nRF52832 DK sleep mode current consumption too high

RE: nRF52832 DK sleep mode current consumption too high

Simonr — Fri, 24 Jan 2025 15:12:04 GMT

I'd recommend that you cut solder bridge SB9 on the nRF52832 DK and power the board over P22 instead of P21, since P21 will result in you powering the entire DK and not just the nRF52832 DK.

I'll try setting this up next week on my end.

Best regards,

Simon

RE: nRF52832 DK sleep mode current consumption too high

James168 — Thu, 23 Jan 2025 09:37:22 GMT

Hi,

1) The setup measurement environment is like the following,

The right black line is connected to DK's P21(-), and the right red line is connected to DK's P21(+).

The DK's P22 is connected nothing.

2) The sampling rate is the highest, and can you set up the environment on your end? Just need nRF 52832 DK, PPK2, and peripheral_uart sample, then you can zoom in/out the waveform to watch more clearly.

As your suggestion, I added more waveform graphs.

zoom in

3) I have changed to use NCS v2.9.0 now.

RE: nRF52832 DK sleep mode current consumption too high

Simonr — Wed, 22 Jan 2025 14:17:59 GMT

1. When a supply is connected to P22 (current measurement header) it will have to power the entire board (debugger MCU included) if you don't also power it using J2. There are voltage detectors on the board that makes P22 only supply the nRF52832 SoC.

P21 is for external supply for the entire DK, and not just the nRF52832.

2. Are you able to get a higher resolution image here, as these spikes seem very low resolution, and hard to tell what is exactly, but if you've powered the board with P21, I assume it's related to some of the other components on the DK, and not the nRF52832 SoC.

Best regards,

Simon

RE: nRF52832 DK sleep mode current consumption too high

James168 — Wed, 22 Jan 2025 09:11:18 GMT

Hi,

1) Do you mean connect USB J2 and External supply P21 at the same time for current measurement? In this connection will has two power supply? I just connect P21 to power monitor for current measurement before.

2) To use the unmodified peripheral_uart sample, keep

CONFIG_SERIAL=y

there is a pulse about 0.9ms interval and peak current value about 3.5mA, which code cause this current period?

If this 3.5mA current can be reduced, the current consumption improves a lot.

The ideal goal is approached the "Online Power Profiler for Bluetooth LE" value.

RE: nRF52832 DK sleep mode current consumption too high

Simonr — Tue, 21 Jan 2025 13:19:27 GMT

1. The DKs should also be powered through the USB connector J2 so that the external supply only needs to power the SoC. Make sure the DK is connected to a power source and is turned on and then you can supply the SoC through the P22 header.

3. The UART peripheral will consume some power, but when the device is not required to be connected to a device over NUS, you can disable the radio and UART peripheral and go into a low power state. What current consumption are you aiming at exactly? There is no way to stay I.E. in single digit µA territory while also maintaining a BLE connection for example.

Best regards,

Simon

RE: nRF52832 DK sleep mode current consumption too high

James168 — Tue, 21 Jan 2025 06:43:27 GMT

Hi,

1) I measured the current with system_off sample,

nRF52832 DK

average current is about 322uA

nRF52840 DK, sw6 in DEFAULT

average about 1.9mA

nRF52840DK, sw6 in nRF ONLY

average current about 5.2uA

2) The setup environment is power monitor connected to the External supply pins of the DKs with 3.0V power,

and I don't have the PPK2 to compare.

3) I do some tests these days, disable the UART can reduce the current consumption, but I need to have both the NUS service AND low power mode.

RE: nRF52832 DK sleep mode current consumption too high

Simonr — Mon, 20 Jan 2025 10:19:21 GMT

So, I talked about this with a colleague, and the reason for the difference between the nRF52832 and the nRF52840 DKs is mainly because there was some improvements to active current made on the nRF52840 DK. In nRF52832 DK, the RX UART will both have the HF clock and DMA bus running.

However, it seems like you are measuring the current consumption of the entire DK here, and not just on the nRF52 SoCs, since we're also seeing the LEDs in the current measurements here.

We recommend that you start out with the device in system OFF mode so we can confirm that the setup is correct on your end. We have the system OFF sample project available here for reference.

Best regards,

Simon

RE: nRF52832 DK sleep mode current consumption too high

James168 — Fri, 17 Jan 2025 01:57:06 GMT

Hi,

nRF52832 DK is

PCA10040

2.1.0

2021.35

nRF52840 DK is

PCA10056

2.0.1

2020.46

RE: nRF52832 DK sleep mode current consumption too high

Simonr — Thu, 16 Jan 2025 14:24:55 GMT

What version of the nRF52832 and nRF52840 DKs are you seeing these numbers on?

Best regards,

Simon

RE: nRF52832 DK sleep mode current consumption too high

James168 — Thu, 16 Jan 2025 10:40:14 GMT

Hi,

1) To test with unmodified code, the LED toggle in 1000ms, so I measure the current when LED off to avoid LED current consumption.

To measure when LED off stage.

2) nRF52832 DK

between the advertising, the minimum and average current still has about 1.2mA

3) nRF52840 DK, SW6 on DEFAULT position

LED1 toggle

measure the current between advertising,

the average current is about 536uA

RE: nRF52832 DK sleep mode current consumption too high

Simonr — Thu, 16 Jan 2025 08:39:17 GMT

Please share how you're doing current measurement. The device is advertising here, so some current consumption is expected. The nRF52840 DK has an nRF ONLY button that isolates the nRF52840 from the IMCU debugger, buttons and LEDs on the DK, while the nRF52832 DK does not have this. I assume the "LED off" you refer to here is referring to the nRF52840 DK and not the nRF52832. Please note that none of the DKs are in sleep mode here and the floor current on the nRF52832 DK is the current consumption for the DK with logging and debugging enabled.

Best regards,

Simon