• State TI Thinks Resolved
  • Locked Locked
  • Replies 11 replies
  • Answers 2 answers
  • Subscribers 64 subscribers
  • Views 513 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

BQ25886EVM: Unstable operation on no battery condition

Satoshi
Guru 19585 points
Part Number: BQ25886EVM
Other Parts Discussed in Thread: BQ25886

When bq25886EVM with no battery connection (Vbus: 5V), 0.4A load became discontinuous switching and 0.5A load became Vsys: 0V. 

Please let me know about three points below;

①Measure to stable operate (continuous switching) on over 0.5A and no battery condition.

②The reason for SW pin occur discontinuous switching

③Correction point for change the continuous switching

 ※I want to know bq25886 detailed operation on no battery condition.

EVM setting condition is below;

・Vbus: 5V, Vreg: 8.2V

・L2 changed to 10μH inductor, ILIM: 1.5A (750Ω), ICHSET (ICO): 1.5A (5.87KΩ)

・Jumper pin: JP1 and JP16 are changed

　⇒JP1: open (LDO no connect), JP16: short (Vreg: 8.2V)

　The other Jumper pin setting and external components are EVM default setting.

・D+ and D- are left short

Waveform is attached below;

※ch1: Vbat , ch2: Vsys , ch3: SW pin 

・0.4A (unstable)

・No load (stable)

Additional information

・When Battery connect, switching was be stable.

・When /CE changed Low to High, SW voltage became continuous switching.

　But Vsys output became 0V (off) if increase current, duty cycle was be limited.

Best regards,

Satoshi

  • 0
    TI__Guru**** 151726 points

    Hi Satoshi-san,

    We have not tested and therefore do not recommend inductors higher than 2.2uH.  Why was the inductor increased to 10uH?

    Regards,

    Jeff

  • 0 in reply to Jeff F
    Guru 19585 points

    Jeff-san

    Thank you for reply,

    The reason of changed 10μH inductor is reduce peak current.

    One customer should to increase inductance because large peak current occur EMC test failed.

    By the way; when inductor changed 2.2μH, output current could output 0.8A (10μH: 0.4A). But switching is discontinuous, too. 

    I summarized additional question below;

    ①Is 10μH inductor strongly no recommend for whatever reason?

    ②Is discontinuous switching mean bq25886's normal spec or any trouble (abnormal)?

    ③If there any way to not occur discontinuous switching, please let me know.

    Best regards,

    Satoshi

  • 0 in reply to Satoshi
    TI__Guru**** 151726 points

    Hi Satoshi-san,

    Regarding 1, the charger's feedback loops are internally compensated for a no higher than 2.2uH.  We cannot warranty operation for larger inductors.

    Regarding 2, the BQ25886 will not normally operate in DCM as it has PFM enabled.  Its I2C siblings operate in DCM with PFM disabled based on our testing. PFM operation is tuned for the inductor being no higher than 2.2uH.

    Regarding 3, No other way to prevent DCM.  However, the recommended way to reduce EMC, in priority,  is 1) add small 0.1uF caps at SYS pin for charge mode and PMID pin for OTG mode, 2) add RC snubber circuit at SW pin 3) add ferrite beads in line with SYS for charge mode and PMID for OTG mode.

    Regards,

    Jeff

  • 0 in reply to Jeff F
    Guru 19585 points

    Jeff-san

    Thank you for advice,

    Customer changed 10μH to 2.2μH inductor, but VBAT was still unstable.

    When no battery test condition, is there measure point by default setting from bq25886EVM?

     Or, please let me know about wrong point for above setting.

    By the way, which is correct for /CE: High or Low?

    Bast regards,

    Satoshi

  • 0 in reply to Satoshi
    TI__Guru**** 151726 points

    Hi Satoshi-san,

    With no battery and /CE=low, BAT and SYS oscillate as the charger thinks the small output capacitors on BAT are the battery.  So, the charger charges up the capacitors, terminates and then after the capacitors discharge, the cycle repeats. IF /CE= high, this will not occur.

    Regards,

    Jeff

  • 0 in reply to Jeff F
    Guru 19585 points

    Jeff-san

    Thank you for reply,

    Customer is confirming to change lower inductance.

    Sorry for additional question, 

    When in the case of battery connecting to suddenly disconnect, Vbat will occur unstable switching.

    (These case cannot to set already /CE: High, because normally battery charging)

    Is this switching OK for NVDC (powerpath) operation?

    Or, is /CE have to change Low to High when detect battery disconnect?   

    Best regards,

    Satoshi

  • 0 in reply to Satoshi
    TI__Guru**** 151726 points

    Satoshi,

    With no battery connected and /CE low, the SYS output will not drop lower than MINSYS (NVDC regulation) and will periodically rise to VBATREG and the charger charges the BAT pin output capacitors.

    Regards,

    Jeff

  • 0 in reply to Jeff F
    Guru 19585 points

    Jeff-san

    Thank you for advice,

    Customer captured waveform for /CE: High and Low condition, with changed 2.2μH inductor.

    But results was not changed regardless of /CE pin setting.

    (No Load ⇒ stable, Any load: unstable(discontinuous) switching and VSYS / VBAT voltage drop when exceed duty limit)

    For measure to stable switching, should BAT pin output capacitor change to small value?

    Or, is there the other measure?

    Best regards,

    Satoshi

  • 0 in reply to Satoshi
    TI__Guru**** 151726 points

    Satoshi-san,

    What is the input current limit setting?  Have you tried this same test on a BQ25886 EVM?

    Regards,

    Jeff

  • 0 in reply to Jeff F
    Guru 19585 points

    Jeff-san

    Input current limit setting is 1.5A (Rilim: 750Ω), I guess that peak input current occur less than 1A by datasheet page 26.

    Is above waveform near than current limit operation?

    Best regards,

    Satoshi

  • 0 in reply to Satoshi
    TI__Guru**** 151726 points

    Satoshi-san,

    The waveforms do not look like the waveforms on d/s page 28.  Input current limit (IINDPM) does not change peak inductor current.  It limits the converter duty cycle, i.e. it power limits the converter.  Can you repeat but include VBUS pin on a plot?  Is the IC entering VINDPM?  

    Regards,

    Jeff