• State Resolved
  • Locked Locked
  • Replies 2 replies
  • Subscribers 62 subscribers
  • Views 122 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

UCC28180: Transient Response

Thomas Egler
Prodigy 20 points
Part Number: UCC28180

Hello,

we are currently using a PFC-Boost-Controller (Continuous Conduction Mode) from a different manufacturer. This PFC is designed for outputs of up to approx. 400W. However, new requirements require sudden load transients of up to 1kW for 200ms. In order to achieve this with the current controller (without the PFC output voltage dropping too much) we have to significantly increase the number of electrolytic capacitors.

For example, at 660uF output capacitor, 230VAC input voltage and a load transient to 1kW, the current topology shows a voltage drop to 345V (from 400V) and a recovery time of approx. 40ms (channel 3 in the figure). The load step in the figure begins at cursor X1 (not shown).

Could you estimate the effect of such a load step with an output capacitance of approx. 660uF, for example with the UCC28180 controller? Are there other components in the portfolio with the same topology (CCM-PFC-Controller, boost diode, external MOSFET, 100kHz and PFC choke) that could perform better than the controller shown in the figure?

If you need further information I will try to provide it.

Thank you and best regards

  • 0
    TI__Mastermind 31740 points

    Hello Thomas, 

    Thank you for your interest in the UCC28180 PFC controller. 

    A load of 1kW for 200ms is a 1kW load for 10 line cycles.  This is a steady-state condition as far as the PFC controller is concerned, so the power stage must be designed to accommodate the maximum current (at minimum input line voltage) for this duration.  Thermal design depends on the duty cycle of the 1kW load level.  

    You mention the normal power level is 400W and a step can go to 1kW (a 600-W increase), but the waveform shows a step from nearly zero power.  
    Because a PFC loop necessarily has a low ~10Hz bandwidth, it is slow to respond to step loads. 
    In your screen-shot, it takes about 20ms for the COMP error voltage to reach a level where an enhanced response is activated.  

    The UCC28180 has a similar response function, triggered at 95% of nominal Vout.  Compare the EDR function of the UCC28180 to that of your existing controller to see if the -180 provides a greater level of current in response to the undershoot from the step.  Be aware that Vout must necessarily drop at least to 95% of nominal before this function is triggered.  Form your screen-shot, nominal Vout is 399V, so Vout will drop to ~379V before EDR is enabled and higher error-amp gain is enabled.
    Vout will still drop further, but not as fast, as the higher gain drives VCOMP higher and delivers more power to the output. 
    I cannot estimate how low Vout will eventually get to before rising again, but based on the 110V/div scale, the slope of Vout falling will start to reduce about ~10ms after the start of the load step so the Vout transient valley may be higher than that shown.  

    A faster loop compensation network (higher bandwidth) would respond sooner than 20ms, but would also increase steady-state THDi.  Recovery time can be shortened by raising the maximum output power capability of the PFC, so that more current can be pumped into the output caps when the load is 1kW.  The down-side of this approach is that the system must be able to withstand a continuous overload at this level, or have an overload detector and a timer. 

    Using the UCC28180 and some slightly wider BW and higher power limit may help reduce how much additional Cout, if any, is needed to meet your transient requirements.  

    Regards,
    Ulrich

  • 0 in reply to Ulrich Goerke
    Prodigy 20 points

    Hello Ulrich,

    thank you for your reply and your thoughts on the topic.
    We tried to make the feedback of the output voltage slightly more aggressive by using a non-linear high precision feedback circuit instead of a simple voltage divider, which gives us an earlier and stronger response from the PFC. As it seems right now, this allows us to save about 20%-30% of the capacity with the same voltage drop.
    Thank you again and have a nice weekend!

    Best regards