nRF52840DK current consumption tests using PPK II

RE: nRF52840DK current consumption tests using PPK II

Håkon Alseth — Thu, 15 Feb 2024 11:55:31 GMT

Hi,

Having several power sources present simultaneously has been improved in newer board revisions, but I would recommend that you only keep one active.

Kind regards,

Håkon

RE: nRF52840DK current consumption tests using PPK II

zazas321 — Thu, 15 Feb 2024 11:34:15 GMT

Sure, but can you confirm if there is a difference between the USB plugged in and USB not plugged in when using nRF only mode? From what I understand there should be no difference?

Anyways, I have performed tests with CONFIG_SERIAL=n

Here are the results:

TEST8

Using Li-Po battery to power the board and having the USB disconnected

SW6: nRF only mode
nRF Power source: VDD
nRF52840 powered by the Li-po battery (3.8V)
PPK II in Ampere meter mode
USB (J2) to disconnected

As you can see form the graph, I can see that the CPU enters system ON sleep mode and wakes up to perform some actions (toggling the GPIO) every 1 second. I guess that makes sense 🙂

TEST9

SW6: nRF only mode
nRF Power source: VDD
nRF52840 powered by the Li-po battery (3.8V)
PPK II in Ampere meter mode
USB (J2) to disconnected

Despite the fact that I have SW6 in nRF only mode, I am measuring the same current consumption as the nRF only mode. Is that because I dont have the USB plugged in which is required to power the interface MCU?

TEST10

SW6: Default mode
nRF Power source: VDD
nRF52840 powered by the Li-po battery (3.8V)
PPK II in Ampere meter mode
USB (J2) to connected

Since I have the USB connected, the current shoots up every 1 second. Does that have to do anything with the Interface CPU being powered?

TEST11

SW6: nRF only mode
nRF Power source: VDD
nRF52840 powered by the Li-po battery (3.8V)
PPK II in Ampere meter mode
USB (J2) to connected

Despite the fact that I have the USB connected, I have the SW6 in nRF only mode so I assume that interface MCU is disabled (opposite as TEST10) is my understanding correct?

RE: nRF52840DK current consumption tests using PPK II

Håkon Alseth — Thu, 15 Feb 2024 10:18:17 GMT

Hi,

NRF_ONLY position should give the most accurate measurement:

https://infocenter.nordicsemi.com/topic/ug_nrf52840_dk/UG/dk/nrf_only_mode.html?cp=5_0_4_7_3_1

Kind regards,

Håkon

RE: nRF52840DK current consumption tests using PPK II

zazas321 — Wed, 14 Feb 2024 14:04:00 GMT

Il repeat the tests with serial turned OFF tommorow.

Additionally, I would like to further understand if there is a difference if I have the USB plugged in when performing measurements or not? I was under the impression that if the switch is set to nRF only there should be no difference if the rest of DK is powered since we only measure nRF only?

Is that right?

RE: nRF52840DK current consumption tests using PPK II

Håkon Alseth — Wed, 14 Feb 2024 13:56:35 GMT

Hi,

[quote user="zazas321"]What would be the easiest method to see the device enter the system ON mode (measuring current in few uA range)? As you can see from all the graphs the device did not seem to have entered proper low power mode even though for some tests I have used nRF only mode? Since my measurements are not taking into account LED current consumption, I would assume to see the device enter ultra low power mode when the LED is turned OFF and the CPU is idling..[/quote]

Have you set CONFIG_SERIAL=n in your prj.conf?

Kind regards,

Håkon

RE: nRF52840DK current consumption tests using PPK II

zazas321 — Wed, 14 Feb 2024 13:27:20 GMT

[quote userid="2115" url="~/f/nordic-q-a/108294/nrf52840dk-current-consumption-tests-using-ppk-ii/468823"]The correct way depends on what the requirements are. The DK is setup to only measure the nRF + what-ever is connected to the GPIOs (except for the buffered LEDs)[/quote]

Ok lets forget the LED's for now (perhaps I will make modification to solder some bridges as you have suggested later)

What would be the easiest method to see the device enter the system ON mode (measuring current in few uA range)? As you can see from all the graphs the device did not seem to have entered proper low power mode even though for some tests I have used nRF only mode? Since my measurements are not taking into account LED current consumption, I would assume to see the device enter ultra low power mode when the LED is turned OFF and the CPU is idling..

Could you provide some steps how to achieve this and perhaps it will get clear to me what was the issue.

RE: nRF52840DK current consumption tests using PPK II

Håkon Alseth — Wed, 14 Feb 2024 13:16:39 GMT

Hi,

When measuring current, it is important that you don't back-power the peripherals/logic around the nRF on the DK itself.

You can find the schematic and layout of the DK on our website, for reference:

https://www.nordicsemi.com/Products/Development-hardware/nRF52840-DK/Download?lang=en#infotabs

I'm combining the answer on questions 1 to 4, as the root-cause is the same:

It is because the LEDs are buffered through transistors, powered on a different voltage net.

[quote user=""]5. Regarding TEST5. Finally able to see the graph that makes some sense. The current increases/decreases at 1second intervals.. However the system does not seem to enter proper System ON low power mode when the task is idling. [/quote]

Here you are sourcing the whole DK, including switches/buffers/etc.

[quote user=""]6. Regarding TEST6 and TEST7 I am not fully understanding why there is no difference whether I have the USB plugged in or plugged out?[/quote]

This is because you are powering the whole board via the external power. Here's the logic for v2.0.x of the DK (which looks like the rev you have):

[quote user=""]7. So what is the truly correct way to measure the nRF52840DK current consumption? I am interested in the current consumption of the nRF52840 and all peripherals connected to it (LED's and whatever other sensors that I plug in and use). For this, I assume I should use nRF only mode to ensure the Interface MCU is cut off. But as I shown in my experiments above, when using nRF only mode the cpu does not seem to run properly - the LED's wont even blink.[/quote]

The correct way depends on what the requirements are. The DK is setup to only measure the nRF + what-ever is connected to the GPIOs (except for the buffered LEDs)

As per yours, it includes the VDD_PER net, which will require that you connect some solder bridges and using the "NRF ONLY" position:

Kind regards,

Håkon