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: TPS61196 not start correctly with 12V

Part Number: TPS61196

Dear Ti Team,

Hi,
I have question about UVLO and problem with startup.
I build device from proposed schematic, everything works perfectly, start nice at around 21V.
But if I would like to go down with supply voltage to 12V I have problems.
When I have resistors
r1 r2 r5
100k 24,9k 100k
still everything works correctly (I increment by 1 voltage thus start  at around 13V).
when I change resistors to
r1 r2 r5
100k 24k,9 82k
nothing happens, even when I rise voltage up to 21V (enable checked, 33k/10k divider enough for 12V, Voltage on UVLO also),
Adding additional capacitor (1uF) refine a little bit more and sometimes TPS start, but sometimes is not acceptable.

Should be used other set of resistors (eg. 10k 2.4k 8.2k ?) or ???

Thank you for help.

  • Hi Bartosz,

        Please use our equation 2 in the datasheet to choose the right resistor for the UVLO configuration. What the resistor you have chosen is not right.

    BR

    Sean

  • I read, this. What  I paste was the last test.

    Ok new one, calculate from equation for

    Vstart Vstop
    10 8,96
    266,666666666666 [k] R1+R5
    37,3655607494394 R2

    So I choose for R5 200k, 33kx2 for R1 and 37.4k for R2

    Starts, but also sometimes (if I "power" twice etc.).

    What I observed when I start them work perfectly if I not discharge (leave for longer time, disconnect from other devices - power supply). Eg. I could block output on power supllay switch of function generator, works till I unplug from PS and Generator.

    It starts quite good (better but not aceptable) if start voltage is set to 14V or higger

    If I exchange R1 to 182k like in proposed schematic, start always from around 14V.

    What should I check ?

    Thank you

    Bartosz Reichel

  • Hi Bartosz,

        The calculate method seems ok, I guess the reason is that it may trigger the current limit with the low input voltage. I think you can check the waveform on "ISNS" pin whether it triggers current limit under low supply voltage.

    BR

    Sean

  • On ISNS I have puls around 200mV then start oscialtion (in 14V case) in 12V case have pulse also around 200mV and nothing,

    but when I start with 14V and goes down to 12V everything looks similar like for 14V.

  • Hi Bartosz,

         There is two suggestion for the test:

        1. Decrease the PWM duty cycle for the test to achieve a light load to check whether it starts normally;

        2. Decrease the LED counts for the test.

        The suggestion that I proposed is to create a light load condition to check whether it triggers something wrong.

    BR

    Sean

         

  • Thank you,

    PWN not help (start from 10%) but decresing count helps (start from half, but works with more, now only test chain to prepare / understoot for final one ). Which parameter should be trim in this case (R7 ?). ISET have around 150k now (and fit to this LED).

    Best Regards

    Bartosz

  • Hi Bartosz,

         What's the LED configuration for your test set up? I'm considering maybe it has triggered maximum duty cycle limitation for low Vin and high Vout.

    Your ISET is too high, and you have set the current to about 32mA for each channel? You 'd better change the output current of each channel above 50mA. Have triy with ISET=98k.

    BR

    Sean

  • OK, I made next test, I back to proposed in DS values (200k/182k/24.9k) for which my test setup work. Exchange resistor to 74.5k to have 65mA and plug one of the finall led string which works under 65mA and 47V,

    nothing happen (old one also with this resistor still works). PWM duty 50%, 5V, 222Hz.

  • Hi Bartosz,

        This is really strange and I don't have much more comments now and please do more test and hope you can find that something that leads to this issue. Anny update please let me know.

    BR

    Sean

  • Thank you,

    next test.

    I build string array with +/- same voltage. Try to start => nothing.

    But when I take half, TPS start, then I take next and next and finally whole string works. Then change to final leds and also strart. But if discharge (eg touch output by finger), nope.

    Where could be a probem ? I try with differ inductor (~ same value, but I start from cheap one then use let say better). Power supply (standard lab ax-6003p) ?

    Any other idea of test will be welcome !

    Best Reagrds

    Bartosz Reichel

  • Hi Bartosz,

        Can you share with us the schematic of your set up? I'd like to check later. Or you can request an EVM from TI.com to evaluate it.

    BR

    Sean

  • Yes, because I have problem with our board I decide to build prototype to easier exchange resistors (THT),

    so I use schemaic from first page of

    Only enable, set by divider 33k/10k and only use one PWM (rest float ...). Ok know that prototype should not have "nice" GND rather "spider" but the question is that could have influence only in one case ? On Mondey I could move resitor values back to our PCB (it have similar schematic), only additional filter for power IN.

  • Hi Bartosz,

        I will check the schematic and reply to you next Monday if I find something wrong. Please reply to me if you have any update. Let's solve this problem together.

    BR

    Sean

  • Thank you!

    Next observation, I measure voltages on end capacitor (LED), Power supply have 25V, when I start nothing happen but when I get 2/3 of the led it start. Then I plug rest and everything works.

    I observed when it stops, whille capacitor have voltage above 25V (28V>)  led start but when capacitor discharge to the value 25V or below nothing happen (rise to Vin). Maybe some protection (should be rather current) level have bad setup ? Will check with additional set of cap (like on DEV kit proposition).

    I will be glad for any ideas.

    Edit: remark, adding capacitor not help (also exchange to other types). But If I plug everything and use only enable have same result. Enable=0 led off, till voltage >30V Enable=1 leds power UP.

    ISNS as earlier (what should I expecten if not work, only one pulse but not exceed 400mV) exchange resistor to let say 0.05ohm ?

    Best Regards

    Bartosz

  • Hi Bartosz,

        Here I share one calculation tool for your reference, you can check the value of each component.

        5340.TPS61196-Calculation-Tool_update.xlsx

    BR

    Sean

  • Thank you,

    I have this calculator. Not sure what I could try, maybe RESR smaller ?!

    Best Regards

    Batosz Reichel

  • Hi Batosz,

        Yes, please have a try. Or you can also increase the supply voltage.

    BR

    Sean

  • I change r7 to 0.05ohm, nothing.

    But I observed that if I start and switch ENABLE 2-5 times it start (voltage rise evry pulse).

    Really not sure where is a bug ?!

  • Hi Bartosz,

        I suggest you request a TPS61196EVM from TI.com and test with your set up. It is really strange and I don't have many thoughts now.

    BR

    Sean