• State TI Thinks Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 1 answer
  • Subscribers 62 subscribers
  • Views 598 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.

TPS61196 LED Driver Stuck in Periodic Fault/Retry Loop

Rob Yost
Prodigy 90 points
Other Parts Discussed in Thread: TPS61196

I am driving 1.08A total out using the TPS61196 with 4 LED string drivers driving one LED string.  I have the drivers in parallel to get the total current needed for the string.  The design seems to work fine except for when I increase the PWM duty cycle on the four string drivers (all connected to the same PWM) from <50% to full-on (100%) suddenly.  As a result, the LED driver appears to begin indicating a fault every 184ms, followed by a retry to start the boost switcher up again, and then a repeat of the same scenario every 184ms.  The fault/retry condition seems to point to a UVLO fault because it's not getting hot, and the ISET pin is not shorted to GND.  The only fault/retry condition left is the UVLO.  However, I have monitored Vin during this scenario and verified it never falls below 10.6V.  The UVLO fault condition is 7VF max Vin falling, which appears to not be occurring.  Thus, I cannot figure out why the chip is doing this. 


Here are the relevant design specs:


Vin = 12V

Vout ~ 36V

Iout ~ 1.08A

Iset = 270mA, FBP = 4.8V, IFBV = 0.7V, Vovp=50V

  • 0
    TI__Intellectual 1780 points
    Hi Rob,

    TPS61196 supports Six Current Sinks, 200-mA Continuous Output, 400-mA Pulse Output for Each String. It means it can support ~< 200mA per channel when duty cycle is lower than ~50%.

    In your application, you are setting 270mA per channel, which is beyond the datasheet spec.

    Instead, I'd recommend using 6 channels, with which per string current would be 180mA (within the spec) and you still get your desired total current of 1.08A.

    Hope this helps. Else feel free to ask followup questions.

    Best,
    -Harini
  • 0 in reply to Harini Sankabattula
    Prodigy 90 points

    Hi Harini,


    You make a good point about the maximum continuous output current.  I tried the driver using 6 parallel drivers instead of 4, and unfortunately it did not rectify the problem.  In actuality, it's a problem related to jumping from a duty cycle less than 50% to full-on (100%).  Jumping from 50% and higher, to 100% does not cause the problem.

    After doing some serious scope probing and investigation, I observed that the output voltage from the boost, does a kind of runaway as soon as the <50% duty cycle is changed to 100%.  I had the OVP set to something fairly high (~80V).  So the output voltage would get up to over 50V and some kind of protection was kicking on, which would then cycle on/off periodically.  I changed the OVP setting to something closer to the maximum output voltage seen (to about 41V).  I then observed that when the output voltage takes off in an uncontrolled fashion the chip clamps the output voltage at the OVP setting for almost 500ms.  Most of the time it recovers before the chip latches off, but there are occasionally cases where the chip latches off and has to have the power cycled before it will work again.  I also observed that the constant current drivers for the strings seem to shut off when this duty cycle switchover occurs.  This may be what's causing the output voltage to take off and then an attempt to recover.  I would not expect the string drivers to shut off upon a change in duty cycle like this.


    This still occurs when using 6 drivers in parallel with the continuous output current set to 180mA. 


    So my next question is: why is the output voltage taking off in an uncontrolled manner when we switch from a <50% duty cycle to 100%?  Why are the string drivers shutting off when a sudden change in duty cycle is imposed?  Is this a normal reaction for the chip?

  • 0 in reply to Rob Yost
    Prodigy 90 points
    Another question: I observed that the REF pin (or reference voltage) rests at 1.4V when the duty cycle is set to >=50%, and 3.1V when the duty cycle is set to <50%. Why is this, and could this be an indication of the issue?
  • 0 in reply to Rob Yost
    TI__Intellectual 1780 points
    Hey Rob,

    Are you using TI EVM? Can you send me your schematic configuration?

    I can get an EVM tomororw and see if I can reproduce these observations on the bench.

    Thank you,
    -Harini
  • 0 in reply to Harini Sankabattula
    Prodigy 90 points
    Harini,

    I can send you a Schematic. We are not using an EVM, we have our own board we designed. How do I send you the Schematic?
  • 0 in reply to Rob Yost
    TI__Intellectual 1780 points
    Hi Rob,

    You can email me to harini.sankabattula@ti.com

    Thank you,
    -Harini