App unusual current draw

[rookie]

I am facing a strange situation with my app. After turning off all my peripherals and switching to the most minimal code, I see quite a lot of current consumption and that too erratic. 

My minimal code is this:

int main(void)
{
    timers_init();
    power_management_init();
    ble_stack_init();
    sd_power_dcdc_mode_set(1);

    for (;;) {
        nrf_pwr_mgmt_run();
    }
}

The corresponding current graph observed is:

For roughly a minute, there is 500-600uA being drawn at quite a high frequency and it then stops. Then there are smaller current draws  and an occasional large spike. 

I don't really know what is causing the initial current draw, and would appreciate some inputs on getting rid of them. Does this current graph look like something I should expect from the code snippet? My app code is actually just what is in the snippet above.

I get a similar behavior with my test app, where I have an adc measurement and twi reads, it seems the app is getting some interrupts causing it to wake up between expected intervals. The advertising interval I have set is 5s and ADC read is 4s, but the current graph shows the following:

The interval at which the current increases is not making sense to me, as it is much lower than either the 5s advertising or 4s adc sampling interval. It settles only after a while. The code snippet is something like this:

static void timers_init(void)
{
    ret_code_t err_code;

    // Initialize timer module
    err_code = app_timer_init();
    APP_ERROR_CHECK(err_code);

    err_code = app_timer_create(&m_saadc_timer_id,
                                APP_TIMER_MODE_REPEATED,
                                saadc_meas_timeout_handler);

    APP_ERROR_CHECK(err_code);
}

static void timer_start(void)
{
    ret_code_t err_code;

    err_code = app_timer_start(m_saadc_timer_id, APP_TIMER_TICKS(4000), NULL);
    APP_ERROR_CHECK(err_code);
}

static void connectable_adv_init(void)
{
    memset(&m_adv_params, 0, sizeof(m_adv_params));

    m_adv_params.type        = BLE_GAP_ADV_TYPE_ADV_IND;
    m_adv_params.p_peer_addr = NULL;                               // Undirected advertisement
    m_adv_params.fp          = BLE_GAP_ADV_FP_ANY;
    m_adv_params.interval    = MSEC_TO_UNITS(5000, UNIT_0_625_MS);
    m_adv_params.timeout     = 0;
}

static void saadc_meas_timeout_handler(void * p_context)
{
    // m_timeout = true;
            saadc_init();
            ret_code_t err_code = nrf_drv_saadc_buffer_convert(m_buffer_pool[0], SAADC_SAMPLES_IN_BUFFER);
            APP_ERROR_CHECK(err_code);
            nrf_drv_saadc_sample();
}


int main(void)
{
    timers_init();
    power_management_init();

    // initialize SoftDevice.
    ble_stack_init();
    sd_power_dcdc_mode_set(1);

    // set up I2C
    twi_init(&m_twi);

    peer_manager_init();
    gap_params_init();
    gatt_init();
    services_init();
    conn_params_init();

    connectable_adv_init();
    timer_start();
    advertising_start();

    // Enter main loop.
    for (;;) {


        nrf_pwr_mgmt_run();

    }
}

How can I debug this?

[Jørgen Holmefjord]

Hi,

5-600 µA sounds like the HFCLK/Timer is running. I would recommend that you try commenting out parts of your application to figure out exactly what peripheral is causing the increased current consumption.

Are you running this on a nRF52 DK or a custom board? If running on a DK, can you upload the project for debugging?

Best regards,
Jørgen

[rookie]

I'm running this on a custom board, I will share a smaller project in a bit (but to produce the first graph it's just the first small snippet). I was going through the sdk_config.h file after you mentioned that the HF clock may be on.

This is my clock setting part of the config file, 

// <e> CLOCK_ENABLED - nrf_drv_clock - CLOCK peripheral driver
//==========================================================
#ifndef CLOCK_ENABLED
#define CLOCK_ENABLED 1
#endif
// <o> CLOCK_CONFIG_XTAL_FREQ  - HF XTAL Frequency

// <0=> Default (64 MHz)

#ifndef CLOCK_CONFIG_XTAL_FREQ
#define CLOCK_CONFIG_XTAL_FREQ 0
#endif

// <o> CLOCK_CONFIG_LF_SRC  - LF Clock Source

// <0=> RC
// <1=> XTAL
// <2=> Synth

#ifndef CLOCK_CONFIG_LF_SRC
#define CLOCK_CONFIG_LF_SRC 1
#endif

[Jørgen Holmefjord]

Unless you have initialized it in your application, it should not be enabled only by setting that in sdk_config.h. The clock might, however, be enabled by other peripherals that require the clock to run.

It looks a bit strange that it produces spikes, and that it suddenly stops. If you could provide a project, that would help a lot.

[rookie]

Hi - I identified this current is coming from a peripheral, not the nrf chip. Closing this ticket. Thanks!