EADOGM LCD interface with STM32

[cumin1091]

Hello All,

I am undergraduate student. I want to interface LCD display EADOGM 16*2
with STM32 controller.
I am using 8 bit interface.
The connection is as below
RS-D7
RW-D8
E-D9
C0-C7 are connected to data lines of LCD.
I could find some display driver and i have modified according to my
requirement.
When i run this application, I could not see anything Displayed.
Can anyone pls suggest where am i wrong

ea-dogm-ex1.c

#include "ea-dogm-ex1.h"
#include <stdio.h>
#define EADOGM162 1


// some special symbol chrs defined
#define EADMSYM_DEG 0b11011111 // degree symbol
#define EADMSYM_DARWL 0b11111011 // double <<
#define EADMSYM_DARWR 0b11111100 // double >>
#define EADMSYM_LT 0b00111100 // less than <
#define EADMSYM_GT 0b00111110 // greater than >
#define EADMSYM_OHM 0b00011110 // ohm symbol

// some command defines
#define EADMCMD_CONTRAST 0b0111000 // contrast command (0b0111xxxx)


// we noticed some issues with GLOBAL on pic24 devices
#ifndef GLOBAL
#define GLOBAL INTR_GLOBAL
#endif




// 2x16
#ifdef EADOGM162
#define EADOGM_ROWS 2
#if EADOGMVDD == 5
#define EADOGM_INIT_BIAS_SET 0x1C
#define EADOGM_INIT_POWER_CONTROL 0x52
#define EADOGM_INIT_FOLLOWER_CONTROL 0x69
#define EADOGM_INIT_CONTRAST_SET 0x74
#else
#define EADOGM_INIT_BIAS_SET 0x14
#define EADOGM_INIT_POWER_CONTROL 0x55
#define EADOGM_INIT_FOLLOWER_CONTROL 0x6D
#define EADOGM_INIT_CONTRAST_SET 0x78
#endif

#define EADOGM_INIT_FS1 0x39
#define EADOGM_INIT_FS2 0x38
#define EADOGM_INIT_CLEAR_DISPLAY 0x01
#define EADOGM_INIT_ENTRY_MODE 0x06
#define EADOGM_COLSPAN 40 // suggested that this be 40 on model 162
#endif




#define EADOGM_CMD_CLR 1
#define EADOGM_CMD_CURSOR_ON 0b00001111
#define EADOGM_CMD_CURSOR_OFF 0b00001100
#define EADOGM_CMD_DISPLAY_ON 0xC
#define EADOGM_CMD_DISPLAY_OFF 0b00001000
#define EADOGM_CMD_DDRAM_ADDR 0x80
#define EADOGM_CMD_CGRAM_ADDR 0x40
#define EADOGM_CMD_SELECT_R0 0b00011000
#define EADOGM_CMD_SELECT_R1 0b00010000
#define EADOGM_CMD_SET_TABLE2 0b00101010

void eaDogM_ini8Bit(void)
{
DisplayOutputDrive(DISPLAY_E);
DisplayOutputDrive(DISPLAY_RW);
DisplayOutputDrive(DISPLAY_RS);
DisplayOutputLow(DISPLAY_E);
DisplayOutputLow(DISPLAY_RS);
DisplayOutputLow(DISPLAY_RW);
}


void eaDogM_WriteChr(char value)
{
DisplayOutputHigh(DISPLAY_RS);
DisplayOutputLow(DISPLAY_RW);
DisplayOutputLow(DISPLAY_E);

DisplayOutputHigh(DISPLAY_E);
Delay(1);
DisplayOutputDrive(value);

DisplayOutputLow(DISPLAY_E);
Delay(1);
}

void eaDogM_WriteCommand(uint8_t cmd)
{
DisplayOutputLow(DISPLAY_RS);
DisplayOutputLow(DISPLAY_RW);
DisplayOutputLow(DISPLAY_E);
// #ifndef EADOGM_NOCSB
//// DisplayOutputLow(EADOGM_PIN_CSB);
// #endif
DisplayOutputHigh(DISPLAY_E);
Delay(1);
// EADOGM_8BIT_PORT(cmd);
DisplayOutputLow(DISPLAY_E);
Delay(1);
// #ifndef EADOGM_NOCSB
// DisplayOutputLow(EADOGM_PIN_CSB);
// #endif
// Delay(1);
}

