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.

BQ77915: Load Removal/Charging to recover from UV fault

Part Number: BQ77915
Other Parts Discussed in Thread: BQ24618

I'm having trouble understanding how the components around the CHG FET are supposed to be selected. In the detailed example, it says load removal is necessary, but then later it says that because attaching a charger is acceptable, a 1-Mohm resistor is ok for R_GS_CHG. However, in the simple example table, it says that a 1-Mohm R_GS_CHG disables load removal for recovery from UV fault. I'm trying to design a system where the battery is a fixed component of the system, so the only way to recover from a UV fault should be attaching a charger.

  • Hi Jeffrey,

    It might be beneficial to walk through the EVM user guide to help you understand this behavior. http://www.ti.com/lit/ug/sluubu2b/sluubu2b.pdf 

    I just ran a quick experiment myself, and I found that with a charger connected to the PACK- and PACK+ pins, I could recover from the UV fault. If you have a load connected instead, you must remove the load to recover. If you would like more details or a step by step description of what I did, please let me know!

    The EVM user guide also features the schematic of the EVM (on page 15). This is the circuit we recommend, so you can use this to help with component selection.

    As for the datasheet, it states that to recover from a UV fault the device requires load removal and the cell voltages to rise above Vuv + hysteresis. The presence of a load is determined by the state of the LD pin. The LD pin is typically connected to PACK-, which can be the negative terminal of the charger or of the load. The value of Rgs_chg is used to determine the fall time for the CHG FET (how fast the FET will turn off). We recommend a value greater than 1 MOhm, typically 3 MOhm. Currently, all of our devices are configured to require load removal for recovery.

    Please let me know if you have any more questions!

  • Hi Shawn,

    What values of R_CHG and R_GS_CHG did you use? Also, what do you mean by "attaching a charger"? I'm using a BQ24618 integrated charger with a stand-alone USB-C PD IC, so the physical attachment will always be there; there will just be a pair of MOSFETs breaking the connection until a USB-C power supply gets plugged in.

    Thank you so much,

    Jeffrey

  • Hi Jeffrey,

    I used the EVM as is. So that would correlate to a R_CHG of 5kOhm and R_GS_CHG of 3MOhm (for exact configuration, see EVM schematic). For my experiment, I used the recommended set up described in Figure 2 of the EVM user guide. The "charger" is just a power supply connected to PACK+ and PACK-. To simulate a load, the power supply is set to a value lower than BAT and to simulate a charger it is set to a value higher than BAT. Here is a step by step of my experiment:

    1. (Optional) Run through EVM quick start sequence (described in EVM user guide) to make sure device is behaving correctly. 
    2. Connect EVM as shown in Figure 2. Add DMM to PACK+ and PACK- to monitor device status (fault mode or normal).
    3. Turn BAT power supply to 18V (3.6V/cell, normal mode). Turn "charger" power supply to 19V (or something above BAT) to simulate charger. Observe that BAT voltage raises to 19V (cells "charging"). Device is in normal mode.
    4. Turn off "charger" power supply. Do not disconnect probes. Observe BAT voltage drop to 18V. PACK voltage at 18V (device still in normal mode).
    5. Lower BAT voltage to 12V (2.4V/cell). Observe PACK connection drop to 0V. Device has entered UV fault mode.
    6. Turn on "charger" power supply to 19V once again. Watch BAT voltage raise and PACK voltage raise to recover from UV fault.

    Hope this makes sense. Please let me know if you have any more questions!

  • Hi Shawn,

    Thanks, that was really helpful! One last question...how does the calculation for R_IN work? The data sheet isn't super clear so I'm not sure how to pick the right value. I'm also not sure how to decide what the cell balancing current should be in the first place haha.

    Thanks,

    Jeffrey

  • Hi Jeffrey,

    I'm glad you found it helpful!

    As for Rin, it is configured based on if you are using the internal cell balancing FETs on the chip or external balancing FETs. We recommend external FETs if you require a cell balancing current greater than 50mA. If you are not very concerned about how quickly the cells balance, using our internal FETs is just fine. For that configuration, use Rini and Cini.