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.

BQ25713: HIDRV pin not zero when no DC input is attached

Part Number: BQ25713
Other Parts Discussed in Thread: EV2400, BQ25710

Hi forum, 

Thank you in advance for your time. 

My design with BQ25713 encounters an issue. When there is only battery attached to the system and no register is set, I measured a voltage at HIDRV1 and HIDRV2 with behavior as shown in Figure 1.    

With no DC input connected, the same voltage is transferred to Vbus which caused the charge status reg to show Adapter Present. 

If I uncheck the EN_LWPWR, disabling the low power mode, HIDRV1 and HIDRV2's behavior change to as shown in Figure 2 and Figure 3. 

  Figure 1 HIDRV1 and HIDRV2 with only battery attached

        Figure 2 and 3, HIDRV1 and HIDRV2 reset period change from 3.58s to 5.24s for different PCBs. 

If I enable HIZ mode, the voltage on HIDRV and Vbus stays around 3.5V. 

Comparing to EVM, HIDRV and VBUS should be zero when there is no DC input. 

I also attached the screen capture of my schematic and layout. 

Please feel free to ask if you spot any error or need more information. 

Thank you

Jerry 

  • Hi Jerry:
       

        Would you please provide detail information below for further analysis:
      

        1,What is the VSYS and VBAT voltage respectively for these test?

        2, How much load do you add on the system?

        3, Do you measure out the Vbus voltage when DC input is disconnected?

    Regards

    Shishuo

  • Hello Shishuo,

    1,What is the VSYS and VBAT voltage respectively for these test?

    ans : In Figure 6, yellow being HIDRV1 and blue being Vbus voltage. Here, low-power mode is enable from power on.  VBAT voltage is stable at the 10.3V. 

     

       

     Figure 6

     Figure 7 VBAT is stable at 10.3V

    2, How much load do you add on the system?

    ans: I tried loading from a few mA to 2A. the shape of the signal that not change. 

        3, Do you measure out the Vbus voltage when DC input is disconnected?

    Ans: I don't measure out the VBus voltage, but I read the reg AC STAT with a mcu as a feedback if the DC is connected. It seems when Vbus >3.2V, AC STAT consider there is a input source.

    Please let me know if you need more info

    Best regard 

    Jerry 

  • Hi Jerry:
      

         The Vbus voltage looks weird with so obvious ripple. You can check your measurement setup to make sure it is grounded properly. 

         For Figure 7, what is the CH2 signal? it looks much different with Fig6.

         What is the work mode of this test? I need to exactly steps how do you come to this waveform from your hardware startup. These will be helpful.

         Note please include switch node information for all the future waveform measurement. HIDRV1 should come with SW1 and HIDRV2 should come with SW2.

    Regards

    Shishuo

  • Hi Shishuo, 

    SW1 looks the same as HIDRV1 in this case, and I connected VBUS and SW1 with a 1 ohm resistor in the schematic. This follows the EVM as JP7 is connected with P1 and P2. So as SW1 toggling, Vbus follows. I confirm that the grounding of my measurement is correct. 

    For figure 7, CH2 is VBAT. It stays at 10.3V with no affect on HIDRV2. 

    The device should be in boost and buck mode as power on. I have not set any register yet. I connected the battery and start reading the registers with EV2400. 

    This behavior, HIDRV2 and SW2 not zero,  is not observed from EVM. 

    I wonder if there is any mistake I made in the schematic creates this issue. 

    Best regard 

    Jerry  

  • Hi Jerry:
       

         Below is our TI EVM  schematic:

         JP7 P1 connected to VIN through 1ohm resistor.

         JP7 P2 connected to IC Vbus Pin,

         JP7 P3 connected to buck boost input terminal.

         Please remove 1 ohm resistor between VBUS and SW1 and have a try again. Let me know if it works normally.

    Regards

    Shishuo

  • Hi Shishuo, 

    Please note on my schematic that the 1 ohm resistor is connecting between DC input and VBUS, not between VBUS and SW1. There will be be power to the charger once I remove it. 

    To follow the EVM schematic strictly, I tried to remove the 1 ohm resistor, then directly connect VBUS to buck boost input terminal (drain of Q1 ). 

