Toroidal core design for LM2576 buck

[manishanand14]

Hello friends

I need to design a 100 micro Henry toroidal core inductor for buck converter using LM2576.

I_out=1A

I have a toroidal core from EPCOS:
N-27 material,
A_L=3500 nH/T²,
Flux density,B_S (25 °C)=500 mT


I am using 20 AWG wire considering 1000 circular mills/A current density.

How many turns of wire shall I use.Please provide me theoretical explanation.

Also someone please provide me a standard AWG Datasheet. I could not find a good one.

 

[exnol]

Hi manishanand14,

Inductor winding :
1. **broken link removed** (Reference)
2. https://www.daycounter.com/Calculators/Inductor-Current-Power-Calculator.phtml (calculator)
3. **broken link removed** (reference)

AWG Conversion :
1. https://www.technick.net/public/code/cp_dpage.php?aiocp_dp=guide_awg_to_metric
2. https://www.engineeringtoolbox.com/wire-gauges-d_419.html

Hope these are useful.

 

[FvM]

In simple words, you can't make a SMPS storage inductor from an ungapped high µr core, e.g. N27 ferrite.

The core will already saturate at low currents. You may want to use the Ferrite calculation tool from Epcos to verify my statement.

 

[mtwieg]

Well given your inputs, you'll need about 5 turns to get close to 100uH. But as FvM states, this will likely saturate the core (though we can't tell unless you give the core area). For storate inductor, ferrites are usually gapped, or you use powdered iron cores.

 

[manishanand14]

So can I use a powdered iron core with low permeability. It has inbuilt air gap.Does EPCOS manufacture powered iron core.

I guess I can also use pot core or E-core of suitable size of same material N-27 ferrite as it has some air gap to avoid saturation.

So where does we use that kind of core if not in inductors for DC-DC converter.

 

[mtwieg]

Ferrite toroids are often used in EMI or RF chokes, where you don't really care about energy storage. I've also used ferrite toroids for high frequency gate drive transformers in the past.

 

[manishanand14]

I guess you have calculated no of turns using formula:N=sqrt[L/A_L​]

 

[mtwieg]

No, for transformers typically you'd calculate turns based on volt-time product in order to avoid saturation, rather than trying to achieve a specific inductance.

 

[manishanand14]

In simple words, you can't make a SMPS storage inductor from an ungapped high µr core, e.g. N27 ferrite.

The core will already saturate at low currents.

Can I use a gapped core like Pot core for the same material i.e. N-27.

 

[FvM]

Yes, preferably with "power" ferrite like N87 or respective materials from other manufacturers. N27 will work too.

 

[manishanand14]

Thanks FVM

I have taken this graph from P-core selection[pg 175] from ferrites and accessories databook.
**broken link removed**



Say I want to design 100uH power inductor for 100-300kHz frequency range. So N-87 material is good for that.

Now need to select the P-core size from the graph.

Using formula n=sqrt(L/Al) I got no. of turns =10{keeping some margin} and I am using 20AWG [0.9mm] enamelled copper wire.

Please tell me which core shall I select from the graph and the process i.e. how will you determine the intersection point in the graph.

What is that 10^0, 10^1,10^2,10^3 etc is given along the y-axis[left hand side]. No of turns in the graph is not very clear.

- - - Updated - - -

Dear mtwieg

No, for transformers typically you'd calculate turns based on volt-time product in order to avoid saturation, rather than trying to achieve a specific inductance.

But this an energy storage inductor.

 

[FvM]

The diagram is about available windings room, not current rating or inductance.

But this an energy storage inductor.

volt-time product calculation still applies. But saturation current calculation is the straightforward way.

Genereally, I repeat my suggestion to look at the Epcos Ferrite software.