//// spi mode
//void eaDogM_WriteChr(char value)
//{
// DisplayOutputHigh(EADOGM_PIN_RS);
// DisplayOutputLow(EADOGM_PIN_CSB);
// eaDogM_outSPI(value);
// DisplayOutputHigh(EADOGM_PIN_CSB);
// Delay(1);
//}
//
//void eaDogM_WriteCommand(uint8_t cmd)
//{
// DisplayOutputLow(DISPLAY_RS);
// DisplayOutputLow(EADOGM_PIN_CSB);
// eaDogM_outSPI(cmd);
// DisplayOutputHigh(EADOGM_PIN_CSB);
// Delay(1);
//}



void eaDogM_Initialize(void)
{



eaDogM_ini8Bit();
//#else
// DisplayOutputDrive(EADOGM_PIN_CSB);
// DisplayOutputDrive(DISPLAY_RS);
// DisplayOutputHigh(EADOGM_PIN_CSB);
// DisplayOutputHigh(DISPLAY_RS);


Delay(200);

eaDogM_WriteCommand(EADOGM_INIT_FS1);
eaDogM_WriteCommand(EADOGM_INIT_BIAS_SET);
eaDogM_WriteCommand(EADOGM_INIT_POWER_CONTROL);
eaDogM_WriteCommand(EADOGM_INIT_FOLLOWER_CONTROL);
eaDogM_WriteCommand(EADOGM_INIT_CONTRAST_SET);
eaDogM_WriteCommand(EADOGM_INIT_FS2);
eaDogM_WriteCommand(EADOGM_INIT_CLEAR_DISPLAY);
eaDogM_WriteCommand(EADOGM_INIT_ENTRY_MODE);
}



// v2.01 moved functions to macros
#define eaDogM_DoubleHeightOff() eaDogM_WriteCommand(0b00101000)
#define eaDogM_Cls() eaDogM_WriteCommand(EADOGM_CMD_CLR)
#define eaDogM_CursorOn() eaDogM_WriteCommand(EADOGM_CMD_CURSOR_ON)
#define eaDogM_CursorOff()
eaDogM_WriteCommand(EADOGM_CMD_CURSOR_OFF)
#define eaDogM_DisplayOff()
eaDogM_WriteCommand(EADOGM_CMD_DISPLAY_OFF)

void eaDogM_DisplayOn()
{
eaDogM_WriteCommand(EADOGM_CMD_DISPLAY_ON);
}

void eaDogM_SetPos(uint8_t r, uint8_t c)
{
uint8_t cmdPos;
cmdPos = EADOGM_CMD_DDRAM_ADDR + (r * EADOGM_COLSPAN) + c;
eaDogM_WriteCommand(cmdPos);
}

void eaDogM_ClearRow(uint8_t r)
{
uint8_t i;
eaDogM_SetPos(r,0);
for(i=0; i<EADOGM_COLSPAN; i++) {
eaDogM_WriteChr(' ');
}
}

void eaDogM_WriteString(char strPtr)
{

eaDogM_WriteChr(strPtr);

}



void eaDogM_WriteIntAtPos(uint8_t r, uint8_t c, uint8_t i)
{

eaDogM_WriteCommand((EADOGM_CMD_DDRAM_ADDR + (r * EADOGM_COLSPAN) +
c));

eaDogM_WriteChr(i/10+'0');
eaDogM_WriteChr(i%10+'0');

}

// this writes a byte to the internal CGRAM (v2.02)
// format for ndx: 00CCCRRR = CCC = character 0 to 7, RRR = row 0 to 7
void eaDogM_WriteByteToCGRAM(char ndx, char data)
{
unsigned int cmd;

cmd = ndx & 0x3f; // mask off upper to bits
cmd = cmd | EADOGM_CMD_CGRAM_ADDR; // set bit cmd bits

eaDogM_WriteCommand(cmd);
eaDogM_WriteChr(data);

// this is done to make sure we are back in data mode
eaDogM_SetPos(0,0);
}


//#endif


static void Delay(int nCount)
{
int index = 0;
for(index = (100 * nCount); index != 0; index--)
{
}
}

void DisplayOutputHigh(uint16_t LCDRegValue)
{
GPIO_WriteBit(GPIOD, LCDRegValue, (BitAction)1);

