• State TI Thinks Resolved
  • Locked Locked
  • Replies 10 replies
  • Answers 4 answers
  • Subscribers 63 subscribers
  • Views 215 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.

BQ24253: Reason for not D+ and D- detection

Satoshi
Guru 19575 points
Part Number: BQ24253
Other Parts Discussed in Thread: BQ25616, , BQ24251EVM-150, BQ24251

One customer is considering bq25243, USB Type-C and Li-Ion battery.

When both USB and Battery connecting, /PG was not assert Low and charge current become about 100mA in any EN1 and EN2 setting. 

(RLIM = 300Ω = 1000mA, RSET = 270Ω = 833mA)

Please let me know about two points below for D+ and D- detection.

①Is there reason and improve idea?

　I think there is a problem with detection of D+ and D-.

　By the way, Battery condition is 3.5V (not become dead battery), EN1 and EN2 toggle is delayed 1sec after USB insert.

②About condition for USB insert and EN1 and EN2 are not toggle, D+ and D- detect was differed on battery connect and no connect. 

　Are above operation normal or D;/D- detection error?

　・Battery is not connect, USB is only connect (Normal)

　　D+ pin: USB insert → 5V (about 110ms) → continue to 500mV 

　　D- pin: USB insert → continue to 500mV on the same timing for D+ become 500mV 

　・Both Battery and USB are connect (Abnormal?)

　　D+ pin: USB insert → 5V (about 110ms) → 500mV (about 200ms) → changed to low 

　　D- pin: USB insert → 500mV on the same timing for D+ become 500mV → changed to low 

Best regards,

Satoshi

  • 0
    TI__Guru*** 149265 points

    Hi Satoshi,

    This was our first charger at D+/D- secondary detection.  It does have some bugs.  However, I have not seen /PG not go low with no battery as long as a valid power source is attached.  I suspect the fault is SDP Good battery and the charger is ending in HiZ. If the ENx lines are toggled, does that fix the issue?  Most customers don't this part because SDP and CDP dead battery ends with current limit = 100mA.  With no battery, I would expected /PG= low, input current limit = 100mA and 45 minute timer per table below:

    Regardless of the detection error, the fix for this is to toggle ENx after D+/D- completes.  

    I recommend moving your customer to BQ25616.

    Regards,

    Jeff

  • 0 in reply to Jeff F
    Guru 19575 points

    Hi Jeff

    Thank you for reply,

    Customer is already toggle ENx after D+/D- complete. (D+/D- complete → 1sec →toggle ENx )

    We hope increase charge current after toggle ENx, but keeping about 100mA.

    ・Is there the other concern point?

    ・Would you let me know in detail points for bug?

    Best regards,

    Satoshi

  • 0 in reply to Satoshi
    TI__Guru*** 149265 points

    Hi Satoshi,

    Toggling per below should allow for either 500mA current setting or ILIM set per ILIM resistor. 

    To be clear, toggling means at least of the ENx lines need to change from original state to different state.

    Regarding the bug for this > 10 year device, I am only speculating that it is a new bug and do not have the resources to investigate further.  But, if can toggle the ENx lines, then the charger has to be in HiZ mode which is supposed to only occur when it detects SDP and good battery.

    Most customers end up not using D+/D- on this device and simply short D+/D- together and use the ENx lines to set input current limit because the 100mA input current limit and HiZ settings above are impractical.

    Regards,

    Jeff

  • 0 in reply to Jeff F
    Guru 19575 points

    Hi Jeff

    I confirming to customer about EN1/EN2 set 1 (Hi-Z) → toggle EN2 to 0,

    Is these setting recommend, correct?

    Additional condition;

    ・All fault condition were not applicable.

    ・Vin waveform found unstable;

    　Vin increase over than 10.5V (VOVP) on 50 ms.

    　→After increase, Vin changed 4.2V〜6.8V.

    　Is this problem cause to D+/D- detection error? 

    　And, is unstable Vin affect of the bq24253 specification (or bug)?

    By the way, customer is already using bq24253, we want to use bq24253 as much as possible.

    Best regards,

    Satoshi

  • 0 in reply to Satoshi
    TI__Guru*** 149265 points

    Hi Satoshi,

    EN1/EN2 set 1 (Hi-Z) → toggle EN2 to 0 sets input current limit to 100mA.  I recommend changing both EN2 and EN1 to 0 for 500mA input currnet limit.

    Regarding unstable VIN up to 10.5V, that will trip OVP and reset the logic.  I suggest after that OVP event that your customer set EN2=EN1=HiZ and then set back to EN2=EN1=0 for 500mA current limit.

    Regards,

    Jeff 

  • 0 in reply to Jeff F
    Guru 19575 points

    Hi Jeff

    Customer tried two points below, but problem was not improved.

    ・D+ and D- short 

    ・EN2=EN1=HiZ and then set back to EN2=EN1=0 for 500mA current limit

    Vin is continue to change 4.2V~6.8V after OVP event, please let me know about the reason for Vin: 4.2V~6.8V. 

    I think that Vin: 4.2V~6.8V is abnormal, is it correct?

    Is this possibility to affect from bug?

    Best regards,

    Satoshi

  • 0 in reply to Satoshi
    TI__Guru*** 149265 points

    Hi Satoshi,

    What is the input source?  Your first post says USB Type C.  The BQ25253 does not have CCx lines and so it can't communicated with USB Type C. .It cannot change its input voltage. The only way the above can happen is if the input source is supply 6.8V and the converter is pulling so much current (but less than the input current limit setting) that the input source can't supply it so the source is pulled down to the VINDPM setting of 4.2V.  Can I get a waveform of VIN, IIN and SYS?

    Regards,

    Jeff

  • 0 in reply to Jeff F
    Guru 19575 points

    Hi Jeff

    Thank you for advice,

    Customer information is update; When 5V bus add to connect 1Ω for avoid overvoltage, EN toggle was normal operation. 

    But waveform occur ringing and IR drop affect to add 1Ωresistor.

    ・500mA condition (EN2: 0 and EN1: 0 )

    ・Zoomed

    I guess these reason is input line, and customer is investigating for the input source.

    By the way, when evaluate bq24251EVM-150 on the same input condition without 1Ω, EN toggle and the other operation was normal.

    Is it possible that only bq25243 affects input behavior?

    （I would like to confirm if the bug or other behavior is only bq25243 or all of bq2524x.）

    Best regards,

    Satoshi

  • 0 in reply to Satoshi
    TI__Guru*** 149265 points

    Hi Satoshi,

    Both parts use the same die with different NVM bits set.   The BQ24251 does have ENx enabled.  BQ24251 input current limit is set per below:

    VINDPM is set by the external resistors (need to confirm).  If adding a series resistor helps, then the issue may be an incompatibility with their power source and the VINDPM regulation loop.  Adding more input capacitance instead of 1 ohm resistor may help.

    Regards,

    Jeff    

  • 0 in reply to Jeff F
    Guru 19575 points

    Hi Jeff

    Thank you for information about bq24251.

    When customer changed that more Cin (10μF) instead of 1Ω resistor, D+ and D- detection were not improved.

    If there update for input side condition, I will feedback to you.

    Best regards,

    Satoshi