Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronic 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.

Ferrite materials used for offline Flyback converter

treez

Advanced Member level 5
Joined
Sep 22, 2008
Messages
7,822
Helped
572
Reputation
1,147
Reaction score
551
Trophy points
1,393
Location
cambridge
Activity points
77,758
Hi,
We are doing the core loss calculation for a 120W Offline BCM Flyback operating at 120VAC input. Its primary peak current is 4.97A. (as attached) This corresponds to a Bpk of 0.368T. (Switching frequency is approx 20kHz at the max load condition and this input volts)

We can see that Ferroxcube 3C92 material is much superior to the 3C90 material (though 3C90 comes with standard gaps)
The TDG TP4E material is said to be equivalent to 3C92 but actually has greater loss.

Can you confirm that the B value in these graphs refers to the peak B (not Bpkpk) with B as a sinusoid)?
Also, since we are using a flyback, then we can divide Bpk by two and then use that value with these graphs (since flyback does not have sinusoidal B , but rather, unipolar B)


Also, can you confirm roughly the cost difference between 3C90 and 3C92?

Ferroxcume 3C92 Ferrite:

Ferroxcume 3C90 Ferrite:

TDG TP4E ferrite (page 38)
 

Attachments

  • Flyback pri and sec currents.jpg
    Flyback pri and sec currents.jpg
    55.9 KB · Views: 9
Last edited:

FvM

Super Moderator
Staff member
Joined
Jan 22, 2008
Messages
48,215
Helped
14,210
Reputation
28,679
Reaction score
12,904
Trophy points
1,393
Location
Bochum, Germany
Activity points
279,308
Also, since we are using a flyback, then we can divide Bpk by two and then use that value with these graphs (since flyback does not have sinusoidal B , but rather, unipolar B)
Surely not divide by two. You can use Ferroxcube SFDT 2010 tool to calculate losses for non sinusoidal B waveform.
 

treez

Advanced Member level 5
Joined
Sep 22, 2008
Messages
7,822
Helped
572
Reputation
1,147
Reaction score
551
Trophy points
1,393
Location
cambridge
Activity points
77,758
Thanks, but SFDT 2010 is only available for windows XP on the ferroxcube site
If you go to Frenetic website, you cant download it, but you can book an interview so they can decide how much you'll pay for it.
I then found a free download here...


....but it wouldnt install...it gets stuck and crashes the computer every time you try.

Surely not divide by two.
I admit it wont be exactly divide by two.......but do you agree the core losses will be significantly less as a flyback is simply unipolar flux?
 
Last edited:

FvM

Super Moderator
Staff member
Joined
Jan 22, 2008
Messages
48,215
Helped
14,210
Reputation
28,679
Reaction score
12,904
Trophy points
1,393
Location
Bochum, Germany
Activity points
279,308
Maybe my fault that I simply ignored the "only Windows XP" comment given with the download at ferroxcube.com Unfortunately the software is just working for me under Windows 10. Otherwise I'd run it in a virtual box.
 

Velkarn

Full Member level 3
Joined
Feb 17, 2014
Messages
151
Helped
79
Reputation
158
Reaction score
81
Trophy points
1,308
Activity points
2,050
there is tool from Epcos - Magnetic Design Tool
 

Easy peasy

Advanced Member level 5
Joined
Aug 15, 2015
Messages
2,917
Helped
1,051
Reputation
2,102
Reaction score
1,076
Trophy points
113
Location
Melbourne
Activity points
16,197
at full power, if the flux excursion is 0 - 325mT, then yes you can treat the delta-B as half of that each way ( ~ 162mT ) for the purposes of heating calcs due to flux & freq - this is not strictly accurate as, as you approach Bmax and the BH curve starts to roll over a bit - the losses here due to flux excursions are a wee bit higher than for the more linear portions of the BH curve, kind regards,
 

treez

Advanced Member level 5
Joined
Sep 22, 2008
Messages
7,822
Helped
572
Reputation
1,147
Reaction score
551
Trophy points
1,393
Location
cambridge
Activity points
77,758
Hi,
Soooo, we wish to use an E42/21/15 transformer for an offline BCM flyback at 30kHz f(sw)
We like 3C90 ferrite material but our peak B is 378mT, so we want to look at 3C95 ferrite instead. However, 3C95 graphs show nothing for core loss at 25kHz, whereas 3C90 graphs specify core loss at 25kHz. How does 3C95 core loss at 25kHz compare with that of 3C90 core loss at 25kHz?

3C90 ferrite:

3C95 ferrite:
 

FvM

Super Moderator
Staff member
Joined
Jan 22, 2008
Messages
48,215
Helped
14,210
Reputation
28,679
Reaction score
12,904
Trophy points
1,393
Location
Bochum, Germany
Activity points
279,308
Here's the result with SFDT2010. The tool doesn't allow a flux above 360 mT for this material, because it considers it as saturation limit.

1618314991885.png

1618314875948.png
 

Easy peasy

Advanced Member level 5
Joined
Aug 15, 2015
Messages
2,917
Helped
1,051
Reputation
2,102
Reaction score
1,076
Trophy points
113
Location
Melbourne
Activity points
16,197
3C95 -> 300mT abs max, else you are into Lmag falling rapidly scenario - esp at 100 deg C

3C90 -> 250mT abs max for same reasons ...
 

LaTeX Commands Quick-Menu:

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Top