Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Core losses are too high for 3kW LLC converter transformer?

Status
Not open for further replies.
T

treez

Guest
Hello,
We are designing a 3kW LLC converter (battery charger) with Vin=425VDC, and Vout = 400vdc, 3kW. (schematic below)
Please give your opinion on the core losses in the PQ3535 ferrite core that we are using as the LLC converter transformer?
(It uses ferroxcube 3C95 ferrite and is centre leg gapped for an AL value of 1000nH/N^2

The circuit is as below. (LTSpice schematic also attached)
In order to get up to 400vdc at the output the switching frequency has to go as low as 80khz.
This gives a peak B of 390mT (but this is considering the “AMIN” value of core area)

Considering the “AE” value of core area, the peak B is 333mT.

Considering the peak B of 390mT, this gives core losses of 16.3 Watts
Considering the peak B of 333mT, this gives core losses of 13.1 Watts.

The above core losses are derived from page 95 of the Ferroxcube “soft ferrites and accessories data handbook 2013”..
https://www.ferroxcube.com/FerroxcubeCorporateReception/datasheet/FXC_HB2013.pdf


Which core area value should we abide by, and are these core loss levels manageable if we mount the core by gap-padding it to the metal enclosure?

(I have always abided by “AMIN” in the past, but for this product we are extremely space constrained.)

The PQ3535 core details are on page 742 of the linked ferroxcube databook.
 

Attachments

  • LLC_FB_400Vout_80khz_PQ3535 txfmr.txt
    7.2 KB · Views: 53
  • LLC converter.jpg
    LLC converter.jpg
    235 KB · Views: 135

Heatsinking a ferrite core transformer?

Hello,
We are wanting to design a 3kw LLC converter, and wish to use a PQ3535-3C95 core. We find that this gives a peak flux density of 350mT when its at 80khz.
This gives core losses of 17W.
This is too much, so do you think we will be able to heatsink the ferrite core by sandwiching its top and bottom surfaces between metal plates? (there would be gap pad between the metal plates and the ferrite core.)

Also, do you know what is the thermal impedance of 3C95 ferrite?

The Ferroxcube databook doesn’t say…

https://www.ferroxcube.com/FerroxcubeCorporateReception/datasheet/FXC_HB2013.pdf
 

Why not just stack two sets of EE cores ?
Not only will that halve the flux density, vastly reducing core loss, it will significantly increase the surface area.
 
  • Like
Reactions: treez

    T

    Points: 2
    Helpful Answer Positive Rating
You can stack E cores but i do not think you can stack a PQ core.
 
  • Like
Reactions: treez

    T

    Points: 2
    Helpful Answer Positive Rating
Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top