PFC Offline Converters -- SEPIC, Cuk, Boundary Conduction Mode Flybacks

Click For Summary
SUMMARY

The discussion focuses on upgrading a non-PFC offline power supply design to include Power Factor Correction (PFC) for European deployment, specifically for a total output power of less than 25W. Participants highlight the advantages and complexities of using SEPIC and Cuk topologies compared to traditional boost-flyback designs. The recommended topology by IC vendors is a boundary-conduction mode flyback followed by multiple buck down-converters, which some participants find cumbersome. Additionally, the conversation touches on the challenges of implementing tapped-inductor buck topology DC-DC converters and the implications of cross-regulation issues.

PREREQUISITES
  • Understanding of PFC (Power Factor Correction) principles
  • Familiarity with SEPIC and Cuk converter topologies
  • Knowledge of boundary-conduction mode flyback designs
  • Experience with buck DC-DC converters and their regulation challenges
NEXT STEPS
  • Research SEPIC and Cuk topology efficiency in low-power applications
  • Explore boundary-conduction mode flyback design techniques
  • Investigate the use of tapped-inductor buck converters for multiple voltage outputs
  • Review Power Integrations and TI reference designs for PFC solutions
USEFUL FOR

Electrical engineers, power supply designers, and professionals involved in low-power electronics seeking to implement PFC in compliance with European regulations.

berkeman
Admin
Messages
69,346
Reaction score
24,678
I'm upgrading a non-PFC (power factor corrected) offline power supply design to include PFC for European deployment. The total output power is less than 25W, and the two output windings are around 20V.

I'm familiar with boost-flyback topologies for isolated PFC supplies, but that seems to have been passed up recently by other topologies to limit the peak intermediate storage capacitor voltages. SEPIC and Cuk topologies would seem to be much more efficient in terms of any capacitor voltages, but they involve extra complexity in terms of switch FETs.

Have any of you worked with offline PFC supplies in the 20W range that use one of these topologies? Can you let me know about the tradeoffs and why you chose the topology that you went with in the end? The Cuk topology seems to have some technical advantages (for the isolated version which I obviously need), but it seems to be a bit overly complicated compared to a SEPIC topology version.

The topology that I'm being recommended by the IC vendors is a boundary-conduction mode flyback followed by multiple buck down-converters, but that seems a bit clumsy compared to the SEPIC and Cuk converter topologies.

On a related question, have you had any experience with tapping multiple voltages off of a tapped-inductor buck topology DC-DC converter? This project is the first time I've heard of this option, and although it sounds okay at first suggestion by the IC vendors, I'm having a bit of a hard time with my initial visualization of the concept. The regulation loop is obviously only closed on one of the output voltages, but it seems like the cross-regulation issues would be pretty significant.

Thanks for your thoughts!
 
Hi Berkeman,

It is surprising that you need PFC for something as low as 25W. As far as I know, EU mandates PFC for powers >= 70W (or was it 75W)? The other question is do you need isolation or not. If not, then use a standard Boost PFC in DCM which will be well suited for this application and then use a second stage that is a DC-DC however, it seems to be a bit cumbersome just by intuition for only 25W.

If the application is something like an LED driver you can use a PFC+isolated flyback all in one. Power Integrations, TI have many solutions and the former may even have a reference design online.

If it is not an LED driver you can also try the buck PFC stage. http://www.ti.com/tool/ucc29910aevm-730

HTH.
 
EU regulations seem to only apply under the 39th harmonic (of 50 HZ). Is it possible to go with a higher frequency?

I don't know, but it's a thought.
 

Similar threads

  • · Replies 20 ·
Replies
20
Views
2K
  • · Replies 10 ·
Replies
10
Views
2K
Replies
8
Views
2K
Replies
12
Views
2K
  • · Replies 9 ·
Replies
9
Views
4K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 14 ·
Replies
14
Views
7K
Replies
2
Views
4K
  • · Replies 21 ·
Replies
21
Views
4K
  • · Replies 8 ·
Replies
8
Views
2K