• State Resolved
  • Locked Locked
  • Replies 9 replies
  • Answers 1 answer
  • Subscribers 62 subscribers
  • Views 509 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.

TPS68470: WLED current issue

Henry Wang3
Genius 3455 points
Part Number: TPS68470


Hi Sir,

We used the WLED is made to flash by using an external STROBE as a trigger.
Among them, there is a PMIC in which current is flowing through the WLED without synchronizing with STROBE.
It seems that the internal LED driver is out of order.
Please tell us the possible causes.
We attached the schematic, WLED current waveform and STROBE waveform as below.
The schematic of OK module and NG module are the same.

Schematics: 2045.TPS68470_schematic.pdf

NG Module waveform:

OK Module waveform:

  • 0
    TI__Genius 10490 points

    Hi,

    Given the 250 mA WLED operating current shown above, I don't see any issues with the WLED boost schematic or components.

    1. Can you please share the I2C commands used to configure the WLED boost and STROBE operation?
    2. If the PMICs are swapped on the OK and NG modules, does the behavior follow the PMIC?

    Thanks,

    Gerard

  • 0 in reply to Henry Wang3
    TI__Genius 10490 points

    Hi,

    Thank you for sharing. I have reviewed the remainder of the schematic and have the following concerns. The current components populated are outside of the recommended operating conditions. Please resolve these in the design and share if the customer still runs into this issue going forward.

    • ANA_OUT output capacitor (C23) should be 1 uF

    • CORE buck output capacitor should be 4.7 uF

    Thanks,

    Gerard

  • 0 in reply to Gerard Copeland
    Genius 3455 points

    Hi Sir,

    The customer has followed your advice to modify these capacitors' values. The NG module has still not be fixed and the LED is always on.

  • 0 in reply to Henry Wang3
    TI__Genius 10490 points

    Hi,

    Since the components were outside of the recommended operating conditions, the device operation is not guaranteed. It's possible the incorrect component values caused an irreversible issue with the device. The customer should implement these changes to their design going forward and let TI know if this issue is seen again in future, corrected builds.

    Thanks,

    Gerard

  • 0 in reply to Gerard Copeland
    Genius 3455 points

    Hi Sir,

    Do you mean the output capacitor close to the TPS68470 can not be outside the recommended value? or total capacitors can not be outside the recommended value in the same power path(ex: ANA_OUT)?

  • +1 in reply to Henry Wang3
    TI__Genius 10490 points

    Hi,

    The output capacitor placed close to the ANA_OUT pin should not be outside the recommended value.

    Thanks,

    Gerard

  • 0 in reply to Gerard Copeland
    Genius 3455 points

    Hi Sir,

    My customer confused that only output capacitor placed close to the power out pin need to meet the recommended value, the total output capacitors value in the same power path it is not to do, could you explain detail with us?

    The I2C commands should not need to make any modification, right?

    7774.IQCapture_Launch_I2Clog_PMIC&OV5675.xlsx

  • 0 in reply to Henry Wang3
    TI__Genius 10490 points

    Hi,

    I assume the other capacitors in the power path are decoupling capacitors for other components. These are usually placed farther from the ANA_OUT pin so there is some trace/via/connector impedance separating these decoupling capacitors from the ANA_OUT output capacitor. This impedance separation lowers the impact on the LDO control loop. Capacitors placed immediately at the ANA_OUT pin have the most impact on the LDO control loop, so this should be within the datasheet recommendation. 

    Regarding the I2C commands, could you please explain the format of the "Data" column?

    For write transactions, I would expect this column to always have 2 bytes - register address and write data. However some commands, like row 9, only have one byte. Do you know the reason?

    # Level Index m:s.ms.us Dur Len Err S/P Addr Record Data
    0 0 0:00.000.000 Capture started
    0 1 0:03.547.051 50.000 us 1 B SP 4C Write Transaction 0D

    Thanks,

    Gerard