}

void DisplayOutputLow(uint16_t LCDRegValue)
{
GPIO_WriteBit(GPIOD, LCDRegValue,(BitAction)0);

}

void DisplayOutputDrive(uint8_t data)
{
GPIO_Write(GPIOC, data);

}







ea-dogm-ex1.h


#define EADOGM163 1 // to use MODEL EADOG163
#define EADOGMVDD 5 // 5v LCD Vdd



#define EADOGM162 1
#define DISPLAY_ROWS 2
#define DISPLAY_VDD 3.3

#define DISPLAY_GPIO_PORT GPIOD
#define DISPLAY_E GPIO_PinSource9
#define DISPLAY_RS GPIO_PinSource7
#define DISPLAY_RW GPIO_PinSource8



#define DISPLAY_INIT_BIAS_SET 0x14
#define DISPLAY_INIT_POWER_CONTROL 0x55
#define DISPLAY_INIT_FOLLOWER_CONTROL 0x6D
#define DISPLAY_INIT_FUNCTION_SET1 0x39
#define DISPLAY_INIT_FUNCTION_SET2 0x38
#define DISPLAY_INIT_CONTRAST_SET 0x78
#define DISPLAY_INIT_CLEAR_DISPLAY 0x01
#define DISPLAY_INIT_ENTRY_MODE 0x06

#define DISPLAY_CMD_CONTRAST_SET 0x70
#define DISPLAY_CMD_DISPLAY_ON 0xC0
#define DISPLAY_CMD_DISPLAY_OFF 0B00001000
#define DISPLAY_CMD_CLR 1
#define DISPLAY_CMD_CURSOR_ON 0xF //Check the value
#define DISPLAY_CMD_CURSOR_OFF 0B00001100
#define DISPLAY_CMD_DDRAM_ADD 0x80
#define DISPLAY_START_LINE1_ADD 0x00
#define DISPLAY_START_LINE2_ADD 0x40

#define DISPLAY_SYM_DEG 0B11011111
#define DISPLAY_SYM_SLASH 0B00101111
#define DISPLAY_SYM_LEFT_PARAN 0B10000010
#define DISPLAY_SYM_RIGHT_PARAN 0B10010010
#define DISPLAY_SYM_MUL 0B01111111
#define DISPLAY_SYM_DOT 0B00000100


#define INPUT_POWER_ON 00
#define INPUT_TEMPERATURE 01
#define INPUT_GENERAL 02
#define START_UP_SCREEN 03
#define MEASUREMENT_SCREEN 04
#define GENERAL_SCREEN 05

#define READ_BUSY_FLAG 0
#define READ_DATA_FROM_LCD 1

void eaDogM_Initialize(void);
void eaDogM_DisplayOn(void);
void eaDogM_WriteChr(char value);
void eaDogM_WriteIntAtPos(uint8_t r, uint8_t c, uint8_t i);
void eaDogM_DisplayOn(void);
void eaDogM_WriteString(char strPtr);
void eaDogM_ClearRow(uint8_t r) ;






Thanks



---------------------------------------
Posted through http://www.EmbeddedRelated.com

[Rich Webb]

On Tue, 13 Nov 2012 09:24:45 -0600, "cumin1091" <29963@embeddedrelated>
wrote:

>Hello All,
>
>I am undergraduate student. I want to interface LCD display EADOGM 16*2
>with STM32 controller.
>I am using 8 bit interface.
>The connection is as below
>RS-D7
>RW-D8
>E-D9
>C0-C7 are connected to data lines of LCD.
>I could find some display driver and i have modified according to my
>requirement.
>When i run this application, I could not see anything Displayed.
>Can anyone pls suggest where am i wrong

I've coded to the EA DOG displays with an STM32F103 controller so I know
that it's (unsurprisingly) do-able.

Proper initialization is where getting character LCD displays up and
running usually runs into difficulties.

I'd recommend starting with your own code, rather than trying to adapt
someone else's boilerplate. The EADOG datasheet together with the
controller's datasheet should have enough pseudocode and command
sequences. <http://www.lcd-module.de/eng/pdf/zubehoer/st7036.pdf>

I notice, for example, that you have both

#define EADOGMVDD 5 // 5v LCD Vdd

and

#define DISPLAY_VDD 3.3

defined. Which is correct? The EADOG's controller wants > 40 msec after
Vdd is stable on a 3.3 V system and 15 msec for a 5 V. I'd check that
first. Are you sure your Delay() function isn't getting optimized out?

