Unaccurate measurments vs nrf connect

[klaudiawro]

Hello i am using the API for an automated production script, one of the steps is measuring power consumption, the API works as intended maybe 50% of the time, the other times it returns values that are very low (0.03 - 0.08 mA) compared to what it should be (around 2.7mA) when the same pcb is measured with nrf connect the correct value is disaplayed, this is the code that is used:

def connect_ppk2_mp():
    stem.usb.setPortEnable(USBHubPort.PPK2)
    ppk2s_connected = PPK2_API.list_devices()
    global ppk2_test

    if len(ppk2s_connected) == 1:
        ppk2_port = ppk2s_connected[0][0]
        ppk2_serial = ppk2s_connected[0][1]
    else:
        print(f"error: none.or.too.many.connected.PPK2: {ppk2s_connected}")
        sys.exit(1)

    ppk2_test = PPK2_API(ppk2_port, exclusive=True)
    
    ppk2_test.ser.reset_input_buffer()
    ppk2_test.ser.reset_output_buffer()
    sleep(0.1)

    ppk2_test.get_modifiers()
    ppk2_test.use_source_meter()
    ppk2_test.set_source_voltage(3000)
    ppk2_test.toggle_DUT_power('ON')
    
    sleep(0.1)
    ppk2_ready_event.set()

def measure_current_ppk2_mp(seconds):
    all_samples = []
    start_time = time.time()

    # Clear buffer to ensure fresh data
    ppk2_test.get_data()
    sleep(0.1)

    # Start measuring and allow stabilization
    ppk2_test.start_measuring()
    time.sleep(0.2)

    end_time = start_time + float(seconds)

    while time.time() < end_time:
        read_data = ppk2_test.get_data()
        if read_data:
            samples, _ = ppk2_test.get_samples(read_data)
            all_samples.extend(samples)  # Keep all data (no [::100] filtering)

    ppk2_test.stop_measuring()

    if all_samples:
        average = sum(all_samples) / len(all_samples)
        print(len(all_samples))
        print(f"{average:.6f} mA")
        return f"{average:.6f}"

    print("error: current measuring failed")
    return "0.000000"

For the record, i am using the PPK2_MP version

[furhadjidda]

you might want to toggle_DUT_power("ON") before starting the measurement. I had missed this in my script before i could get it to work.

[MrKevinWeiss]

Did this solve your issue? @furhadjidda if not I would maybe look at #50 in case it is something with the corrupted metadata readings.