S32K MCU Processor Becomes Shorted

cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

S32K MCU Processor Becomes Shorted

463 Views
S32kUser
Contributor I

We are using FS32K144UAT0VLLT processors and experiencing problems with the MCU processor developing an internal short. It develops a short internally within the MCU processor and draws excessive current from the 5V regulator. The current draw can be as high as 1.0A. (When the MCU processor is replaced, the new processor does not have the short and our board functions properly again.) The processor receives it's power via a 12V power supply, which goes through a LM2676 regulator down to 8V. The 8V goes to other components and then goes through a uA7805 regulator down to 5V. This 5V is powering the MCU processor. The failure doesn't occur during extended steady operation.  This internal short is usually detected after our board has been powered off and then back on.  VDDA, VDD1, VDD2, VDD3, VDD4, & VREFH are all tied together to the supply voltage.

Any ideas of what to check for to help diagnose the problem?
Is this an indication of supply voltage going above 5.8V?
What are some things that would damage the processor to have an internal short?
Is this an indication of excessive load currents on the processor?  (Every time we power up, we turn on 60mA of LED's plus normal IO operations at the same time.  If we exceed IOHT, Output high current total for all ports, of 100mA, would it cause this damage?)

0 Kudos
1 Reply

447 Views
danielmartynek
NXP TechSupport
NXP TechSupport

Hello @S32kUser,

If the MCU gets shortout, there are probably some operational conditions out of the specs.
Can you confirm it follows the absolute maximum ratings.

Can you please share the schematic of the board?
Is it possible that some I/O pins can be powered before the MCU (VDD) is powered?
If the voltage at I/O is higher than VDD, the MCU is then powered by the injection current through the pins (clamped to VDD).
The injection current must be limited to +-3mA per a pin.

 

Regards,

Daniel

0 Kudos