Modifications of the nRF7002DK to be compatible with an external MCU for SPI at 3.3V

[HolmanB]

Hello!

I am currently using the nRF7002DK, and I am interested in utilizing the nRF5340 MCU as a slave SPI with an external MCU that supports 3.3V SPI voltage levels. My goal is to establish SPI communication between the nRF5340 MCU (used in the nRF7002DK) and the external MCU, utilizing different pins than those used for QSPI communication with the nRF7002 IC. This is necessary as there will be two separate SPI/QSPI buses.

Could you provide guidance on how to modify the nRF7002DK to apply a 3.3V supply to the nRF5340 MCU? This adjustment would make the nRF5340 compatible with the external MCU's SPI voltage levels, allowing seamless communication without the risk of damage.

Thank you and best regards,

Holman.

[Simonr]

Hi Holman

I assume you're aware that the on the nRF52/53 series devices, the GPIO voltage always follows the VDD. Does 3.0V not suffice in your case? As I think supplying the board with 5V should let you use the voltage follower on the DK. Alternatively you can disable the voltage follower so the 3V BUCK regulator is the only thing used which should supply the whole board with 3V exactly. Please see the nRF7002DK user guide for more details.

What are the specs of this external SPI device you want to connect exactly?

Best regards,

Simon

[Saxman58]

The problem, Simon, is that unlike every other Nordic board I've seen that all run on 3.3V, the nRF7002DK has a Vdd of 1.8V.  That makes it difficult to interface with UART, SPI or I2C to other boards that have 3.3V logic.

I assume the 1.8V voltage was selected due to the problem with running the nRF5340 QSPI (to the nRF7002) at 96 MHz with the higher voltage.  Two of the errata for the nRF5340 state that QSPI doesn't work right at 96 MHz with > 2.0V, and that it would also really degrade the radio performance.  The recommended workaround is to either use 1.7 to 2.0V Vdd, or a max 48 MHz QSPI clock.  See:

https://docs.nordicsemi.com/bundle/errata_nRF5340_Rev1/page/ERR/nRF5340/Rev1/latest/anomaly_340_136.html#anomaly_340_136

and

docs.nordicsemi.com/.../anomaly_340_133.html

[Simonr]

You're correct Glen, and I understand the predicament here. Unfortunately there is not much to do about this for external devices/boards not supporting lower voltages. Sorry about the inconvenience. Please let me know if there is any action points you want us to note here. We are aware that this is not optimal, but it is the way of things.

Best regards,

Simon

[Bard]

Can the VIO_REF pin be used to set a higher voltage (if ok with lower QSPI speed)? From the nRF7002_DK_User_Guide: "For the 5 V domain, there is one fixed 1.8 V buck regulator and one voltage follower regulator that follows the VIO_REF voltage"