 in our project, and we conducted a test where we reversed the polarity of the battery. In this scenario, the board should not be damaged and should function properly once the fault is rectified. When this happens, the MCU (S32) is powered off because we have an LDO and reverse battery protection in place, which results in the VCC of the MCU floating. However, the GND of the MCU is connected to the positive terminal of the battery. In this state, we have 12 ADC channels where, under reverse battery conditions, a negative voltage is applied to the MCU's ADC pins. This situation causes an injection current of approximately -60mA. Interestingly, the MCU does not get damaged despite the datasheet indicating that the maximum sum of injection currents should be less than 30mA. Could this be because the MCU is in an off state, or is the injection occurring through an alternate path instead of the clamp diode (such as the body diode in the last push-pull stage of the MCU)? We tested the MCU under these harsh conditions for several hours, and it did not sustain any damage, which is quite puzzling for us.
Thanks.
#S32K148
#ADC
#injection current
Hi,
having injected current over limits doesn't mean to have an instant silicon damage. There could be small leakages which can increase over time.
It doesn't matter if injecting current flows through clamp diode or body diode of the last push-pull stage, or on MCU state. Definitely, you should avoid this situation.
I suggest to limit the injected current by resistors.
Alternatively, you may use the common trick of reverse power supply protection with nmos and resistor. Here is an example from S32K3 T-BOX:
Best regards,
Pavel
