Core loss for ETD49 Full Bridge SMPS?

[cupoftea]

Hi,
We are doing Full Bridge SMPS, Split coil sec, 380VDC to 175Vdc, 500W, 70kHz

One uses ETD49, N87 with 34:23 = Np/Ns, gap 0.1mm
Bpk = +/- 0.286T
Pri = 2 strands 7/0.22 TEX-ELZ (2 layers of 17 turns in series for 34 turns, sandwiching the 2 secs)
Sec = 2 strands 7/0.14 TEX-ELZ (1 layer of 23 turns for each sec)


Other One uses ETD54,N87 with with 36:24 = Np/Ns, gap 0.1mm
Bpk = +/- 0.186T
Pri = 2 pllel layers of 36 turns of 7/0.23 TEX-ELZ sandwiching the two secs
Sec = 1 layer of 2 strands of 7/0.15 TEX-ELZ for each sec.



Do you think +/-0.286T is too much for core loss?

 

[Easy peasy]

hahaha - at 70kHz, +/- 0.3T - the core will heat the wdgs quite nicely - do you have a fan mounted on top of the transformer ? this should cure your ills - say at least a 3W fan

--- Updated Mar 25, 2022 ---

BTW if you drop the Np from 34T to 36 T - all else being equal the peak B will drop by the same amount

in this case to 0.27 T

 

[cupoftea]

at 70kHz, +/- 0.3T - the core will heat the wdgs quite nicely -

Thanks, do you think it will be ok though, as we are using Litz wire, as in the top post?

 

[Easy peasy]

as lot as you fit the fan you might get away with it ...

 

[cupoftea]

Hi,
Along similar lines.....
We now wish to do a 2kW Full Bridge LLC converter off 380Vdc at 100kHz.
We wish to use a PQ5050 core from TDK, with N87 material.

The B in the core is +/-0.137T (21 turn primary, Kapton gapped)

The below shows that this means a power loss in the core of 160kW/m^3. The core volume is 37230e-9 m^3. So that’s a core loss of 6W. Would you agree that this is too much?

N87 ferrtie material:

https://www.tdk-electronics.tdk.com/download/528882/71e02c7b9384de1331b3f625ce4b2123/pdf-n87.pdf

PQ5050 TDK core set:

https://www.tdk-electronics.tdk.com/inf/80/db/fer/pq_50_50.pdf