• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 64 subscribers
  • Views 723 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.

BQ25120: Power output issue, Not stable on battery

Johann van Niekerk
Prodigy 50 points
Part Number: BQ25120


Hello

I have designed a small wearable with the BQ25120AYFPR and BQ27421YZFR-G1A. The device works as it starts up at 1v8 and can be programmed and charges the battery with default hardware settings.. 

I have Three Issues 

1) The communication is very fragile, the comms will fail 3- 5 times and had to write a retry driver to continuously poll the device.. I suspect this might be i2C interference related but Need to mention it for the second problem. 

2) The SYS_VDD and GPS_VDD do not want to work at any voltage above 3V( Battery at 3.92) it becomes extremely noises and eventually drops and restarts at 1v8. When these outputs are used with 5v supply they are more stable but still drop out when programmed to go over to 3.0v( Target 3.2v). 

3) The GPS can run at 2.7v so I proceeded to test the GPS with the output set to 2.9v. Issue when I have the device plugged into USB everything works as described above but when I unplug the USB the device does not seamlessly failover to Battery it resets and has to wait 10-15 seconds for MCU to reconfigure charger/output before the voltage is available at 2.9v. 

Max current needed in bursts is 150mah but continues is no more than 40mah

Below is the setup 

  • 0
    TI__Genius 12485 points

    Hi Johann,

    The schematic on the BQ25120A looks ok. I have a few questions to help debug this-

    1. Communication very fragile- What is the pull-up resistor being used on the system? Have you tried lowering the resistors to make it a bit stable? Also are the pullups being tied to VSYS?

    2. On the noise- do you happen to have a scope capture of this behavior? Have you taken a look at PMID which is the input of the DC-DC converter on battery only?

    3. This very much seems to indicative of a transient response and how the battery responds to it. Have you looked at battery pin (close to the device) to see if it dips when a load is applied? To diagnose this- i would look at PMID/ VBAT and VSYS when this behavior happens. Also you could place a large cap on VBAT pin  of BQ25120A to see if this behavior disappears indicating either a higher Z battery or connection. What is the resistance of the fuse? 

    I am also adding my colleagues that support the BQ27421 to see if they have any thoughts as well. 

    Regards,

    Gautham

  • 0 in reply to Gautham
    Prodigy 50 points

    Hi Gautham 

    Thank you for the quick response please see comments in blue 

    1. Communication very fragile- What is the pull-up resistor being used on the system? Have you tried lowering the resistors to make it a bit stable? Also are the pullups being tied to VSYS?

    They are tied to SYS_VDD via 10K have also tried 4K7, I did not try weaker than 10K should I ? there is about 5 devices on that Bus but all can be seen and talked to but its only the BQ charge chip that gives comms issues

    2. On the noise- do you happen to have a scope capture of this behavior? Have you taken a look at PMID which is the input of the DC-DC converter on battery only?

    I did put it on the scope, and could not see any obvious oscillation, I did not get a screenshot but will try and get one. ( Hard with the COVID lockdown ) is the PMID tied internally to the DC-DC converter? I only have it bypassed externally as I never intended to run my output voltage higher then batt voltage 

    3. This very much seems to indicative of a transient response and how the battery responds to it. Have you looked at battery pin (close to the device) to see if it dips when a load is applied? To diagnose this- i would look at PMID/ VBAT and VSYS when this behavior happens. Also you could place a large cap on VBAT pin  of BQ25120A to see if this behavior disappears indicating either a higher Z battery or connection. What is the resistance of the fuse? 

    I have replaced the battery with a 3.6v power supply for a short time just to test if it happens on the supply and the same happens. With regards to the VBAT pin, Yes this is a concern, on the PCB layout I did not have space for a cap since the track from SRX out from BQ27421 to BQ25120A is 2mm in length. I have changed the input cap to the BQ24 from a 1uF to 10uF to try and make up for it. the fuse rating is 2.8Ohm but iv removed the fuse and replaced it with solder bridge to test. Same issue. 

