• State
  • Replies 5 replies
  • Subscribers 63 subscribers
  • Views 100 views
  • Users 0 members are here
Support feedback
Options
Options
Related

TPS1H100-Q1: Damage Issue

Joyce Gao
Intellectual 2870 points
Part Number: TPS1H100-Q1

Tool/software:

Hi expert,

Customer using TPS1H100A-Q1 and now they met the failure issue. When power on the HS_PW+ (24V, VS input), some of the devices directly damaged and VS pin was shorted to GND pin, and some device also tested short between VS and IN. I am suspecting some GND connections and thermal design, are there any potential risks for such damage issue?

Here is the partial SCH. HS_PW+ and LS _W- are 24V input power lines. GND label is same as MCU side GND, that R393 is bypassed the GND network, is this risky when power on? 

Joyce

  • 0
    TI__Expert 7701 points

    Hi Joyce,

    So, TPAD_GND is IC GND before the ground network. TPAD_GND seems to be the same as GND - the MCU ground. LS_PW- is also ground, system ground after the ground network. Please let me know if my understanding is correct here. 

    So, the test is short to ground on the output, right? Why is the customer testing a short between the VS and IN pins? 

    This device is rated for a short to ground test at 12V not 24V. It is tested towards the AEC-Q100-012 standard. Please see the following for more information on the test conditions (specific resistance and inductance).  

    AEC-Q100-012 Short-Circuit Reliability Test Results for Smart Power Switches (Rev. A)

    What is the application? 

    Thanks,

    Rishika Patel 

  • 0 in reply to Rishika Patel
    TI__Intellectual 2870 points

    Hi Rishika,

    Rishika Patel said:
    TPAD_GND seems to be the same as GND - the MCU ground.

    ——Correct, through R393 (0ohm resistor). 

    Rishika Patel said:
    LS_PW- is also ground, system ground after the ground network.

    ——Correct, and LS_PW- is the primary very first ground of the system. MCU GND also comes from this line but through choke. As below. POWER+ equals to HS_PW+, and POWER- equals to LS_PW- (connection lines from same 24V source), after the filter there comes the GND for MCU.

    The datasheet shows the device GND pin through ground network to GND, and MCU directly connected to GND.

    But at this customer design, TPS1H100 GND connects directly to MCU GND without ground network, and TPS1H100 GND also goes through the ground network to the system primary very first GND (LS_PW-). Is there any potential effects?

    Rishika Patel said:
    the test is short to ground on the output, right?

    ——No. There's no specific test, just power the system 24V on, and TPS1H100 will have chance to get damaged. And the damage result is pin8 shorted to pin2 (VS shorted to GND), and pin3 shorted to pin8 (IN shorted to VS). These are the results, not the test condition.

    Rishika Patel said:
    What is the application? 

    ——Battery pack relays controllers.

    Joyce

  • 0 in reply to Joyce Gao
    TI__Expert 7701 points

    Hi Joyce,

    I am working on this, I will get back to you soon. Sorry!

    Thanks,

    Rishika Patel 

  • 0 in reply to Rishika Patel
    TI__Intellectual 2870 points

    Hi Rishika,

    Please help check internally if anyone familiar with this device, customer is facing line down...

    Joyce

  • 0 in reply to Joyce Gao
    TI__Expert 7701 points

    Hi Joyce,

    Yes, the IC ground and the system ground having significantly different potentials can cause damage to the device. Does the customer have any waveforms we can take a look at? What is on the load when the damage occurs? 

    Thanks,

    Rishika Patel