Support changes for USB C female version of the board?

[Jeff-Hackford]

First off, this project is awesome and I can not wait to see what future features bring! Great job!

Secondly, I think I f*cked up... When researching this project I found the USB-C version of this board that offered a slightly larger screen, more gpio pins, and a sleeker body profile (I plan on adding a battery to the board and didnt want the male USB dongle sticking out). Now I am having trouble getting the sd card to read ( 1x107 memory error with a 4GB known good card ) and can't seem to reconcile the differences in boards. Any help would be appreciated but I know I did this to myself. Lol

[Jstudner]

howdy! that is actually a really neat board, and from a quick overview it should run the software just fine. that being said you are going to have to do some code modification. Not too bad though. You will need to update the LVGL screen stuff fully for the new screen size, no way around that. The Tools section options are likely different for this model, also check the SD card pinout. The 1x107 error also might not be an SD card failure, (though if the pins are different that would happen) but it could also be a different psram setup, which would cause that too.
I will take a more in depth look at this tomorrow night, but try the above out based on the waveshare docs you linked (top of the page right side there is a blue "expand" button that usually hides the pinout and such. It also looks like the SD card might use SPI, which would require some updates to the sd library. as I said ill look more in depth tomorrow, but you are not doomed at least. the board also has a bunch of features like motion sense, and a touch screen, which you could defiantly use.

[Jeff-Hackford]

I have not also updated the screen pins. I can look into that and the SD card library in the next few hours. Thanks again for your response!

[Jeff-Hackford]

Hmm, I am still having trouble with the 1.64 S3 board. I can't seem to get the screen and SD card reader to work at the same time. Individually, yes, but not together. Any additional help is much appreciated.

[Jstudner]

Are you getting any specific errors or are they just individually not working? what are you changing when testing between the two (as in are you just running demo code or is it the full nomad system)?

[Jeff-Hackford]

When running the full nomad system I can tweak the Pin assignments and get it to compile with my board but they I get the above error. I am using these tools settings in the IDE:

I can run a lvgl demo and get the screen to work (colors are off but one problem at a time) or I can run an SD_MMC demo and get the card info read and printed to the monitor, but I can't seem to reconcile the two when running on the nomad code. This is my first micro-controller project. I'm sorry if some of these problems are easily solved.

[Jstudner]

Easily solved problems are the best problems, gimme a sec ima take a peak

[Jstudner]

ok, the color issue on the screen is probably due to it being amoled, we can leave that for later but should be a simple tweak in lvgl. In the nomad code under the pile of #includes do you still have #include "SD_MMC.h" If so thats likely the issue. check the demo for sd card and see if it uses sd.h instead. If so it might be as easy as renaming every SD_MMC line to just SD. (prolly not but here's to hopeing). If you haven't redesigned the screen UI for that board yet ill throw one together tomorrow for you to use.

[Jeff-Hackford]

The SD card demo that is working appears to be using SDMMC_U..?

And gives this output when compiled:

I tried renaming the SD_MMC instances to just SD in the nomad code but then I get many different compile errors that don't make much sense to me.

[Jstudner]

ok, going over this some more it looks like SD_MMC totally does do SPI (I somehow missed that). but for now lets try doing it like this, remove #include SD_MMC.h and replace it with both #include <SPI.h> and #include <SD.h> those should handle it better. Then we have to update some code to match that change with the specific pins:
Replace This:

//OLD
if (!SD_MMC.begin()) {
    request->send(500, "text/plain", "SD card not available.");
    Serial.println("[ADMIN] SD card not mounted.");
    return;
}

With This:

//NEW
if (!SD.begin(SD_CS_PIN)) {
    request->send(500, "text/plain", "SD card not available.");
    Serial.println("[ADMIN] SD card not mounted.");
    return;
}

Then replace this:

//OLD
bool tryRecoverSDCard() {                          
    Serial.println("[SD] Attempting recovery…");
    SD_MMC.end();          // unmount
    delay(1000);           // give hardware a breather

    if (!SD_MMC.begin("/sdcard", true, false, SD_FREQ_KHZ, 12)) {
        Serial.println("[SD] Recovery failed.");
        return false;
    }
    Serial.println("[SD] Recovery OK.");
    return true;
}

With:

//NEW
bool tryRecoverSDCard() {                          
    Serial.println("[SD] Attempting recovery…");
    SD.end();              // unmount
    delay(1000);           // give hardware a breather

    if (!SD.begin(SD_CS_PIN)) {
        Serial.println("[SD] Recovery failed.");
        return false;
    }
    Serial.println("[SD] Recovery OK.");
    return true;
}

And finally replace this:

//OLD
// Initialize SD card
Serial.println("Initializing SD Card...");
if (!SD_MMC.setPins(SD_CLK_PIN, SD_CMD_PIN, SD_D0_PIN, SD_D1_PIN, SD_D2_PIN, SD_D3_PIN)) {
    Serial.println("ERROR: SDMMC Pin configuration failed!");
    return;
}

if (!SD_MMC.begin("/sdcard", true, false, SDMMC_FREQ_DEFAULT, 12)) {
    Serial.println("ERROR: SDMMC Card initialization failed.");
    return;
}

With:

//NEW
Serial.println("Initializing SD Card (SPI)...");
SPI.begin(SCK_PIN, MISO_PIN, MOSI_PIN, SD_CS_PIN);

if (!SD.begin(SD_CS_PIN)) {
    Serial.println("ERROR: SD Card initialization failed.");
    return;
}

Pretty sure that will work for your system, you will also need to remove the pin definitions at the top and replace them with

//NEW
#define SD_CS_PIN   39
#define SD_MOSI_PIN 39
#define SD_MISO_PIN 40
#define SD_SCLK_PIN 41

Let me know how that goes, hopefully it should fix the SD issue and we can move to the screen!

[dvanco]

@Jeff-Hackford - did you happen to find a suitable case with SD access for this board? I've got one of them as well and would like to mess with it.

[Jeff-Hackford]

Unfortunately I could not get a version of this code to work on this board, but I started from scratch and with some help from chatGPT I was able to get some bare bones code in place to serve a basic list of all mp4 files at the root level of the SD over a Wi-Fi access point. If you want I can upload that code but it is very basic and doesn't even utilize the screen.