• State
  • Locked Locked
  • Replies 8 replies
  • Subscribers 62 subscribers
  • Views 429 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.

Original question:

BQ24070: The chip thinks there is a battery connected and in fast charge mode (According to STAT bits) even when no battery...

BQ24070: The chip thinks there is a battery connected and in fast charge mode (According to STAT bits) even when no battery is connected. More insights .... and ISET1 may be damaged?

Jerry Molnar
Prodigy 210 points

Part Number: BQ24070

Tool/software:

Hi Garrett,

This is a follow up to my previous thread.
I still have the original board with the fault. Now we are bringing up pre-production units and and now I've seen this problem appear in one of those...so a red flag is started to wave in my head!

I did a detailed test yesterday and have some new info.
Some of what I saw yesterday contradicts what I reported in the previous discussion about a month ago. Let's go with this new info (I can't explain the discrepancy but it may have to do with having the battery inserted at power up of the chip or not)

To recap what the failure modes are:
- With no battery connected, the status bits indicate that the chip is in fast charge mode. The BATT pins are static and at 4.2V - not cycling as per 8.4.3.5 of the datasheet which states.

- I ran the test you suggested - I took a deeply discharged battery, ~2.3V, and had that connected to the BQ24070. The system load was not present (Switched off), and a 5V bench supply was powering the BQ24070.

The charger behaved correctly - started at a charging current of ~110ma and the status pins showed that the chip was in precharge.
At approx 2.9V battery voltage the charging current went to the desired 1A and the status pins showed the fast charge state.

All was going well until the end.
When the voltage at ISET1 was in the threshold range for turning off the charger (~250mV), it did not turn off the charging current.
The voltage at ISET1 got down below 200mV before I gave up. At that point the charging current was still ~100ma.

So my conclusion was that the mechanism to shut down the charging current based on the voltage at ISET1 was not working anymore....perhaps the circuitry at the ISET1 pin was damaged.

Questions:

We have a 0.1uF cap between BAT (Pins 5 and 6) and ISET1 as per the recommendation in 9.2.2 -
"A 0.22-μF capacitor is connected between BAT and ISET1 to improve operation at low charge currents."

Q1: Have you ever seen battery load transient voltages coupling into ISET1 through this cap?
We do have large (3-4A) current spikes coming out of the 18650 battery which may cause momentary droops in the battery voltage....which if steep enough in terms of being step changes in voltage could couple through that cap.
(Note that these high currents do not pass through the BQ24070 - we have a "high current path" from the battery in parallel with the chip).

I will be doing some lab measurements today on what the load transients may look like on the battery and will report back.

Q2: What is the impact of not using the 0.1uF cap between BAT and ISET1.

Q3 - One more thing I should mention is that in our application we always have the 5V adapter plugged in and when we need to perform an operational cycle, we switch to battery by disabling the BQ24070 and enabling our parallel, high current path for the battery to supply the load.
Can you see any possibility that as the chip is disabled, a negative step voltage occurs on the battery and there is a resulting negative going spike (below 0V) at ISET1?

Thanks and Best Regards

Jerry Molnar

  • 0
    TI__Mastermind 34260 points

    Hi Jerry, 

    Thank you for the great detail regarding this issue you are observing. We are looking into it and will get back to you with an update. 

    Best Regards,

    Garrett 

  • 0 in reply to Garrett Kreger
    TI__Mastermind 34260 points

    Hi Jerry, 

    Please allow me to provide an initial update.  

    Jerry Molnar said:
    Q1: Have you ever seen battery load transient voltages coupling into ISET1 through this cap?

    I do not have knowledge of this sort of issue on this part, but I will also preface the typical use case of this part does not include an external high current path. As you point out current spiking causing momentary change in battery voltage could be a factor.  

    Jerry Molnar said:
    What is the impact of not using the 0.1uF cap between BAT and ISET1.

    The capacitor can be removed. In general a 0.22uF capacitor between BAT and ISET1 is recommended when set charge current is less that 300mA. Given you operate at 1A programmed charge current this capacitor is not necessary.  

    I am continuing to investigate the theory you mention regarding load transients causing negative voltage at ISET1, but moving forward I recommend removing this capacitor between BAT and ISET1 from your design to eliminate any potential risk due to its presence. 

