This post is older than 2 years and might not be relevant anymore
More Info: Consider searching for newer posts

temperature and current consumption

Hi, all great guys,

We are using NRF51822 as the MCU,and met a very strange problem about the current consumption in our application. Normally,the lowest standby current of our application is about 50uA, but we found the lowest current of some PCBAs are higher than 100uA, and finally we found the strange problem, if we just heat the 51822 chip,the current will drop to 50uA, this will happen even only when we just put a finger on the chip to heat it,now the indoor temperature is about 15 degree C . I want to say that not all the chips have the same problem, but still this problem is very important, which will greatly affect the standby time of our product. We want to know that why this problem happened ,and how we can eleminate it whatever from the aspect of design or manufactering proceduce. Thanks a lot.

  • Hi Hunter,

    You may want to provide some more detail about your board, schematic, layout.

    50-100uA was with the nRF51 in deep sleep mode ? It's pretty high if it's just the nRF51 in sleep mode. Usually it should be 3uA or less. So I suspect something else is drawing current and the parasitic current consumption is changed when you touch the board with a finger.

  • Hi,Hung Bui , We have located the problem probably, before we have a GPIO setting to NOPULL and it have been disconnected from anything,so it was floating, if we didnt set that GPIO, it seems the current came back to normal. However,some chips dont have this problem ,I am not sure why, and I am also curious that why the current is related to the temperature. Thanks a lot.

  • Did you set the pin to output or input ? We need to clarify it was the issue with the GPIO pin. When in input mode, I don't think the pin would cause any current draw except if you enable GPIOTE PIN interrupt, but the current consumption should be nearly 1mA as the 16MHz will be kept running.

    In your tests, how many chips have this problem and how many doesn't have ? Could you send the schematic, layout and a photo of the test board(s) ?

    Could you reproduce the issue on a Nordic Development Kit ?

Related