• State
  • Locked Locked
  • Replies 11 replies
  • Subscribers 63 subscribers
  • Views 2133 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.

power factor correction

yeshwanth kumar
Prodigy 245 points
Other Parts Discussed in Thread: TPS92210

how to improve Power factor, for led driver, am using tps92210,am getting 0.87 as PF ,can any one suggest me

  • 0
    TI__Mastermind 28655 points
    You have to figure out what the cause of the 87% power factor is. Is this being run in peak current mode control? Constant On Time control would be used for PFC but peak current mode as the PMP6023 works also. Power factor is above 0.9 for the entire input voltage range of PMP6023. If you look at the input current waveform it can suggest what the probable cause it, is it phase shifted, is it distorted, etc? Does the power factor drop as the input voltage is increased?

    Some causes:
    Reducing the output power of PMP6023 without reducing the EMI and PFC capacitor values, this will cause phase shift.
    Bandwidth near or above 60 Hz, this will cause distortion.
    Bad layout causing noise coupling, etc.
  • 0 in reply to Irwin Nederbragt
    Prodigy 245 points

    sir,

     but my knowledge is nill on pmp6023 which u mentioned here am using just tps92210 in constant on time mode control only, and u have pointed out phase shift delay because of reducing in output power with out emi cap values yes we have solved that problem and we have achieved 0.95 power factor,but sir can u tell me how exactly we need to design emi filters based on output power, 

    note and please tell why u are notifying abt pmp06023 where can i use that in IC in my circuit please tell me purpose of that IC

  • 0 in reply to yeshwanth kumar
    TI__Mastermind 28655 points
    PMP6023 is a 0-10 dimming TPS92210 reference design. If you already have a board running then you probably don't need it though you can use the report for additional information.

    Conducted EMI is a measure of current returning back to the AC source. If you double the power of a discontinuous or critical conduction mode flyback the peak currents will double (being general). This adds 6dB to the amplitude which requires 6dB more filtering. When you design a discontinuous mode flyback or critical conduction mode flyback with universal input it becomes more difficult to provide the needed filtering to pass conducted EMI limits and not cause too much phase shift at high input voltage. The peak currents are high at low line and the larger capacitor needed to pass EMI for low line causes phase shift at high line. The PFC capacitor is part of the EMI reduction as well and also affects phase shift.

    You design the EMI filter to pass some level of EMI, FCC15 Class B, CISPR 22 or some other limit determined by the end application, this determines input capacitor sizes. Measuring the fundamental switching frequency and it's harmonics determines whether the filtering is appropriate, too large or too small.
  • 0 in reply to Irwin Nederbragt
    Prodigy 245 points
    sir,
    how ur telling the emi is increased by 6db, when power doubled???
  • 0 in reply to Irwin Nederbragt
    Prodigy 245 points
    its on constant ON time control,nd as i increase voltage(input ) power factor reducing sir,
    also help me out in reducing harmonics
  • 0 in reply to yeshwanth kumar
    TI__Mastermind 28655 points

    In generality if you leave the switching frequency the same, reduce the transformer inductance to get similar duty cycles at twice the power the peak currents will be twice as high or 6dB higher.  If the wave shape remains the same the harmonics will be the same ratio regardless of power level.  This means there will be 6dB higher harmonics reflected back to the input which would require 6dB more filtering.  This is a generality, many other things change in a design that doubles the output power.

  • 0 in reply to yeshwanth kumar
    TI__Mastermind 28655 points
    It is likely that the power factor reduces because the current phase lead increases as the input voltage increases. It can also reduce for other reasons such as instability, loop bandwidth, etc.

    Harmonics are caused by distortion. Generally the line capacitors wouldn't affect this but if you have large capacitor values for the PFC and any filtering on the output side (rectified AC side) of the rectifier bridge you can get discontinuous currents which can increase harmonic content.
  • 0 in reply to Irwin Nederbragt
    Prodigy 245 points

    sir, once please clarify me design of filter across AC side and after diode brdige rectifier, am unable catch out the design

  • 0 in reply to yeshwanth kumar
    TI__Mastermind 28655 points

    The PFC capacitor is on the rectified AC side of the rectifier bridge, the EMI X-capacitor can be before or after the rectifier bridge, if it is after it does not need to be X2 rated (X2 rated is typical for universal input 85-265).  The capacitors on the input side of the rectifier bridge can be discharged by the connections to the line and neutral (either line or neutral can have negative current to discharge the capacitor).  If it is on the rectified AC side the diodes block the line or neutral from discharging the capacitors.  If the power circuit draws less than is needed to discharge the line and PFC capacitor it will hold up (not get to zero) causing a discontinuity.  This will add to harmonic distortion.  A power factor corrected circuit tends to draw much less current when the AC line is low meaning there isn't much current draw from the circuit to discharge the capacitors on the rectified AC side including the PFC capacitor.

  • 0 in reply to Irwin Nederbragt
    Prodigy 245 points

    sir you have told mehow harmoics occurs, but i need to know from hoe to design the fiter to reduce harmonics.

  • 0 in reply to yeshwanth kumar
    Prodigy 245 points
    sir, can u share me ur mail id, i want send my schematic