Posted on May 17, 2011 at 13:20When a ''Read ID'' of the Samsung K9F1208U0C is done in STM32F103VE,
the expected waveform is:
CLE ___----_____________________
ALE ___________----_____________
However, the observed waveform is:
CLE ___----___-____--___________
ALE ___________----_____________
K9F1208U0C does not respond to the request due to the CLE issues.
Other observations:
When only CLE operations are done in a loop, the observed waveforms are clean.
When only ALE operations are done in a loop, the observed waveforms are clean.
When a CLE operation is followed by an ALE operation, the waveform is not clean.
There is a work-around that seems to work, but it has not been tested for other operations yet.
The Standard Peripheral Library sample code programs the GPIOD, Pin 11 (CLE) as GPIO_Mode_AF_PP and the FSMC_NAND_ReadID function uses the following sequence:
*(__IO uint8_t *)(Bank_NAND_ADDR | CMD_AREA) = 0x90;
*(__IO uint8_t *)(Bank_NAND_ADDR | ADDR_AREA) = 0x00;
Instead, if the pin is configured as GPIO_Mode_Out_PP, and the sequence is modified as follows:
GPIO_SetBits(GPIOD, GPIO_Pin_11);
*(__IO uint8_t *)(Bank_NAND_ADDR | CMD_AREA) = 0x90;
GPIO_ResetBits(GPIOD, GPIO_Pin_11);
*(__IO uint8_t *)(Bank_NAND_ADDR | ADDR_AREA) = 0x00;
the waveforms look like the expected waveforms and K9F1208U0C responds correctly.
Essentially, we are controlling the waveform from software, and the ReadID operation works correctly. This suggests that it is not a board related issue, but some issue related to FSMC.
Is there any FSMC mis-configuration that could cause this issue ? Has anyone observed such an issue ?
- Yantra
