• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Subscribers 64 subscribers
  • Views 551 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.

TPS65270: Relation for inrush current and Loop compensation mismatch

Satoshi
Guru 19645 points
Part Number: TPS65270

Datasheet P16 is described "fzff is smaller than soft start equivalent frequency (1 / Tss)".

When fzff is much higher than 1/Tss, is these condition possibility to occur high inrush current on startup?

I think that high fzff occur below Vout "jump", and occur unexpected high current. 

【Background】

On the evaluating based on EVM and changed Vout1:3.3V/2A (tss: 624μs), Vout2: 5A/3A (tss: 1890μs), 500kHz, and 10μH. 

The other difference is Cff: 22pF (smaller than default value).

When Cout set 100μF, both Vout1 and Vout2 occur high current startup and overcurrent protection. (Cout: 22μF is normal operation)  

I hope to confirm large Cff value (ex. 2.2nF) have good effect on suppress large current or not. 

  

Best regards,

Satoshi

  • 0
    TI__Genius 15135 points

    Hi, Satoshi 

    When Cout is increased, we should increase Css/Tss to reduce inrush current. 

    Large Cff value also can help to suppress inrush current too, but i think Css will be more effective. 

  • 0 in reply to Zhao Ma
    Guru 19645 points

    Hi Zhao

    Thank you for reply,

    I think that inrush current is calculated below, but Cout: 100μF setting have enough margin from current limit.

    I = Cout × Vout / Tss

    From the above and below E2E thread (relation for Css and Cff), I feel to need Cff value for decrease the other reason of startup inrush current.

    https://e2e.ti.com/support/power-management-group/power-management/f/power-management-forum/437224/tps65270-loop-compensation-external-component-values

    Additional question, please let me know rise time for Vout "jump" (during soft start function ON).

    I will calculate inrush current on Vout "jump" timing.

    Best regards,

    Satoshi

  • +1 in reply to Satoshi
    TI__Genius 15135 points

    Hi, Satoshi 

    I think Vout jump is caused by fast SS rising in the beginning of startup, IC switching to make FB catch up with SS voltage and it has a little overshoot, so we see the Vout Jump. 

    Enlarging Css can make SS rise slowly in the beginning of startup. 

    Cff also can help to reduce the Vout jump, Cff can feedforward the Vout change to FB pin, then make IC switching pulses slowly, which will reduce the Vout Jump.  

  • 0 in reply to Zhao Ma
    Guru 19645 points

    Hi Zhao

    Thank you for reply, I understood.

    When Cff changed to 2.2nF on Vout1:3.3V (tss: 1600μs), Vout2: 5A (tss:5920μs) , startup current was not reach overcurrent and normal operation.

        

    And, keep Cff: 2.2nF condition with changed fast soft start time: Vout1:3.3V (tss: 624μs), Vout2: 5A (tss: 1890μs), overcurrent was occur again.

    ※Both output current are 500mA.

          

    Above fast tss are customer's request spec, I need to adjust Cff value for decrease inrush current.

    Maybe last question, please let me know about two points below for Cff setting.

    ・Calculation for inrush current is not only I = Cout × Vout / Tss, please let me know about the equation for total inrush current (on during Vout jump and after Vout jump timing).

    Especially I want to know the relation for Cff decrease Vout and inrush current jump, and soft start is affect to reduce inrush on Vout jump timing or not.

    ・Is there maximum value of Cff?

    Best regards,

    Satoshi

  • 0 in reply to Satoshi
    TI__Expert 7520 points

    Hi Satoshi

    Zhao is looking into this and will feedback to you soon.

    BR

    Ruby

  • +1 in reply to Satoshi
    TI__Genius 15135 points

    Hi, Satoshi 

    1. the inrush/input current I=Cout×Vout/Tss + Iload. We need to make I smaller than OC limit, and leave some margin. 

    2. There is no limitation for Cff in theory, but when Cff is too big, IC will be easy to be disturbed by noise. 

    Suggest Cff<=10nF. 

    When enlarging Cff, suggest enlarging Css at the same time.