Go Down

Topic: SPIFlash library - Now works with Winbond/Microchip/Spansion/Cypress (Read 38601 times) previous topic - next topic


@liolau: The Nano is just a matter of testing. It uses the same ATMega328 µC as in the Uno so it should work with no issues.


Hi all, just pushed through a major bug-fix release v3.0.1

@hanyazou identified a major bug (detailed here in issue 102) which has been fixed with v3.0.1

v3.0.1 also adds support for the S25FL127S from Spansion/Cypress.

As always, you can find this version on Github here --> SPIFlash Library for Arduino v3.0.1
A ZIP file is also attached to the first post on this thread. The easiest way, as always, is to open up Library Manager on your Arduino IDE and update the library to v3.0.1 :)



Does this library support QSPI ? I have wired W25Q128JV in QSPI mode and was wondering if I could try that with this llibrary


I get some error on save to flash. flash.writeAnything with char variables in struct fails.

I have two structs

Code: [Select]
struct Configuration {
  uint8_t tempHigh;         // store higher value for temperature
  uint8_t tempLow;          // store lower value for temperature
  bool parkingMode;         // store Parking Battery value boolean
  uint8_t chargingRange;    // store charging range status
  uint8_t chargingHigh;     // store charging higher value
  uint8_t chargingLow;      // store charging lower value
  boolean sendData;         // store send data flag
} configuration;

struct NetworkConfig {
  char ssid[50];            // Store network ssid from client
  char pwd[50];             // Store network password from client
  char userid[50];          // Store identification ID from client
} networkConfig;

set some values on both struct, and save to flash:

Code: [Select]
  if (flash.writeAnything(0, configuration)) {
    Serial.println ("Data write successful");
  } else {
    Serial.println ("Data write failed");

  if (flash.writeAnything(1, networkConfig)) {
    Serial.println ("Data write successful");
  } else {
    Serial.println ("Data write failed");
  flash.readAnything(0, configuration);
  flash.readAnything(1, networkConfig);


Serial Monitor:
Data write successful
Data write failed

All the values from struct "configuration" are OK.
Some has an Idea please.



Hi - thanks for a fantastic library! I know this is a pretty big ask but maybe someone here can help out. I'd love to use this library on an STM32 MCU in the Particle ecosystem. Particle is a fork of Arduino/Wiring, in my understanding, so I don't think a ton would need to change but it's hard to tell. Has anyone tried this or perhaps have any advice for where I could start? Thank you!


is there a conflict between EEPROM (mega2560 internal flash) and SPIFLASH Lib?
I tried to adjust the bootloader, then I noticed that the data on the winbond is gone, and the bootloader is not working as expected.

I tried this one.

Any idea?


@a7ashh14: I'm afraid it does not yet support QSPI. What µC are you using with your W25Q128JV?

@supscientist: I've yet to play around with the Particle/Electron ecosystem. Feel free to play around with the code in the 'stable' branch of the library on Github and see if you can make it compatible with the Particle ecosystem. :)

@mysource: What pins are you connecting your Flash memory to? I don't think that the conflict is between the EEPROM and the Flash - its the probably because you use the SPI bus to program your Mega. If somehow, during the programming the Slave Select pin (that you have your flash connected to) is a) pulled low or b) the voltage drops and the µC enters Brown-out mode - then the data that's being sent into your Mega might a) get piped into your Flash memory as well or b) be interrupted unexpectedly. This could very easily cause corruption of the data on both the Mega's internal flash and the external SPI Flash you are using.


@supscientist As a place to start, take a look at how Particle's SPI Library works and see how you can replace the calls to the Arduino SPI Library with calls to the Particle SPI Library in FLASHIO.cpp. Every function in SPIFlash is built on top of the private functions in FLASHIO.cpp that make calls to the SPI Library - so the majority of your work should be done with this step. You might also have to modify some defines in defines.h.


Hi Marzogh,

Thanks for creating SPIFlash -- I can see a tremendous amount of work has gone into it.

I am new to SPI Flash memory so I may be asking a stupid question but here goes.  My basic question is: does SPIFlash support the "JV" series of chips from Winbond, specifically the W25Q32JVSSIQ and W25Q64JVSSIQ?  Or is there something about the JV series that makes them incompatible, and I should try to find the BV of FV series mentioned in the wiki?

The details:

I got some Winbond chips from DigiKey: W25Q32JVSSIQ and W25Q64JVSSIQ.  I soldered them on some SMT breakout boards from Adafruit and wired them up to an Adafruit Feather M0.  The pinouts are:

3V --> 8 (VCC)
GND --> 4 (GND)
SCK --> 6 (CLK)
MOSI --> 5 (DI / IO0)
MISO --> 2 (DO / IO1)
GPIO #11 --> 1 (/CS)

At first, I did not hook up pins 3 (IO2) or 7 (IO3) to anything since it seemed they were not needed in this configuration.

I modified the readWriteString example sketch to tell it to use pin 11 for chip select in the SPIFlash constructor:

Code: [Select]

//SPIFlash flash;
SPIFlash flash(11);

However, when I run I get an endless stream of this error message:

Unable to Enable Writing to chip.
Please make sure the HOLD & WRITEPROTECT pins are pulled up to VCC

According to the Winbond data sheet, the JV series of chips do not have HOLD and WRITEPROTECT pins:


Unlike the FV series, which state that pin 3 is both /WP and IO2, and pin 7 is both HOLD and IO3, the JV series just calls them IO2 and IO3.  Looks like write protection is enabled/disabled only via the status registers.

I tried various combinations of tying pins 3 and/or 7 to Vdd and Vcc, but always got the same error.  Then, based on the error message, I tried putting pullup resistors between 3 and 7 and Vcc, with values of 5.1K and then 100K, but I admit I'm just guessing there.  Same result.

So ... am I out of luck?  Is this chip series not supported? 

Should I add pullup resistors between pins 3 and/or 7 to Vcc?  If so, any suggestion on what values?



Hi ManlyEchidna, I'm afraid I have not tested the JV series of flash memory with the library. I've just quickly glanced over the datasheet you've linked to and I don't see any reason why they shouldn't work. The Error message that you are getting is only meant for chips with HOLD or WP pins, so, if you don't have them, you can ignore it.

Could you please run the following steps and let me know what output you get?

- Make sure #define RUNDIAGNOSTIC is uncommented in your SPIFlash.h file and save it.
- Upload the Diagnostics.ino sketch to your Feather M0 after setting
Code: [Select]

SPIFlash flash(11);

- Paste the serial output from the sketch in your reply.

Let's see if we can get the library working with your chips :)


Hey there! I'm having a hard time getting SPIFlash to work with a Spansion S25FL127S (MCU is an ATMega328p with Uno bootloader running at 16 MHz, 3.3v). I keep getting various errors, including 0x0B and 0x06; I suspect a wiring issue, however everything seems OK. Power and ground are OK, SPI lines are correct, CS is pulled high with a 10kohm resistor, and write protect and hold are both tied high. Any idea what's going wrong? Other devices on that same SPI bus work fine. (SPIFlash still errors out even with nothing else on the SPI bus.)


@compucat have you tried removing the pull up on the CS pin?


Yup, I've tried with CS pulled up, no pullup, write protect and hold floating, tied high, tied low. Nothing seems to work. Only change I've made to the FlashDiagnostics code is to set an alternative CS pin. Do you have a working schematic for an ATMega328p+Spansion setup?

EDIT: I've also tried wiring it to a stock Arduino Uno with resistors as level shifters, no dice. I might try again with a fresh chip just to rule out bad silicon.

EDIT 2: Tried a fresh chip, no dice.

Go Up