I'm building a battery charger circuit with BQ24278. The battery draws 500mA minimum and 2.5A maximum current; therefore, I must prevent IC from overheating. I printed the board onto single sided copper plate with ironing method and unfortunately there is no opportunity to drill the copper plate since I don't have a proper drill equipment. So, considering my restriction, is there a practical way to cool down the chip? I already have DC fan and heatsink for TO-92 package; however, since I have burned out 2 ICs already, I would like to apply a solution by getting ideas of other people.

I connected AGND and PGND under the bq24278, and put as much copper as I can. I attach the datasheet and the circuit diagram.


the circuit: enter image description here

datasheet: http://www.ti.com/lit/ds/symlink/bq24278.pdf

  • \$\begingroup\$ What did you do to burn those chips? They should have an internal thermal regulation that should work quite well... are you sure you didn't do something else to destroy them? And are you sure they are original parts? \$\endgroup\$ – PlasmaHH Feb 23 '15 at 14:54
  • \$\begingroup\$ They have burned after 10 minutes of perfect operation. The battery was charging, LDO, and converter have been working as they were supposed to be. Every single connections have been check almost 5 times or more. And yes, I bought the chip directly the website of TI. @PlasmaHH \$\endgroup\$ – cryptokovski Feb 23 '15 at 15:04
  • \$\begingroup\$ Actually, burning down is just a comment. The chip was warmed up dramatically, and suddenly started not to work. That is the problem. @PlasmaHH \$\endgroup\$ – cryptokovski Feb 23 '15 at 15:05
  • 1
    \$\begingroup\$ Check the datasheet again if everything is correct and pay extra attention to the thermal regulation section. It should work just fine without any extra cooling. \$\endgroup\$ – PlasmaHH Feb 23 '15 at 15:07
  • \$\begingroup\$ well, TS pin drops down to zero volt, which means that IC stops charging automatically. It is weird that the IC is quite cold in this state. I couldn't manage to make it work anyway. Something mysterious I cannot identify seems to prevent IC start charging. @PlasmaHH \$\endgroup\$ – cryptokovski Feb 23 '15 at 16:03

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.