MC68HC908AZ60A FLASH LOADING FAILURE

[LIN_JASON]

Dear Support Team,

I am writing to seek your expert advice regarding a recurring issue we are experiencing with the MC68HC908AZ60A microcontroller in our rail transportation products.

These products have been deployed in the field for over a decade, with some units exceeding 20 years of service. Acknowledging that the datasheet specifies a minimum data retention time of 10 years for the on-chip FLASH memory, we proactively performed a firmware update on all devices in 2023.

However, in June 2025 (a period of relatively high ambient temperature), we encountered a batch of failures where multiple MCUs could not bootload the program from the internal FLASH. Our failure analysis yielded the following observations:

1. Oscilloscope measurements of the MCU's RST pin showed that it was being pulled low repeatedly at intervals of approximately 32.85 ms, with a pulse width of about 5 µs, preventing successful startup.

2. Using a Multilink Universal programmer, we attempted to read the memory content in the address range 0x0E00~0x8000. The readback data was consistently 0xAD across this entire segment.

3. The faulty devices were recovered by reprogramming the firmware using the emulator. After reprogramming, the units functioned normally and could successfully boot from the FLASH.

Based on the described symptoms, we would appreciate your assistance in identifying the root cause of this failure.

Furthermore, we have several specific questions regarding this issue:

1. Data Retention Interpretation: Could you please clarify the definition of the "10-year data retention" specification? Does this imply that the memory cells are guaranteed to retain data for a minimum of 10 years under specified conditions, after which data loss may occur? Or does it mean the chip reaches its end of life and becomes unusable after this period? If the data is only read and not erased, does reprogramming the device effectively "reset" this retention timer for another 10-year period?

2. Security Mode Mechanism: The datasheet mentions that a failure in Security Mode could lead to similar symptoms. Could you elaborate on the principle of the Security Mode and how its failure might manifest as the observed bootloops and specific data readback (0xAD)?

3. Known Issues & Case Studies: Are there any documented similar failure cases specifically for the MC68HC908AZ60A or other microcontrollers within the HC08 family? We would be grateful if you could list any typical failure modes related to FLASH memory and their recommended solutions.

4. Recommended Solutions & Alternatives: Considering the constraint that a PCB redesign is not feasible at this stage, what would be your recommended course of action?

   （1）Is a pin-to-pin compatible replacement MCU available that offers better long-term reliability or enhanced FLASH retention?

   （2）Is it possible to mitigate this issue or extend the effective data retention time through a specific firmware update strategy? Thank you for your time and patience in addressing our concerns.

We greatly value your technical expertise and look forward to your insights.

Best regards.

#MC68HC908 #HC08

[Pablo_Ramos]

Hi @LIN_JASON,

I'm reviewing the issue. In the meantime, could you help me with the following questions?

If you read from address ranges other than 0x0E00–0x8000, is the read value still 0xAD?

Is there any block protection affecting the address range 0x0E00–0x8000?

Best Regards,
Pablo

[LIN_JASON]

Hi Pablo,

Please see the attached files. I have sent you the data readbacks (from address 0x0000 to 0xFFFF) of the Flash memory for both the normal and faulty boards. You can use Notepad++ to compare the data at each address.

Best regards,
JASON

#HC08

[Pablo_Ramos]

Hi @LIN_JASON,

I’ve sent you a private message through the community.

Please let me know if you're unable to see it.

Best Regards,

Pablo

[LIN_JASON]

HI Pablo，

Could you help answer each point according to the questions raised?

Best Regards,

JASON

[Pablo_Ramos]

Hello,

I sent a private message; would you please help me check it and confirm on it the information? 

Regards, 

Pablo