Magnetics

Magnetics covers transformers and inductors

• https://coefs.uncc.edu/mnoras/files/2013/03/Transformer-and-Inductor-Design-Handbook_Chapter_2.pdf
• https://coefs.uncc.edu/mnoras/files/2013/03/Transformer-and-Inductor-Design-Handbook_Chapter_5.pdf
• https://www.youtube.com/watch?v=xA8HalLJ92c
• https://training.ti.com/understanding-basics-flyback-converter
• https://www.youtube.com/user/katkimshow/videos

Transformer

These are just a pain to understand

Larger transformers increase hysteresis loss.

The correct frequency is critical. Ideally, the frequency should be as low as possible without saturating the transformer. Higher frequency increases hysteresis loss.

Power

From this document, they have a great table showing cores listed for power at various frequencies. They only go up to 250kHz. But it looks like P and R materials actually reach peak performance around 500kHz. The power curve seems to flatten off, so that increase might not actually buy much, but maybe another 10%. A 110v 15A wall outlet can only provide up to about 1500W. Looking for values in the chart, the cores that match that are not terribly expensive. Less than $10, possibly as little as $1. I need to do research, but it simplifies the problem of getting high power cores. Though at the cost of needing a high frequency switch.

Looking at this one the volume is about 80kmm^2.

This is a great resource. I should look at push-pull circuit and try to understand the large equation in the middle describing the WaAc of a transformer.

Materials

• Laminated Iron Alloyws: 1.5-2.0T 50Hz-1KHz
• Powdered Iron: 0.6-0.8T 1KHz - 100KHz
• Ferrites: 0.25-0.5T 20KHz - 20MHz

Higher tesla is associated with higher core losses, but also higher saturation.

DIY Tests

• epoxy and magnetite
• epoxy and ferrite (Fe2O3)
• laminated steel plates
• plaster as binder
• steel shavings: these are cheap and might provide a core closer to electrical steel

If I had access to a kiln I could try making soft ferrites by mixing Fe2O3 and ZnO and heating to 1500F.

I need a way to measure

• Permeability
• Hysterisis

I may also need a way to measure and quantify saturation.

Notes from "Transformer and Inductor Design Handbook"

1

Hysteresis produces core loss.

Two forces; magnetomotive force (mmf) and magnetizing force (H).

Units

• Magnetomotive: Gilberts
• Magnetizing: Gilberts/cm = Oersteds
• Flux Density: Gauss
• Resistance to flux: Reluctance

Magnetomotive Force

 mmf = 0.4πNI
 N: number of turns
 I: current

Magnetizing Force

 H = mmf / MPL
 MPL: magnetic path length

Flux Density

 B =μH
 μ: permeability of core

2

Core materials

• silicon steel
• nickel iron (permalloy)
• cobalt iron (permendur)
• amorphous metallic alloys
• ferrites

Two primary magnetic losses; hysteresis and eddy currents.

Silicon steel is more stable (than just steel) over time and reduces eddy currents. Good in the lower range. High stauration flux density. M6 grain-oriented steel is particularly good.

Nickle-Iron (Permalloys). Supermalloy is a good variant.

Powder iron cores are cheaper. Require more turns.