Mains input filter for Boost PFC.......Filter resonance

[cupoftea]

Hi,
As you know, a mains input filter for Boost PFC should contain no resonance frequencies anywhere near the Booster’s switching frequency. And this includes the resonance frequency of the line inductance with the input X2 filter capacitor.

The attached shows five Boost PFC’s…alike in all but the mains input filter. (LTspice and PDF)

B has significantly lower inductance than A (but the same capacitance), and yet its input current is more attenuated than A……this is because A has a sub-resonance of L10,L11,C8 which is at the switching frequency of 60kHz.

C has the same filter components as B, but just in different order. Due to this, it actually attenuates higher frequencies less than B.

E has significantly less filter inductor values than D, but gives greater attenuation of the input current due to resonance in D at f(sw)

The dominant resonance frequency in the mains input filter, is that one which comes from the sum of all of the filter inductors, and the downstream most X2 (“ex 2”) capacitor. (given that all X2 caps are the same value).

Can we say that in overhead mains line supplied Boost PFC’s, the resonance from line inductance and the upstream-most X2 capacitor is significant, but in underground-cable supplied PFC’s, that resonance is not significant as the distributed capacitance of the mains cable defeats it?

 

[scopeprobe]

Hi,
As you know, a mains input filter for Boost PFC should contain no resonance frequencies anywhere near the Booster’s switching frequency. And this includes the resonance frequency of the line inductance with the input X2 filter capacitor.

The attached shows five Boost PFC’s…alike in all but the mains input filter. (LTspice and PDF)

B has significantly lower inductance than A (but the same capacitance), and yet its input current is more attenuated than A……this is because A has a sub-resonance of L10,L11,C8 which is at the switching frequency of 60kHz.

C has the same filter components as B, but just in different order. Due to this, it actually attenuates higher frequencies less than B.

E has significantly less filter inductor values than D, but gives greater attenuation of the input current due to resonance in D at f(sw)

The dominant resonance frequency in the mains input filter, is that one which comes from the sum of all of the filter inductors, and the downstream most X2 (“ex 2”) capacitor. (given that all X2 caps are the same value).

Can we say that in overhead mains line supplied Boost PFC’s, the resonance from line inductance and the upstream-most X2 capacitor is significant, but in underground-cable supplied PFC’s, that resonance is not significant as the distributed capacitance of the mains cable defeats it?

I suppose the question to be asked here is what are you trying to filter here and by how much to meet your requirement? If your have to meet any form of compliance the LISNs used during the conformity test is the load as far as your emissions is concerned regardless of anything else. Brief simlation of your first filter shows resonant peaks @ 32kHz & 83kHz with no attenuation at 60kHz. If 60kHz is what your trying to filter then you likely need more inductance than the values selected.