S32K3 HSE

[ALXY_WANG1]

Currently, we are using the S32K344 + FS6500 chip in combination with it for the VCU controller. This project is the first to use the HSE module to meet the information security requirements. The used RTD version is 1.0.0, and the HSE version is 0.2.1.0. Currently, the project has entered the small-scale verification stage. Currently, there is a problem:
Occasionally, after power-on, the VCU cannot start. Testing revealed that the power output of the SBC is normal after power-on, but the LED indicator controlled by the MCU does not work. Testing found that the SPI of the SBC did not send communication after power-on, and the REST pin continuously emitted periodic reset signals with a period of approximately 200ms, pulling low for about 8ms. Currently, we suspect that it gets stuck in the HSE part after power-on and does not reach the SBC initialization, resulting in a timeout reset within 256ms.
However, we have discovered an unexplainable phenomenon. When we disconnect the reset pins of the MCU and SBC, the controller recovers. After recovery, if it is reconnected, the problem cannot be reproduced. Or if a debugger is attached, it also recovers. After recovery, the fault cannot be reproduced either. Currently, we are unable to explain this phenomenon.

[guoweisun]

我这里的support SBC的。

从SBC的角度看您的问题以及debug的建议：

1：如果首次看门狗刷新不正确，FS45/FS65 直接复位，FS45/FS65 再次进入 INIT_FS 模式，并在 256 毫秒内请求首次正确看门狗刷新。

如果首次看门狗在 256 毫秒内未刷新，FS45/FS65 直接复位，FS45/FS65 再次进入 INIT_FS 模式，并在 256 毫秒内请求首次正确看门狗刷新。

此时MCU的reset接在SBC 的RSTB pin在刚开始上电的256ms以内喂狗不对或者喂狗超时 RSTB 都会被拉低一个pulse，会直接reset MCU。

2：

a:当我们断开 MCU 和 SBC 的复位引脚时，控制器恢复--》控制器恢复如何理解？

我这边测试看到假如是您断开RSTB那么SBC 可以给MCU最长供电大概1.3s，然后再不喂狗 后续的output power就没有了

也就是说在这1.3s以内您的控制器程序跑起来可以应答SBC了。

b:恢复后，如果重新连接，问题无法重现->你说的重新连接是说的从新连接RSTB吗?

您需要在250ms以内把MCU的程序跑起来这样MCU 就不会被reset。或者就是启动HSE之前把SBC watchdog关掉，完成以后再打开

[ALXY_WANG1]

目前该问题大概率是MCU启动停留在HSE内核，应用核并未运行，MCU相关的技术人员是否可以指导一下，给点建设性意见，感谢

[ALXY_WANG1]

启动HSE之前把SBC watchdog关掉，这个的具体有哪些实现方式，因为目前分析大概率MCU软件并未运行

[guoweisun]

 WD_WINDOW register 这个寄存器里设置WD_WINDOW_3:0 bits as ‘1111’--》 1024ms我觉得差不多够你更新HSE了 这个可以在任何模式下设置比较随意，如果要彻底关掉必须要在INIT 的阶段操作上述的寄存器 WD_WINDOW_3:0 为‘0000’

[ALXY_WANG1]

a:断开MCU 和 SBC 的复位引脚后，控制器可正常运行，应用程序运行正常
b：重新连接是将原先断开的MCU 和 SBC 的复位引脚中间的二极管重新焊接上，焊接上无法再复现原来的问题
所以目前怀疑的可能是在Boot代码启动前代码卡在HSE耗时较长，导致没有在250ms内运行。

[guoweisun]

好的有问题再联系！