• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 63 subscribers
  • Views 548 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.

GSM pulse vs bq24616?

Zhang Hui
Intellectual 625 points
Other Parts Discussed in Thread: BQ24616

Hello,

We are doing production test and need some urgent confirmation on charging algorithm:

  1. In “Charge Termination, Recharge, and Safety Timer” of SLUSA49B bq24616, the termination is determined by two conditions: VRECH reached and current small than ITERM.
  2. The buck converter regulates the duty cycle according to VFB.

Both are easy to determine with steady load. However, if we regularly discharge high current pulses per 100ms, then there is chance at the moment of sampling, the voltage is lower due to discharging.

What is the voltage and current sampling algorithm used in the charger? Would 100ms period pulse dipping (for example GSM like, etc.) cause problem?

Please instruct us on the details of (1) and (2)’s algorithm used by the charger.

 

Hui

  • 0
    TI__Genius 11480 points
    Hi Zhang,

    I do not think transients at the battery will be an issue. It is an analog control loop, that will be able to handle a load transient if properly compensated. As long as you use a 100k resistor as the bottom resistor in the Vfb voltage divider, and a 22pF feedforward capacitor, I think it will be okay since most of the compensation is done internally. I would use the EVM schematic as a reference.

    In regards to 2, the buck converter for the bq24616 works the same as any buck converter. It samples the output voltage from Vfb, and will generate an error signal, and compare that signal to a ramp wave. That will then be used to generate a PWM signal. More info can be found here :

    www.ti.com/.../slyt293.pdf
  • 0 in reply to David Wiest Jr
    Intellectual 625 points
    Thanks for point 1.
    For point 2, do you mean that since the frequency is already compensated as in point 1, so it equally would be fine?

    So Basically the divider+22pf feedforeward capacitor solves both?
  • 0 in reply to Zhang Hui
    TI__Genius 11480 points
    I believe so. Typically, proper compensation will create stable load transients.

    Are you using the power path feature of the IC? If you are doing this, the adapter will supply the load transients (unless they overload the adapter, then DPM will kick in and the battery will supplement the current). If there is no adapter, then the BATFET will be on and connect the system directly to the battery - in which case you will not have to worry about charging.

    If you are just connecting straight to the battery- I think it would still work but am not so sure. However, you may come into some problems with the battery overheating and lifetime due to the constant drain/recharge coming with the pulses. It is situations like this that the power path topologies were created for.

    If any issues crop up during testing, just let me know and I can help you troubleshoot.