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.

TPS40210 boost converter

Other Parts Discussed in Thread: TPS40210, TPS40211, TL431

Hello,

I designed a boost converter which generates an output voltage of 24 V from a source voltage of 5.0 to 8.4 V.
The converter has to drive a motor with a maximum motor current of 2.2 A rms.
Because of a very high current during the starting phase of the motor, the current immediately reaches the overcurrent level of the converter. So the converter doesn’t work.
Is it possible to limit the output current before reaching the overcurrent level?
Can anyone explain me the function of the “current mode controlling?

Thanks in advance for your answer

  • Klaus,

     

    Peak Current Mode Control (the style of Pulse Width Modulation used by the TPS40210 controller) compares the control voltage (COMP) to the sensed switch current, terminating the switching pulse when the sensed switching current exceeds the level programmed by the COMP voltage.

    It is possible to clamp the COMP voltage at a level that will limit the current below the current limit, though tolerances make this a very difficult prospect as the offset voltage (COMP voltage that corrisponds to 0A of current) is not tightly controlled.

    There are 3 additional possibilities that would likely work for your case:

     

    1) Use a TPS40211 controller to build a constant current source (similar to what is used by an LED driver) and programm a current of 2.6A, then use an external Op-amp and TL431 shunt regulator to build a second voltage loop to reduce the source current when the voltage exceeds 24V.

        Note: limiting the current into a motor during start-up can cause it to stall, so you'll need to make sure the programmed current is high enough to prevent motor-stall.

    2) Increase the current limit and the ouptut capacitance and reduce the control loop bandwidth  to absorb the start-up surge current without forcing the power stage into current limit.  This may require a significant amount of capacitance and slowing of the bandwidth, so it may be undesirable.

    3) Filter the ISNS value beyond 1 switching cycle.  This will eliminate the pulse by pulse current mode control and require compensating a voltage mode boost converter (more difficult than current mode control) but can provide a programmable time constant to an average current limit rather than the peak current limit the controller was designed with.  The time constant would need to be longer than the start-up surge to ensure it filtered the surge out. 

      The TPS40210 will run in voltage mode control with a voltage ramp equal to VDD/20

  • Hello Mr. Miller,

    Thank you very much for your helpful support.

    I think my existing application is already a voltage mode converter.

    The used components are:

    Inductor:   10 µH, In 15 A, Isat 21.5 A, Rdc 6.9 Ohms
    Diode:       MBRS540    ( 5 A )
    HEXFET:  IPD088N04 ( 8.8 mOhm, total gate charge 28 nC )
    C out:        470 µF add. 4.7 µF ceramic
    Shunt resistor:  10 mOhm
    Shunt divider:    51.1 k, 1.5 k
    R ISNS:     1 k
    C ISNS.     330 pF ( delay time 0.33 µs )

    input voltage:   5.0 ~ 8.4 V
    output voltage:  24 V
    load:  2.2 Amps rms max.
    switching frequency:  300 kHz

    Can you confirm the application is a voltage mode converter?
    The ISNS filter is too small, isn't it?
    What is the correct compensation?
    Is there a possibility to limit the input current at 10 Amps max.?
    Is it possible to avoid the overcurrent mode during the starting phase of the drive?
    Please, can you explain the term “The TPS40210 will run in voltage mode control with a voltage ramp equal to VDD/20”.

    Thanks in advance

    Klaus

  • Klaus,

    1) The current limit of the TPS40210 controller will limit the input current via the peak switch current.  The peak switch current will be the input current + 1/2 the ripple current.  Setting the "shunt resistor" (resistor in series with the source of the MOSFET) to 10mOhms would give you an input current limit of 12-18A  We can set an input current limit of 10A, though the accuracy might be a little loose

    2) I calculate about 1.3A of peak to peak inductor current at low line (5V input) which calculates to a peak switch current of about 11.3A.  (10A input current wont support 2.2A @ 24V when Vin = 5V)

    3) The TPS40210 controller uses peak current mode control with integral slope compensation.  The ISNS voltage is amplified (x5.6) and added to the oscillator ramp (VDD/20) and compared to the COMP voltage to generate the PWM control.   In your application, the rising slope of the current sense is 10mOhms * 5V / 10uH = 5mV/us, amplified by 5.6 this provides a slope of 28mV / us at the PWM comparator or 93mV per switching period.  With VDD at 5V, the oscillator ramp is 250mV per switching cycle.  The result is a hybrid between voltage mode control and current mode control.

    4) The ISNS filter right now is 1/10 the switching period and about 1/5th the ON time.  It is passing more than 99% of the peak inductor current value to the ISNS pin.  If you want peak current control and peak current limit, it's fine.  If you want voltage mode control and average current limit, it needs to be increased to several switching cycles, but this will also prevent the current limit from being able to function as an input current limit.

    5) Please, can you explain the term “The TPS40210 will run in voltage mode control with a voltage ramp equal to VDD/20”. - Above, I have described the normal opertation of the TPS40210 controller.  the ISNS pin tracks the switch current and the control voltage (Vcomp) controls the sum of the oscillator voltage and the peak sensed switch current.  If the ISNS filter is increased to several switching periods, the filter removes the peak current value from the ISNS voltage and the control voltage (Vcomp) controls the duty cycle via only the oscillator ramp voltage.  Emulating voltage mode conttrol loop.

    For compensation, I would recommend starting with a target bandwidth of about 10kHz and use the compensation equations from the datasheet (you'll need the ESR of that 470uF capacitor)  Unfortunately, the characteristics of the boost topology don't allow for as fast a control loop as the BUCK where loop bandwidths of 1/5 - 1/8 the switching frequency are common.