MCU restarts when entering DeepSleep

[TaTa_2447651]

I am developing a product using the CYBLE-416045-02 module.
When nothing needs to work, both CM4 / CM0 + CPUs are DeepSleep using Cy_SysPm_DeepSleep.

In rare cases, the MCU will restart in Cy_SysPm_DeepSleep.
I found that it restarts at the following points of EnterDeepSleepRam.

/ * The CPU enters Deep Sleep mode upon execution of WFI / WFE * /
SCB_SCR | = SCB_SCR_SLEEPDEEP_Msk;

I have modified what is in KBA229335.

The frequency of occurrence varies depending on the individual.
It does not reproduce at all in one individual, but it reproduces 100% in another individual.

Are there any other similar reports?
Also, please give me some advice on how to solve this problem.

[Rakshith · Accepted solution]

Hi @TaTa_2447651, 

As the issue cannot be recreated in all the devices, I believe this could be a hardware issue. Is the board for the 2 modules that you mentioned different? If so, can you please try a swap test with the two devices that you have and let me know your observations?

Please try to program the same project into the PSoC 63 BLE Prototyping Kit and let me know if you are able to observe the same issue.

[TaTa_2447651]

Hi Rakshith.

I swapped the bad device and the normal device MCU.
Then, the problem started to occur on the originally normal device. So I think it's an MCU problem.

In addition, after a lot of research, I found that setting the core voltage solved the problem.
I have problems setting the core voltage to a 1.1V LDO.
I have no problem setting the core voltage to 1.1V SIMO Buck.
Also, I have no problem setting it to 0.9V LDO, 0.9V SIMO Buck.

Is there anything that could be the cause from these things?
Please give me some advice.

Thanks and Regards.

[Rakshith]

Hi @TaTa_2447651, 

I swapped the bad device and the normal device MCU.
Then, the problem started to occur on the originally normal device. So I think it's an MCU problem.

I am not sure I understood this. You replaced the bad MCU with a new MCU on the bad board. You were able to observe the issue with the new MCU too. So the issue can be isolated to the board right?

Could you please share the schematic design of your board so that we can have a look? Additionally, can you please let us know if you are disabling or making any power configuration change in the firmware before transitioning to deep sleep? 

[TaTa_2447651]

Hi Rakshith.

I will explain again.
I replaced the MCU as shown in the picture below.
As a result, there was a problem with the device that was originally good.
The device that originally had the problem is no longer the problem.
From this result I thought there was a problem with the MCU.

Also, on the device with the problem, I changed the core voltage setting to BUCK, then the problem disappeared.

Thanks and Regards.

[Rakshith]

Hi @TaTa_2447651, 

Thank you very much for explaining in detail. I now understand the procedure followed. Could you please share the schematic design of your board and let me know if you are disabling or making any power configuration change in the firmware before transitioning to low power mode? 

[TaTa_2447651]

Hi Rakshith.

The circuit diagram of the board is a product specification. So it cannot be shared. I'm sorry.

Instead, I share the development environment of the PSoC Creator that has the problem.

This firmware will still have problems even after removing as much functionality as possible from the product source code.
The green and red LEDs will flash for 30 seconds after the power is turned on. After 30 seconds, the red LED will flash 3 times and then enter Deep Sleep.
I haven't disabled or changed the power configuration before entering DeepSleep.

If something goes wrong, the last red LED will not flash and the MCU will freeze or restart.

When building, please do a release build.
This firmware also works with the CY8CPROTO-063-BLE board.

Thanks and Regards.

[TaTa_2447651]

Hi Rakshith.

I appreciate your advice.

I tried it with the PSoC63 BLE prototyping kit, but it didn't reproduce the problem.

We also imagine it's a hardware issue.
So, as advised, I will do a swap test of the MCU on the problematic device and the normal device.
I will contact you as soon as I know the result.

Thanks and Regards.