Diodes across filter inductors in offline LED drivers

6 posts / 0 new


Why don't you simply avoid overvoltage problems due to LC resonance when UK mains is switched on at the mains voltage peak by putting diodes across the  filter inductors, as in this thread...


Which particular PI LED driver solution you're referring to?

any of them that are offline power factor corrected flybacks. -because they will all suffer LC ringing of the filter.

Sorry to ask, I often try to download DERs in UK, and powerint site is extremely slow, do you know why?

If youre referring to the emi filter inductor and not the power inductor in a buck or buck-boost, adding a didoe across the filter inductor will limit the effectiveness of the filter in rejecting emi noise. The noise can now pass easily thru the diode if its conducting and allow high frequency noise to pass even if its reverse bias.

I admit that high frequency noise can pass through the diode capacitance...but an additional , small inductor can be additionally used to quell that higher frequency noise. The compbination of the bigger filter inductor with diode across it, and the smaller high frequency filter inductor is far less hardware than all those RC dampers, resistors across inductors, and active inrush networks...do you agree?

I guess you can try it.   When diode is conducting, its like bypassing the emi filter, and Im not sure how you can get away with that unless by increasing the other series inductance. The RC dampers not only damp the internal emi filter but also the emi filter in the dimmer and line inductances.