• State Resolved
  • Locked Locked
  • Replies 12 replies
  • Subscribers 64 subscribers
  • Views 2577 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.

LMR23630 / 24V_input hot power line connect

ACGUY
Guru 11170 points
Other Parts Discussed in Thread: LMR23630

Hi all,

I have issue at LMR23630.
When 24V_hot power line connected to Vin, device is be damaged that shorted SW-port to GND.

Does this devise have limitation about rise time for Vin?

Regards,
Nagai

  • 0
    TI__Mastermind 19891 points
    Hello Yoshitaka-san,

    Is it possible that the input voltage is ringing up above the absolute maximum rating of the part?
    If you are using long power leads (inductance) and only ceramic input capacitance (without much ESR) and you hot plug the supply, the input voltage at the IC can overshoot. If the overshoot is above the absolute maximum rating of the part, then damage can occur.

    See if you can capture a scope shot of the input voltage across the IC VIN and PGND pins when the failure occurs.

    Also, please attach the schematic and layout for review.

    Cheers,
    Denislav
  • 0 in reply to Denislav Petkov
    Guru 11170 points
    Hello Denislav-san,

    Thank you for response.
    I work in distributor at Japan as an FAE.
    I will to request that provide the schematic and waveform to our customer.
    I drived out competitors (Rohm and LTC) from their circuit board that I got D-IN.
    Currently, customer work situation is final evaluetion.
    Please help us.

    Overshoot wasn't over the absolute-maximum-rating. That was 36V peak.
    At hot connect, when connected GND-line first that device doesn't have damage.
    If connected Hot-line first, Do you think the possibility of device corruption?

    Regards,
    Nagai
  • 0 in reply to ACGUY
    TI__Mastermind 19891 points
    Hello Yoshitaka-san,

    Is the input capacitance ceramic only?
    If they place an electrolytic capacitor with some ESR in parallel with the ceramic, do they still have the same problem at hotplug?

    Regards,
    Denislav
  • 0
    TI__Intellectual 2790 points

    Hi Yoshitaka san,

    We will try to duplicate this issue in lab. So pls let know more detailed test conditions ctm did.such as what's the load? And is there bulk electrolytic cap at the input side? 

  • 0 in reply to Luke Pan
    Guru 11170 points

    Hi Luke-san, Denislav-san,

    Thank you for your cooperations.

    Q : Input capacitor

    A : 4.7-uF x 2 ceramic caps .

    Q : Load

    A : Electric load, CC-mode, 2A.

    5-V reset circuit is connected already.

    I attached simple circuit diagram. Please see that.

    I look forward your comments.

    Regards,

    Nagai

    LMR23630_Hot-line-connect.pdf

  • 0 in reply to ACGUY
    TI__Intellectual 2790 points

    Yoshitaka san,

    I checked on bench, and see a large ringing on VIN , as high as 44V. This voltage may destory the device potentionally. The ringing amplitude depends on the speed the hot line is pluged, the faser, the larger ringing.

  • 0 in reply to Luke Pan
    Guru 11170 points

    Luke-san,

    Thank you for update.
    May I request additional experiment?
    Before connect the hot-power-line that please pull-up the Vout by 10-kohm to floating 5V.
    Customer talked that they seems to be damaged by connection order.

    [Vout_5V_pull-up -> 24V -> GND]
    Please experiment the above procedure connect.
    Please shorten the connection interval as much as possible.

    Kind regards,
    Nagai

  • 0 in reply to ACGUY
    TI__Intellectual 2790 points

    Yoshitaka san,

    Did ctm use FPWM version or PFM version devices? For FPWM version, we used to see the VIN is boosted to super high voltage as output is connected to 5V and input is floating.

  • 0 in reply to Luke Pan
    Guru 11170 points

    Luke-san,

    Thank you for your confirmation work. That is very important information to me. Our customer is using the PFM device.

    Regards,

    Nagai

  • 0 in reply to Luke Pan
    Guru 11170 points

    Hi Team,

    Customer adopted LMR23630 to new product design.
    Without your help, we couldn't have succeeded.
    Thank you for your many support and many cooperation.

    Many thanks,
    Nagai

  • 0 in reply to ACGUY
    TI__Intellectual 2790 points
    Yoshitaka san,
    When the Vout is weakly pulled to 5V, the input voltage will be charged to 5V through the body diode of the HS FET. Then the device is active as the VIN is over its UVLO threshold. You will see the FEN turn on/off actions at such condition. Then hot plug the device to DC supply, the large ringing voltage (as high as 44V) may forced on either HS or LS FET, which may burn out the device.

    If Vout is is not pulled high to 5V, then the device won't be active. There is a delay time taht Vout start to ramp up after VIN is ready (refer to the waveform I attached above at earlier time). So even it's hot plugged to the DC supply with large VIN ring voltage, the HS/LS won't swich during this delay time. So the ringing voltage is forced on HS and LS FET in series, so every single FET only suffer half the ringing voltage, so it won't burn out the device.

    I think that's why different power on sequence will result in different results.
  • 0 in reply to Luke Pan
    TI__Intellectual 2790 points

    The soultion is to set the system UVLO is a bit higher than 5V throgh the EN Pin with EN divider resistors.