• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 2 answers
  • Subscribers 63 subscribers
  • Views 396 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.

TPS23754: TPS23754 Vc large capacitor

Roy Chou1
Genius 16120 points
Part Number: TPS23754

Hi All,

As I know, Vc cap is charged by TPS23754 internal LDO. When it charge to 15V, internal LDO will turn off, the Vc cap stored power start up to gate signal and switch the MOSFET. When the MOSFET is switching, the transformer will generate voltage to charge the Vc cap.But when Vc cap is too small, the stored power from capacitor couldn't drive the MOSFET for three duty cycle and couldn't generate voltage to charge the Vc capacitor and Vc voltage will run to UVLO. Like image below.

 When I enlarge the capacitor value, then issue could be fixed. But original capacitor value I used is 22uF. The stored power will "Q=CV=22uF*15*0.5=165uC"

The MOSFET's Qg I used is 9nC. Normally the stored power could drive the MOSFET completely, but the result is wrong. Is IC has another function will use the power the Vc cap stored?

Roy

  • 0
    TI__Genius 16120 points

    Vgate means MOSFET "Vds"

    Vsec means 12V that transformer generate to charge Vc cap

  • 0 in reply to Roy Chou1
    TI__Mastermind 21320 points

    Hello Roy,

    Thank you for your question. The VC pin is also internally connected to Vb through a regulator. Increasing the VC cap is a typical design debug step, so it is not surprising. I would also suggest including a 1uF cap in parallel to bypass the gate drive if it is not already present. 

    Regards,

    Michael Pahl

  • 0 in reply to Michael Pahl
    TI__Genius 16120 points

    Hi Michael,

    Thank you for your reply, but I also have some question about the power that IC need initially drive the MOSFET and VB. I arrange some measurement below. Thank you for your help.

     Roy

  • 0 in reply to Roy Chou1
    TI__Mastermind 21320 points

    Hello,

    I am not sure where the 10mS time limit comes from. The time highlighted in the picture above the 10mS arrow is indicating when the cap is discharged, not charging. 

    The datasheet specifies a longer bootstrap time: 

    This would give rise to a larger VC cap needed. 

    Regards,

    Michael Pahl

  • 0 in reply to Michael Pahl
    TI__Genius 16120 points

    Hi Michael,

    This specification is Vc from 0V-15V. I also measure the customer board VC waveform below. When I use 22uF. The bootstrap start-up time is only 39ms and it doesn't meet the specification and it's non-linear. Because VC is charged by constant current and why it charge quickly to 15V and non-linear?

    And I also add the Vc cap from 22uF to 44uF to charge more energy. and its start up time is 78ms. VC could normally work. do you have any idea about why  Vc waveform is non-linear because I also measure the EVM420 board VC waveform and shows linear

    EVM board meausement, 22uF;81.4ms charge times

    Thank you for your help.

  • 0 in reply to Roy Chou1
    TI__Genius 16120 points

    Hi Michael,

    I find the issue. The VC capacitor we use is MLCC material, the capacitor value might be changed when voltage get high, so the charge curve is non-linear and let stored energy get lower and I change the cap from MLCC to EVM board'cap. It could normally work and meet the specification. Thank you for your help.

    Regards,

    Roy