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.

BQ24780S: BQ24780S dead time query

Part Number: BQ24780S
Other Parts Discussed in Thread: BQ24780,

Hi

My Customer measured the dead time of BQ24780 and found that it is much larger than the typical value of 20ns .

The value measured on the  board is 80ns. I do n’t know what caused the dead time to be too long.


We use 24v power supply, and the charging current is set to 1A, which is a 3s2p lithium battery.

The dead time of my test is from the falling edge of Lodry to the rising edge of Hidry.

Attached the waveform as below.

Waiting for your reply.

Thanks

Star

  • Hi Star,

    I assume that the attached captures are on the customer's board,, correct?  I will set this up on the EVM with 24V input and see if I get a similar result.

    Have they observed similar behavior on multiple boards or just on one board?  Can they try different battery voltages and different charging current levels to see if this pattern changes?  One possibility is that they are right at the transition from DCM to CCM switching.  During DCM there is a period where both switches are off.  If it happened that they were right on the border between DCM and CCM switching, there could be a gap like what you show as a result.  They could verify this by setting the current one value higher and one value lower and seeing if the width of the gap changes.  If the gap gets wider as the current setting gets lower and disappears when the current setting goes higher, then it is being caused by DCM switching.

    Regards,

    Steve

  • Hi Star,

    I tried this in the lab today with 24V adapter and 12V battery.  At 1A charging current, I got a similar gap as you show,

    but when I increased to 1.5A charging current, the gap reduced to about 20 nS dead time on both upswing and downswing.

    This isn't really dead time, it is DCM switching that turns off both switches when inductor current gets close to 0A.  The 1A charging setting under these conditions is right on the transition from DCM to CCM, so increasing the current gets it into full CCM switching that shows the expected dead time.

    Regards,

    Steve

     

  • Neither the rising edge nor the falling edge is not monotonic , any comments  ?

  • Hi Star,

    This is a property of DCM switching.  The fundamental frequency (600 kHz, 800 kHz or 1 MHz as set) is still present but when the inductor current crosses 0A, both HIDRV and LODRV are low, which results in the irregular pattern and the extra gap that looks similar to dead time.  Increasing the charging current will bring the BQ24780s into CCM switching, which will have the regular period you are expecting.

    Regards,

    Steve