Power profiler regulates DK pin to 1.5V

[youti]

Question/Problem Statement

For evaluation we decided to do some current measurements using the power profiler kit. As a benchmark run, we wanted to measure the current on nRF52832 DK board with the setup described below. Somehow, the 832 doesn't turn on, because VDD_DUT is regulated to 1.5V. Why is this? How could we fix this to ensure VDD_DUT follows our supply voltage? We clearly missed something, but I cannot figure out what it is.

Setup

• External Supply ranging from 1.8V - 3.3V;
• External Debugger connected to the power profiler;
• Power profiler configured as:
  • Power select (SW4) set to external;
  • External supply set in the range 1.8V - 3.3V;
  • DUT select (SW2) set to DK;
• nRF52832 DK configured as:
  • Supply set in the range 1.8 - 3.3V (same supply as for the ppk);
  • SB5, SB6, SB7, SB8 are open, since those LEDs provided a current path to the nRF52832 chip, resulting in leakage current and the device not turning off;
  • SB9 open;
  • SB33, SB34, SB35 are open, not necessary when the USB is not plugged in, but otherwise the 3.3V regulator (U3) would overrule the 1.8V supply at P21;
  • SB12 closed to bypass the input diode;

Results

Now for the results,

• The 832 chip on the DK simply won't turn on. Whatever supply voltage we select in the range 1.8V to 3.3V the power profiler regulates the pin going to the DK to 1.54V. The current runs only through resistor R2, the other switches (U3 and U4) are open, hence no parallel path.
• The voltage drop across R2 + 1.54V = supply voltage.
• The power profiler is available and can be read with the nRF Connect SDK app.
• The atmega chip is in sleep mode, when pressing the reset, we see an increase in current of 5mA.
• A jumper across P22 enables the 832 and we see an increase of 7 mA in current (just a simple loop is programmed to the 832, nothing else);
• Insertion of the USB to the DK increases the supply current demand with a lot, think in the range 90mA;

Thanks in advance for your help!

[Stian Røed Hafskjold]

Hi, I will try out your setup here. Which PPK and DK version are you using?

Did you connect a USB cable to the PPK? Seems like this is needed when using the external regulator mode.

Are you plugging the PPK on top of the DK? In general it is better to keep the DK and PPK separated and power the DK from the DUT output header. Then you don't have to connect the external supply to both the DK and the PPK, which I'm thinking can cause problems. You must isolate the chip on the nRF52 DK so you don't power the atmega chip. You can try to power directly on the VDD nRF domain on P20 from the PPK.

[youti]

Hi Stian! The ppk is the nRF6707 and the development kit is PCA10040. With the USB it works better, I indeed managed to forget about it. Still, it doesn't work as intended. We tested it again on two DKs, with the ppk mounted on top and as external device. With these setups, the the results are different. To the DKs we uploaded the ble_app_beacon and the results are as below. The first plot shows the ppk on top of the DK, with SW2 set to DK of course. The second plot shows the plot when connecting the DK to the ppk via external DUT and P20 on the DK. We also connected the DK to the same supply as the ppk to see what happens, this is the third plot. As you can see there is still a small offset that I believe shouldn't be there.