ESP32 Wrover-E - internal SPI connected flash *always* 3.3V?

[mirkomirko]

I'm using the ESP32 Wrover-E with additional flash (16MiB) and overlooked GPIO12 being a strapping pin configuring whether the internally SPI connected flash is powered by 1.8V or 3.3V.

GPIO12 is used as MISO on our board, whereas the ESP32 is the master and the partner is slave. Hence, the slave is driving MISO and pulling it high, causing the ESP32 on our board to configure 1.8V for the SPI flash - and hence malfunction.

Is it save to assume that *any* ESP32 Wrover-E module has *only* flash powered via 3.3V? If yes, I would just burn the eFuse via `espefuse set_flash_voltage 3.3V` during manufacturing and stop bothering.

However, since it's an irreversible operation, I wonder about side-effects / whether there might be Wrover-E modules which have only 1.8V compatible flash built-in?

Thanks!

mirko

[ESP_Sprite]

The datasheet specifies that module as having a 3.3V flash/psram, so you can assume that specific model of module will always have that. (If we ever release an 1.8V version, it would get a different name, like ESP32-Wrover-EL or something; we did something like that in the past when we moved from 1.8V to 3.3V flash.)

[mirkomirko]

Thank you!

Could you shed some light on why IO12 is actually used as a strapping pin on the wrover module? Why isn't the eFuse burned by default, as one of the main purposes of the wrover-module is to have additional *internal* flash? (internal as in: part of the module; I know from the bare-metal SoC's perspective it's still external). Is there any use case with the wrover module to power the VCC_SDIO with 1.8V?

[ESP_Sprite]

Possibly simply to be safe: it's always possible to burn that eFuse as the end user, but it's impossible to *un*burn it.