BlueNRG-345 custom board repeatedly resets due to Hardware_Code = 0x01

[RGord]

I am working on a custom board with a BlueNRG-345 microcontroller, with the BlueNRG-LP Development kit 1.2.0 and using WiSE Studio as the development environment. I have taken the BLE Beacon project for the STEVAL-IDB011V1 and converted it to run on the BlueNRG-345 as described in section 6.2 of the Development Kit User Manual. The custom board is based on the reference design but it does not have an LSE so I have replaced the CONFIG_HW_LS_XTAL defined symbol with CONFIG_HS_LS_RO. I have commented out the call to HAL_PWR_MNGR_Request() so that I can run it under the debugger.

The project runs happily on an STEVAL-IDB011V1 but when I run it on the custom board it goes into a tight reset loop. The reason it is resetting is that the user code function hci_hardware_error_event() is being called with Hardware_Code = 0x01 and hence it calls NVIC_SystemReset(). If I change the code so that it only resets when Hardware_Code == 0x03, then the beacon functionality does work, at least for a while.

I have a separate project with the target application code and pins properly configured to match the custom hardware. This still suffers from the above problem but otherwise works fine. In this project I have avoided getting Hardware_Code = 0x01 by commenting out the calls to HAL_VTIMER_Tick() and BLE_STACK_Tick(). This appears not to stop the BLE Beacon function from working, which is fine so far as that is the only BLE function I need but it would be a problem if other BLE functions were needed in the future.

The bluenrg_lp_events.h files says that "Hardware_Code Error code 0x01 and 0x02 are errors generally caused by hardware issue on the PCB; another possible cause is a slow crystal startup." What hardware issues might be causing this or what might I have done wrong in the software?

[Laurent LOUAZON]

Hello,

This could be related to Xtal selected for example.

Feel free to share with me your schematics (at least BlueNRg part) and Xtal full reference.

Quick tests could be

• Use LSI

If you are using a 32kHz xtal and problem is related to this, you can quickly built your application with CONFIG_HW_LS_RO pre processing option for using LSI

If this solves the problem, it means problem is related to 32kHz LSE

• Increase HSE start up time

If problem is related to HSE start up tie you can do a test pushing this value to maximum.

#define HS_STARTUP_TIME 320 // 780 us is the default setup and you can try chnage from 320 to 778 (1900 us). It will just impact a bit power consumption.

.If this solve the problem, it means problem is related to 32MHz HSE

Regards

Laurent

Regards

Laurent

[RGord]

Hi Laurent

Thanks for responding.

• Use LSI

The board does not have an LSE so I am already using LSI by building with CONFIG_HW_LS_RO. (There is a typo in my original post.)

• Increase HSE start up time

I think I tried this before but I have tried it again anyway and it made no difference. I am still seeing hci_hardware_error_event() being called with Hardware_Code = 0x01 causing a reset loop.

This is the BlueNRG-345 part of the schematic. Hopefully this is clear enough for you to read.

The HSE, Y2, is an ECS-320-8-37CKM-TR.

Regards

Rob

[Laurent LOUAZON]

Hi Rob,

I see some problem here.

C24 should not be a cap but a 10uH coil.

Also recommended value for C32 is 220nF.

Regards

Laurent

[RGord]

Hi Laurent

Thanks for speedy response. I've passed this information on to our hardware engineer and hopefully this will solve our problem.

Kind regards

Rob

[RGord]

Hi Laurent

It turns out that our hardware guys had spotted the C24 cap/coil error already as without this change the board would not work at all. They have now changed cap C32 to 220nF and this cures the problem I was having with the firmware going into a reset loop.

Thanks very much for your help with this issue.

Kind regards

Rob

[Grzegorz.Bednarski]

I assume mute option is ok when i solved problem?