Core losses in LLC converter resonant inductor?

[cupoftea]

Hi,
Doing 4kW Full Bridge LLC converter resonant inductor of 24.9uH.
ETD59 with 1mm gap for AL=508.
(7 turns of 3 strands of TEX-ELZ 7/0.3mm)

100kHz.
Irms is 11.9A
Ipk is 17.2A. (+/-17.2A)

Bpk is 0.166T. Will the core losses be too excessive?

There are fans but not mounted right next to this inductor.

I can reduce the Bpk to 0.142T by using the 1.5mm gap ETD59 (AL = 381) and winding 8 turns.....but i thought the bigger gap will give more fringing flux problems?

--- Updated Apr 5, 2022 ---

On the same type of subject...
Hi,

We need ETD59 in N87 material ungapped.

However, ungapped is only available in 3C90
At 100kHz and 200mT, N87 is at 350kW/m^3, whereas 3C90 is at 450kW/m^3
That would mean the 3C90 ETD59 core dissipating 23W, but the N87 ETD59 dissipating 17.9W.

Would you say, with those dissipations, that even with a fan right on it, either 3C90 or N87 would go over the Curie temperature, and the transformer coils all go short circuit?

3C90

https://elnamagnetics.com/wp-content/uploads/library/Ferroxcube-Materials/3C90_Material_Specification.pdf

N87

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

 

[cupoftea]

Incidentally, pg 20 of the below...

https://www.infineon.com/dgdl/Infineon-Evaluationboard_EVAL_3K3W_LLC_HB_CFD7-ApplicationNotes-v01_00-EN.pdf?fileId=5546d4626cb27db2016d3a60583725dc

...shows an LLC resonant inductor with >0.6T of flux change in it at f(sw). (it runs at 70kHz). It does have a fan right by it, (as shown on page 12) but i woudlnt have thought ferrite was conductive enough to be well served by a fan being nearby.