• State
  • Locked Locked
  • Replies 2 replies
  • Subscribers 63 subscribers
  • Views 1602 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.

BQ24250 without Battery

Michael Allwright
Prodigy 240 points
Other Parts Discussed in Thread: BQ24250

Hi,

I'm seeing somewhat strange behaviour on the BQ24250 without a battery, that I'm trying to get a better understanding of. My main aim at the moment is to get a voltage to appear at the Vsys pin without a battery. I have attached my circuit as a reference.

When the system is first booted VSYSTEM is around 3.6V and the boost converter U4 is disabled. As expected, this provides approximately 3.6V at the input terminal (pin 19) of the BQ24250. This is below all values of the BQ24250 VPDM (min. 4.20V in Host mode, 4.36V default in SA mode). In any case LDO is powered at around 3.6V and the I2C interface is active. Using the microcontroller, I then set BQ24250_VREG_IN to high, and verify that input to the BQ24250 (as well as LDO) rises to around 4.5V.

I then put the device into host mode by issuing a watchdog reset and attempt to configure it. What I'm seeing is really unusual however. When I write 0000xxx1 to register 2 (at 0x01) (100mA ILIM and HIZ mode enable) the device accepts the configuration and the value is updated. When I reset the watchdog, I can verify that the watchdog fault and enable flag in register 1 (0x00) are set to 0 and 1 respectively.

However when I try to disable HIZ mode and set the input limit to 500mA - writing 0010xxx0 to register 2 (0x01), the value that I read back is x110xxx0, HZ_MODE disabled but input current limit set to "External ILIM resistor" mode! Furthermore, with these settings, it seems that the watchdog will no longer stay enabled. Reading register 1 (0x00) immediately after the reset of the watchdog (0x04, WD_EN = 1), returns 0x04 as expected. However rereading this register approximately 150ms later gives 0x00. No fault, watchdog disabled?

An additional note, I have attached an additional 10k resistor between BQ24250_TS and ground, as the 10k thermistor is not present in the circuit when the battery is disconnected - this confuses the BQ24250.

My questions are as follows:

1. What is the behavior of the BQ24250 with no battery, when VIN is below VPDM but just above the maximum UVLO voltage of 3.5V. Should the current draw stay around 0?

2. What is required / what conditions must be met in order to see a voltage at VSYS without a battery connected?

3. Why would the device not allow me to enable the watchdog when HZ_MODE is disabled, and ILIM is set to 500mA?

Please let me know if you need additional information such as oscilloscope traces or register dumps after given configurations.

All the best and thank you for your help in advance!

  • 0
    TI__Expert 7730 points

    Hello Michael,

    To answer your questions:

    1. When VIN is below VIN_DPM, there should be no current drawing from the input.

    2. VIN needs to be above VIN_DPM to see the stable signal at VSYS. Note without the battery, VSYS will be a square wave between 3.9V and the programmed Vbat_reg. Also, when short VDPM to ground, the default VIN_DPM is 4.68V instead of 4.36V. The datasheet is being updated to reflect the change.

    3. The device should allow you to program the ILIM to 500mA no matter the HIZ_MODE is enabled or disabled. In fact I just used thebq24250 EVM and verified it on bench. What you saw looked like the registers were reset to default values after the watch dog timer expired. Within ~50 seconds after you write to the register, if there's no other command is sent to the host, the watch dog timer will expire and all register values will reset to default. For bq24250, the WD_EN is default to 0 and the ILIM is default to external. However, it doesn't make sense if the value changed after 150ms. Can you disable watch dog timer by writing WD_EN to 0 and see if you can program ILIM to 500mA and hold the value for more than 1 mins?

    Thanks,

    Wenjia

  • 0 in reply to Wenjia Liu
    Prodigy 240 points

    Hi Wenjia,

    I added some code to my test-bench today that would trigger the watchdog reset every 10secs to ensure no timeouts were occurring. I also checked several nodes with the oscilloscope and found the source of the register(s) resetting -  Whenever I enable the BQ24250 with functional parameters (VDPM 4.2V, Not HIZ) the inrush current on the IN pin seems to crash the boost regulator U4 (shown in my schematic above).

    Booted
    SET_IN_HIZ
    2682ms - I2C: TX(0x6A) {0x01}
    2684ms - I2C: RX(0x6A) {0xEC}
    2684ms - I2C: TX(0x6A) {0x01, 0x0D}
    2693ms - I2C: TX(0x6A) {0x01}
    2699ms - I2C: RX(0x6A) {0x8D}

    BQ24250_IN ON // enables the boost converter

    SET VDPM TO 4.2V

    4450ms - I2C: TX(0x6A) {0x04}
    4452ms - I2C: RX(0x6A) {0x0A}
    4454ms - I2C: TX(0x6A) {0x04, 0x08}

    RST WATCHDOG

    10002ms - I2C: TX(0x6A) {0x00, 0x40}
    10004ms - I2C: TX(0x6A) {0x00}
    10006ms - I2C: RX(0x6A) {0x40}
    Register(0)
    11171ms - I2C: TX(0x6A) {0x00}
    11171ms - I2C: RX(0x6A) {0x40} // Watchdog enabled, no errors

    Register(0)

    12546ms - I2C: TX(0x6A) {0x00}
    12548ms - I2C: RX(0x6A) {0x40} // Watchdog enabled, no errors

    SET_IN_L100 // Set 100mA mode (no HIZ)

    16633ms - I2C: TX(0x6A) {0x01}
    16635ms - I2C: RX(0x6A) {0x8D}
    16637ms - I2C: TX(0x6A) {0x01, 0x0C}
    16646ms - I2C: TX(0x6A) {0x01}
    16652ms - I2C: RX(0x6A) {0xEC}

    RST WATCHDOG

    20002ms - I2C: TX(0x6A) {0x00, 0x40}
    20004ms - I2C: TX(0x6A) {0x00}
    20006ms - I2C: RX(0x6A) {0x40}
    Register(0)
    21698ms - I2C: TX(0x6A) {0x00}
    21700ms - I2C: RX(0x6A) {0x00} // Watchdog disabled, no errors

    Register(0)

    23169ms - I2C: TX(0x6A) {0x00}
    23171ms - I2C: RX(0x6A) {0x00} // Watchdog disabled, no errors


    As can be seen above (and similar experiments), the moment the BQ24250 is taken out of the HIZ mode, the registers are reset. Monitoring IN (pin 19) we see this sudden drop exactly occurring as the device is set to 100mA input mode.

    CH1 is Vsystem (stable), CH2 is the output of the boost converter and the input to the BQ24250 (unstable when taken out of HIZ mode)

    Since this thread has somewhat diverged from the origin question / topic, I am continuing this discussion in a new thread at: http://e2e.ti.com/support/power_management/battery_management/f/179/t/375305.aspx