    But It dose not seem to do anything to the situation. 

    Best regard

    Jerry 

  • Hi Jerry:
       

         Sorry for the misunderstanding. It should be fine if 1ohm is between Vbus and Vin.

         Would you please include the SW information in the waveform for further analysis? 

    Regards

    Shishuo

  • Hi shishuo,

    Sure, let me update the figure as soon as I get back to office on Monday.

    Thank you

    Jerry

  • Hi Jerry:
        

         Thanks Jerry, please probe VIN, SW1, HIDRV1,BTST1 waveform under both no system load and 2A system load for analysis purpose. 

    Regards

    Shishuo

  • Hi Shishuo, 

    The updated waveforms have been attached.

    Ch1 yellow is BTST1, CH2 blue is Vin, CH3 purple is SW1, CH4 green is HIDRV1.   

    there is no different behavior observed between zero load and 2A load. 

    And the frequency of the event, shown in the BTST zoom out, is the same for zero load and 2A load. 

    Best regard 

    HIDRV no zero Photos.zip

    Jerry  

  • Hi Shishuo, 

    Is there any update on the issue?

    I have 2 designs with BQ25713, one with number AG4J on the package which works fine. the other one is AE61. 

    I dont know if the serial number has any connection with the malfunctioning.

    Would this issue cause by assembly? 

    Looking forward to your reply. 

    thank you 

    Jerry  

  • Hi Jerry:
       

         From the waveform, it looks like your HIDRV1 is just any offset 1V offset adding over the SW1 which means that you HIDRV1 is not driving normally. Do you try to replace a new part to see if this issue consistent? You may also need to replace the FETs also.

    Regard

    Shishuo

  • Hi Shishuo, 

    Sorry for the confusion on the photo i provided. There is no offset between SW1(purple) and HIDRV1(green) if you look at the CH3 high and CH4 high on the right side. 

    I offset the display so I can show you both of the waveforms.  

    I have not replace any part since it is very difficult for this packaging with the thermal pad below. 

    Thank you 

    Jerry 

  • Hi Shishuo, 

    Just want to do some update. I replaced the part but the problem persist. 

    It seems to be there is some charge accumulated over the path to cause the voltage build up at the input. 

    I place a 10k ohm resistor to between the input and ground, the voltage drops to 0.8V stable, as well as HIDRV1, VDDA is not being turned on. 

    The accumulated current is about 80~50uA, Does this value ring any bell?

    Best regards, 

    Jerry  

  • Hi Jerry:
        

        Thanks for letting me know. Do you have any other silicon on the system besides our BQ25710? I am trying to figure out the leakage current source. Under low power mode(disable HIDRV1 and VDDA), our part should not contribute any leakage current you mentioned as charge accumulation. Thanks.

    Regards

    Shishuo

  • Hi Shishuo, 

    There is no other IC besides BQ25713. In fact, I only connect one diode to the output of BQ25713 which ensures nothing flows backward, and the charge accumulation persists.

    Under low power mode, HIDRV1 should not be on at all. The only time I have seen HIDRV1 and HIDRV2 on are pass-through mode, but the IC does not power up in PTM neither. 

    What I observe at start up is that the input voltage first rises up, then the VDDA starts to turn on and HIDRVs begins to toggle. 

    last, I set BQ25713 in HIZ mode in my previous PCB which works fine, HIDRV1 and HIDRV2 is 1V, the input stays at around 1V as well. 

    But in the new version, HIDRV and input quickly rise to 5V and 7V respectively. 

    Hope this HIZ comparison would give you more information. 

    So far, I had

    1. replace the IC but the issue persists,

    2. compared the schematic and layout of both functioning and malfunctioning version of PCB,

    3. probe the resistance on both versions of PCBA and bare PCB respectively but found no short, open or big discrepancy 

    Do you think if there one or more peripheral capacitors/resistor with wrong value would cause this issue?  

    Thank you 

    Jerry 

  • HI Shishuo, 

    Finally, I found there is a protection diode at the very beginning of the the PCB, it has a 1mA reverse current. 

    By replacing that diode, it fixed the issue. 

    Thank you for your help, much appreciated. 

    Jerry