Inductor design using a ferrite

[bigbrain1]

Hi,

I'm trying to design a buck converter but ran into some trouble trying to design the inductor. I wanted to make my own using a ferrite core. The maximum current of the psu including the current ripple is 1.2A and the value of my inductor is 330uH for my specific design. I calculate the number of turn which yields me 100. I then calculate the inductance factor (AL) the relative permeability (μr) for this "imaginary core" which gives me about 33nH and 51 respectively.

So now I look for a core that has an inductance factor that is greater than 33nH.

I choose the N30 material which gives me an inductance factor of 2.77μH and from this recalculate the required number of turns (which yields around 11) and relative permeability with this particular core and it yields me μr=4303 which is very close to the value in the datasheet

But when I compare the final values for the magnetic field calculated with all of those variables, I get 2 different values and I do not know what I am missing here. I feel like I should get the exact same magnetic field but it looks like the factor μr*N is not the same for the "imaginary core" and the real core.

I want to make sure I do not saturate the core, this is why i calculate the magnetic field which has to be <0.2T at 100°C, i chose this value for the design for security margin reason.

The details for the calculations are in the images.

Thanks

 

[bigbrain1]

Here is the datasheet, I do not know how to include 2 images per post so sorry

 

[bigbrain1]

alright so with a little bit of research and elbow grease, I'll have the pleasure to answer myself like a fool . I figured that you need to take an inductance factor that is less than what you calculate otherwise you're going saturate the core, even though it would be less of a hassle because less turn would be required, this is not how it works , and with my understanding it is because the factor N (required for the minimum inductance) does not decrease as much as the factor µr increase (required for maximum magnetic induction field).

Also tried to figure the thermodynamics of the core and the wires and I get less than a degree of temperature rise, so I'm going to test all of this in the next days.

I am probably not going to update you on the results unless some of you are really interested, let me know.

 

[L Demon]

Your common ferrite core will go handle the frequencies pretty easily, at least handle up to 200khz without core hysteresis.

You won't saturate the core that way. In PWM it is an excessive duty cycles that will try to saturate the core.

The rule of thumb would be to not exceed 50% on a pulse trans. I'm feeling pretty safe saying the same rules apply here. Once you exceed 50% duty you are pulsing your way into saturation. Doesn't matter how big your core is.

 

[bigbrain1]

Thank you L Demon,

From my learnings, the core will saturate from an excessive flux, and the saturation will generally not depends whether or not the PWM is higher than 50% (but it is of course all related because of the Ton). The reason is if Bmax>(L*Ioutmax)/(Ae*N) or
Bmax=((V*Ton)/(Ae+N)) + Φdc, then you not going to saturate the core. Also, the bigger the core the less likely you are going to saturate, since B=Φ/Ae.

I see in your message that it has to not exceed 50% on a pulse trans, I am unfamiliar with the design method, maybe I am misrepresenting what you are trying to say.

Can you provide a reference document for this particular design method.

my best regards