Ncp1653-Ncp1653a 2kw pfc- 4kw pfc is not working. It does not produce pwm. The design is wrong..

I also made fan7930b pfc circuit. Current control is very precise. 380volt output directly crashes when it can be loaded. It does not produce pwm.
Is there a running 2kw pfc circuit? I will combine with self oscillator smps. If you have a suggestion, I want to make scheme, pcb, layout.


  • 81902862_2686620618053747_8358130538405429248_o.jpg
    78.2 KB · Views: 220
  • fannnnnnnn.png
    224.7 KB · Views: 207
  • 83224216_2705184489530693_6370282793208905728_n.jpg
    36 KB · Views: 203
  • Schematic_smps-pfc_Sheet-1_20190424001902.pdf
    86.9 KB · Views: 121
Take into account that this forum doesn't like line related equipment, so warn with mods.

Most Motorola (On) PWM circuits are a true thrash. Starting with the MC34063 and 78S40 to the 34162 PFC are thrash, too hardly to repai or impossible. MC34063 faulted I saw thousands including one that catched fire!.

Why not to try STM products that are by far much more reliable : I repair lots of industrial grade SMPS's with STM PFC's, like the well known L4981 (Lambda used it many time)? Or the L6561 newest?
Last edited:
Sincerely I don't know, but I am completely sure that 600W yes. Never I saw a stand alone PFC unit. I repair industrial equipment (PSU's and voltage/frequency converters) but never saw a PFC alone.

In fact, to scale it up you will need a bulkier inductor, capacitor(s), MOSFET's and its drivers, and lowwer current sense resistor, but the logic is the same.

in my opinion, I would use integrated MOSFET's drivers like MIC44xx devices in place of those complementary pairs, they have less delay, lower output impedance (Some of them have 6A peak capability) and internal protections (Undervoltage, etc.).
Last edited: