• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 63 subscribers
  • Views 401 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.

TPS2492: Simulation model gate sinking current in hard short-circuit scenario

Konstantin E.
Prodigy 30 points
Part Number: TPS2492
Other Parts Discussed in Thread: LM5069

Good evening Rakesh, Good evening fellow forum users,

We're considering the use of the TPS2492 for our design and are simulating it under various test conditions with the PSPice model provided on your website (in PSpice for TI).

One of them is a hard output short-circuit (with the parasitics we know from previous tests). During these simulations I'm seeing a behavior of the constant power engine that I can't explain myself from the datasheet:

During the output short-circuit the voltage V(VCC-Sense) exceeds 300mV (so we're far above the overcurrent setpoint) with a V(PROG) of 0.4V (so the lower end of what you recommend). Despite these conditions that indicate a heavy output short-circuit, the  TPS2492 Gate sourcing current is less than 20 mA (increasing but very slowly and not fast enough), in contrast to the 75-250 mA sinking capability given in the datasheet. This slow turn-off puts too much thermal stress on our Mosfet and if the behavior would be like this in reality, would not allow us to use the 2492.

I have seen in some other posts that for time-critical and fast-switching overcurrent protection applications you usually recommend the LM5069 (which in fact works in my simulation) but unfortunately it is not an option for us right now since the stock situation is as it is ;)

If you could advise on the reason for this behavior i'd be super thankful. We could test the circuit on a real test-board under the conditions as in the simulation however I'd only invest the time if there are any indicators for different behavior as in the simulation.

I've attached the schematic of the simulation and the relevant curves (Mosfet Drain Current, TPS2492 Gate sinking current & TPS2492 VCC-Sense voltage). I tested if it could be a violation of UVEN threshold voltage or VCC voltage, but with both of them right in spec for the 75-250mA gate sinking case, I'm seeing the < 20 mA.  Please do let me know if I can provide any additional Info.

Schematic_Probes_1.pdfSchematic_Probes_2.pdf

Cheers in advance for the help!
Konstantin

  • 0
    TI__Guru** 105361 points

    Hi Konstantin,

    Welcome to E2E!

    Short-circuit response of TPS249x and LM5069 is almost same. It looks, R18 in your schematic is not allowing GATE to discharge. Please use npn based soft-start circuit as shown below to facilitate local fast discharge path for the C4.

    Please share once your schematic is ready for review and use design calculator to check the SOA margin of the design. https://www.ti.com/lit/zip/slvc033 

    Best Regards,

    Rakesh

  • 0 in reply to Rakesh
    Prodigy 30 points

    Hi Rakesh,

    Thanks for the warm welcome and the fast reply.

    That's good news to me that response should be similar to the LM5069.

    I checked your suggestion, removed R18 and C4 and resimulated. The result is almost the same as before (In the previous simulation I was checking the current flowing out of R18 before as I suspected this could be an issue). The attachment shows the updated results.

    Since we're checking the current at the Gate Pin of the TPS2492 directly, independently on the dv/dt control circuit we should see a higher gate sinking current.  If the current flowing out of R18 during turn-off is significant, that would result in lower current flowing out of the Mosfet Gate but this is not the case in the circuit here as R18 limits the currents to an almost negligible level.

    Do you have any other explanations of what is happening here and do you trust the simulation to behave as the chip would in reality in the short circuit scenario?


    Schematic_Update_no_dvdt_control.pdf

    Thank you for your help, appreciate it!

    Konstantin

  • 0 in reply to Konstantin E.
    TI__Guru** 105361 points

    Hi Konstantin,

    I have verified the model at my end. It looks the fast pull-down strength is not modelled correctly. We will take action on it.

    Please proceed with either LM5069 or TPS249x and share us the schematic, design calculator for review.

    Best Regards,

    Rakesh

  • 0 in reply to Rakesh
    Prodigy 30 points

    Hi Rakesh,

    That's great to hear. In that case we will follow up with hardware tests and I can get back to you with results if you're interested.
    In the meanwhile could you let me know please once there is an update of the simulation model?

    Thank you for offering the review I'll definitely get back to that offer.

    Best regards,
    Konstantin

  • 0 in reply to Konstantin E.
    TI__Guru** 105361 points

    Hi Konstantin,

    Unfortunately, the model revision for this device is in low priority now due to other development work. I will let you know.

    Best Regards,

    Rakesh