Input line filter for high power factor TM Flyback

[ysba]

Hi,

Does anyone have some info about which is the best input line filter for a high power factor flyback converter operating at transition mode (variable frequency)?

Thanks.

 

[goldsmith]

Hi,

Does anyone have some info about which is the best input line filter for a high power factor flyback converter operating at transition mode (variable frequency)?

Thanks.

Hi ysba
Your question is absolutely unfair !
What you mean by high power factor ? where is your schematic ? how much power ? what kind of specifications . your information are very low .



Merry Christmas + Best Wishes
Goldsmith

 

[BradtheRad]

Are we talking about values for a plain choke-and-capacitor filter? To allow you to draw smooth current from a DC supply?

If so, then use the formulae for reactive impedance. This will get you in the vicinity. The effect is a second order low-pass filter.

XL = 2 Pi f L

XC = 1 / ( 2 Pi f C )

Where X is your load impedance. For this value, calculate supply V divided by Amperes drawn during the switch-On time.

For frequency, take a much slower speed, say 1/10 of your operating frequency. The aim is to inhibit resonant action.

Example, operating frequency 20 kHz, supply 10V, draw is 1A during switch-On.

Solve inductor value in this equation: 10 = 2 Pi 2000 L
Therefore L = 800 uH

Solve capacitor value in this equation: 10 = 1 / ( 2 Pi 2000 C )
Therefore C = 8 uF

The left-hand simulation uses the above values.
The right-hand shows what happens when you use too small a capacitor.

 

[ysba]

Hi, sorry for the few information...

I'm working on a circuit like this:



It's a 50W power supply, fed by 90~240 V AC mains. It is a flyback converter controlled by an active PFC IC. I'm actually using Fairchild's FAN7527B instead of ST's L6561.

Since this circuit has a low input capacitance (470nF), it works limiting the inductor's current according to the instantaneous AC line value. It is also transition mode operated, thus the operation frequency is variable.

My question is: How this control technique affects the input line filter design?

Thanks!

 

[mtwieg]

Having a variable frequency certainly complicates things. Keep in mind that with this type of converter, the largest ripple currents will coincide with the lowest switching frequency. Thus after applying the filter that ripple will be very dominant. You can almost ignore the ripple current near the zero crossings.

There's no real trick to it, you just need to find an appropriate cutoff frequency in order to keep the PF high, then determine the appropriate L/C ratio to get small enough size without impacting efficiency too much.