--
Rich Webb Norfolk, VA

[cumin1091]

Thanks for the reply

I have not set any optimization level, currently its 0.
And I am using 3.3 Volts.
i have written my own code as suggested.
And I am following the Pseudocode in ST datasheet for initilazation.

But still i could not find any difference.
And my delay is gretaer than 40 msec.

[Simon Clubley]

On 2012-11-15, cumin1091 <29963@embeddedrelated> wrote:
>
> Thanks for the reply
>
> I have not set any optimization level, currently its 0.

That's nowhere near good enough.

The only way to be sure is to look at the generated code and make sure
the statements in Delay() are present and are in the code path (either
as a call to Delay() or placed inline by the compiler).

If Delay() is been removed (and not replaced with inline code) then try
making the loop counter volatile.

> And I am using 3.3 Volts.
> i have written my own code as suggested.
> And I am following the Pseudocode in ST datasheet for initilazation.
>
> But still i could not find any difference.
> And my delay is gretaer than 40 msec.
>

You don't know that until after you have looked at the generated code
and estimated how many clock cycles each pass through the delay loop
will take.

Simon.

--
Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP
Microsoft: Bringing you 1980s technology to a 21st century world

[Rich Webb]

On Thu, 15 Nov 2012 12:34:16 +0000 (UTC), Simon Clubley
<clubley@remove_me.eisner.decus.org-Earth.UFP> wrote:

>On 2012-11-15, cumin1091 <29963@embeddedrelated> wrote:
>>
>> Thanks for the reply
>>
>> I have not set any optimization level, currently its 0.
>
>That's nowhere near good enough.
>
>The only way to be sure is to look at the generated code and make sure
>the statements in Delay() are present and are in the code path (either
>as a call to Delay() or placed inline by the compiler).
>
>If Delay() is been removed (and not replaced with inline code) then try
>making the loop counter volatile.
>
>> And I am using 3.3 Volts.
>> i have written my own code as suggested.
>> And I am following the Pseudocode in ST datasheet for initilazation.
>>
>> But still i could not find any difference.
>> And my delay is gretaer than 40 msec.
>>
>
>You don't know that until after you have looked at the generated code
>and estimated how many clock cycles each pass through the delay loop
>will take.

Even then, I wouldn't (and don't, in my own work) believe it until I've
seen the results on a scope or logic analyzer. In addition to the
initial delay, there are (as I'm sure you know, this is for the OP)
constraints on setup and hold times, and on minimum pulse widths.

A recommended approach is to write and verify the initialization routine
from the ST7036 flowchart one step at a time, making sure that all of
the timing and instruction sequencing is good at each step.

[Mark Borgerson]

In article <k82ng8$k1l$1...@dont-email.me>,
clubley@remove_me.eisner.decus.org-Earth.UFP says...

>
> On 2012-11-15, cumin1091 <29963@embeddedrelated> wrote:
> >
> > Thanks for the reply
> >
> > I have not set any optimization level, currently its 0.
>
> That's nowhere near good enough.
>
> The only way to be sure is to look at the generated code and make sure
> the statements in Delay() are present and are in the code path (either
> as a call to Delay() or placed inline by the compiler).
>
> If Delay() is been removed (and not replaced with inline code) then try
> making the loop counter volatile.
>
> > And I am using 3.3 Volts.
> > i have written my own code as suggested.
> > And I am following the Pseudocode in ST datasheet for initilazation.
> >
> > But still i could not find any difference.
> > And my delay is gretaer than 40 msec.
> >
>
> You don't know that until after you have looked at the generated code
> and estimated how many clock cycles each pass through the delay loop
> will take.
>

This is the kind of problem that begs for the use of a few test output
bits and an oscilloscope. IIRC, you can find USB scopes that display on
your PC for under $200.

Until you can put a probe on various signal lines to the LCD, you don't
really know if you've forgotten to turn on the clock to one of the
peripheral ports. I've had problems of that nature on the STM32 quite
often. I'll cut and paste code for a peripheral, but forget to change
one of the clock or port assignments.

Mark Borgerson