Black horizontal lines when SDRAM is used. Why?

[Exit0815]

Hello,

another day, another problem...

I just started with a custom board. But i am using parts which are already working on boards from ST so i thought it will be easy to adapt.

I started a new project and if i use internal RAM for a touchgfx box everything is fine.

As soon as i add the SDRAM to the linker file i have a lot of lines (black or just not read from sdram correctly?).

For SDRAM the MT48LC4M32B2 with 16MB is used like in a lot of disco or eval boards.

MCU is running at 200MHz

My FMC / SDRAM Settings:

#define REFRESH_COUNT                    		 ((uint32_t)0x0603)   /* SDRAM refresh counter */
#define SDRAM_TIMEOUT                            ((uint32_t)0xFFFF)
#define SDRAM_MODEREG_BURST_LENGTH_1             ((uint16_t)0x0000)
#define SDRAM_MODEREG_BURST_LENGTH_2             ((uint16_t)0x0001)
#define SDRAM_MODEREG_BURST_LENGTH_4             ((uint16_t)0x0002)
#define SDRAM_MODEREG_BURST_LENGTH_8             ((uint16_t)0x0004)
#define SDRAM_MODEREG_BURST_TYPE_SEQUENTIAL      ((uint16_t)0x0000)
#define SDRAM_MODEREG_BURST_TYPE_INTERLEAVED     ((uint16_t)0x0008)
#define SDRAM_MODEREG_CAS_LATENCY_2              ((uint16_t)0x0020)
#define SDRAM_MODEREG_CAS_LATENCY_3              ((uint16_t)0x0030)
#define SDRAM_MODEREG_OPERATING_MODE_STANDARD    ((uint16_t)0x0000)
#define SDRAM_MODEREG_WRITEBURST_MODE_PROGRAMMED ((uint16_t)0x0000)
#define SDRAM_MODEREG_WRITEBURST_MODE_SINGLE     ((uint16_t)0x0200)

FMC_SDRAM_CommandTypeDef command;
 
	/* Program the SDRAM external device */
	BSP_SDRAM_Initialization_Sequence(&hsdram1, &command);
 
	//Deactivate speculative/cache access to first FMC Bank to save FMC bandwidth
	FMC_Bank1->BTCR[0] = 0x000030D2;

And the Init sequence:

static void BSP_SDRAM_Initialization_Sequence(SDRAM_HandleTypeDef *hsdram, FMC_SDRAM_CommandTypeDef *Command)
{
 __IO uint32_t tmpmrd = 0;
 
    /* Step 1: Configure a clock configuration enable command */
    Command->CommandMode            = FMC_SDRAM_CMD_CLK_ENABLE;
    Command->CommandTarget          =  FMC_SDRAM_CMD_TARGET_BANK1;
    Command->AutoRefreshNumber      = 1;
    Command->ModeRegisterDefinition = 0;
 
    /* Send the command */
    HAL_SDRAM_SendCommand(hsdram, Command, SDRAM_TIMEOUT);
 
    /* Step 2: Insert 100 us minimum delay */
    /* Inserted delay is equal to 1 ms due to systick time base unit (ms) */
    HAL_Delay(1);
 
    /* Step 3: Configure a PALL (precharge all) command */
    Command->CommandMode            = FMC_SDRAM_CMD_PALL;
    Command->CommandTarget          = FMC_SDRAM_CMD_TARGET_BANK1;
    Command->AutoRefreshNumber      = 1;
    Command->ModeRegisterDefinition = 0;
 
    /* Send the command */
    HAL_SDRAM_SendCommand(hsdram, Command, SDRAM_TIMEOUT);
 
    /* Step 4: Configure an Auto Refresh command */
    Command->CommandMode            = FMC_SDRAM_CMD_AUTOREFRESH_MODE;
    Command->CommandTarget          = FMC_SDRAM_CMD_TARGET_BANK1;
    Command->AutoRefreshNumber      = 8;
    Command->ModeRegisterDefinition = 0;
 
    /* Send the command */
    HAL_SDRAM_SendCommand(hsdram, Command, SDRAM_TIMEOUT);
 
    /* Step 5: Program the external memory mode register */
    tmpmrd = (uint32_t)SDRAM_MODEREG_BURST_LENGTH_1          | \
             SDRAM_MODEREG_BURST_TYPE_SEQUENTIAL   | \
             SDRAM_MODEREG_CAS_LATENCY_3           | \
             SDRAM_MODEREG_OPERATING_MODE_STANDARD | \
             SDRAM_MODEREG_WRITEBURST_MODE_SINGLE;
 
    Command->CommandMode            = FMC_SDRAM_CMD_LOAD_MODE;
    Command->CommandTarget          = FMC_SDRAM_CMD_TARGET_BANK1;
    Command->AutoRefreshNumber      = 1;
    Command->ModeRegisterDefinition = tmpmrd;
 
    /* Send the command */
    HAL_SDRAM_SendCommand(hsdram, Command, SDRAM_TIMEOUT);
 
    /* Step 6: Set the refresh rate counter */
    /* Set the device refresh rate */
    HAL_SDRAM_ProgramRefreshRate(hsdram, REFRESH_COUNT);
 
}

Here are the SDRAM Settings in MX and the pictures of the display with and without SDRAM active.

What did i forget?

Can it be a not proper soldered connection? Should this not end in a hardfault?

Thanks a lot for helping.

[Tesla DeLorean]

>>Can it be a not proper soldered connection? Should this not end in a hardfault?

If the SDRAM isn't working properly you typically won't get a Hard Fault unless you executing code from it.

You could write and validate data patterns into the memory to confirm the read-back content is correct, and maintained over time.

[Exit0815]

Thanks for your answer.

I changed Data and Byte endable to 16 bit instead of 32 and now it is working.

But i expect a lower performance, right?

How can i validate the best way?

Read the 0xC0000000 in debug memory?

[Tesla DeLorean]

Use pointers

Fill the entire memory space with pseudo-random data sequence you can index, repeat and test.

People often use walking or counting patterns, you write the whole memory, then read it back and compare with expected content. I tend to use more complex patterns as it stresses the memory (data bits) / address decoder (address bits, row, columns)

[Exit0815]

Thank you.

I will try the weekend and post results.

[Exit0815]

@Community member​ 

So i just found some time and i checked the 0xC0000000 just quick with the memory function in debug mode.

Am i right that the values should be the same one, if 32Bit or 16Bit?

Then there is a difference and something is wrong.

I am not sure if all pins are 100% connected from MCU to the SDRAM.

Does that look like it could be wiring issue or is it a setting problem?

The only change i did is the "Data" and "Byte enable" from 16 Bit to 32 Bit in CubeMX.

Thank you.

[Exit0815 · Accepted solution]

OK, this happens when not every connection between MCU and Flash is connected.

Reworked PCB and it works now.

In the unlikely event someone else has the same problem, check the connections