• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 62 subscribers
  • Views 498 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

LM27313: Unstable 12v output when driving display backlight.

Matthew Alley
Prodigy 10 points
Part Number: LM27313


I am having an issue with the LM27313 Boost Converter. I have a circuit configured to match the typical application circuit illustrated in the datasheet. However, my output voltage is not stable when my load changes. I have the Boost Converter powering the backlight in a LCD display. When using PWM to change the brightness of the display I get an unstable output voltage, audible ringing, and visible artifacts on the display. The PWM I am using is 10kHz with a 50% duty cycle. I am seeing about a 2v drop in the output voltage. When using a 100% duty cycle I have a clean 12v output without any issues. Is this normal behavior for this part? If so, is there a recommended Boost Converter that can operate at these frequencies?

  • 0
    TI__Mastermind 20090 points

    Hi Alley,

    What's the load current(input current of the LCD display) when the LCD is on? And could you send me the part number of the Cout used in your circuit? I'd like to check whether Cout with bigger effective capacitance works.

    Regards

    Lei

  • 0 in reply to Lei Zhong
    Prodigy 10 points

    Hi Lei,

    The load current is typical 195mA, max 310mA. The Cout part number is GRM219R61E475MA73D.

    I've tried increasing the capacitance to 10uF and did not see an improvement.

    Any recommendations would be appreciated.

  • +1 in reply to Matthew Alley
    TI__Mastermind 20090 points

    Hi Alley,

    Thanks for providing the information.

    There should be 2 reasons that result in the 2V drop.

    • The effective Cout capacitance is two small. For the part you provided, we can see that at 12V DC the effective capacitance remains 20%, that is, Cout = 0.94µF only.

             

    • The max 310mA load current exceeds the max guaranteed Iout already. The min guaranteed peak current(Isw) is 0.8A. Using this Isw and 82% efficiency, the max Iout is around 200mA. With so, when the max 310mA load current coming, the Vout will be pulled down.

    Go back to the application with pulse Iload, the most effective way to decrease the Vout drop is to increasing the Cout capacitance. I believe bigger capacitance will take effect, for example, 5µF corresponds to less than 1V drop. Please pay attention that all the capacitance mentioned here is the effective value at the specified DC voltage(i.e., 12V). As for how big should the Cout be, it dependents on the percentage of max load current and the acceptable Vout drop.

    Just let me know if you see any concerns. 

    Regards

    Lei

  • 0 in reply to Lei Zhong
    Prodigy 10 points

    Hi Lei,

    First off, thank you for the details. The DC Bias information has been very helpful. I switched to a 10uF tantalum capacitor and my display artifacts and ringing issues went away.

    Would you mind sharing the equation or technique for estimating the expected Vout drop? With the new capacitor I went from over a 2V drop to about a 0.7V drop. I am interested in seeing if I can reduce this drop further. The display's specifications lists a permissive input ripple voltage of 100mVp-p.

    -Matthew

  • 0 in reply to Matthew Alley
    TI__Mastermind 20090 points

    Hi Alley,

    It's hard to calculate the Vout drop accurately. For a rough estimation, you may refer to the following equation.

    Regards

    Lei