• State TI Thinks Resolved
  • Locked Locked
  • Replies 7 replies
  • Answers 2 answers
  • Subscribers 63 subscribers
  • Views 926 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.

Original question:

UCC28056: About Ton and switching frequency

UCC28056: Switching frequency of UCC28056

UA&A_CodyDowling
Expert 1006 points
Part Number: UCC28056
Other Parts Discussed in Thread: UCC28180

Hello,

I noticed the comment about offering a way to adjust the switching frequency, but at the time it wasn't mature enough at the time to share. Has any progress been made on this? 

  • 0
    Expert 1006 points

    Another question. It appears that the calculator gives different plots for the same operating criteria (Switching Frequency at Minimum AC Input Voltage).

    The top plot has the switching frequency gathered around 100kHz, the bottom one has a much wider dispersion but they are labeled the same.


  • 0 in reply to UA&A_CodyDowling
    TI__Mastermind 27145 points

    Hi,

    The bottom plot is showing the switching frequency profile over maximum AC line. It is a typo in the UCC28056 calculator. It is possible to adjust the switching frequency profile depending on the inductance chosen. This would move the operating point at which the controller drops into DCM operation and the operating point at which the controller operates in burst mode as well. What is the switching frequency range your design needs to operate?

    Best Regards,

    Ben Lough

  • 0 in reply to Benjamin Lough
    TI__Mastermind 27145 points

    Hi,

    To provide further context, the recommended inductance uses the full voltage range of COMP (0V to 5V with 5V representing maximum load). By reducing the inductance value, this changes the COMP voltage that would correspond to max load. For example, if the inductance is is reduced by 25%, the COMP voltage will only reach 3.75V at maximum load. This means the on time when in transition mode will be smaller than if the recommended inductance was used.. This also means the load point at which the controller drops into DCM operation will be larger and also the load point at which the converter transitions into burst mode will be larger. 

    In the UCC28056 design calculator, the switching frequency is plotted for various values of COMP voltage. This information can be utilized to determine an appropriate inductance if a minimum switching frequency requirement for your application.

    Best Regards,

    Ben Lough

  • 0 in reply to Benjamin Lough
    Expert 1006 points

    Hi Ben,

    Thank you for your help. That makes sense.

    Regarding the switching frequency, we are targeting 200kHz nominal switching frequency. I don't quite follow what will happen at light load. Do the controller continue to switch at the same frequency, but just emits "packets" of switching events?


    How can I set the minimum frequency of the converter and still meet the power specifications?

  • 0 in reply to UA&A_CodyDowling
    TI__Mastermind 27145 points

    Hi,

    In burst mode, what will happen is the controller will switch with a set on time until the COMP voltage falls below the VBSTFall threshold. As the output voltage begins to droop, the COMP voltage will begin to rise. Once COMP becomes greater than VBSTRise, the controller will start switching again with the set on time. As the output voltage rises above the nominal set point, COMP starts to fall again and when COMP falls below the VBSTFall threshold, stops switching. 

    For 200kHz nominal switching frequency, is this at full load, mininum input? This might be difficult to achieve with a transition mode PFC unfortunately. Would CCM PFC be acceptable? CCM PFC such as UCC28180 would allow you to directly program the switching frequency to meet your specification.

    Best Regards,

    Ben Lough

  • 0 in reply to Benjamin Lough
    Expert 1006 points

    Hi Ben,

    Thanks for your reply and your explanation. 


    We considered the UCC28180 but, since this is used in a powered speaker, the low-audible noise spec was very important to us. We also avoid the auxiliary supply requirements of the UCC28180. We also valued the low-quiescent current requirements of the UCC28056. We need to pass the "<1W when powered off" energy efficiency requirements.

    We were considering a NCP1654 controller from On-Semi (https://www.onsemi.com/pub/Collateral/NCP1654-D.PDF), which offers a 200kHz version. We expected at minimum output, max output level we would probably drop down below <90kHz. Is that reasonable?

    The audio amplifier is only rated at 150Watts continuous, but needs to sustain peaks of 650W for 20-30ms. 

  • 0 in reply to UA&A_CodyDowling
    TI__Mastermind 27145 points

    Hi,

    Thanks for the additional info. I reviewed the NCP1654 datasheet and it is also a CCM PFC similar to the UCC28180. CCM PFC controllers would have a fixed frequency that is either selectable or programmable. 

    If I could summarize the difference between continuous conduction mode PFC and transition mode PFC, I would say transition mode operates at the boundary between continuous and discontinuous conduction. A fixed on time is used and the off time varies over the 1/2 line cycle. The benefit if TM PFC is that because the inductor current reduces to 0A before the next MOSFET on time, the boost diode is zero current switched and the MOSFET sees valley switching. TM PFC generally has low switching losses but high conduction loss due to the large current ripple. CCM PFC uses a fixed frequency and varies the duty cycle over the 1/2 line cycle to achieve power factor correction. In CCM, the inductor current does not drop to 0A before the next MOSFET on time so the MOSFET is hard switched and the boost diode needs good reverse recovery performance. Switching losses are higher in CCM PFC but the current ripple is much smaller compared to TM PFC so conduction loss is less. 

    For UCC28056 to satisfy 650W peak power, the controller will likely operate in DCM at around 110kHz or so over the AC input range at 150W output. The peak current in DCM will be lower however. Below is a slide deck containing an animation showing the inductor current as UCC28056 transitions from TM to DCM as the load decreases.

    UCC28056 One Slide.pptx

    Regarding UCC28180 vs the NCP device, the UCC28180 does have some circuitry to help avoid audible noise as well. While the UCC28180/NCP device don't have the good standby power performance of UCC28056, the UCC28180 has a shutdown mechanism by pulling VSENSE to ground. This approach is common for AC/DC designs where the PFC controllers that don't have good standby power performance. The downstream DC/DC converter would just need to be able to hand the wider input voltage when at light load. The pinouts are very similar between the two devices. It may be possible to do a co-layout between the two to compare performance.

    Best Regards,

    Ben Lough