About PFC Choke Coil

[fethiyeli]

Hi, I need to design 3kW single CCM boost PFC but as i have seen in PFC circuits, PFC choke coil is generally put on AC side before the rectifier. In reference design and textbooks, i have always seen that PFC choke coil was put on DC side after the rectifier.

May i ask the advantages and disadvantages of the both usage of the PFC choke coil ? I know how to design it on DC side.

Is the way of calculation of AC side PFC choke coil similar to DC side coil?

Thank you.

 

[scopeprobe]

PFC Correction via a Boost converter is a DC/DC Topology which is AFTER the bridge rectifier. Inductors on the AC side are normally for filtering purposes. A PFC correction IC typically modulates the PWM to the mosfet to track the Haversine.

 

[D.A.(Tony)Stewart]

here is a 2.5 kW 4-level flying capacitor multilevel (FCML) GaN FET totem-pole PFC converter bridge > 99% efficient https://epc-co.com/epc/GaNTalk/Post...400-V-Rectifier-Using-200-V-eGaN-sup-sup-FETs

 

[FvM]

Chokes on AC side are generally found in bidirectional PFC circuits (also called active front end). The shown 4-level bridge is a relative complex example of this PFC type, a simple one would be a basic H-bridge. An unidirectional PFC with choke on the AC side is e.g. the Vienna rectifier.

 

[Easy peasy]

ha ha, 8 gate driver ckts, 6 needing isolaiton, required on the flying cap version, versus 1 gnd referenced on the classic booster.

 

[D.A.(Tony)Stewart]

ha ha, 8 gate driver ckts, 6 needing isolaiton, required on the flying cap version, versus 1 gnd referenced on the classic booster.

Yes and the classic 1 cap with an IGBT would be 3x heat rise

 

[Easy peasy]

.... if you choose an igbt, SiC mosfet and diode give 97% with a standard B rect @ 230Vac

 

[fethiyeli]

Hi, this is what i am talking about. I know designing with other topology and using GaN, SiC. However, we need to choose cheaper way to do it.

 

[scopeprobe]

Hi, this is what i am talking about. I know designing with other topology and using GaN, SiC. However, we need to choose cheaper way to do it.

View attachment 171917

I must be honest i can't really see the benefit of doing that, most bridge rectifiers are relatively slow which means you'd probably end up building the bridges through discrete parts to get faster diodes. With that you now require 3 more diodes than a standard Boost all of which could cause EMI issues and therefore most likely all require damping networks etc.... Now factor in the Vf of the diodes has now been doubled due to 2 diodes in the output the losses go up which causes heat issues. Looks like a can of worms to me, will be interesting to see what the others think.

 

[Easy peasy]

you can do the same thing with a lot less diodes ....

 

[fethiyeli]

The schematic above i have given was taken from Toshiba PFC app note. I have also seen this type of PFC in air conditioner circuits.

I need to implement about 3kW PFC circuit but i have some doubts about which topology to choose. The cost is the main thing to consider for our side. I am thinking to use single CCM boost topology having PFC choke coil on DC side.

I just wanna get your ideas about topology choose. Thank you.

 

[scopeprobe]

It could be done with a standard boost but would require careful component selection and thermal management. i’d probably have a look at the Texas Instruments Interleaved CCM PFC controllers to see if it would meet the requirements. It uses the standard boost pfc but distributes the power between 2 smaller converters which may help with heat, EMC and costs. Have a look at the UCC28070A

 

[fethiyeli]

Hi, PFC will probably be controlled by MCU but i will take a look analog PFC controllers as you said.

I have seen a PFC app note from TI and they were using single CCM boost PFC for 3.5kW. However, they used 2 mosfet and 2 output diode for current handling in that app note.

MCU will be already used for Inverter. Therefore, using analog controller may add additional cost up.

 

[scopeprobe]

Hi, PFC will probably be controlled by MCU but i will take a look analog PFC controllers as you said.

I have seen a PFC app note from TI and they were using single CCM boost PFC for 3.5kW. However, they used 2 mosfet and 2 output diode for current handling in that app note.

MCU will be already used for Inverter. Therefore, using analog controller may add additional cost up.

you’ll have to weigh up the pros and cons of the various options. More parts may not necessarily be more cost with all things considered. Interleaving will add more parts but they could be lower spec and you’ll get higher volumes which will help in bulk buying. Also the filtering requirements will be less which will bring these costs down. You may be able to do it in your micro assuming youhave the bandwidth.

 

[RCinFLA]

Chokes on AC side before rectifier are for EMI backfeed suppression. PFC switcher is not totally clean for high frequency current EMI.

 

[fethiyeli]

I have seen some air conditioner outdoor unit electronic boards where external AC side PFC choke coils were used. There are only 2 boards using internal DC side PFC choke coils implemented on the PCB i have seen. Another one was interleaved topology. Rest of them are single CCM boost type and have external AC side PFC choke coil.