    I am reviewing the design to get new boards made with other issues detected so just want to get to the most likely cause of this issue and address it in the next revision. the Cap on the VBAT pin is on the list.

    More Info, The battery was manufactured to out spec and should be able to deliver 150mah continuously.  

    I am also adding my colleagues that support the BQ27421 to see if they have any thoughts as well. 

  • 0 in reply to Gautham
    Prodigy 50 points

    The BQ25120 SYS (SYS_VDD) and LDO (GPS_VDD) output voltage is set to 2.9V.

     

    With the LDO output off the SYS voltage is clean. On turning it on we get the following:

    Battery voltage in Blue. SYS_VDD in Green

     

    Battery voltage in Blue. LDO (GPS_VDD) in Green

    LDO (GPS_VDD) will switch after a few seconds, presumably due to chip watchdog reset.


     

     

     

  • 0 in reply to Johann van Niekerk
    TI__Genius 12485 points

    Hi Johann,

    Thanks for the scope captures. I have a few suggestions below- 

    1. Communication very fragile- What is the pull-up resistor being used on the system? Have you tried lowering the resistors to make it a bit stable? Also are the pullups being tied to VSYS?

    They are tied to SYS_VDD via 10K have also tried 4K7, I did not try weaker than 10K should I ? there is about 5 devices on that Bus but all can be seen and talked to but its only the BQ charge chip that gives comms issues

    GR: No the 4k7 would be the right direction to strengthen the comms. We could take a quick look at the I2C lines to see if there is anything suspicious. Also I am assuming /CD pin goes high before I2C transactions on battery only mode and MRREC actions are not inadvertently triggered. 

    2. On the noise- do you happen to have a scope capture of this behavior? Have you taken a look at PMID which is the input of the DC-DC converter on battery only?

    I did put it on the scope, and could not see any obvious oscillation, I did not get a screenshot but will try and get one. ( Hard with the COVID lockdown ) is the PMID tied internally to the DC-DC converter? I only have it bypassed externally as I never intended to run my output voltage higher then batt voltage

    GR : That is correct. the DC-DC gets it input from the PMID pin. PMID is an OR between VIN or VBAT. 

    3. This very much seems to indicative of a transient response and how the battery responds to it. Have you looked at battery pin (close to the device) to see if it dips when a load is applied? To diagnose this- i would look at PMID/ VBAT and VSYS when this behavior happens. Also you could place a large cap on VBAT pin  of BQ25120A to see if this behavior disappears indicating either a higher Z battery or connection. What is the resistance of the fuse? 

    I have replaced the battery with a 3.6v power supply for a short time just to test if it happens on the supply and the same happens. With regards to the VBAT pin, Yes this is a concern, on the PCB layout I did not have space for a cap since the track from SRX out from BQ27421 to BQ25120A is 2mm in length. I have changed the input cap to the BQ24 from a 1uF to 10uF to try and make up for it. the fuse rating is 2.8Ohm but iv removed the fuse and replaced it with solder bridge to test. Same issue. 

    GR: I have another recommendation- can you write the BUVLO bit to be 2.2V? As expected- we see a sharp dip on BAT below 3V which could be triggering BUVLO and indication that we shut down to prevent deep-discharge. Reg- 0x09, Bits B2 to B0 - 111. Also on the fuse- I am assuming both the power and GND connections are removed. 

     

    I am reviewing the design to get new boards made with other issues detected so just want to get to the most likely cause of this issue and address it in the next revision. the Cap on the VBAT pin is on the list.

    More Info, The battery was manufactured to out spec and should be able to deliver 150mah continuously. 

    GR: Do you have a datasheet for this battery? Do you have a typical impedance of it?

  • 0 in reply to Gautham
    TI__Mastermind 21590 points

    Johann,

    Checking if you have any updates here?