nRF54 with nPM2100 and nPM1300 Power Consumption Issues

[joey]

HI,

Test Setup

• MCU: nRF54L15  PDK

• SDK: nCS 3.0.0

• Example: Blinky (modified)

Modified Code

main.c:

#include <stdio.h>
#include <zephyr/kernel.h>
#include <zephyr/drivers/gpio.h>

/* 1000 msec = 1 sec */
#define SLEEP_TIME_MS   10000

/* The devicetree node identifier for the "led0" alias. */
// #define LED0_NODE DT_ALIAS(led0)

/*
 * A build error on this line means your board is unsupported.
 * See the sample documentation for information on how to fix this.
 */
// static const struct gpio_dt_spec led = GPIO_DT_SPEC_GET(LED0_NODE, gpios);

int main(void)
{
	// int ret;
	// bool led_state = true;

	// if (!gpio_is_ready_dt(&led)) {
	// 	return 0;
	// }

	// ret = gpio_pin_configure_dt(&led, GPIO_OUTPUT_ACTIVE);
	// if (ret < 0) {
	// 	return 0;
	// }

	while (1) {
		// ret = gpio_pin_toggle_dt(&led);
		// if (ret < 0) {
		// 	return 0;
		// }

		// led_state = !led_state;
		// printf("LED state: %s\n", led_state ? "ON" : "OFF");
		k_msleep(SLEEP_TIME_MS);
	}
	return 0;
}

prj.conf:

CONFIG_GPIO=y

CONFIG_CONSOLE=n
CONFIG_UART_CONSOLE=n

CONFIG_SERIAL=n

CONFIG_PM_DEVICE=y
CONFIG_PM_DEVICE_RUNTIME=y

overlay: Empty

Test Configuration

The firmware is flashed to nRF54L15 PDK. The outputs of the nPM1300 in the PDK and the nPM2100EK are respectively connected to a power analyzer for testing. For the nPM2100EK, the I2C is not connected to the nRF54 PDK, and the output is directly connected. 

Both,  then the power analyzer  supplies power to the PDK's P6 interface for nRF54.

Observed Behavior

With nPM1300 Power Supply:

• Only current peaks at each SLEEP_TIME_MS interval

• System remains in sleep mode the rest of the time

• 

With nPM2100 Power Supply:

• Current peaks at each SLEEP_TIME_MS interval (same as nPM1300)

• Additional small current spikes every 500ms

• 

Common Observation (Both Power Supplies):

• Very small current spikes approximately every 8ms

• 

Questions

1. Why is there this difference between nPM1300 and nPM2100?

   • What causes the additional 500ms current spikes only with nPM2100?

2. What could be generating the 500ms current spikes observed with nPM2100?

   • What internal mechanism or task might be running at this interval?

3. What causes the approximately 8ms small current spikes observed with both power supplies?

   • Is this related to nRF54 internal charging or some periodic operation?

Thanks!

[Benjamin]

Hi,

In very light load condition the boost refresh cycle (PFM mode frequency) would be in this time range.

The smaller 8 ms spikes you see are almost certainly from internal periodic wake‑ups of clocks/interrupts/peripherals that are enabled by default.

Regards,
Benjamin

[joey]

Benjamin said:

In very light load condition the boost refresh cycle (PFM mode frequency) would be in this time range.

HI，

the first question,

Under what load current does the npm2100 boost output enter this mode? When the current is continuously 600 μA, these spaced current pulses do not appear, but when the load current is only a few microamps, they do occur.

Thanks

[Ressa]

Hi,

The feedback from PMIC team was that under "light load condition this spikes would appear" and seems like is compatible with your observation. I have not received exact load current range that this spikes occur but based on your measurements seems like for less than 300,400 uA you would see it and for more than that they will be hidden under actual load current. 

Best regards,
Ressa

[joey]

Thank you for your reply. I have also posted a thread about this issue.

Originally, my entire circuit only draws around 10 µA of current, but due to the presence of pulse currents, the average current has increased significantly. Are there any suggestions on how to reduce this power consumption?

devzone.nordicsemi.com/.../the-current-consumption-of-npm2100

[Ressa]

Please have look at this page as well and try the recommended caps with exact part number as the quality of caps can also play important rule. 
Selecting output capacitors for the boost regulator

[joey]

HI，

Originally, the capacitors for VBAT and VINT were 100uF and 22uF respectively. As suggested in the document, I increased them to approximately 150uF and 120uF, but the average current monitored by power consumption analysis showed no significant change. Does the capacitor value need to be increased even further?

[Ressa]

I don't think you need more than that. have you tried to have some nF caps as well in addition to the bulk ones? this is my own idea that due to lower internal resistance for lower caps those might help better with small spikes.
Also according to the last answer from other ticket you have created, it seems there will be spikes anyway and those will not be fully suppressed. But rather managing them to not impact hugely the whole system is the way to go. 

[joey]

Although I understand it cannot be completely eliminated, I want to minimize this current value as much as possible. I have tried increasing the capacitance, including adding capacitors in the nF range, but it seems to have no effect.

[Ressa]

Hi,
I understand your intention, but if the previously shared graph regarding the relation of caps and peak currents didn't help much, I suppose that is how it behaves in boost mode. Also as PMIC team mentioned in the other ticket, these spikes can happen even if there is no load. I would suggest to investigate how much your application can tolerate the spikes both in terms of average current and also possible RF systems that might be affected. If the requirement are met with these spikes then you can move on.