FatFS example on SDK 14.2: initialization error (disk_state = 1)

[Sébastien Schiesser]

Hi

Freshly downloaded the SDK14.2 and trying to start a new development based on the fatfs example.

I connected a small Sparkfun SD card breakout to the SPI pin on my pca10040 (actually rigado BMD-300 evaluation board which should be fully compatible). My connections are:

- P0.22 --> CS

- P0.23 --> DI/MOSI

- P0.24 --> DO/MISO

- P0.25 --> SCK

(+ VCC and GND)

I also tried to change the pullup settings on MISO to all three possible configurations.

Result: initialization error (disk_state = 1).

The analyzer shows that the SPI master tries to communicate with the SD card, which does not answer at all (see attached picture).

I double- even triple-checked the wiring, but cannot tell what could be wrong.

[Edvin]

Hello,

Are you sure you aren't mixing up the MISO and MOSI pins? If I am not mistaken, DI is MOSI. Hence, shouldn't the green wire be connected to pin P0.24, and not P0.23?I might be mistaken, it is not too easy to see the wires on the SD card reader in the picture.

If that is not the issue, what sort of SD card do you use? Can you plug it into a computer and try to reformat it to FATFS32?

It should be fairly easy, and it can be done something like this: https://www.lifewire.com/format-sd-card-using-windows-4128719

Best regards,
Edvin

[Sébastien Schiesser]

Hi,

Not completely done with this issue! Since a couple of days, the SD card initialization returns STA_NOINIT after the disk_initialize function (from SDK_14.2.0\examples\fatfs). More precisely, the block_dev_handler gets the result NRF_BLOCK_DEV_RESULT_IO_ERROR.

I tried different setups from what I've read on this site:

• set NRF_GPIO_PIN_NOPULL on sdk_config.h
• use two different FAT32-formatted cards
• supply the breakout with 5V or 3.3V
• try the example projects on SDK 12.2 (without change), 14.2 (with NOPULL setting) and 15.0 (without change)

I hope this time I didn't mix up the wires, but as on the picture:

• P0.25 --> SCK
• P0.24 --> DO
• P0.23 --> DI
• P0.22 --> CS

Can you help me on what could produce this IO error?

Thanks,

Sébastien

[Edvin]

Sorry for the late reply. I have been out of office.

Could you try to attach your project, so that I can try to replicate the issue?

 

Best regards,

Edvin

[Sébastien Schiesser]

Directly taken from the example folder...

fatfs.zip

[Edvin]

Hello,

I just did the test on the SDK15 version of the fatfs example, and it works here. Can you please try to change the cables? Maybe one of them are defect? And if you have, can you try another sd-card reader?

 

What changed between the time when it worked and now? Maybe you can do another logic analyze like in your original post?

 

BR,

Edvin

[Sébastien Schiesser]

Could it be because of the card size? I formatted one 4 GB and one 16 GB microSD, and the small one works. BTW it was also my change between the first (working) time and now...

Are there some limitations? Did I miss the information?

[Edvin]

I think that the card size should not be a problem. Can you try to format the card using a computer?

I don't have an SD-card reader for my computer right now, so I don't remember what format it should be, but follow this guide, and try a few different file systems. I think FAT or FAT32 should work. If not, try the other alternatives as well.

 

Best regards,

Edvin

[Sébastien Schiesser]

On the 16 GB card I got 2 different errors (randomly): 1 (low level I/O) and 13 (no valid FAT volume). So I made a slow formatting (FAT32, since FAT is not available for size > 4GB) and re-soldered the header pins of the card breakout.

...now it seems to work...

I cannot tell what was the cause (probably both) since the 4GB card did always work. Hopefully I'm done with this issue now!

Thanks for being patient ;-) Best, Sébastien

[Edvin]

Glad to hear that it worked out

--

Edvin

[NickH]

Did you get this working as I have the same problem and have tried the same things that you list...