• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 3 answers
  • Subscribers 63 subscribers
  • Views 342 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.

Original question:

LP873220-Q1: how to program the OTP memory

LP873220-Q1: LP873220RHDRQ1 OTP

Julia Ju
Intellectual 360 points
Part Number: LP873220-Q1
Other Parts Discussed in Thread: AWR1642, LP873220

LP873220RHDRQ1 is used to power up AWR1642, and there are some questions need help:

 

1. What is the progress if we need TI OTP factory program?

2.At present, we configure the voltage value by I2C , then enabling EN; but there are some occasional non-working phenomena during many times test.

   Is there a configuration process guide with OTP initially null for reference?

 

3.If TI can provide the driver code of I2C for reference?

Here is my schematic diagram:

Thanks!

  • 0
    TI__Expert 5580 points
    Hi Julia,

    1. Process for the custom OTP needs business case approval from product line. Please contact your local TI representative to begin discussion regarding this.

    2. Please see www.ti.com/.../snvu582a.pdf for reference for programming.
    My first advice is to check that there is enough time between NRST set high and I2C writings.

    3. Please see programming guide.
  • 0 in reply to Jari Niemela13
    Intellectual 360 points

    Thanks for your advice！

    Yes，we found that the OTP register seems to be volatile during debugging: the register restores to the factory settings after power off and restart, but if the restart interval is short, the voltage is not low to the threshold value, and the register can be saved.

    The document only guides the reference value of the OTP register. Is there a reference C code driver for LP873220?

    We would like to refer to the official driver for processing, thanks!

  • 0 in reply to Julia Ju
    TI__Expert 4155 points
    Julia,

    There is no driver for the LP873220. The device has all of the settings programmed in at the factory, and any changes need to be made to the registers via I2C after the device is powered on.

    We also have versions of this device that are designed to be configured at startup. The LP873300 and LP873200 have all of the rails disabled by default so that the settings can all be modified before any of the rails turn on.

    www.ti.com/.../LP8733-Q1
  • 0 in reply to Jari Niemela13
    Intellectual 360 points

    We currently follow below steps：

        Pull down EN à

        Manual reset the chip (RESET register SW_RESET bit set 1) à

        Waiting RESET_REG_INT bit to be 0 à

        Configure the output voltage (only configure BUCK0, BUCK1, LDO0, LDO1 output, others remain OTP default value) à

        Pull up EN

    There are still occasional chips that cannot output the corresponding voltage correctly, while the EN will be pulled down by the chip itself.

    Whether it will be affected if step 4 (Clear Interrupts) is not operated？

    What interrupts should clear if “Clear Interrupts”is a must? How to clean up?

  • 0 in reply to Julia Ju
    TI__Expert 4155 points
    Julia,

    Please let me clear some things up.

    1. You cannot "wait" for the RESET_REG_INT bit to be set to 0. When the registers are reset to the default values this bit will be set to 1, and can only be set to 0 by writing a 1 to that bit.

    2. If you are using an LP873220, then the interrupts will not behave the same way as the LP873200, which is designed for user configuration. The configuration guide Jari linked to is focused around the LP873300 and LP873200. On these devices the nINT pin can be monitored to let the user know when I2C communication is ready. The device is only ready for I2C communication after the register reset interrupt is generated. On the LP873220 the RESET_REG_MASK is set so the nINT pin will not behave this way.

    3. There must be something else that is causing the EN pin to be pulled down. The device itself can't pull this pin down because it is not an output pin.

    4. If the interrupts are not cleared, the device will continue to work properly, however you would not be able to tell if the interrupt was generated a second time. For the LP873300 and LP873200 this is important so that the MCU can tell when the registers have been reset, but again the LP873220 will not behave this way.

    If you are changing all of the output voltages then I would highly recommend switching from the LP873220 to the LP873200.
  • 0 in reply to Nick Jakse49
    Intellectual 360 points

    In the document  SNVU582A.pdf, it is mentioned that “2.Wait for the nINT line to be set low”.

    If nINT is a must in hardware to extract out , so that MCU is OK to detect whether the chip is ready?

     

    What does it mean of below table:

     

    If SNVU582A.pdf is only for LP8873300 and LP873200, can you share LP873220D related document?

     

    What’s the main difference between LP873220 and LP873200? What benefit will bring to us if use LP873200?

     

    In addition to the above questions, I need to solve the problem urgently that why there are the accidental chips which cannot output the corresponding voltage correctly.

     

    Thanks！

     

  • 0 in reply to Julia Ju
    TI__Expert 4155 points

    Julia,

    I am not able to see the table you posted. What document is it from, and what is the figure #?

    The main difference between the LP873220 and LP873200 is that the LP873220 has pre-programmed sequencing and voltage levels, while the LP873200 is designed to be configured by the user and by default does not have any rails enabled. Both can be configured using I2C, but one is specifically designed for it, while the other is not. The LP873220 does not have a configuration guide like the LP873200.

    How long are you waiting between powering on the device and sending the I2C commands? The nINT pin on the LP873200 is designed to let the MCU know the device is ready for I2C communication, but if you are waiting a sufficient amount of time before trying to communicate, then that shouldn't be an issue.

    Are you able to read something from the device before you start sending I2C commands to confirm the I2C communication is ready?

    Also, are the chips that aren't working having problems with a particular rail or all of them? Are the voltages just going back to the default values or are they a completely different value?