• State TI Thinks Resolved
  • Locked Locked
  • Replies 17 replies
  • Answers 2 answers
  • Subscribers 30 subscribers
  • Views 885 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

MCF8316A: I2C questions

Jason Ciou
Expert 3600 points
Part Number: MCF8316A
Other Parts Discussed in Thread: MCT8316A

Hi team,

When the customer debugs the MCF8316A, it is found that the I2C always returns -7, and the measured I2C waveform is detailed in the attachment.

Configure as follows:

1) I2C: 0x60

2) data length: 32bit

3) CRC: No

Programming logic: According to the format of Table 7-10, the data of each EEPROM address (take 0x00000080 address as an example) is as follows:

address 0x00000080 data is 0xC0, 0x10, 0x00, 0x80, 0x20, 0x8C, 0x63, 0x44 Every time you write an address corresponding to 8 bytes of data, it will delay 100ms

Excuse me, is the programming logic correct? If it is not correct, how to modify it?

  • 0
    TI__Mastermind 18916 points

    Hi Jason, 

    We will look into this and aim to provide a response by early next week 

    Best Regards, 
    Andrew 

  • 0 in reply to Andrew Liu
    TI__Expert 3600 points

    Hi Andrew,

    Customer has scheduled for the solution of replacement PMD1000, please help asap, thanks.

    Customer confirm that DDP can not write data to MCF8316A with the function of I2C_MasterWrite or I2C_PolledMasterWrite.
    The attached file is the driver for MCF8316A.
    Could you please confirm if the driver is OK?

    MCF8316A drivers.7z

    Jason

  • 0 in reply to Jason Ciou
    TI__Genius 9220 points

    Hi Jason,

    Could you please confirm whether the customer is using the RTM or APL version of the MCF8316A?

    • The APL version's default I2C address is 0x60
    • The RTM version's default I2C address is 0x01

    It looks like they are referencing the APL version of the datasheet, so if they switched over to using an RTM device, they would need to change the slave address to 0x01 according to the latest datasheet:

    Regards,
    Eric C.

  • 0 in reply to Eric Chen
    TI__Expert 3600 points

    Hi Eric,

    Change address to "0x02", DDP can not write data to EEPROM of MCF8316A via I2C(always show non-ack in 9th clk, please see the attached picture).
    How to confirm if MCF8316A is in sleep mode? If in sleep mode, how to wake up from sleep mode?

    Jason

  • 0 in reply to Jason Ciou
    TI__Mastermind 18916 points

    Hi Jason, 

    We can aim to provide a follow-up response early next week from Eric. 

    In the meantime, wanted to drop in one quick comment regarding the SLEEP mode configuration 

    • it's configured by the SPEED pin (see description, and table sections)

    Best Regards, 
    Andrew 

  • 0 in reply to Andrew Liu
    TI__Expert 3600 points

    Hi Andrew,

    Thanks for your comments.

    Update my customer questions.

    Configure SPEED pin to control with PWM and set High for SPEED while data is wrote to EEPROM of MCF8316A.
    I2C always show non-ack, please see the attched picture for address 0xC0/0x02.
    Any settings need confirm or any solutions to resolve this phenomenon?

    Jason

  • 0 in reply to Jason Ciou
    TI__Mastermind 24415 points

    Jason,

    We will get back to you tomorrow.

    Regards,

    Vishnu

  • 0 in reply to Vishnu Balaraj
    TI__Genius 9220 points

    Hi Jason,

    Could you please help confirm the following information with the customer so we can better support the issue:

    • Is the customer using the MCT8316A EVM or their own board design?
      • If this is on their own PCB, could you ask them to share the schematic?
    • Could they please take a picture of the device package marking so that we can confirm whether they are using the RTM or APL device?
    • What I2C clock frequency is the customer using?
      • If possible, do they have a logic analyzer for capturing the I2C SDA/SCL? It would be much easier to decode the I2C communication.

    Assuming the customer did not reprogrammed the EEPROM of the device (since they're getting NACKs), the factory default setting on the MCF8316A configures DEV_MODE = 0b (standby mode). This means the device will NOT enter SLEEP mode even if the SPEED pin is 0V. In fact, the customer should set the SPEED pin to 0V on power-up to prevent the MCF8316A from attempting to spin the motor immediately.

    After they power up the device, could you please ask them to measure the following regulator output to ensure they are powered up properly:

    • Buck regulator voltage Vbk output (BUCK_SEL = 01b, 5V by default)
    • AVDD (3.3V)
    • DVDD (1.5V)

    Thanks,
    Eric C.

  • 0 in reply to Eric Chen
    TI__Expert 3600 points

    HI Eric,

    Please see below comments from customer.

