How to separate nrf52dk target and probe power bus?

[butch]

On the nrf52dk development kit, I want to power the target nrf52 from an external supply without powering the interface mcu (the debugging probe). How do I wire that? I don't want to measure the current, or debug, I just want to separate the power busses.

My best guess is to 1) turn the "Power" slide switch off 2) cut SB9 and 3) connect my power supply to net Vdd_nRF (on either pin 1 of P20 or pin 1 of P22.) Should that work?

(Fig. 2 "Power Supply Circuitry" for the nrf52DK shows that net Vdd_nRF is connected to net VDD by R6 but looking at the board, R6 is not populated.)

Fig. 2 also shows that SW6 also switches net V5V to net VBUS, but I don't know for sure which of those nets is the power supply to the interface mcu and which is power from the USB cable. Alternatively, if I leave the PowerSwitch in the On position, I could connect my power supply to "External supply" connector P21 if I am then careful to not connect the USB cable? (And the debugger probe would not be powered?)

(A related question is: could I connect DebugOut and DebugIn with a JTAG cable and still use the separately powered debugging probe to flash and debug the separately powered target? I haven't studied that much, but it looks like that is not possible, the cable would be parallel to the hardwired SWD and UART nets between the two.)

[Jørgen Holmefjord]

Hi,

You can download the hardware files for the nRF52-DK here. This includes all schematics for the board. On sheet two of the schematics, you can see that the interface MCU is powered by VDD through SB29. Cutting SB29 will disconnect the interface MCU. You can then power your board through External Supply (P21).

There is no JTAG, but you can connect an external debugger that support the SWD interface to the Debug in port (P18), to flash and debug the nRF52832 IC.

Best regards,

Jørgen

[butch]

Thanks, now I can figure it out.

Here is what I just tested: 1) turn Power switch to off position 2) apply external power to pin Vdd_nRF (do not cut SB9, that doesn't work.) It seems to work (the target flashes the LED's but the interface mcu does not flash its LED.) But I suppose the mcu might still be powered, so I will take your suggestion.

I also would like to suggest an improvement for your next DK: make the target and the interface mcu more easily separable, as on some dev boards for TI MSP430, where there is a clear jumper between the two halves. And I think you can isolate the power supplies but still use the probe half of the board to debug the target half of the board via SWD.

[butch]

Looking at the schematic, the LED's on the radio are powered from VDD. So my first solution probably was powering the radio, but the LED's had no supply and so did not flash. So another suggestion for the next DK: at least one LED that is powered by VDD_nRF.

(I could just buy a separate radio module. But I bought a BMD-300 having neither an LED nor a 32khz crystal. And I bought a BLE Nano whose nRF51 has a different ADC. Which is why I am trying to isolate the radio on the DK.)

[Jørgen Holmefjord]

I'm not sure what you mean by radio, do you mean the nRF52832? This can be isolated from both interface MCU, crystal and LEDs by cutting the solderbridges, as described in the link i posted above. You should not power the board using NRF_VDD, this pin is for current measurement. The reason why no leds are powered by this is that the current measurement should not be affected by other components on the DK. I'm also not sure I understand why you want to separate the two chips. The interface MCU is mainly used for debugging/programming the nRF52832 chip through the SWD interface and as a UART to USB bridge.

[butch]

1. Yes, radio == nrf52832. 2) It doesn't matter what the original purpose is, if the pin VDD_NRF of the board is on the net that includes the pin VDD of the nrf52, then it can be used to power the nrf52. 3) whether the LED's draw current depends on whether the app turns them on, so they they COULD be powered from VDD_NRF without much affecting measured current of the nrf52 itself. And the DK could just as well have solder bridges for the LEDs to choose their power supply? 4) The reason I want to isolate the nrf52 power is that I am testing an ultra-low, solar, power supply ranging from 1.8 to 2.4 volts typical in low light of 200lux, with the app duty-cycling the radio, and I don't want any other chips draining power. Its not a big deal, I am not trying to be critical, I very much like the product and the open communication, thanks for your informative and helpful response.