• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Subscribers 64 subscribers
  • Views 464 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.

TPS1641EVM: Startup with load applied - unexpected behavior

Jim Brinkhurst1
Mastermind 44680 points
Part Number: TPS1641EVM
Other Parts Discussed in Thread: TPS1641

Hi experts

We are working with TPS1641 evaluation board and getting unexpected results. Can someone review the information below and provide debug guidance?

I am using the TPS14612 side with all jumpers set at default.

There are other issues I’m currently analyzing, but one concern is startup into a load.

Below is a startup into a 1A (24ohm) load. As you can see it is hiccupping. Voltage on the IDLY pin seem to indicate the fault timer is started. A 1A load is well below the Ilimit currently set to 1.8A. Experimentally, I can start up at loads <0.5A. Can you explain this?

Thanks for your help!

Jim B

  • 0
    TI__Guru* 79181 points

    Hi Jim,

    I think device is hitting thermal shutdown due to high power dissipation across internal FET . During startup of eFuse due to high VDS and IDS(load current ) stress thus causing high power dissipation will cause FET junction temp to rise and when temp reaches near a threshold device will turn off. 

    That IDLY pulse is due to retry timer implementation. After thermal shutdown device will retry after a timer. 

    This is expected behavior of device. 

    Regards

    Kunal Goel

  • 0 in reply to Kunal Goel
    TI__Mastermind 44680 points

    Hi Kunal

    What is the recommended change if we want to start up into a 24 ohm (1A at 24V input voltage) load? 

    Will changing the slew rate setting help?

    Or do we need to select a different TI product?

    Thanks!

    Jim B

  • 0 in reply to Jim Brinkhurst1
    TI__Guru* 79181 points

    Hi Jim,

    Are they using CC load? I will recommend using CR load because actual loads are more like CR. Increasing slew rate by decreasing dvdt cap can help. 

    What is the actual application load? I dont think it will be a load that will draw 1A at 0V. Every load will have some turn on threshold.

    Regards

    Kunal Goel

  • 0 in reply to Kunal Goel
    TI__Mastermind 44680 points

    Hi Kunal

    As shown in the plot above, the test load is resistive, 24 ohms. In the image the current only ramps up to around 200-250mA (voltage around 5.25V)  before it starts to fall.

    I will get more details on the actual application load.

    Jim B

  • 0 in reply to Kunal Goel
    Prodigy 10 points

    Hello, Kunal - thanks for helping out.  As the scope shot in the original message indicated, it was at 24Vin, 24ohm resistive load. 

    Regardless, what I am seeing is that an aborted startup can occur even at no load if the input voltage is high enough.  Attached is a waveform I just captured at 32V (unsuccessful) and 31V (successful), both no load.  This does support your idea that the part is protecting itself under the high Vds x Id.  As well, if I reduced the Vin to 18V, startup is successful with the 24 ohms where it failed previously at 24Vin.

    So, it goes back to Jim's question what adjustments can be made to keep the device well within its SOA?  That is, what limitations should be imposed on Cout, cap on the Cdvdt pin, etc?

    Failed at 32V.

    Passed at 31V

  • +1 in reply to first_name last_name
    TI__Guru* 79181 points

    Hi ,

    Design calculator on product page can be used to predict whether startup will be successful or not based on VIN,COUT,RLOAD,CDVDT cap. Please let me know if any concern. 

    Regards

    Kunal Goel