• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Answers 1 answer
  • Subscribers 63 subscribers
  • Views 532 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.

BQ24610: BQ says it's charging, even after disconnecting AC adapter

M K
Prodigy 50 points
Part Number: BQ24610

Hi All!

Please take a look at my particular problem with weird behaviour of my circuit.

I have a 6 cell Li-ion package. My target charging voltage is set to 24.66V (with VFB pin), fast charge current set to 1.06A (with ISET1) and termination current to 0.15A (with ISET2).

After plugging the battery all seems fine: STAT1 is OFF (high), PG is OFF (high). After connecting my power source (26VDC, but after schottky diode charger really sees ~25.5VDC) still everything is as expected: PG goes ON (low), STAT1 goes ON (low), STAT2 goes OFF (high) - charger reports that it is indeed charging. I can see it in measured current (~100mA flowing INTO the cells).

Now what is wrong: I disconnect the power source and charger still reports charging - STAT1 stays ON (low), PG stays ON (low)! Basically BQ chip is charging my battery with it's own energy. Why it did not notice that power adapter was disconnected? Simple - because on VCC pin it still sees voltage higher that on SRN ( 23.8V vs 23.0V). Where did this higher voltage came from? It seems that a DC-DC boost converter suddenly emerged, consisting of 2 mosfets and a inductor. And it is BQ chip that drives them (checked on HIDRV and LODRV pins).

Charger seems to be stuck (latched) in this state. I can manually restore 'not charging' state by eg. applying significant load to output, or by shorting out thermistor, and hopefully by shorting CE to GND.

Clearly I missed something fundamental here. Let me show you my schematics:

Please help me understand why this is occuring. Or at least give some hint how to patch this problem. This is 2nd prototype of this board, well into the project and I somehow didn't notice this weird behaviour before. Is BQ24610 even suitable for my product? I'm starting to pull my hair...

Cheers!

Mateusz

  • 0
    TI__Mastermind 32286 points

    Hello Mateusz,

    I recommend following the schematic checklist:BQ24610_BQ24650_Schematic_Checklist (15).pdf

    You don't have a BATFET, If you are using a power path you need a BATFET.

    When the adapter is removed, the system waits until VCC drops back to within 200 mV above SRN to switch from the adapter back to the battery. The break-before-make logic still keeps 10 μs dead time. The ACDRV is pulled up to VCC and the BATDRV pin is set to ACN-6V by an internal regulator to turn on P-channel BATFET, connecting the battery to the system.

    Since there is no BATFET, There is no wait time.

    Best Regards,

    Christian.

  • 0 in reply to Christian Moyer
    Prodigy 50 points

    Hi Christian,

    Thank you for your quick response! I run through this checklist and only deviations I saw, is that I have a schottky diode from Vbat to Vsystem. So I soldered it out and re-run all the tests with the same effect.

    Indeed I don't have a BATFET, by design. I think the main issue here is boiled down to this:

    When the adapter is removed, the system waits until VCC drops back to within 200 mV above SRN

    Well, it never drops. What I mean by that:

    - connect battery - BQ sees VCC = 21.97V

    - connect power adapter - BQ sees VCC = 25.29V

    - disconnect power adapter - BQ sees VCC = 22.75V

    So as I understand it, charger doesn't even know that is should react to it.

    I was tinkering some more, tried to lower the regulated battery voltage with VFB resistor divider. I lowered it by ~1V, and it seemed to start working correctly, i.e. this PG signal actually dissapeared after disconnecting the power adapter. But then I switched to another battery pack, which was slightly more discharged, and problem was there again - charger was hang up 'charging', even without power adapter.

    One more thing I noticed:

    - when charger is actually charging, with adapter connected, mosfet keys are switched with frequency ~568kHz

    - when charger is not really charging, with adapter disconnected, mosfet keys are switched with frequency ~186kHz, about 3 times lower

    Maybe this is an indicator of some fault?

    Cheers!

  • 0 in reply to M K
    TI__Mastermind 32286 points

    Hello Mateusz,

    Is this with a system load attached?

    Do you see any difference with/without system load?

    Best Regards,

    Christian.

  • 0 in reply to Christian Moyer
    Prodigy 50 points

    Hi Christian,

    I observed this problem with no load (device sleep, current in tens of uA), and with low load (device on, but not actively working, ~150mA).

    But: when in this 'hung-up' state, when charger is latched in charging state, when I apply significant load (stepper motor), charger 'un-latches', and returns to proper state (PG off, STAT1 off, STAT2 off).

    Hope this is helpful.

    Side note: I tried to replicate problem on EVM kit, but accidentally fried it. Will try again when new kit arrives.

    Cheers!

  • 0 in reply to M K
    TI__Mastermind 32286 points

    Hello Mateusz,

    M K said:
    Side note: I tried to replicate problem on EVM kit, but accidentally fried it. Will try again when new kit arrives.

    Okay, Please let me know the results of your testing with the EVM kit.

    Best Regards,

    Christian.

  • 0 in reply to Christian Moyer
    Prodigy 50 points

    Hi Christian,

    I took EVM kit (schematics for reference here) and reworked it so it is resembles our own design. Quick note: EVM schematic has a lot of components, but those without values are DNP). What I did:

    - disconnected ACDRV and BATDRV control paths,

    - Q1 and Q5 replaced with our schottkys (D1102, D1103 respectively)

    - shorted Q2 and R9

    - replaced step-down coil (L1, 6.8uH) with our own (22uH)

    - replaced all res. dividers to match exactly our design (ISET1, ISET2, VFB)

    - tied ACSET to VREF

    - removed some of 10uF capacitors, so charger sees rougly the same capacitance of 'SYS' node as in our design

    - replaced bootstrap diode D1 (ZLLS350) with our diode (SS14), also, wired it so it is connected directly to BTST pin, just as in datasheet fig. 24 and in design checklist (EVM design differs from datasheet in this manner)

    Basically, EVM altered in this manner should be electrically identical to our board. And yet, EVM works correctly, i.e. charger correctly recognizes disconnection of power adapter, and turns of PG signal. Therefore I cannot replicate our problem with EVM.

    We are debating right now on how to proceed, but if you have any idea on what to check, test, measure, please let me know.

    Cheers!

  • 0 in reply to M K
    TI__Mastermind 32286 points

    Hello Mateusz,

     

    Another thing to check is to confirm that you followed the PCB layout guidelines:BQ24610 Layout guidelines

    Best Regards,

    Christian.

  • +1 in reply to Christian Moyer
    Prodigy 50 points

    Hi Christian,

    I found what I believe was the root cause of problems, although I cannot really explaine the mechanics.

    It turns out there was a packaging error at our supplier, and what supposed to be a current sense 10mOhm resistor, actually was 100mOhm resistor. I only discovered it while trying to manually assemble battery charging part of our PCB and noticed the label sticker with wrong value.

    So basically both ISET1 and ISET2 was set to value x10 lower. I don't understand how set current values are connected to improper charging signalling, but I guess it somehow is.

    After further testing there are still some issues with our circuit: sporadic short transients of PG signal, and supprisingly early switch over from fastcharge current regulation phase to voltage regulation phase (should happen at 24.6V, happens event before reaching 23.5V). But for these, I will start another thread for clarity.

    Thanks for all your help and support!

    Cheers!