Post Go back to editing

Somethimes the Buck DCDC will not work and starts heating

Category: Hardware
Product Number: LTC3114

Hi Experts,

we design a Buck DCDC with these caratteristics:

- Vin = 15V +/- 10%

- Vout = 8.5V

- Iout = 500mA max

Following the schematic part of the DCDC.

After this DCDC we have some LDOs and capacitors.

If we power up the board, sometimes the DCDC not work and the main voltage is shorted, the chip starts heating.

We have produced several boards for an important project and we need to fix this issue.

Can you help us.

Thank, have a nice day.

  • Hello,

    I tried simulating your circuit and it appears the feedback resistor value in the schematic may be incorrect. Based on my analysis, it should be 750 kΩ rather than 75 kΩ. When simulating with the current value, the loop shows signs of instability, suggesting that the VC‑pin compensation components may need further optimization.

    Before making further changes, I recommend first double‑checking the layout against the datasheet recommendations. A good approach would be to start from our demo board, modify it to match your design parameters (particularly the feedback network and VC‑pin components), and then evaluate whether any layout‑related improvements are required. This can also help validate the selected component values.

    Please note that LTpowerCAD is available for this part, and when I enter your design parameters it recommends different feedback values, compensation components, and inductor selection. If you have not already done so, it would be worthwhile to optimize these values first, especially given the observed instability. The datasheet also provides the necessary design equations for component selection and compensation; however, the final values typically require tuning in hardware.

    Lastly, I recommend measuring the efficiency on your board to ensure the design is not dissipating excessive power, when doing so remove any downstream LDO, try to use e-Load or resistive load to validate your board.

  • Hi tonie,

    thanks for your help.

    Yes the feedback was wrong, we fix it by changing the 100k feedback resistor to 10k. 

    Unfortunately we don't have the eval board.

    I check the LT power cad design and I saw the difference in the VC-pin component.

    Our problem is generated when the input voltage rise smoothly.

    In addition to that we have several capacitors after the DCDC, so we have inrush current.