• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 2 answers
  • Subscribers 49 subscribers
  • Views 540 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

Original question:

TLV6700: Unused pin check

TLV6700: Schematic review

user5003739
Prodigy 60 points
Part Number: TLV6700
Other Parts Discussed in Thread: TPS2121, , TLV6703

Hi,

Please help to review schematic about TPS2121 + TLV6700 application.

VDCIN_ACC is a power source operating in 12V or 6~8.4V, our customer want to use TPS2121 + TLV6700 to get both OVP and UVP.

We assume 3 scenario (OVP, Normal, UVP) as attachment.

The OVP is setting at about 15.427V, UVP is at about 5.483V.

Please help to check is this structure can meet our application or not, thanks~

ET5X_TI_solution_for_UVP_20190424.pptx

  • 0
    TI__Mastermind 24820 points

    Sorry for the delay, we will provide feedback first thing tomorrow morning.

    Chuck

  • 0
    TI__Guru 54726 points

    Hello User,

    Looks good to us. You have taken into account the built-in hysteresis, which is good.

    One thing to watch is what may happen if there are momentary transients on the supply, such as a fast over or under voltage event (such as switching loads). Would a quick "glitch" cause a false shutdown? Would the starting of the switcher cause a supply dip (I assume it has soft-start enabled). Start-up behavior over various input voltages and turn-on/turn-off times should be verified on the prototype device.

  • 0 in reply to Paul Grohe
    Prodigy 60 points

    Hi Paul,

    Thanks for review, and other question :

    1. Due to we had verified on EVM board, but got a pulse on pin OV2. 

    Try to change R3548 from 49.9K to 10K, can reduce the time of pulse.

    Please refer to the measurement report as attachment, is there any risk about this change?

    2. We want to change the protect level to 6.2V. It's the value about our power pack low voltage level.

    But the protect level would have some tolerance, when about 6.2V pass to system, our PMIC will try to power on, and if the current not enough, it would be shut down.

    Maybe it will try to reboot infinite loop..

    Please refer to attachment last 2 pages, can see the oscillate will cause we cannot avoid the reboot.

    Is there any idea about this ? or is there any component has larger threshold range ?

     

    Thanks

    Chanah Wu

    UVP_OVP_0513a.pptx

  • 0 in reply to user5003739
    TI__Guru 54726 points
    Hello Chanah,

    Decreasing R3548 will slightly decrease the MOSFET turn-off time. Turn-off time will be dominated by the time constant of the pull-up and the MOSFET gate capacitance. Gate cap is about 1000pF.

    I believe the "glitch" you are seeing is the TLV6700 start-up delay. See note 2 in the TLV6700 datasheet:

    "During power on, VDD must exceed 1.8 V for 450 μs (max) before the output is in a correct state."

    The TLV6700 is a micropower device, so it is slow. It takes a while for the internal bias circuits to reach their operating points when power is first restored. The last two slides show about a 50-60us glitch, which is within the range for this. Before power-up, the output is High-Z.

    I'm not sure what is causing the "oscillation" - need to look into that more. Kinda hard to tell from the photos. But you may want to check with the Power group about the TPS2121. Are you pulling OV2 down too low at the lower voltage?
  • 0 in reply to Paul Grohe
    Prodigy 60 points

    TLV6703 threshold.xlsDear Paul,

    Step1. The protect level would have some tolerance, when about 6.2V pass to system, our PMIC will try to power on with 6.2V battery input.

    Step2. During the device power on, the low battery(6.2V) may drop to under 6.14V. When drop to under 6.14V, then the threshold voltage is under  396 mV. The PTS2121 will not pass power to system.

    Step3. When no loading, the battery will recover to 6.2V. Then the PTS2121 will pass power to system.

    So It will cycle(Step 1->2->3->1->2-3.....) to cause oscillation.

    Due to the TLV6703 threshold voltage range is very small. How do we avoid this issue?

    Thanks.

    By Small

  • 0 in reply to user6010589
    Prodigy 60 points

    Dear Paul,

    Do you have any suggestion?

    Thanks.

    Small

  • 0 in reply to user6010589
    TI__Guru 54726 points

    Hello Small,

    This is what I was warning about.... If your supply has a high impedance (a few ohms), the voltage will drop as current is drawn. Peak currents are drawn as the switcher starts-up and bypass caps are charged. This causes a momentary droop in the supply.

    If it drops to 6.14V and stays, then the threshold will need to be re-adjusted lower than 6.14V by adjusting the value of R3545. You can add external hysteresis on top of the built-in hysteresis by adding a feedback resistor from the output to the positive input (classic comparator hysteresis). The resistor value should be in the 330k to 1Meg range.

    You can try paralleling a large capacitor ~1uF in parallel with R3545 to create a time constant. It will take longer to initially start-up (or detect fast under-voltages!), but the start-up "glitches" should not cause false tripping. The feedback resistor mentioned above cannot be used if using this method as it would negate the hysteresis feedback.

    You can also add large capacitors across the input to provide a bulk-storage to supply the switcher start-up currents. These caps should be before any power switch so that they are continuously charged. This will lower the effective battery impedance and lessen the voltage drops due to the transients. Does your switcher have a soft-start feature? This is one of the purposes of soft-start.