HolyIoT nRF54L15 dev kit using 2mA with system_off sample

[Tom McGuinness]

Hi,

I'm looking for some guidance.

I'm using a HolyIoT nRF54L15 dev kit to evaluate their nRF54L15 module.

I am currently looking power consumption. 

I have tried the boards/nordic/system_off sample on my official nRF54L15 dev kit and it reaches the expected 1µA.

When I try the same sample on the HolyIoT board, it consumes about 2mA, which tells me it's not going to sleep!

I've reached out to Holyiot Technology asking them for a sample project, but figured I could try and figure this out myself in the mean time.

Can you offer me any guidance on what steps to try? 

I've tried a custom overlay

&uart20  {
	status = "disabled";
};

&uart30 {
	status = "disabled";
};

&pwm20 {
	status = "disabled";
};

&spi00 {
	status = "disabled";
};

&adc {
	status = "disabled";
};

and I've also tried turning off a bunch of stuff. This is my prj.conf file

CONFIG_PM_DEVICE=y
CONFIG_GPIO=n
CONFIG_CRC=n
CONFIG_POWEROFF=y
CONFIG_HWINFO=n
CONFIG_GPIO_WAKEUP_ENABLE=n

CONFIG_SERIAL=n
CONFIG_CONSOLE=n
CONFIG_USE_SEGGER_RTT=n
CONFIG_LOG=n

CONFIG_PM_CPU_OPS=y
CONFIG_PM_DEVICE_RUNTIME=y
CONFIG_PM_DEVICE_POWER_DOMAIN=y

# From Power Down sample
CONFIG_USB_DEVICE_STACK=n
CONFIG_DEBUG_OPTIMIZATIONS=n
CONFIG_DEBUG_THREAD_INFO=n

CONFIG_RAM_POWER_DOWN_LIBRARY=y

CONFIG_LTO=y
CONFIG_ISR_TABLES_LOCAL_DECLARATION=y

Nothing happens. Consumption measured by my PPK II sits around 2mA.

[Mike Austin (LPI)]

Its clearly a hardware "issue" with the HolyIOT module, as you've got confirmation of the System OFF current using the nRF54-DK using the same firmware.  I have "issue" in quotes, as it might just be that you aren't configuring the hardware you have correctly with your firmware.

A couple of ideas to look into:

• What board info are you using when you build your firmware?  I can see there is a HolyIOT board file for their nRF52850 module, but doesn't look like there is one for the hardware you are using.
• You may need to get a schematic of their module, and take a look at how they are configuring buttons and LEDs.  It might be that their LEDs or buttons are configured in the opposite sense to the nRF54-DK, which means the system_off example is actually turning some GPIO on when they should be off.
• There is a high probability that the GPIO that the HolyIOT module is using for the LED(s) and Button(s) are not the same ones as the nRF54-DK is, so where the example code is putting GPIO into the correct low power state, it may actually be turning GPIO on with your HolyIOT unit
• You will need CONFIG_GPIO=y in order to correctly configure GPIO states

A schematic will certainly help you get a better idea, and would enable you to put together a more suitable board file for the module you are using.

I noticed on your Reddit post that people are suggesting the LDO as a potential source of excessive current draw.  Not sure this is your problem though.  Even a rubbish LDO regulator will have a quiescent current in the uA range, and if your MCU isn't drawing current (or drawing very little), then its really only the quiescent current that you'll be seeing if the MCU truly is in System OFF.

The above are the sorts of issues  you'll see people have had to resolve when trying to get System OFF currents down to the target values on some of the other nRF processors, particularly on custom boards (myself included)

Cheers,

Mike

[Tom McGuinness]

Thank you.

I'm trying to get my hands on more information from HolyIoT, but they are being slow to respond. 

I'll update when I know more!

Tom

[JONATHAN LL]

Hi Tom, 

I see that there is a both manufacturer and EU person contact info, from this listing: https://www.aliexpress.com/item/1005008993181489.html#nav-description 

Manufacturer information
Name:Shenzhen holyiot technology Co.,LTD
Address:Rm.308, 310, Bldg 9. Wangtang industrial Park xinwei Vil., xili Town. Nanshan Dist. Shenzhen China 518055
Email:[email protected]
Phone:[phone]

EU responsible person information
Name:Humiss International B.V.
Address:Joop Geesinkweg 701，Amsterdam-Duivendrecht , the Netherlands
Email:[email protected]
Phone:[phone]

I am also not able to find any more detailed information or schematics here for this so I would try to double down on asking them to share this info. 

Regards,
Jonathan