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.

BQ25790: Switch mode transformation

Part Number: BQ25790

Hi Team

Customer find the BQ25790 will work in buck converter mode during fast charge stage when the configuration is 1.5MHZ/2s battery.

I preproduced the phenomenon on our EVM. Set up: Vin 9V, 2S battery, 1.5M frequency.

it will works at 1.5M Hz after entering fast charge.  and converter's frequency would transform to 750MHz gradually with the Battery voltage rose. and finally worked in buck-boost mode with the Vbat rose continue.  

There are some question hope can get your support

1、Does it normal works at buck mode with 750Mhz frequency when configuration is 1.5MHZ?  Because I don't noticed there is description about this phenomenon on datasheet

2、if it works normal, Can you explain why it happen, and the principle of conversion between these modes_1.5M buck,750M buck, buck-boost.

  • Hi Harry,

    Attached is a training presentation that explains this feature on slides 10 and 11.  When VBUS is close SYS/BAT, the charger interleaves between buck and boost to keep the efficiency higher.  

    Regards,

    JeffBQ2579X_webinar_slides_2020.pdf

  • Harry,

    Due to the interleaving, the effective switching frequency is still 1.5MHz.

    Regards,

    Jeff

  • Hi Jeff

    Thanks for your sharing.

    I got it that the effective switching frequency is still 1.5MHz, but if do you noticed the waveform I put, there is a duration which works on buck mode with 750Mhz, no interleaving. This confused customer.

  • Harry,

    This is expected behavior.  I can't share the exact details since they are patent pending.  In summary, when VBUS is close to VBAT, instead of immediately interleaving with potential unstable switching pulses and therefore erratic system ripple, the converter starts skipping pulses for brief periods.  This can appear as 1/2 switching frequency for brief periods as your middle plot above shows.  Eventually, the converter starts interleaving both buck and boost. 

    Regards,

    Jeff

  • Hi Jeff

    Thanks a lot for your support.

    There is another question hope can get your support.  because of this brief periods, when we do inductor selection, which frequency should be follow?

    For example, if customer configures the work frequency at 1.5MHz , because there is brief periods which works at 750Mhz.  So which one , 1u or 2.2u, they need choose? and in order to the maximum ripple current calculation, which frequency should we choose? 

  • Hi Harry,

    The ripple current, and therefore SYS ripple, will increase during the periods of 750MHz switching as we can see from the inductor current peaks above, which double.  One of the reasons we recommend 5 x 10uF on the SYS output of the charger is to reduce the SYS ripple.  If your customer needs even less ripple, then I recommend adding more capacitance at SYS.  We have very little data on sustained operation at 750kHz with 2.2uH inductor. 

    Regards,

    Jeff

  • Hi Jeff

    Many thanks again.

    In datasheet shown as blow, it says: Please note that the 1.5 MHz switching frequency only works with the 1μH inductor and the 750 kHz switching frequency only works with the 2.2μH inductor.  t even customer's configuration is1.5Mhz. it still has a period which work at 750MHz frequency.  So which one customer should choose?

    And, about 1u and 2.2u,  which delta_I do you choose  when you calculate the value?

    Thanks very much

  • Hi Harry,

    You still use the 1uH with 1.5MHz and 2.2uH with 750Hz.  

    Regards,

    Jeff

  • We have 1.5Mhz and 750Mhz at one charge cycle.  So you mean the value of inductor depends on the configuration?

  • Harry,

    No.  The customer must accept that for brief period the inductor current and therefore the SYS voltage ripple will be higher.  The SYS voltage ripple can be reduced by adding more capacitance.

    Regards,

    Jeff

  • Thanks Jeff

    How does we get 1u and 2u for 1.5M and 750K?  L*▲i/▲t=V_inductor ?

  • Harry,

    You do not change inductors.  The dI increases for brief period but does not affect the internal control loops of the charger.   

    Regards,

    Jeff

  • Harry,

    Those equations are for when the converter is in deep continuous conduction mode (CCM) and operating under the control of the analog feedback loop.  Once the converter realizes VBUS is close to VBAT, a digital control loop starts managing the pulses.  So there is no concern for analog loop stability but there will be higher ripple, which can be mitigated with more SYS capacitance.

    Regards,

    Jeff