• Locked Locked
  • Replies 5 replies
  • Subscribers 63 subscribers
  • Views 2329 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.

LM3478 12V to 24V Boost application

Gary Perez
Prodigy 120 points
Other Parts Discussed in Thread: LM3478, LM3481

Hi, 

i am having trouble with my application. Built as shown on the attached schematic, my input power supply (which is rated at 3A) goes into shutdown mode when I try to turn on the power. i believe the transistor M1 is always on. the transistor being used is a IRFB4710. I have also tried various gate resistors.

I also noticed the input filter RC note.  Does this RC filter also tie to the inductor or should that side be directly connected to Vin? What are some good starting point values for the low pass RC filter?

Thank you , 

Gary

  • TI__Expert 6245 points
    Hello Gary,

    If your power MOSFET is left on, it is likely that your power supply current limits below ~12 A, the current necessary trip current limit, and that your power supply ramps slowly. The LM3478 does not have a duty factor limit, so if output voltage cannot be achieved duty factor can become 100%. If the power supply is current limiting below the current limit for the LM3478, the circuit can get stuck. If the power supply can provide more power, the circuit should hit current limit, shutting down the power MOSFET and allow startup to retry. If very high duty factor is not desirable in your circuit, the LM3481 limits duty factor eliminating getting stuck at very high duty factor.

    The note on the WEBENCH design points out that the startup of the LM3478 can be delayed by adding an RC filter to the LM3478’s VIN pin, increasing the amount of time between when power is applied to the system’s input and when the LM3478 starts up. The filter should only be connected the LM3478’s VIN and not be in series with other parts of your circuit. You would be using the LM3478’s internal UVLO to delay startup. During this time, the output charges through the diode, called D1 in your schematic, allowing operation to start with much lower inductor current.

    Regards, Robert
  • in reply to Robert Blattner
    Prodigy 120 points
    Hello Robert,

    thank you very much for the help. I will try and scope the turn on time of my switching power supply. I currently have a layout for the LM3478 IC.

    -Gary
  • in reply to Robert Blattner
    Prodigy 120 points
    Hello Robert,

    one more quick question. I have followed the above suggestions and after applying power to the chip after a time delay, the supply does not go into a short. but the output remains equal to the input. i dont think the switcher is switching. Is my time delay too long?

    thanks,

    Gary
  • in reply to Gary Perez
    TI__Expert 6245 points
    Hello Gary,

    The most likely problem is that the resistance in your filter is too high. VIN supplies power to the LM4378 so if the resistance is too large, VIN may never reach a working level. I would check the voltage at the VIN pin and if voltage remains very low, try a smaller value resistor. Once the supply is able to start, I would adjust capacitance to give the delay that you want. I would try to limit drop across the resistor to less than 1 V after the RC delay.

    Regards, Robert
  • in reply to Robert Blattner
    Prodigy 120 points

    Hi Robert, 

    Again thank you for discussing this part with me. 

    I was able to not have the LM3478 get stuck when powered on. 

    The output voltage jumps to 24V then drops down to the input voltage there after with the power supply current climbing very slowly. 

    I have a built in 20mA load on the output. 

    At which point, I power down at 0.5A.

    I will keep on testing. Please feel free to offer any suggestions.  

    Thanks again, 

    Gary