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.

bq24753A - Change in Charge Current When Supply Changes - Why?

Other Parts Discussed in Thread: BQ24753

Charge current is set to 400 mA.

At Vcc of 14V, charge current is stable at 400 mA.

When Vcc increases to 16V, charge current drops to 280 mA.

I'm read the datasheets, and re-reading the datasheets but it's still not clear to me why this is happening.  I was hoping somebody else might have experienced this behavior and could shed some light.

Thanks for any suggestions!

Chris... 

  • Some more information:

    It appears that the charge current tracks the battery voltage.

    For example, Vbat starts at 10.1 Volts (Li-Poly 3 cell) and charge current is set at 400 mA.  The actual starting charge current is measured to be 240 mA and climbs to about 320 mA as the battery charges to 12.6 Volts.  Supply voltage is 14.4 Volts.  I assumed that charge current would be fairly linear over the charge cycle.  Is that a wrong assumption?

    Changes in supply voltage, from 14.0 to 16.0 Volts, also appears to impact charge current adversely.   Why?

    Any suggestions would greatly be appreciated,
    Chris...

     

     

     

  • Sorry to ask a stupid question, but, where are you measuring the charge current?

  • Not stupid at all.

    Followig the recommendation in the data sheet.  There is a sense resistor that comes off BTST called Rsr.  The current is monitored by the charger via SRP & SNP but we also monitor this same sense resistor via the micro and display the value near real-time and log the value during the charge cylce.

    I have attached two graphs.  One showing a change in supply voltage and the other with a constant supply voltage.  The battery charge current is set to 400 mA.  Note on the first graph that for a charged battery (12.31 v) the charge current starts good at 400 mA but takes a big dump around 16 volts.  Also note that with a discharged battery (10.02 v), the chage current starts at around 240 mA and takes a big jump around the 15.4 v mark but never exceeds 300 mA.  Why, I'm not sure.

  • Looks like my first graph didn't post.  Reposting.

     

      

  • Do you have any load connected to the system pin?

    Could it be that for hi Vin the load is taking more current and triggering the AC adapter current limit? What is your programmed AC adapter current limit?

  • First, let me say thanks for the repsonse.

    Not sure what you mean by "system pin".  PVCC is connected to the input  FETS (Q1 & Q2 per the data sheet) with no load if that's what you mean.

    AC adapter current is set to 600 mA.  This current is being monitored via the power supply and never exceeds 400 mA.

    I suspect the issue is something to do with noise on the sense resistor.  We are using 0.01 ohm instead of 0.1 ohm but the datasheet indicated that would be fine.

    What's strange is that charge current appears to be dependant on battery voltage.  For a low battery, the starting charge current is 240 mA for a medium charged battery the charge current starts at 370 mA.  Why would that be?

  • Some more info concerning the first graph tittled "Battery Charging".  Note the large decrease in system power (actually charge current) near the 20:24 mark.

    Supply Voltage = 14.1
    High Charge Current =  (5W/14.1V) = 354 mA, Low Charge Current (3.7W/14.1V) = 262 mA.

    Per the datasheet, the values we are using for:
    L1 = 15 uH
    C11 = 10 uF
    C12 = 10 uF
    Rsr = 0.10 ohm

    We changed C12 from 10 uF to 4.7 uF and it eliminated the ~100 mA dump during the charge cycle.  Why did this work?

    Must be noise across Rsr, right?  This must be influencing the synchronous-buck PWM duty cycle is the only thing I can think of.  Is that a correct assumption?

    What is still happeing is our programmed charge current of 400 mA is starting at 240 mA for Vbat of 10.02 Volts.  Why?

    The charge current appears to track the battery voltage.  Why?

    Thanks again for any suggestions!

    Chris...

     

     

     

  • Some more info.  We reduced the Charge Current (SRSET) to 150 mA by changing R11 (data sheet).

    Reconnected the board to Supply Voltage and varied it between 14.1 to 16.5 Volts.  The Charge Current remained steady at around 150 mA.  This is a very good sign.

    The assumption is that this reduced the noise on the sense resistor Rsr.  Agree?

    My next step is to attempt is to capture the noise across Rsr.

    Also, if anybody knows where to get the TI evaluation board HPA207 for the bq24753, it would be greatly appreciated.  It appears that TI discontinued this kit about a year ago. 

    Chris...

     

     

  • Would you check it with an EVM?  Please compare your circuit with EVM schematic.

    After that, measure Vsrp-srn and Vacp-acn. And compare the with the charge current setting. 

    What is this application? And, what is the annual volume for this project?

  • Hello Wang5577,

    Thanks for the reply and interest.

    If by "an EVM", you mean evaluation module - I've tried that road.  The HPA207 has been discontinued, at least that's what I was told by TI.  So either I work with the board I have, or fabricate the HPA207 from scratch.  Any other ideas?  I would love to get my hands on a HPA207 eval board.

    I've scoped Vsrp-Srn (voltage acros Rsr) and p-p is around 40mV with noise spikes going up to ~90 mV p-p.  Is that normal?

    The design of the circuit almost identically matches with the example in the datasheet (page 3) with two expections.  L1= 15 mH & Rsr = 0.1 Ohms.  But both of these changers are within the spec of the datasheet.

    Thanks,
    Chris...

     

     

  • 40mVp-p of Vsrp-Srn is not normal. Do you put enough cap between inductor L1 and sense resister Rsr?

  • This begs the questions, what is normal?  Any details would greatly be appreciated. :)

    We did change C12 from 10uF to 4.7uF which seemed to help some.

    Note the we are using a larger sense resistor so one would expect larger voltages.  Correct?

    The cap across Rsr is 0.1 uF per the data sheet.  Should this be increased due to the larger sense resistor?  We thought about doing this but ...

    Thanks again for the response,
    Chris...

  • if the charge curent is 0.4A, sense resistor is 0.1ohm, the DC voltage is 40mV. But, you said, the AC voltage is also 40mV. It is too big compare with DC voltage. So, I said, "it is not normal."  For a large and high frequency ripple, the internal OPAMP can not filter the AC compoents out . So, it may trigger current limit early.

    Please change sense resistor from 0.1ohm to 0.02ohm. It should resolve the issue.