• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 65 subscribers
  • Views 587 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.

bq24086 current issue

Rob Mitrunen
Expert 1685 points
Other Parts Discussed in Thread: BQ24086, BQ24610, BQ24133, BQ24072, BQ24087

My customer asks:

"I use bq24086 for a charger and am having trouble with deeply discharged batteries.  Everything works fine on normally discharged batteries, but the system must be able to recover from a deeply discharged state (i.e. after the product has sat on a store shelf for months, or if the user depletes it and does not use it for a period of time).  So that’s what I’m testing.

 

My battery terminal voltage is 2.7V and I have disconnected any load on the circuit.  I have a current meter (set at 1A range for minimal resistance) in line with the battery to monitor charging current.  I’ve got an oscilloscope looking at battery voltage.   The charge current on the 24086 is set at 650mA. 

 

Without the charging voltage plugged in, the current meter shows -50uA (small discharge into the circuit).  It changes to 73uA (small battery charge current) when I attach 5V charging voltage.  The battery voltage remains at 2.73V.  If I switch in the circuit, the battery current changes to -50mA to power the circuit and the voltage starts dropping.  If I disconnect the battery, the charger enters the battery absent state (observed on the scope).

 

So the bq24086 is staying very resistive.  It won’t supply any current to the application circuit, and I don’t think it’s outputting enough voltage to deactivate the low battery protection in the Li-Ion pack.

 

Is there anything I can do to force the 24086 to output more so that I can get my system running (enough that the micro can run in low power mode and keep everything off until the battery charges above 3V) and have the battery charge?  I’m going to leave the charger running over the weekend to see if the battery will eventually get high enough.

 

Update:

 

This is a show stopper and has four different products on hold as a result because they all use the same circuit. 

I changed the charger to “LDO” mode (removed the TMR resistor) and bypassed my selectable charge rate so RSET is now 1k for about 450mA.

The main circuit attaches to the battery charger board through a connector, so I can easily check if just the charger works.  It does.

 

When I attach the main circuit, the charger output latches at just under 2V and oscillates a bit (see the scope shot attached).  I disconnected some circuits on the main board and when I remove the connection to our GSM module, the charger will work.  There are two 47uF ceramics on the GSM module that are connected directly to the battery.

 

I assume that the inrush current to these caps is too much and trips the “short circuit” protection in the charger – this is consistent with the charger datasheet that specs V(scind) around 1.8V.  If I jiggle the charger input so it connects and disconnects quickly (before the battery discharges back to 0), then charging will actually start – so again I assume I’m just avoiding the major inrush and getting past the short circuit detection in the charger.

 

I can also get the circuit into a mode where the output of the charger is stuck at about 2.98V and no current flows into the battery – I assume this is the battery protection circuit (which is specced to come on around 3V) turning off so no current can charge the battery (this is about the V(lowv) spec of the charger).

 

I’m guess-and-testing some different values for the resistor between OUT and BAT to see if that makes a difference.   Not yet.

I’m wondering if I could add a bit of resistance right at OUT on the charger to limit the rush current.  Or even an inductor.  I cannot afford any resistance from the battery to the main circuit, but from the charger to the battery would be okay.  Since I charge at 500mA, I’m not sure if this would screw up the charge sequence – probably not as I think it would sense before the resistance and still fully charge the battery since CV mode is last so the drop on the R would reduce.

 

Any ideas?"

 

 

  • 0
    TI__Mastermind 31120 points

    The bq24086 only puts out C/10 in precharge mode and with a 10k on the ISET that is about 10mA of precharge current.

    The Short circuit current for battery voltages less than 1V is 15mA.

    I believe your issue is a load on the Battery when it is fully discharged.  Your system should be smart enough not to load the battery until it is up to the weak battery threshold which is where the system can run on the battery energy and in fast charger, above 3V, where the battery is in fast charge.

     

    Your GSM pulse is probably keeping the battery from charging.

    You probably need a power path part that will allow running the system on a discharged battery directly from the source.

    Look at bq2403x and bq24072 for linear chargers...bq24133 and bq24610 for switchers.

     

    One could add a external supplement circuit with a resistor in series with a P-CH FET connected between the input and OUT with the gate tied to the stat1 pin.  When the IC is in charge mode it would pull on the FET and supply the supplemental current (Vin-Vout)/R and when the charge terminated the stat1 would go high and FET would turn off.  Must not disable termination.  The risk the customer has to assume is that the advantage of the pre-charge circuit to detect a partially shorted cell, stay in precharge and have a charger fault after 30 minutes is over-ridden by the supplemental current, which may be a safety concern.

  • 0 in reply to Charles Mauney
    TI__Mastermind 32370 points

    Hi Charles,

    Thank you for replying back to this

    The customer tried it with the bq24087, and it works just fine, he did not find any issues. But the customer prefers to use the bq24086 instead because of the NTC input.  

    What are the differences between the bq24086 and bq24087?

     

    Also i suggested to the customer to do some testing on the STAT pin and he comes up with following results:

    No GSM:

    STAT1: low

    STAT2: high

    Charging current 440mA

    With GSM:

    When charge does not start: STAT1 and STAT2 are floating

    When charge starts, STAT1 low, STAT2 high

     

    I looked at the datasheet on page17 table 03, when STAT1 and STAT2 are HZ (OFF)>>>battery short/fault absent. And when STAT1= off and STAT2= on >>> done (termination enabled only). Unless the ON means the LED in ON and the pin is LOW and most likly that is the case...

    What is done mean here? I think it indicates that the battery is fully charged but the state of the battery is under charged!

     

    The suspected scenario is on the 100uF cap on the front end of the GSM. As the GSM plug into the charger it draws a lot of current to charge the cap and the charger detects a short circuit.

    But the question, why this is happening only on the bq24086 not on the bq24087?

     

    we are still debuging, any comment on this is appreciated.

     

     

     

    Thank you again for your help on this!

    Tahar

  • 0 in reply to Tahar Allag
    TI__Mastermind 31120 points

    Problem solved.  Once the schematic was received, it was noticed that the TS pin was pulled up to OUT instead of Vin.  The TS pin being out of range caused the issue.