• State Resolved
  • Locked Locked
  • Replies 1 reply
  • Subscribers 61 subscribers
  • Views 477 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.

why pulse skip happens in charger

Bin Chen36084
Intellectual 2295 points
Other Parts Discussed in Thread: BQ24745

post for Katsuhiro Kanao in bq24745/47 applications. It also applies for other charger.

Customer would like to use 9.3V +/-0.2V adapter to charge 2-cell battery (8.4V) and observed higher ripple than normal when input voltage close to output.

what happens and why?

  • 0
    TI__Intellectual 2295 points

    The ripple increase and we might also hear noise is because the charger enters pulse-skip mode when input voltage close to output voltage. All switching mode buck-converter charger will  rely on pulse-skip mode to achieve higher effective duty-cycle, thus the battery can be fully charged when adapter voltage is close to battery regulation voltage.

    We can share some math and test results here for better understanding.

    From the measurement on bq24745EVM, the duty range right before pulse skipping is D=(0.9328 min; 0.9364 mean; 0.9397 max). Using mean value, here we can get the input voltage for 8.4V output is 8.4V/0.9364=8.97V. The two sensing resistor is 10mOhm each and Rdson of Q1/Q2 is 35mOhm,  and Rdson of Q4 is 13mOhm, plus the inductor DCR is 16mOhm, totally we have 2*10+2*35=90mOhm at input side and 13+16=29mOhm at output side, 0.8A charge current will roughly results in 0.8*0.9364*90 + 29*0.8= 90.6mV. So required input voltage need to be 9.06V without pulse skipping effect kicks in.

    Set input voltage to be 9.06V, I did not see any pulse skip and output voltage instable (20mV range) even the charge current goes to be 896mA. The boundary is 9.016V input, then pulse skip happens.

    Pulse skip is due to max duty the IC can have (considering the dead time limitation--> 30ns min) and the low side MOSFET has 80ns typ minimum on-time, the switching period T=1/300kHz=3.33us and 20% accuracy results in 2.778us ; so it’s roughly to be (2.778us-2*30ns-80ns)/2.778us=94.9%)


    From the above calculation, you can see the customer setting is very close to the condition that pulse skipping might happen. The voltage drop on effective resistance increases so pulse skipping more likely happens at higher current and/or high Rds_on MOSFET and high DCR inductor etc.