• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 1 answer
  • Subscribers 66 subscribers
  • Views 1476 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.

LM26484: LM26484 - Problems with 3.3v supply

Shlomi Yehezkia
Mastermind 6150 points
Part Number: LM26484

Hi Team,

LM26484.docxPlease help with the below problem.

TI support is required as I in pre-production phase.

 

I have a problem with 3.3V supply on my boards:

(1)    Burnt on 2 boards

(2)    On working board we see ~0.5V ripple (182820.jpg file)

 

Observations:

(1)    When reducing the input voltage from 5V to ~3.8V the ripple is significantly reduced (182804.jpg file)

(2)    Other 1.8V, 1.35V and 1.0V voltages generated by the same DC-DC works ok.

(3)    The output current is few hundreds mA.

(4)    This is revision of board that worked ok at similar conditions (no changes in power supply section).

 

Attached relevant scheme section & relevant layout section.

Thanks,

Shlomi

 

  • 0
    TI__Intellectual 810 points
    Hi Shlomi,

    The behavior is expected, in the case where the output voltage ripple is significantly reduced due to lowering the input voltage from 5V to 3.8V. This is in part due to the fact that you are reducing the inductor ripple current. The ESR of the output capacitors paired with the inductor ripple current will add to the output voltage ripple. Can you add a small 0.1uF low ESR capacitor in parrallel to C222 and provide me with scopeshots? Your layout looks fine, your output and input capacitors are tightly coupled to the IC.

    Regards,
    Raymundo Hernandez-Toscano
  • 0 in reply to Raymundo Hernandez-Toscano
    Mastermind 6150 points

    Hi, 

    The behaviour is very strange as the ripple is not dropped gradually with reduction of the input voltage.

    It drops drastically from ~0.5V to ~0.1V while Vin passes 3.9-3.8V.

     

    Tried your proposal, it didn't help.

    The ripple raised to ~0.6V.

     

    Need TI support ASAP.

     

    Thanks,

    Shlomi

     

  • 0 in reply to Shlomi Yehezkia
    Prodigy 120 points
    I'm working with Shlomi on this problem.
    Adding more information:
    (1) Replaced all the peripheral components of the 3.3V buck, but it didn't help
    (2) We observe ~100mV ripple @ 5KHz on the 1.8V, generated by the BUCK2 of LM26484SQ (U14).
    We have additional LM26484SQ on this board, supplying 1.35V and 1.0V. We observe ~100mV ripple @ 5KHz on those supplies as well.
    According to the output voltage ripple plots in the datasheet we expected ripple of ~10mV(ptp).
  • 0 in reply to Shlomi Yehezkia
    TI__Intellectual 810 points
    Shlomi,

    This is odd, I would not expect a drastic change in the ripple voltage due to a minor 100mV adjustment to the input voltage. Furthermore, how are you taking your measurements? What is the output load current for each rail? Can you provide me scopeshots of the SW node, input voltage, and output voltage for each rail? Although, you mentioned the load current was a few hundred milliamps, I want to make sure the device is actually operating in PWM mode. Also, what does your ground plane look like? If there is any stray inductance due to your return path it will reflect on your output. Can you provide me with information regarding the capacitors or did you use the recommended?

    Regards,
    Raymundo Hernandez
  • 0 in reply to Raymundo Hernandez-Toscano
    TI__Intellectual 810 points
    Igal,

    Thanks, for the additional information.

    Regards,
    Raymundo Hernandez
  • 0 in reply to Raymundo Hernandez-Toscano
    Prodigy 120 points

    Hi Raymundo,

    I've uploaded tests and scheme sheet of the both 2 PMUs we have on the board (U14, U15).

    POWER_EN1 enables 1.0V output, POWER_EN2 enables 1.8V output and POWER_EN3 enables 1.35V and 3.3V outputs.

    I've done VIN and VOUT tests while sequentially turning on the POWER_EN signals (described in file names).

    As you can see:

    (1) It seems that there is some problem with 3.3V buck

    (2) The ripple on the VIN rises as more DC-DCs are enabled.

    I can't share the PCB layout here, but will be glad to make a TeamViewer debug session with you.

    We have assembled 70 boards and must make them functional ASAP.

    There is an additional batch of 200 boards waiting for assembly.

    Thanks,

    Igal

    DC-DC.zip

  • 0 in reply to Igal Kronshtein
    TI__Intellectual 810 points
    Hi Igal,

    I will follow up with you via email.

    Regards,
    Raymundo Hernandez