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.

TPS53667: Read value of 8Ch

Part Number: TPS53667

Hi,

Would you give me advice?
Our customers are considering TPS53667 + CSD95490 at 6Phase.
The actual load current does not match the read value of 8Ch.
Are there any settings to check?
Please let me know if there is any necessary information.

TPS53667_Value.pdf

Best regards,
Yuto Sakai

  • Hi Yuto,

    The first thing I would check on your design is the IMON resistor value. Does the populated resistor match the calculated value? Is the calculated resistor value correct (i.e. double check your math)?

    I suggest this because assuming the correct value was placed on the PCB, VIMAX (0.85V) would correspond to a current of 377A. That's pushing past the capabilities of a six phase design and seems unlikely. 

    After checking this resistors I would then double check the initial IMAX settings. If you're using the pin-strapping options make sure the correct resistors are populated on the F-IMAX pin. For NVM settings, make sure the pre-programmed IMAX value is correct. The wrong IMAX value will affect the scaling seen on IMON and could affect your results.

    Once those options get straightened out, you can also use the IOUT_CAL_OFFSET register to further improve your current sense accuracy. Simply apply a known load current. Read the READ_IOUT register and subtract the register value from the applied current to get your offset error. Then write the correct offset value into the register.

    Example: 50A load is applied, 49A is read. 49A - 50A = -1A = offset error. Since the error is -1A an offset of +1A should be written to IOUT_CAL_OFFSET to cancel our the error.

    Let me know if you have any questions.

    Cheers,

    Carmen

  • Hi Carmen-san,

    Thank you for your reply.
    If IMON resistor value differs by 1k from calculation, can you judge how much the accuracy of 8Ch will change (37.1k → 36.1k)?

    Also, I checked the voltage on the CSPx pin, but is the expected value of VCSPx 1.7 V or more?
    Since the voltage value became 1.7 V or less, I would like to confirm.

    Best regards,
    Yuto Sakai
  • Hi Yuto,

    A 1kΩ difference wouldn't cause all the issues you're seeing. Have you checked the F-IMAX divider to make sure it's setting the correct IMAX value? Is the IMAX register reading correct? I'd also check the OFFSET register too just to make sure it's initially 0 like it's supposed to be.

    As for CSP pins they should sit at 1.7V DC so that's expected. The value of those pins will change at 5mV/A, ±2.5%. Typically to see those small changes the scope would need to be AC coupled.

    Cheers,

    Carmen
  • Hi Carmen-san

    Thank you for your surpport.
    I'm sorry to bother you over and over. 
    Our customers are designing with RF = 20 kΩ, RIMAX = 11 kΩ, RIMON = 37.1 kΩ.
    Also, RLSET of CSD95490 is 110 kΩ.
    The calculated value of IMAX is 164.5 A, will be the expected value of DAh be A5?
    Please gime me advice if you have anything else to worry about.

    Best regards,
    Yuto Sakai

  • Hey Yuto,

    Can you share a schematic with me? Your resistor values all check out for the config and IMON resistors so everything should be getting set properly. The A5h value in the IMAX register is correct as well. 

    If you're using 150nH inductors on the power stages then RLSET is right too.

    I'd like to look over the schematic and see if anything else looks incorrect.

    Thanks,

    Carmen

  • Carmen-san,

    Thank you for your surpport.
    Would you please contact me at the address below so I would like to send the circuit diagram?
     
     
    Best regards,
    Yuto Sakai

  • Hi Yuto,

    Since we're discussing your schematic/layout in a private thread I'm going to close this thread.

    Cheers,

    Carmen