Peak current when the nRF52 is in the SYSTEM ON mode

RE: Peak current when the nRF52 is in the SYSTEM ON mode

Sylvain — Fri, 07 Sep 2018 10:14:38 GMT

Thank you Sigurd !

RE: Peak current when the nRF52 is in the SYSTEM ON mode

Sigurd — Fri, 07 Sep 2018 09:28:57 GMT

Hi,

[quote userid="68596" url="~/f/nordic-q-a/38000/peak-current-when-the-nrf52-is-in-the-system-on-mode/147640"]I guess the noise comes from the USB power supply as you previously mention.[/quote]

Yes, this looks alot better.

[quote userid="68596" url="~/f/nordic-q-a/38000/peak-current-when-the-nrf52-is-in-the-system-on-mode/147640"]As you can see on the graph, I have some peaks at a frequency of approximately 38 Hz. Could be the internal regulator?[/quote]

Yes, 38 Hz is consistent with refresh spikes. So yes, this is the refresh spikes of the internal regulator.

RE: Peak current when the nRF52 is in the SYSTEM ON mode

Sylvain — Thu, 06 Sep 2018 14:44:03 GMT

I performed measures with external power supply and now I have good results. I guess the noise comes from the USB power supply as you previously mention.

As you can see on the graph, I have some peaks at a frequency of approximately 38 Hz. Could be the internal regulator?

Thanks a lot for your support.

Sylvain.

RE: Peak current when the nRF52 is in the SYSTEM ON mode

Sigurd — Thu, 06 Sep 2018 12:12:42 GMT

I also tested the main.c you posted with a KeySight N6705B also, and it didn’t show these peaks.

There is something called refresh spikes in the internal regulator, but that is not what is shown in your plot. The behavior you see is not normal.

It could maybe be USB noise. Could you do a test where you power the board with “External supply”(P21) or with a battery instead.

RE: Peak current when the nRF52 is in the SYSTEM ON mode

Sylvain — Wed, 05 Sep 2018 14:24:31 GMT

Unfortunately, I think the dynamic of the PPK is not wide enough to measure the same behaviour as me. Furthermore, the slow sampling frequency probably smooths the signal.

I have the same behaviour as previously. My sampling rate is 1 Mbps. I expect a sort of refreshment from the nRF52 to maintain this System On mode. I think is like that with the CC2642 ...

I'm just asking myself if this behaviour is normal because it changes the average consumption ...

RE: Peak current when the nRF52 is in the SYSTEM ON mode

Sigurd — Tue, 04 Sep 2018 15:15:02 GMT

The sampling frequency for the PPK is 77 kHz.

RE: Peak current when the nRF52 is in the SYSTEM ON mode

Sylvain — Tue, 04 Sep 2018 14:48:12 GMT

I'll perfom new measurements in the following days to be sure of the test condition.

What's the sampling frequency of the Nordic PPK ?

Yes, I saw the same result on the DMM7510.

RE: Peak current when the nRF52 is in the SYSTEM ON mode

Sigurd — Mon, 03 Sep 2018 16:58:10 GMT

I tested your code on a nRF52840-PDK, and it didn't show any of these peaks here.

Image from Nordic PPK:

* If the PDK is connected to a USB-hub, could you try to connect it directly to the computer?

* Have you connected any external sensors to the PDK ?

* I see that you have plotted the measurement in Matlab or something similar, did you see the same peaks directly on the Tektronix DMM7510 display also?

RE: Peak current when the nRF52 is in the SYSTEM ON mode

Sylvain — Mon, 03 Sep 2018 16:11:06 GMT

I will check to post the main.c, but uncomfortable because it is part of a whole project. I don't know if it'll work.devzone.nordicsemi.com/.../7077.main-_2D00_-Copy.c

- I used a USB supply with nRF only switch ON.
- Tektronix DMM7510
- Yes I measure the current through P22

RE: Peak current when the nRF52 is in the SYSTEM ON mode

Sigurd — Mon, 03 Sep 2018 15:42:44 GMT

Hi,

Could you post the whole main.c file ?

How are you powering the board? USB or external supply?

What device are you using to measure the current?

Are you measuring on P22? Could you cut the PCB track shorting solder bridge SB40 and SB41 ?

RE: Peak current when the nRF52 is in the SYSTEM ON mode

Sylvain — Mon, 03 Sep 2018 11:41:57 GMT

This is a broadcaster so there're no connection.

Yes, same result with beacon example. In fact, I use the same principle ...

