Inductor design

[Gowda1]

Hey Guys,
I am designing an Inductor for Ipk-pk= 50A 20u f=85Khz. I went through lot of cores and decided to do with ETD cores. But the problem is I am designing it for very core loss so i decided to keep the maximum flux density of 100mT. But i did not find any ETD cores with this range, i could only get the peak flux density of 330mT (if my formula was right  B pk= Inductance x peak current divided by (turns x cross sectional area of the core)) ...I just want to know what are all the ways to get low peak flux density... Awaiting your answers.. Thank you

[Benta]

Larger core?

[T3sl4co1l]

Turns sets flux.  Air gap sets current flow.  The ratio of the two is defined as inductance (1 H == 1 Vs / A).  Flux density of course is flux / A_e.

L = N^2 mu_eff mu_0 A_e / l_e

If l_g = 0, mu_eff = mu_r, the relative permeability of the core material itself.

As l_g goes up, mu_eff falls.  If l_g were infinite (i.e., no core at all), mu_eff --> 1.

Approximately, mu_eff = l_e / (l_e / mu_r + l_g)

You'll probably find best results with mu_eff around 20-60.  This is a practical range for copper wire windings on typical core types.


You may not have many options for commercial ETD (or any other shape) core, in terms of off-the-shelf gapped core sets.  Gaps can be custom-ordered, but unless you're buying a pallet of them, you're probably better off adjusting normal (ungapped) cores.  The center peg can be ground down with carbide abrasives, or the core can be shimmed.  Mind that for shimming, the air gap counts twice: once in the middle and once in the outer legs.  So if you need e.g. l_g = 1mm, use 0.5mm shims.

Tim

[dietert1]

Unless you are a student, you could try Würth 7443763540220. It's a power inductor 22 uH for up to 40 A DC and usable up to 9 MHz. Maybe it serves the purpose. Those inductors are very special and really i have no idea how they make them: a multiturn coil of flat wire with very small coil diameter, amazing.

Regards, Dieter