    Best Regards,

    Garrett 

  • 0 in reply to Garrett Kreger
    Prodigy 210 points

    Hi Garrett,

    Thank you for your reply.
    We will plan to remove that capacitor. I had misinterpreted the "useful at low current" as useful when the current is tailing off from the 1A......but I'm still going to feel better if I can find a tangible root cause in the lab as to what caused the failure mode.

    Question:
    Q1: Given that it wasn't switching to "fully charged" when the voltage at ISET1 crossed the ~250mV switching threshold range, are there any other inputs that are "involved" in making that "fully charged" decision?

    Q2: In our system when the high current path is switched on, the battery feeds a boost converter that has an output voltage of 5.5V. I did do some tests with a weakish battery that was being asked to rapidly (FET switched load) go from 50ma load to about 4A on the 5.5V rail. When the battery came under heavy load the voltage dropped some and you can see the expected negative (below 0V) excursion on ISET1. The biggest negative excursion I got was -160mV.
    That is still less than the absolute max pin rating of -300mV, but could that be enough to damage something on the voltage comparison circuit at ISET1?

    It's not possible to see the detail below, but is there a way to send the file (if you'd like to see it). The green trace has the -160mV  transient. 200uS per division. Approx 200mV drop in battery voltage under load (the blue trace at the top).

    Thanks and Best Regards
    Jerry

  • 0 in reply to Jerry Molnar
    TI__Mastermind 34260 points

    Hi Jerry, 

    Please see my comments below. 

    Jerry Molnar said:
    Q1: Given that it wasn't switching to "fully charged" when the voltage at ISET1 crossed the ~250mV switching threshold range, are there any other inputs that are "involved" in making that "fully charged" decision?

    ISET1 voltage is the only input for termination. Just to verify you do pull MODE pin high in your system correct? 

    Further research is necessary regarding Q2. 

    Best Regards,

    Garrett 

  • 0 in reply to Garrett Kreger
    Prodigy 210 points

    Hi Garrett,


    Yes, confirming that MODE is connected to the 5V coming in from the AC-5VDC wall adapter that provides the charging current.

    Best Regards
    Jerry

  • 0 in reply to Jerry Molnar
    TI__Mastermind 34260 points

    Hi Jerry, 

    Thank you for confirming. In that case yes ISET1 voltage falling below typical 250mV termination threshold (min 230mV, max 270mV) is the input for termination. 

    Let me also restate something we discussed in the previous thread. TMR pin being pulled up to VREF is a method to disable termination. Therefore, this is also an input "involved" in making the termination decision. With that said in your schematic TMR is connected to GND via a resistor so TMR is not expected to be the cause of the observed no termination behavior. 

    Did you happen to measure TMR pin voltage during charging on this new unit? Reviewing the original thread you reported that unit had a lower than expected TMR voltage. 

    Best Regards,

    Garrett 

  • 0 in reply to Garrett Kreger
    Prodigy 210 points

    Hi Garrett,

    Re TMR - as you probably saw, the original board has TMR sitting at 0.71 volts...definitely not the 2.5V we expect. I will have to try to measure the more recent failure. It's still in a unit being used by another development team...it may take a few days to get access to it.

    In that previous thread I saw that you mentioned that DPPM could be causing TMR to be messed up. Even though DPPM seemed OK on that unit, could pulling excessive current spikes (e.g. 4-5A) from the OUT pins when there is no 5V input to power the charger (i.e. running purely on battery) cause something in the DPPM system to become damaged?

    Thanks!
    Jerry

  • 0 in reply to Jerry Molnar
    TI__Mastermind 34260 points

    Hi Jerry, 

    Please allow me to try to clarify my point on DPPM from the previous thread.

    On a normal (i.e. non damaged) IC DPPM operation or thermal regulation operation would cause TMR voltage to reduce below expected 2.5V. Note (9) on datasheet page 7 is one place where this is mentioned. In the previous thread I was attempting to say TMR = 0.71V is unexpected because given VOUT = 4.4V the device did not appear to be in DPPM operation. 

    Jerry Molnar said:
    could pulling excessive current spikes (e.g. 4-5A) from the OUT pins when there is no 5V input to power the charger (i.e. running purely on battery) cause something in the DPPM system to become damaged?

    I would not expect this. DPPM operation is specific to when input supply is present. There is no DPPM operation when running purely on battery. Similarly there is no thermal protection for battery discharge state, only when input is present. 

    No rush, but it will be helpful to know TMR voltage on this new unit when available. 

    Best Regards,

    Garrett