• State
  • Locked Locked
  • Replies 1 reply
  • Subscribers 62 subscribers
  • Views 107 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

TPS65313-Q1: Conditions for operation without MCU

Lee DongJae
Prodigy 20 points
Part Number: TPS65313-Q1
Other Parts Discussed in Thread: TPS65313-EVM, LP2985

Tool/software:

Hello, I am planning to use the TPS65313-Q1 as a PMIC.

I am currently trying to use the TPS65313-Q1 to generate a 1.2V supply for operating your 28377S and 28377D MCUs.

I understand that the TPS65313-Q1 is an SPI-based PMIC. However, in order to use SPI, the MCU needs to be powered first — which requires the 1.2V output from the PMIC. Therefore, I assumed that the PMIC should be able to output 3.3V, 1.2V, and 5V even without SPI communication, at least during initial power-up.

To test this, I also purchased the TPS65313-EVM and used it for comparison.

I created a simple custom board to check the output voltages of the PMIC, referencing the schematic of the EVM.
Since I am not using any external power source, I connected Exsup, Ext_Vsen1, and Ext_Vsen2 to GND.
For the WAKE signal, I used your LP2985 regulator to generate 5V and applied it to the WAKE pin. (In the EVM, I observed that 12V is momentarily applied and then removed, but in my case, I’m applying 5V continuously. The resistor and capacitor values are the same as on the EVM.)

Regarding the VIO pin, I removed all jumpers going to the MCU on the EVM and observed that output voltages still appeared to be generated, which leads me to believe that even with VIO = 0V, the device can still generate outputs.

In this setup, I applied 12V to my custom board and verified that the WAKE pin receives 5V.
I also confirmed that the bootstrap capacitor between Boot1 and PH1/PH1A is charged to 4.4V. However, since PH1 and PH1A remain at 0V, it seems like no switching is occurring.
As a result, Buck1 is not turning on, and VREG shows a strange triangular waveform around 3V.

So here are my questions:

  1. Is it impossible to operate this device without SPI communication?

  2. Are there any differences in the default register or bit values between the EVM's preloaded TPS65313-Q1 and newly purchased standalone TPS65313-Q1 devices?

  3. Or could it be that a fault occurred early on, and without SPI access, I am unable to clear the fault, which is why Buck1 is not starting?

Thank you in advance for your support.

  • 0
    TI__Mastermind 21170 points

    Hi Lee,

    1. You can operate this PMIC even without SPI by using the wake pin. And yes you also don't need VIO for the regulators to turn on as the VIO is powering the digital functions of the PMIC such as SPI.
    2. There should be no differences. You can read the device ID and revision with the RD_DEV_REV and RD_DEV_ID SPI commands.
    3. There could be some fault happening but they shouldn't interfere with the device powering on. 

    Lee DongJae said:
    VREG shows a strange triangular waveform around 3V

    This could be the cause for the PMIC not powering on as if this rail is not operating properly it can't power up. The VREG output should be higher than 3V. Please see the capacitor recommendations for this regulator in sections 12.2.2.6 and 12.2.2.8:

    The AVIN pin is used as the supply pin for the VREG regulator. TI recommends using a 2.2-μF, 50-V, X7R-type
    ceramic capacitor close to the AVIN pin. A 100-nF, 50-V, X7R-type ceramic capacitor is recommended close to
    the VIN_SAFE pin.

    The device has a linear regulator to supply the gate drives of each regulator. A 2.2-μF, 16-V, X7R-type ceramic
    capacitor is recommended on the VREG pin.

    Would it be possible for you to share the schematic?

     

    regards,

    Niko