Here's the basic test code :

static void advertising_init(void){
    uint32_t      err_code;
    int8_t tx_power = 0;
    ble_gap_conn_sec_mode_t sec_mode;
    
    ble_advdata_manuf_data_t manuf_specific_data;
    ble_advdata_t advdata;
    
    BLE_GAP_CONN_SEC_MODE_SET_OPEN(&sec_mode);
    err_code = sd_ble_gap_device_name_set(&sec_mode,
                                          (const uint8_t *) DEVICE_NAME,
                                          strlen(DEVICE_NAME));
    APP_ERROR_CHECK(err_code);

    err_code = sd_ble_gap_tx_power_set(tx_power);//Set the radio's transmit power in dBm.
        
    sd_power_dcdc_mode_set(1);
    
    manuf_specific_data.company_identifier = APP_COMPANY_IDENTIFIER;
    manuf_specific_data.data.p_data = (uint8_t *) BLE_data;
    manuf_specific_data.data.size   = sizeof(BLE_data);

    // Build and set advertising data.
    memset(&advdata, 0, sizeof(advdata));

    advdata.name_type             = BLE_ADVDATA_FULL_NAME;
    //advdata.short_name_len        = strlen(DEVICE_NAME)-3;
    advdata.flags                 = BLE_GAP_ADV_FLAG_BR_EDR_NOT_SUPPORTED;
    advdata.p_manuf_specific_data = &manuf_specific_data;

    err_code = ble_advdata_set(&advdata, NULL);
    APP_ERROR_CHECK(err_code);
    // Initialize advertising parameters (used when starting advertising).
    memset(&m_adv_params, 0, sizeof(m_adv_params));

    m_adv_params.type            = BLE_GAP_ADV_TYPE_ADV_NONCONN_IND;//BLE_GAP_ADV_TYPE_EXTENDED_NONCONNECTABLE_NONSCANNABLE_UNDIRECTED;//
    m_adv_params.p_peer_addr     = NULL;    // Undirected advertisement.
    m_adv_params.fp              = BLE_GAP_ADV_FP_ANY;
    m_adv_params.interval        = NON_CONNECTABLE_ADV_INTERVAL;
    m_adv_params.timeout         = 0;       // Never time out.
}

/**@brief Function for initializing the BLE stack.
 *
 * @details Initializes the SoftDevice and the BLE event interrupt.
 */
static void ble_stack_init(void){

    ret_code_t err_code;

    err_code = nrf_sdh_enable_request();
    APP_ERROR_CHECK(err_code);

    // Configure the BLE stack using the default settings.
    // Fetch the start address of the application RAM.
    uint32_t ram_start = 0;
    err_code = nrf_sdh_ble_default_cfg_set(APP_BLE_CONN_CFG_TAG, &ram_start);
    APP_ERROR_CHECK(err_code);

    // Enable BLE stack.
    err_code = nrf_sdh_ble_enable(&ram_start);
    APP_ERROR_CHECK(err_code);
}

static void advertising_start(void){
    ret_code_t err_code;

    err_code = sd_ble_gap_adv_start(&m_adv_params, APP_BLE_CONN_CFG_TAG);
    APP_ERROR_CHECK(err_code);
}

int main(void)
{    
  
  ble_stack_init();
  advertising_init();
  advertising_start();

while (1){
    
            power_manage();
    }
}

RE: Peak current when the nRF52 is in the SYSTEM ON mode

Sigurd — Fri, 31 Aug 2018 14:30:01 GMT

It looks like spikes casued by interrupts.

Are you able to connect to the device with a BLE central?

[quote]I use my own code.[/quote]

Do you see the same spikes with the BLE SDK examples?

RE: Peak current when the nRF52 is in the SYSTEM ON mode

Sylvain — Thu, 30 Aug 2018 13:05:12 GMT

I use my own code.

I simply configure the Advertising and then advertise each second. On infinite loop side I use the power_manage(); function to manage the power. That's all.

Is it possible that when the MCU is in System On mode it performs a refresh memory ?

RE: Peak current when the nRF52 is in the SYSTEM ON mode

Sigurd — Thu, 30 Aug 2018 12:49:29 GMT

Ok, I understand. These other peaks at around 3-4 mA could be some interrupt that is triggered, the CPU will then wake up, and then go back to idle mode again after the interrupt has been served.

What example in the SDK are you testing?

RE: Peak current when the nRF52 is in the SYSTEM ON mode

Sylvain — Thu, 30 Aug 2018 12:00:03 GMT

Sigurd,

I know how the System ON mode and how the advertising work. You don't answer my question. Maybe it was not clear. So, my question is about the peak that occur when my mcu is in System On mode. So, I talk about the peaks after the data transmission. In resume, why I measure peaks between 31.5 and 31.88 s and also 32.2 and 32.5 (MCU is in System On mode between these intervals).

RE: Peak current when the nRF52 is in the SYSTEM ON mode

Sigurd — Thu, 30 Aug 2018 11:05:44 GMT

Hi,

The SoftDevice will continue to advertise in System ON mode. You can read more about System ON mode here. What you see in the plot is first the pre-processing and HFXO ramp, then you have 3 radio TX peaks when you advertise on the 3 advertising channels. You can also see this using our Online Power Profiler.