Unexpected Current Spike Between Advertising Events in Direct Advertising Sample (nRF54L15)

RE: Unexpected Current Spike Between Advertising Events in Direct Advertising Sample (nRF54L15)

Einar Thorsrud — Wed, 28 May 2025 14:45:38 GMT

Hi Muhammed,

The peak looks higher than I would have expected, but it depends on several factors and measurement setup. I am not able to see from the image, but it may be very short in time so that the total energy is what you would expect.

RE: Unexpected Current Spike Between Advertising Events in Direct Advertising Sample (nRF54L15)

Muhammed — Wed, 28 May 2025 03:46:14 GMT

Hi Einar,

[quote userid="7377" url="~/f/nordic-q-a/121664/unexpected-current-spike-between-advertising-events-in-direct-advertising-sample-nrf54l15/536774"]That looks like CPU processing. It looks higher than I woudl have expectedx, but that could be due to measurment setup or configuration. Can you try to get a better plot showing more details of it?[/quote]

This is the oscilloscope image for better view.

Measurement setup

using keysight current probe for the measurement. Connected to P6 of nRF54L15DK. Disabled UART and Debug interface using board configurator.

There is no application modification rather the i mentioned in the previous post.

Thanks
Muhammed

RE: Unexpected Current Spike Between Advertising Events in Direct Advertising Sample (nRF54L15)

Einar Thorsrud — Fri, 23 May 2025 12:24:40 GMT

Hi Muhammed,

[quote user="Basheer"]Is there any other peripheral running in this example code[/quote]

There are quite a few used by the BLE stack, but the only peripherals runing continiously would be the GRTC and the UART for logging (before it has been disabled).

[quote user="Basheer"]The spike in 1Hz is expected, That is advertising spikes. However we are worried about the spikes (~2mA) in between the adv interval. The first plot which i shared in the first post is the zoomed-in view of the spikes of in between the adv interval. [/quote]

I am not clear on which spikes this were, but I saw a current consumption floor (but it will not be flash, as the power suply is DCDC, so the current draw is not constant).

[quote user="Basheer"]However after this configuration we got a big spike (~30mA) before the all advertisement. Please find the scope image below.[/quote]

That looks like CPU processing. It looks higher than I woudl have expectedx, but that could be due to measurment setup or configuration. Can you try to get a better plot showing more details of it?

[quote user="Basheer"]This spike occurs before all advertisement. Is it because of wakeup from the sleep? If not how to eliminate it?[/quote]

Yes, I expect so. The pattern matches expected CPU processing, starting HFXO etc, when the device wakes up for the BLE event.

RE: Unexpected Current Spike Between Advertising Events in Direct Advertising Sample (nRF54L15)

Muhammed — Thu, 22 May 2025 09:03:17 GMT

Hi Einar,

Thanks for the quick response.

I have used the sample application provided Nordic SDK v3.0.0 which is "Bluetooh Direct Advertising". In this code i have modified the advertising interval and packet length only. In the non advertising time the system is going to sleep mode as per the code. Please find code snippet below.

Is there any other peripheral running in this example code? I didn't enabled anything apart from the above two modifications.
The spike in 1Hz is expected, That is advertising spikes. However we are worried about the spikes (~2mA) in between the adv interval. The first plot which i shared in the first post is the zoomed-in view of the spikes of in between the adv interval.
Disabled the UART as you suggested, Current reduced
- Added CONFIG_SERIAL=n in the project configuration.
However after this configuration we got a big spike (~30mA) before the all advertisement. Please find the scope image below.

This spike occurs before all advertisement. Is it because of wakeup from the sleep? If not how to eliminate it?

Thanks,

Muhammed

RE: Unexpected Current Spike Between Advertising Events in Direct Advertising Sample (nRF54L15)

Einar Thorsrud — Wed, 21 May 2025 12:11:19 GMT

Hi Muhammed,

[quote user=""]What is causing this periodic current spike?[/quote]

The spikes I see at 1 Hz is expected given that that is the advertising interval, so the spikes are expected.

[quote user=""]Is it related to any background system task, RAM retention, clock calibration, or BLE stack behavior?[/quote]

I assuem the problem here is the high idle current (the current consumtpion between the spikes caused by advertising)? This looks like more ocnstant curernt consuption (though you wil see high frequency spikes due to DC/DC operation).

[quote user=""]How can we suppress or eliminate these spikes to optimize current between advertising events?[/quote]

Disable any peripherals or resources that is not needed durign sleep, paritcularily any that requier the high frequech clock, such as UART or other serial interraces. There are some typicall reason for high current consumption, and often it is caused by UART logging. Do you have UART enabled? If so, you should disable it. See Power optimization recommendations. If that does not help, we need to look a bit further, and would need more information about your application.

PS: I suspect that some information may have been lost through processing of AI. If so, please post the original question without AI processing and further replies without AI as it is difficult to know which information is correct and what is AI hallucination.