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.

BQ24308: Datasheet typo and other purposes of IC

Part Number: BQ24308
Other Parts Discussed in Thread: TINA-TI

Hi,

1. For your datasheet here - https://www.ti.com/lit/ds/symlink/bq24308.pdf - is there a typo on page 14? I believe the PMOS is flipped, since the current as shown is going through the Drain to source direction.

2. Also, is your product compatible for the cases where a power supply may not be always connected (like in a battery scenario)?

3. Furthermore, can this product serve as a lithium cell protection circuit? Or should a separate protection IC be used? Any suggestions for such ICs are welcome.

4. For your example with the external PFET for reverse polarity protection, what would the voltage drop at the output be if the input power source is 5V? Seems your application example diagrams plot VIN (the input to the IC) not the input power source - so I wanted to better understand this.

5. Is there a spice model of this part for TINA-TI?

Thanks.

  • Hi Aziz,

    I think the PMOS is placed in the right direction. It is to protect in the case where VIN is connected with reverse polarity.

    If the VIN is not supplied, the output of the device will switch off because VIN is below UVLO.

    A separate protection IC should be used for a battery protection circuit. There are a few different single cell battery protection ICs shown here: https://www.ti.com/power-management/battery-management/protectors/overview.html

    There is no spice model available for this device.

    Best regards,

    Matt

  • Hi Matt,

    Maybe I'm missing something - Since the current of PMOS should flow from source to drain (to protect against the body diode from turning on), then how can the PMOS source be tied to the VIN pin? Can you elaborate more on this?

    Thanks.

  • Hi Aziz,

    When the VDC input is negative, the FET body diode is off, reverse biased.

    When the VDC input is positive, when it is first connected the FET body diode allows the IN pin to receive a voltage from VDC of the adapter.

    When the device has recognized the power and set PGATE "low", the FET QEXT turns on and VDC is connected to IN.  

    Best regards,

    Matt

  • Hi Matt,

    Thanks for clarifying - I made a simple PCB with your component but am having trouble replicating the expected behavior. My original design had the Pfet in the wrong direction but I reversed it during assembly. I noticed that when pfet was the in the incorrect direction, (as shown below), the circuit did behave correctly only when the voltage was positive. When it went negative, the power supply detected a short (likely due to the body diode of the fet and likely in agreement with your characterization above). 

    When I corrected the pfet polarity, I noticed that my output voltage (VBUS_5V was always hovering around 0V regardless of whether the power supply input (TP3) was positive or negative.

    Specifically, for the correct pfet direction, when the power supply input was 5V, I saw:

    VDrain = 4.92V

    Gate: 1.73V

    Source: 4.92V

    And VOUT/VBUS_5V [TP4]: ~0V.

    When the power supply is switched to -5V, we see:

    VDrain: -4.92V

    VGate: -0.56V

    VSource: -0.5V

    VOUT/VBUS_5V[TP4] = -0.78

    Note that VDD below is my LiPo battery which I disconnected in several tests but saw similar results.

    I am experimenting with ZVP3306A transistors but also tried SI2307CDS-T1-GE3.

    It appears that when the pfet is oriented in the correct direction, reverse voltage is blocked (which is good). However, in the forward direction, the voltage is not appearing on the output terminal of the BQ IC even though the correct voltage is on the input terminal (so that suggests the transistor is working correctly and BQ is controlling it correctly). I tried this with several of my PCBs so this likely some other design issue.

    What could I be doing wrong? 

    Thanks for your help!

  • Hi Aziz,

    I am not sure why the output would be disabled in this case. Is there any chance there is an over-current condition on the output (this would also disable the output)?

    Matt

  • Hi Matt,

    In short, there is no over-current condition occurring. I've been doing some experiments have been unable to replicate your results with your schematic. I need to determine whether your product will work for my circuit or go with another one - so any advice would be greatly appreciated.

    My goal is to get your component working with an external PMOS so that we get the additional reverse polarity protection. 

    Here are some experiments repeated with multiple boards:

    1. The most peculiar observation is that the PGATE terminal seems to be critical - if it is floating (even in the case of no external PFET), the BQ component does not activate, contrary to your datasheet.

    2. I see the same behavior as (1) when I connect the PFET in the correct configuration shown in your schematic. 

    3. If I put the PFET in correct direction and put either a 0ohm or 1K ohm resistor between gate and power supply (not in agreement with your datasheet), the component works for positive voltages. But, the reverse polarity protection is not working.

    4. I was curious and experimented with a BJT and found interesting results. If instead I use a PNP (BJT) transistor (TIP30CGOS), and tie the collector to the power supply, base terminal to the PGATE pin, and the emitter terminal to the BQ Input voltage terminal, the circuit works correctly (even when applying a negative input voltage). But, the output terminal of BQ is 4.3V, which would be too low of a voltage drop. So even though this observation is interesting, it's not going to work for me. 

    5. If I remove the transistor completely and short together PGATE BQ terminal, VIN BQ terminal, and 5V input supply, the circuit does work when the input supply is positive, but obviously not with negative voltages.

    6. My most successful experiment used a Schottky diode (1N5817) (instead of a transistor) and tied the PGATE BQ terminal, VIN BQ terminal, and diode cathode together. It works in forward and reverse voltage directions. However, voltage drop to the output terminal of BQ is large (VOUT=4.58V) due to the voltage drop on the diode from a ~ 100mA charging current. So, this is an ok solution but not preferred.

    Can you elaborate on why we are seeing these strange scenarios (specifically 1, 2, 3) and how to replicate your correct results (including what FETS have worked).

    I am spending a lot of time guessing what is wrong and need to make a call on whether we can proceed with this component in our design or not.

    Thanks.

  • Hi Aziz,

    I think the Si2343DS was listed as the PFET on the EVM back when it was still available. I am not sure I understand the results of your testing but I will take a closer look tomorrow and see if I can provide an explanation.

    Best regards,

    Matt

  • Hi Matt,

    Let's simplify this and remove the external FET. I am unable to replicate your result if I leave PGATE floating (i.e. without tying the PGATE to the supply voltage). Can you clarify why is that the case?

  • Hi Aziz,

    Have you seen this document? https://www.ti.com/lit/ug/sluu291b/sluu291b.pdf  See Page 12 where is shows which PFET is used. It uses a 100k pullup resistor on the PGATE pin also. I am not sure what happens if PGATE is left floating (the designer for this device is no longer around, so it is difficult to look into the architecture). 

    Regards,

    Matt