    Is the customer using the MCT8316A EVM or their own board design?
    --> Our own board design

    If this is on their own PCB, could you please share the schematic?
    --> Please see the attached schematic(Replace PMD1000 SCHEMATIC_Core.pdf)

    Could they please take a picture of the device package marking so that we can confirm whether they are using the RTM or APL device?
    --> Please see the attached picture(MCF8316A picture.png)

    What I2C clock frequency is the customer using?
    --> I2C clk frequency is 100kHz

    If possible, do they have a logic analyzer for capturing the I2C SDA/SCL? It would be much easier to decode the I2C communication.
    --> We measured by oscilloscope, please see the attached picture(no ack with 0x02 address.jpg and no ack with 0xC0 address.jpg)

    Buck regulator voltage Vbk output (BUCK_SEL = 01b, 5V by default)
    --> 1.1918V

    AVDD (3.3V)
    --> 3.28V

    DVDD (1.5V)
    --> 1.37V

    Replace PMD1000 SCHEMATIC_Core.pdf

    Jason

  • 0 in reply to Jason Ciou
    TI__Genius 9220 points

    Hi Jason,

    The customer is using a APL version of the MCF8316A, so the original 0x60 should be used as the default I2C slave address. 

    Vbk output of 1.19V indicates that the buck regulator is not configured correctly. Vbk should range from 3.3 to 5.7V.

    Looking at the schematic, L8 inductor is not populated (NC) and R472 is 0Ω. If using resistor mode, R472 needs to be 22Ω in order for the buck regulator to be functional.

    Please refer to table 7-1 in the datasheet for recommended settings for the buck regulator:

    The internal circuitry in MCF8316A is powered from the buck regulator output, so please ask the customer to ensure that the buck regulator is configured correctly and outputting nominal voltage before attempting to communicate with the device via I2C.

    Regards,
    Eric C.

  • 0 in reply to Eric Chen
    TI__Expert 3600 points

    Hi Eric,

    update customer questions, please see below.

    Data(follow Table8.3 and config SPEED "PWM Input") can be wrote via I2C, but CW still can not work.
    SPEED pin can measure PWM waveform, but OUT_A/B/C no output.
    Do you have data of EEPROM similar to our hardware design or followed our hardware design?

    Jason

  • 0 in reply to Jason Ciou
    TI__Genius 9220 points

    Hi Jason,

    Could you please ask the customer to follow the Essential Controls section in the MCF8316A Tuning Guide to configure the device to spin up the motor?

    The motor parameters R, L, and Ke need to be either entered or measured using the device's built in MPET algorithm in order for the FOC algorithm to operate properly.

    If the device still fails to drive OUT_A/B/C properly, the device might be hitting a fault condition. They can read back the GATE_DRIVER_FAULT_STATUS and CONTROLLER_FAULT_STATUS  registers to find out what fault is triggering when attempting to spin up the motor.

    Regards,
    Eric C.

  • 0 in reply to Eric Chen
    TI__Expert 3600 points

    Hi Eric,

    Customer would like to have a call with you, could you help?

    Jason

  • 0 in reply to Eric Chen
    TI__Expert 3600 points

    Hi Eric,

    Customer would like to have a call with you, could you help?

    Jason

  • 0 in reply to Jason Ciou
    TI__Mastermind 24415 points

    Eric will get back to you shortly. 

  • 0 in reply to Vishnu Balaraj
    TI__Expert 3600 points

    Hi Vishnu,

    Update customer question, thanks.

    1. CW can not work
    2. Out A/B/C output wrong waveform (see the attached picture)
    3. Address is 0xC0, configuration data (see the attached txt file)
    4. When is available for our AE to have a con-call with customer?

    Configuration Data for MCF8316A.txt
    Fullscreen
    1
    2
    3
    4
    5
    6
    7
    8
    9
    10
    11
    12
    13
    14
    15
    16
    17
    18
    19
    /********************** I2C data word format ******************************/
    /* Target_ID R/W Control_Word Data CRC_8 */
    /* A6~A0 W0 CW23~CW0 D15/D31/D63~D0(LSB first) C7~C0 */
    /***********************************************************************/
    /********************** 24-bit control word format *****************************/
    /* CW23 CW22 CW21~CW20 CW19~CW16 CW15~CW12 CW11~CW0 */
    /* OP_R/W CRC_EN DLEN MEM_SEC MEM_PAGE MEM_ADDR */
    /* 1: R 1: enable 00: 16bit 4bit 4bit 12bit */
    /* 0: W 0: disable 01: 32bit */
    /* 10: 64bit */
    /* 11: reserve */
    /*************************************************************************/
    static DataFormat_32bit gRegisterAddressTable[RegisterMax] =
    {
    //Control Data+Address+Data(4 Bytes)
    {0x10, 0x00, 0x80, 0x20, 0x8C, 0x63, 0x44}, // 0x80 : init speed detect config
    {0x10, 0x00, 0x82, 0x64, 0xF0, 0x3A, 0x28}, // 0x82 : reverse driver config
    XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

    Jason

  • 0 in reply to Jason Ciou
    TI__Genius 9220 points

    Hi Jason,

    As discussed in email, we'll continue to support this issue offline and on the scheduled call.

    We may need to simply ask the customer to load in the pre-configured register settings first before spinning up the CW.

    Regards,
    Eric C.