• State
  • Locked Locked
  • Replies 4 replies
  • Subscribers 63 subscribers
  • Views 360 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.

LM74810EVM: Hot-Plug Inrush Current

Peter Davis1
Intellectual 420 points
Part Number: LM74810EVM

Tool/software:

Hello, we are using the LM74810MDRRR as a hot-plug/ideal diode controller. We are running into the following issue:


1. When EN/UVLO is pulled low (disabled), we are seeing significant current when we hot-plug a 50V source.
2. We have series resistors on all of our capacitors that touch the input rail, but we see the IC get damaged upon plug-in. It appears there is a current path through either the C/A pins to the charge pump cap hooked up to C67. See the image attached

When we look at this on the Evaluation board, we also see much higher current when hot-plugging a device. On the order of 3-4A after change the series resistor to the caps to a 100Ohm.
Is there internal capacitance to the IC that we need to consider?
Is there a way to limit this inrush?
Can we add series resistor to the A/C traces without affecting the ideal diode operation, such as adding 100Ohm?

Thank you,

Peter Davis

  • 0
    Intellectual 420 points

    I will also add, that I've confirmed no voltage on the output side (V_SYSTEM), thus it is inrush solely due to the IC and input side.

  • 0 in reply to Peter Davis1
    TI__Mastermind 45610 points

    Hi Peter,

    There isn't really any current path internal to IC that can do so. As you mentioned the IC is damaged, after that anything can happen.

    While hot-plugging 50V, are you violating the abs-max by any chance? Is there any spike crossing 70V? 

    It is not recommended to add series resistor in A and C path.

    You can iteratively remove the capacitors connected to input rail and see which capacitor is causing the inrush.

    Regards,

    Shiven Dhir

  • 0 in reply to Shiven Dhir
    Intellectual 420 points

    Hi Shiven,

    We aren't violating the absolute max voltage.
    Understood on the series resistor not being recommended
    We've isolated all input capacitors and still see significant inrush current. I have also simulated this using TI's spice model and am seeing significant inrush, though in a slightly different outcome.

    When Vinput to the LM74810 is < 25V I see expected inrush of just the capacitors, however, when I increase voltage beyond 25V and particularly at 50V I start to see voltage show up on the output of my Mosfet which causes inrush on the order of 60A. Can you confirm if you see this behavior in your simulations? Is there an issue with the models, or is this real?

    In the Simulation plot:
    Blue = Input Voltage
    Red = Input Current
    Green = Vgs of the blocking FET U2 | Showingthe FET is staying off but there is ~20V on the source and gate of the FET



  • 0 in reply to Peter Davis1
    TI__Mastermind 45610 points

    Hi Peter,

    I simulated same design as yours. Please see if there is any difference.

    I don't see my VOUT rising. With EN low, VOUT can't go high. There is no significant leakage that can charge VOUT to such high level in real use case as well.

    Possibly your chosen FETs have some leakage modelled. Can you change the FETs and try again?

    Regards,

    Shiven Dhir