AT25SF041B - Confirm on chip marking

Hi,

I have a new prototype board with the  AT25SF041B-MAHD-T, which is not responding to ID enquiries - nor any other SPI commands. The datasheet does not mention the marking on the chip, and I want to check that there wasn't a mix-up at our CM.

It looks like the marking is

AC4

HDG

1AA

Is this correct?
Simon.

  • Hi Simon,
    Before I address the SPI issue that you mentioned, let me clarify on the marking on the chip,
    The first row 1 = Device Family Code + Device Density Code + Device Revision (option) : AC4: 4 means 4 Mbit.
    Row 2 = Class Code, Operating Voltage Code + Assembly Location Code (see ordering info in datasheet):
    HDG --> “G” specific location (internal information); Operating Voltage Code "D" = 2.5V - 3.6V.
    Row 3 = Date Code + Series Number : 1AA --> “1” means “the last digit of year”, “AA” means “the trace code for each assembly lot (assigned and maintained by assembly vendor)”

    Now, regarding the issue you have in reading the Mfr ID Code:
    You say: "it is not responding to ID enquiries - nor any other SPI commands".
    >>>
    In this type of situation, I would first try to confirm that your host/MCU and associated code / driver for SPI is really functioning in terms of writing and reading a known tested SPI device. After this is confirmed, make sure that the SPI frequency (start with a lower frequency) is set within the limits, set SPI Mode (try Mode = 0 to start with).

    Study carefully, how this SPI transaction works =>
    See attached timing diagram and refer datasheet for your part for more details.

    Basically, the clock from your host MCU (you are sending commands from) should continue until the SPI Flash has completed sending back the ID to MCU.
    Also, just for testing on the scope, try to send 90h(8 bits) + 24 bits of zeros + 16 bits for ID (CLK count required = 48)in a loop.
    So, you need a total of 48 clock cycles.
    For every loop, toggle the CS.

    In a nutshell, the following bit sequence should be used =>
    The SPI Read Mfr ID instruction is initiated by driving the CS pin low and shifting the instruction code 90h ( confirm this command code for your device from datasheet ) followed by a 24-bit address (A23-A0) of 000000h. If the 24-bit address is initially set to 000001h, the Device ID will be read first.

    Note that there will be nothing (high impedance) on the SO line until the transmission of the instruction code 90h and 24-bit address (A23-A0) of 000000h are finished. That is 8+24 = 32 bits of silence. After that, the ID will appear on the SO line.

    Best regards,
    Renesas Electronics Online Support

  • Thank you for the reply. Your description of the code structure makes sense, and builds confidence that the CM didn't use the wrong part.

    Is the family code correct for this part? Since this morning we were able to check an older/other boards with 'same' part, but these have "A4" (rather than "AC4").

    On reading ID, I have been using test tools to send the correct sequence. We will continue to investigate.

  • Hi Simon,

    I have forwarded your inquiry regarding the part marking to our NOR Flash Product group. I will let you know as soon as I get some clarification on this.

    Please note: Company has a holiday next week. So, make take some time.

    Best regards
    Renesas Electronics Online Support

  • Hi Simon,
    Code ‘A4’ would be the AT25DF041B

    Another example would be code ‘B4’ which would indicate AT25FF041A as the primary device.

    Code ‘AC4’ is confirmed to be correct for the AT25SF041B primary device, and series ‘B’ version.

    Best regards,
    Renesas Electronics Online Support

  • Hello Sarpa,

    Thank you for confirming the part markings, this closes our question.

    I was eventually able to get the parts to ID and Program in the end,
    Simon.