• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 62 subscribers
  • Views 148 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.

TLV803: TLV803S - Not working as expected

Coratron
Prodigy 10 points
Part Number: TLV803
Other Parts Discussed in Thread: , TPS3760

Hi,

I hope you will be able to help me with this problem:

Description:

  • I am having issues with the TLV803S. I am using it to monitor the voltage of a Li-Ion battery and shut the output down if it falls below the threshold.

Problem:

  • The /RESET output pin is always low. I have measured the voltage between GND and VDD as 4V. This should make the /RESET pin go high due to the pullup resistor. It remains low.

Tests:

  • I have bridged Q5 just in case it was a GND problem. The switch SW1 is definitely open.
  • +BATT does output power from the battery when I remove the TLV803S from the circuit - so it is the culprit.
  • I have tried a TLV803R instead and it does work if I poke with my finger (some coupling? - all the solder joints are solid), but once the voltage falls below the threshold, even if it increases again the /RESET pin stays low (and never goes high again). It seems to have some sort of latching behaviour.
  • I have tried reducing R98 from 430k to 100k - it made no difference.
  • I have tried another TLV803S in case the first one was damaged - it made no difference.

I am growing a bit suspicious about the spec as well. The new TLV803E which is the new generation has an interesting first line in the datasheet:

Ensured? Isn't it ensured in the TLV803S?

Kind Regards

  • 0
    Prodigy 10 points

    More test:

    • Added a 100nF capacitor from /RESET to GND.
    • Four distinct phases are observed depending on the VDD voltage - see o.scope screenshots.
    • Channel 2 (purple is +BATT), Channel 1 (yellow) is nRESET
    • Channel 2 ref is the safe as channel 1 ref (marker overlap)

    VDD < 2.93V
    No pulse observed on nRESET


    2.93V < VDD < 3.04V
    Short Pulse observed (periodic)


    3.04V < VDD < 3.56V
    Long pulse observed, short activation of +BATT (periodic)


    VDD > 3.56V
    Full activation of +BATT (single event)


  • 0 in reply to Coratron
    TI__Expert 4000 points

    Hi Coratron,

    Thanks for your question and providing more test results.  After adding capacitor it seems the /reset output pulled up gradually. I'll be looking into why you experienced short  and long periodic pulses when VDD between 2.93V - 3.04V, and 3.04V - 3.56V.

    It seems to me TLV803S working correctly, but the reset is not being pulled up. Can you please check if that the case after removing Q27? If that the case I think you may need to change R49 and R91 values. 

     

    Best Regards,

    Sila Atalar 

  • 0 in reply to Sila Atalar
    Prodigy 10 points

    Thank you for your reply Sila.

    Removing Q29 produces the expected behaviour on /RESET - so it's the interaction that is causing the issue.

    I provisioned R91 to be able to ramp up the load so I will try that next with much higher values. I am still trying to understand why the TLV803 is unable to remain stable since VDD does not seem to be falling below 1.1V and driving the gate of Q29 should not pose a problem. I will keep doing some tests and report back but if you have any other ideas let me know in the meantime.

    Regards

  • 0 in reply to Sila Atalar
    Prodigy 10 points

    I have found the issue and this chip may not be suitable for the intended purpose. The voltage drop that can be seen in channel 2 (purple) corresponds to a drop on VDD due to ON transient of the load on +BATT. This drop on VDD goes below 3V (TLV803S triggers at 3V) and this engages the /RESET output, which then disables +BATT and keeps the system in this periodic resest. Adding a 330uF tantalum capacitor on VDD has not made a difference.

    Perhaps an RC network from /RESET to the Q29 gate could do the trick.

    If you know a TI chip that can do what I am after please let me know. I want a battery cutoff at around 3V with some hysteresis to avoid the current situation. The observed voltage drop is about 200us in duration.

    Kind Regards

  • 0 in reply to Coratron
    TI__Expert 4000 points

    Hi Coratron, 

    Yes, I can help you with recommending another TI product may fit more with your design, which is TPS3760. 


    You can set the sense delay time so that reset won't be able affecting from the voltage drop happen in sense delay time period, so that you can set it up more than 200us. And TPS3760 has adjustable threshold option so you can set it up for 3V.

    Please let me know if further help needed. 

    Best Regars,

    Sila Atalar