370uA current draw with all peripherals idle

My exact chip is N52810-QFAAD1-2145FM.

I have pared down my hardware and software to the bare minimum: The hardware is essentially just the reference circuit with DC/DC regulator and external LF crystal, plus a very simple custom breakout board I am using for development. The schematics are attached as PDFs below, and the full KiCad files are in this repository that I also link below. Note that the KiCad files include a lot of other components, but I have one board I have tortured by removing all extra components, so they are not present at the moment.

The software, which is included in the "power_debug" branch of this repository, is extremely simple: it enables the DC/DC regulator and both external crystals, then enters either sleep mode (Using WFI) or shutdown mode. No peripherals are enabled (I explicitly disabled some in testing, but that made no difference). I am also not using a Softdevice.

When entering Idle mode using WFI, I see a consistent current draw of 370uA (Measured with a multimeter on the Current pins of the breakout board). When entering System OFF mode with a full shutdown, I see less than 1uA (Which is exactly what I would expect).

Because I see the expected low current in System OFF mode, I suspect that the problem does NOT lie with my circuitry. I assume that if the problem was external to the NRF52 chip, then the current draw would remain high when the chip is OFF.

The problem persists across at least the 3 boards I have tested, so if it's a faulty chip, then it would have to be part of a larger faulty batch.

I have checked the errata for Rev 2 of this chip, and there are two which result in increased current draw: Static 400uA current after SAADC is disabled, and Intermittent extra current consumption when going to sleep. I would not expect either of these to apply in this simple test case, but just to try it, I applied the workarounds and saw no change.

How can I find the cause of this current draw? Did I miss something obvious in the documentation? I would appreciate any ideas of suggestions for further troubleshooting steps. This project is entirely open-source, so I will gladly provide any more information that would be helpful.

7802.circuit_schematic.pdf1134.breakout_board_schematic.pdf

  • Update: I have made progress! When I do not enable the external 64MHz crystal, I get the expected (lower) current draw of ~1uA. I now realize there must be something wrong with the circuitry of my 64MHz crystal. The reason I got the low current draw in System OFF mode is because the HFCLK is disabled in OFF mode.

    If anyone has suggestions for how I can identify and fix the issue with my circuitry, I would love to hear them! For what it's worth, the 32.768KHz crystal works perfectly fine, and does not have a large current draw.

Related