[BUG] SSD1306 I2C display does not work

[jonathanmlang]

Did you test the latest bugfix-2.1.x code?

Yes, and the problem still exists.

Bug Description

I uncommented #define MKS_12864OLED_SSD1306 in configuration.h
Added pin definitions in the board file, pins_CREALITY_V4.h
#define I2C_SCL_PIN PB10
#define I2C_SDA_PIN PB11

Compiled and flahsed fine,
Blank display.

Bug Timeline

Not sure

Expected behavior

Expected the display to show the marlin interface

Actual behavior

Nothing happens.

Steps to Reproduce

I uncommented #define MKS_12864OLED_SSD1306 in configuration.h
Added pin definitions in the board file, pins_CREALITY_V4.h
#define I2C_SCL_PIN PB10
#define I2C_SDA_PIN PB11

Version of Marlin Firmware

02010300

Printer model

Creality ender 3 with custom display wired to exp3 cable

Electronics

Creality v4.2.2 stm32 pcb

Add-ons

No response

Bed Leveling

None

Your Slicer

None

Host Software

None

Don't forget to include

• A ZIP file containing your Configuration.h and Configuration_adv.h.

Additional information & file uploads

Configuration.zip

[ellensp]

MKS_12864OLED_SSD1306 is not a I2C display by default, it is a SPI display. Yes the SSD1306 does i2c, if its wired that way. and the firmware is setup for it.

MKS_12864OLED_SSD1306 needs to the appropriate SPI pins

[ellensp]

If your playing with a generic i2c SSD1306 module, you need #define U8GLIB_SSD1306 not MKS_12864OLED_SSD1306

[jonathanmlang]

Hello thanks for replying, Ive tried #define U8GLIB_SSD1306 but still no luck, no output at the pins as proven by an oscilloscope.

[dbuezas]

Same here, I'm using an skr3 ez board on an Ultimaker 2.
I see the i2c clock for the i2c epprom on startup but no clock at all if I disable it (not a single pulse). I did manually set the i2c pins since disabling eeprom skips defining them.

I tried changing pins and software/hw i2c. It looks like Marlin isn't even trying to talk to the display at all.

Update: my case was unrelated. The skr3 uses software i2c and that isn't supported by u8glib

[ellensp]

having a play

using a creality 4.2.2.controller and a 0.96" SSD1306 I2C module

I wired it up

  /**             LCD PORT
   *              ------
   *        PC6  | 1  2 | PB2
   *  (SDL) PB10 | 3  4 | PB11 (SDA)                               
   *        PB14   5  6 | PB13
   *        PB12 | 7  8 | PB15
   *         GND | 9 10 | 5V
   *              ------
   */

I added this code to Marlin/src/MarlinCore.cpp after SETUP_LOG("setup() completed.");

  SERIAL_ECHOLN("I2C scanner. Scanning ...");
  Wire.setSDA(PB11);
  Wire.setSCL(PB10);
  Wire.begin();
  for (byte i = 1; i < 120; i++)
  {
    hal.watchdog_refresh();
    Wire.beginTransmission (i);
    if (Wire.endTransmission () == 0)
      {
      SERIAL_ECHOLN("Found address: ",i);
      delay (1);
      }
  }
  SERIAL_ECHOLN ("Done.");

And #include <Wire.h> after #include <math.h>

Marlin boots and says

I2C scanner. Scanning ...
Found address: 60
Done.

60 is 3C, the expected address of the LCD, so this confirms the address and the wiring is OK

[ellensp]

I'm getting clock and data... cant say if its the correct data. (this is well after the I2C scanner)
Edit, this was due to my i2c scanner code initializing Wire. Signals are not visible otherwise

[dbuezas]

Good idea with the i2c scan. I tried the same but with Software i2c and get this:

Screen disconnected

10:18:16.948 > I2C scanner. Scanning ...
10:18:16.957 > Found address: 80
10:18:16.962 > Done.

That should be the onboard EEPROM

Screen connected

10:18:16.948 > I2C scanner. Scanning ...
10:18:16.949 > Found address: 3
10:18:16.954 > Found address: 60
10:18:16.957 > Found address: 80
10:18:16.959 > Found address: 96
10:18:16.961 > Found address: 112
10:18:16.962 > Done.

So my wiring should be correct.

[ellensp]

Odd, removing my I2C scanner code you dont get any output, as reported

but just add adding

  Wire.setSDA(PB11);
  Wire.setSCL(PB10);
  Wire.begin();

above SETUP_RUN(ui.init()); in Marlin/src/MarlinCore.cpp and you get a display!!

Ignore the power bank, its just a weight to hold the wires in place.
This is far form a solution, more a hack, But my brain has quit for the day, will have to continue this later

[ellensp]

@jonathanmlang

I purpose this as a fix. (basically it rearranges marlinui.cpp to setup the i2c pins before setting up the display, and changes the pin variable type)

diff --git a/Marlin/src/lcd/marlinui.cpp b/Marlin/src/lcd/marlinui.cpp
index 463d406110..fa511b4882 100644
--- a/Marlin/src/lcd/marlinui.cpp
+++ b/Marlin/src/lcd/marlinui.cpp
@@ -218,6 +218,10 @@ constexpr uint8_t epps = ENCODER_PULSES_PER_STEP;
 
 void MarlinUI::init() {
 
+  #if HAS_U8GLIB_I2C_OLED && PINS_EXIST(I2C_SCL, I2C_SDA) && DISABLED(SOFT_I2C_EEPROM)
+    Wire.begin(uint8_t(I2C_SDA_PIN), uint8_t(I2C_SCL_PIN));
+  #endif
+
   init_lcd();
 
   #if HAS_DIGITAL_BUTTONS
@@ -274,10 +278,6 @@ void MarlinUI::init() {
     slow_buttons = 0;
   #endif
 
-  #if HAS_U8GLIB_I2C_OLED && PINS_EXIST(I2C_SCL, I2C_SDA) && DISABLED(SOFT_I2C_EEPROM)
-    Wire.begin(int(I2C_SDA_PIN), int(I2C_SCL_PIN));
-  #endif
-
   update_buttons();
 
   TERN_(HAS_ENCODER_ACTION, encoderDiff = 0);

[ellensp]

Ive created a PR to fix this, so closing this issue

[github-actions]

This issue has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue for related bugs.