scan high power consumption

[schef]

Hi.

I scan devices with sd_ble_gap_scan_start command and on start it consumes around 6mA.

here are my parameters:

#define SCAN_INTERVAL 0x00A0 /**< Determines scan interval in units of 0.625 millisecond. */
#define SCAN_WINDOW 0x0050 /**< Determines scan window in units of 0.625 millisecond. */
#define SCAN_TIMEOUT 0

ble_gap_scan_params_t const m_scan_params =
{
.active = 1,
.use_whitelist = 0,
.adv_dir_report = 0,
.interval = SCAN_INTERVAL,
.window = SCAN_WINDOW,
.timeout = SCAN_TIMEOUT,
};

Is this normal?

[Susheel Nuguru]

The average power consumption of the RADIO itself as receiver (scanning) is around 5.5mA as shown here.

With 100ms scan interval and 50ms scan window, you are keeping the RADIO on half of the time. So just the RADIO will consume an average current of 5.5/2 = 2.75mA. And this is assuming your device is not doing anything else.

But obviously you will need clocks and few peripherals and CPU core to do some pre and post processing of this incoming packets.

Scanning is very power consuming process, that is the reason in older BLE specifications only mobile phones were doing this due to larger batteries. If you have a scanner on your product with smaller batteries, you should scan for as less time as possible otherwise it will drain your battery much faster.

You asked if the numbers you see are normal?

Yes, they look normal.

[schef]

Here are some results of scanning that could be helpfull to someone:

Scanning parameter: interval 2000ms, window 60ms.

With 2000 sampling maximum was 2.091mA , minimum 21uA and average 275uA.

My testing board consumed 25uA so just the scanning action draws around 250uA.


Results of scanning Advertising device with maximum advertisement of 20ms on a distance of 1m with 577 sampling (20 min):

 - 4 sampling sa 4 packets

 - 16 sampling sa 3 packets

 - 281 sampling sa 2 packets

 - 238 sampling sa 1 packets

 - 39 sampling sa 0 packets


U 14.7% cases we get a packet inside 4 sec and in 85.3% inside 2 sec.

It never occured that i got 2 samplings in a row with 0 packets.