• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 63 subscribers
  • Views 895 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.

bq24650: Layout consideration for charge current larger than 8A

Darren DeBattista
Prodigy 200 points
Part Number: BQ24650

Hi,

I'm designing a LiFePO4 solar battery charger using the bq24650. As the title says, I need to achieve currents larger than 8A (max 10A). My current layout is a near exact replica of the EVM board layout with the difference that I have components only on one side of the PCB. My question is, with the EVM layout and using the same components, can the charger deliver 8A-10A charge current while keeping the mosfets at reasonable temperatures?

I already have a previous version of this charger which had the mosfet case temperature rise to 84C at a charge current of 5A. This previous layout had some bad layout techniques (Inductor pads extending underneath the inductor, mosfet to inductor track connected using parallel multi layer tracks, mosfet pads connected using thermal relieve etc... basically, all things that the datasheets lists on page 28 as Layout Guidlines were ignored). This made me think that there maybe an issue of loop stability in the charger too. I measured the mosfet drain/inductor voltage and there is no ringing. I also measured the rise/fall times and they are the same as seen in the datasheet. I did see some ringing in the inductor current though. I manually removed the extra copper from underneath the inductor and the current ringing decreased considerably but the rise in mosfet temperature remained. 

Before sending the PCB to be manufactured, I would like to know if the design is even possible and if so, if there are other consideration to be taken. As a side note, I was thinking about using the BSC072N03LD mosfet instead of the Si7288DP due to it having a smaller RDSon value and a better FOM value too. Using the Ploss calculation given in the bq24650 datasheet, this mosfet should have roughly 1W dissipation at 10A as opposed to about 2W with the Si7288DP. 

Again, is the design I'm making even possible or is it necessary to use single mosfets instead of a dual mosfet package (or even using two paralleled mosfets) and are there other considerations I need to take care about? 

Thanks

  • 0
    TI__Intellectual 1980 points

    Hello,

    The charger could support 8-10A application.

    For the MOS, because the charger doesn't integrate MOS, so the temperature of MOS doesn't depend on the charger. You need to choose a suitable MOS(suitable current, Rdson etc.) for you application.

    Regards

  • 0 in reply to Kerry Song
    Prodigy 200 points

    Hello,

    Thank you for your reply. My question was mostly regarding the BQ24650EVM board since I am basing my design and layout on it. Does the BQ24650EVM boards support 8A (or even 10A) charging by just changing the inductor, output caps and shunt resistor or does the mosfet need to be changed also? If the mosfet need to be changed, is a dual mosfet package (like the Si7288 used in the EVM) with a better RDSon and FOM enough or is a different package needed (making the EVM board not good for 8A/10A applications) due to the possible need of a better heat sinking area?

    At the end, I want to know if with the concept of using a dual package mosfet like the one used in the EVM boards and the layout as is in the EVM board is possible to charge at 8A/10A.

    Thanks

  • 0 in reply to Darren DeBattista
    TI__Intellectual 1980 points

    Hello,

    For the current question, recommend read the MOS's spec in it's datasheet to get the information. Just the package do not have a big influence on the performance.

    Regards