LTC4162EUFD-FST becomes very hot at start-up

Hi,

our customer has designed a prototype of a charger of 8-cell LiFePO4 battery based on the design files of a demokit DC2038A and selected the PN: LTC4162EUFD-FST. After the start of the battery charger it becomes very hot and blows. 

Is it ok to use this PN for the charging an 8-cell battery? How to avoid a critical temperaure increase?

Also what is a difference between LTC4162EUFD-FST and LTC4162EUFD-FAD? Maybe we should take -FAD version? 

  • Hello,

    It is fine to use 8 LiFePO4 cells with the LTC4162. Make sure your current limit is set to no more than 3.2A which is the max allowed for the part.

    LTC4162 is a monolithic buck, so there will be heat to an extent. However, I have never heard of any catastrophic heating issues with this part causing it to blow up. If that is the case, you likely need a more thermally-dissipating layout to help get the heat out. Make sure you are appropriately via-ing heat from the GND paddle into a GND plane.

    The -FAD is i2c adjustable and defaults to the -FST settings. There is no difference by default.

    Regards,

    Zack

  • Hi, many thanks, Zack!

    Looks that the thermal dissipation is an issue: we have reviewed the layout and found that the prototype PCB has thin copper underneath the part (35 um instead of used in demoboard 70 um). This is however is not mentioned in the datasheet.   

  • Hi Dmitry,

    This is more of a general guideline when laying out monolithic regulators, or really anything that dissipates significant power.

    We also have some inner layers that are filled with pours and help to spread the heat throughout the board. With these thermally coupled (through vias) to the top and bottom PCB planes, you can get the heat out into the air with some effectiveness.

    There will always be some heat, so take any steps you can take to mitigate it and decide how much you can tolerate.

    Regards,
    Zack