• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Subscribers 63 subscribers
  • Views 1031 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.

Switch-mode vs linear battery charger selection.

Oscar Gomez Fuente
Expert 2066 points
Other Parts Discussed in Thread: BQ24297, BQ24266, BQ24392, BQ25890, BQ40Z50-R1

Hi everyone,

I'm trying to choose my battery charger (and fuel gauge to monitor de capacity, but I'm going to leave this issue aside for the moment). 

1º.- So first of all, I need to choose the battery I'm going to use. The application is a portable device with a Li-Ion battery of 5000 mA, so If I choose a charger capable of supplying around 2A, It takes me around 3,5 hours to charge the battery, I'm right?

2º.- I know the voltage input to charge the battery is going to be 5 Volts DC. And system output has to be between 3.3Vdc - 4.2Vdc.

3º.- Now, I need to choose from a switch-mode vs a battery charger. I know the benefits of a switch-mode, they minimise the power dissipation but consume more space in PCB than the linear chargers, as well as, the switch-mode charges are noisier than linear chargers but at the same time, I need to be careful with EMI because I need to certificate the final product, so for this issue it'll be better a linear charger.

But in my case, I need to charge a 5000 mA battery, so I need to charge the battery in around 4h, so I'm going to need a high rate current so If I use a linear charger is going to heat a lot (I remember that I'm going to use an input voltage of 5V, so I don't have experience if the system is going to heat too much or not. Anyone has checked something similar), and the heat is bad for batteries and electronic in general. 

I would like that the charger was self-sufficient, that it's to say, I don't need an i2c port to configure any setting. I would like that charger was able to be connected to any typical charger, similar to the phone chargers.

Any suggestion?

Best regards.

  • 0
    TI__Guru**** 167196 points
    Hi Oscar,

    Regarding 1, the rule of thumb is charge time~= mAhr/Ichrg *1.3. So, charge time = 5Ahr/2A*1.3 = 3.25Hrs

    Regarding 2, a power path charger can provide a SYS voltage = higher(VMINSYS, VBAT) when input power is applied. When input power is not applied, V(SYS) = V(BAT) - ISYS*Rdson-batfet.

    Regarding 3, for charge currents higher than 1.5A, we recommend a switching charger due to the high power dissipation of the linear charger when charging a fully discharged battery. For example, (5V-3.0V)*1.5A = 3W of heat that the package and PCB must dissipate until the battery charges.

    At such high charge currents, it is generally recommended to use I2C to enumerate a USB power source and then change the charger's input current limit accordingly, so that the USB source is not forced into overcurrent. In addition the I2C communication allows the host to monitor the charger's status and fault registers. An I2C charger such as bq24297 has D+/D- that detect USB power sources at plug in and set the input current limit initially (as low as 100mA for SDP port or as high as the value per resistor on the ILIM pin) but it is expecting the host to reset the input current limit after USB enumeration. If you are not using a USB power source and only wall adapters with fixed output current (referred to as DCPs) then you could use the standalone (non I2C) bq24266.
  • 0 in reply to Jeff F
    Expert 2066 points
    Hi Jeff,

    Thank you very much for your help and for your quick reply.

    Regarding : "It is generally recommended to use I2C to enumerate a USB power source and then change the charger's input current limit accordingly, so that the USB source is not forced into overcurrent. In addition the I2C communication allows the host to monitor the charger's status and fault registers"

    1º.- I forgot to tell you that I'm going to use a linux platform, so Do you know if there are drivers for controle this device? If there isn't a driver to control the battery charger this task will be more tedious for me.

    2º.- I would like to understand what's happen If I connect any standalone battery caherger to a USB 2.0 that it's capable of give for example 1A. If I have the battery-charger configured with a ILIM = 1.5A, Could it damage the USB port?

    3º.- If I have a I2C USB charger, Is it necesary to have a fuel gauge to control de status of the charge too? Or each device has his function.

    Best regards.
  • 0 in reply to Oscar Gomez Fuente
    TI__Guru**** 167196 points

    Regarding 1, drivers are available at https://github.com/tibms .  Keep in mind that the charger does not have a processor, only a state machine that makes decisions based on register settings that can be changed by the host through I2C.  Some register settings change themselves based on events as explained in the datasheet.

    Regarding 2, in standalone mode, all of our I2C chargers with D+/D- pins select no higher than the current allowed per the  USB BCP 1.2 spec.  For chargers without D+/D-, some customers add another USB communication chip, like bq24392, to determine the current per the USB BCP 1.2 spec and they set the charger's input current limit accordingly. One common complaint about the BPC 1.2 spec is that when an SDP is detected, only 100mA of current is allowed.  The spec assumes that USB enumeration will occur shortly after initial detection and report to the host that the current can increase the charger's current limit.  If the host does not perform enumeration and the current stays at 100mA, the system may not be able to start up and/or the battery will charge very slowly.  All of our chargers have VINDPM feature which monitors the charger's input voltage for drooping to a fixed threshold, indicating power source overcurrent, and prevents further drooping by regulating the input voltage to the threshold.  It is generally recommend that VINDPM be used as last resort to prevent a power source crash.  Our most recent charger, bq25890, has an Input Current Optimization (ICO) feature that automatically resets the charger's input current limit to the a value that will not allow the VINDPM to trip a second time. 

    Regarding 3, the chargers I have previously mentioned do not have fuel gauge functionality. The bq25890 has a 7-bit ADC that can reads charge voltage and current and temperature but does not provide a true gauging function. TI's battery gauge team provides BQ40Z50-R1 which integrates gauging and charging.  

  • 0 in reply to Jeff F
    Expert 2066 points

    Ok,

    1º.- Then If I'm going to use an as a charger a PC or something like that I need to accomplish with the USB BCP 1.2 spec, I'm right?

    2º.- I have another question: If the HW platform where is connected the bus ic2 of the charger BQ24297 is off, and I want to charge the battery while my system is off, I can't. So this is a problem because I need to charge my system when the HW platform is off.

    Best regards.

  • 0 in reply to Oscar Gomez Fuente
    TI__Guru**** 167196 points
    Hi Oscar,

    If you want to claim USB compliance then you need to follow USB BCP 1.2. In standalone (no I2C) mode, the bq24297 charges using the default parameters: VBATREG=4.2V, ICHRG=2.0A if input current limit allows. Input current limit is determined by D+/D- determination with the maximum being clamped by the value set by resistor on the ILIM pin.