• State Resolved
  • Locked Locked
  • Replies 13 replies
  • Answers 3 answers
  • Subscribers 63 subscribers
  • Views 1157 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.

LMR14050: LMR14050: Part getting too hot.

Ty12199
Prodigy 160 points
Part Number: LMR14050

So I am thinking maybe I need to go to a higher power part, but just want to check if there is anything I can do otherwise. I have a design that takes in 20V from a battery and drops it down to 12V to a 3.5A blower fan.

So the part is getting very hot and I was thinking I wasn't that close to the 5A max, but maybe because of the voltages I am using it dissipates too much power. I also only have a 2 layer board and didn't realize they were assuming you use 4 layers.

Layout.

  • 0
    TI__Mastermind 34745 points

    Hi,

    The PCB layout design is not optimized for thermal, even a 5A regulator will also be very hot with this design:

    Pls arrange it as below if you want to re-design it:

  • 0 in reply to Andy Chen
    Prodigy 160 points

    Thank you i make those changes on my next prototype and I was looking at maybe adding a heat sink to the back side for good measure.

  • 0 in reply to Andy Chen
    Prodigy 160 points

    So I made the changes and just put together a new board today and it is behaving weirdly. So the output goes to a 12V brush less blower and the blower starts to spin up and maybe just a bit before it gets to full speed it and all of a sudden the output voltage drops to about 5V. Any clue what would cause that? The previous one didn't do that. I did change the feed back resistors to 150K and 10K instead because that is what your web tool used. I also had a programmable resistor IC on lower feedback resistor, but I have a way to bypass it and doing that didn't make a difference.

  • 0 in reply to Ty12199
    TI__Mastermind 34745 points

    Hi,

    First please check your circuit with Electric load or resistor load to see what's maximum constant load current it can drive stably, then check it with transient load like 0.5A to 5A at 1A/us to see if loop regulation is stable.

    Then, drive 12V BLDC blower with 12V DC power supply, check the startup current waveform to see the peak and duration.

    If the peak is below 5A~6A, then it  can be driven by converter board directly, otherwise you may need to increase cap of 12V output.

    B R

    Andy

  • 0 in reply to Andy Chen
    Prodigy 160 points

    So the web designer said I should have a 820pF capacitor across R10&R11. I decided to take that off and it is working correctly now. I don't know why the tool recommends things that aren't good for the design. It is still running hot so I will have to put a heat-sink on the back of the board for now.

  • 0 in reply to Ty12199
    TI__Mastermind 34745 points

    Hi 

    The Webench design based on ceramic output capacitor, the feedforward cap is not required if you use high ESR output capacitor.

    B R

    Andy 

  • 0 in reply to Andy Chen
    Prodigy 160 points

    Ah I should have known it would be tricky that way. Unfortunately, I can't say I am a pro-power designer. I was doing this for volunteer work and originally I was hoping to not have to use DC-DC converters, but I couldn't find a way out of it.

    Thanks for the help.

  • 0 in reply to Ty12199
    Prodigy 160 points

    So I did some calculations based off of my measurements of the board and it is using about 5 watts. The web-bench designer stated it should be about 3 watts. So is there anything that I might be missing causing that?

  • 0 in reply to Ty12199
    TI__Mastermind 34745 points

    Hi

    What do you mean "it is using about 5W " or the 3W stated by Webench?

    B R

    Andy

  • 0 in reply to Andy Chen
    Prodigy 160 points

    The converter is using about 5 watts of power when I measure power going in vs. what is going out.  Now the Webench said it should use about 3 Watts. So I just trying to understand the disconnect.

    Thanks

  • 0 in reply to Ty12199
    TI__Mastermind 34745 points

    Hi 

    Because of high thermal resistance of your PCB layout, LMR14050 RDS(on) becomes higher on your application, the current rating of inductor maybe also different from Webench design, these items will reduce conversion efficiency, so the total input power maybe higher than Webench result.

    B R

    Andy

  • 0 in reply to Andy Chen
    Prodigy 160 points

    I keep forgetting to get back to you on this. I wanted to show you the updated layout. So do you think there is still issues with it? I'm just curious what I could do better int he future. For now I put a heat sink on the back of the board.

  • 0 in reply to Ty12199
    TI__Mastermind 34745 points

    Hi,

    Please modify it as below:

    B R

    Andy