#66🚀 Why Most Engineers are failing measuring the Rac of Magnetics
SuperPowers for Electrical Engineers 🦸🏻♂️🦸🏼♀️
👋 Hello, friends! Dr. Molina here 👨🔧
Welcome to Dr.Molina Newsletter, where I break down Magnetic design for you every week, in
4 minutesthe time needed.
Before moving to the technical part, I can already announce that we have hired a great engineer, who, before being part of Frenetic, he was a customer. He is a senior specialist in Power Electronics, and now, he will work to build Better Magnetics at Frenetic. These kinds of things are the ones that make my day.
Why Most Engineers are failing to measure the Rac of Magnetics
This week, I wrote a post about my opinion of core losses.
In my opinion, Core loss estimation is not the problem in the current state of the art of magnetic optimization.
If you are in contact with manufacturers, they update their material parameters when they have considerably improved the recipe of their products. I have worked with Ferroxcube and TDK materials for a long time. My experience talking with Ferroxcube is excellent (Thank you, Ferroxcube, for your support 🤗.).They are charming, and the material information they provide is good.
In the picture, you can see the core loss in function of the temperature. Of course, you are dead if you want to find the values in the tables.
However, suppose you use the raw data or any software to get their Steinmetz coefficients and estimate losses considering the magnetic field and the temperature with iGSE. In that case, you can believe these losses (consider an average error of 10%).
But, the estimation of the winding losses is another story. Here, we enter the wild wild west.
There are a lot of reasons, but today, I will explain to you the basic ones.
You can measure the core losses separately (with low current values).
But, You can´t measure the winding losses without the core directly. The main reason is that you can´t measure the resistance of the wires to the high-frequency harmonics, what we use to call, Rac.
When you measure the impedance of a magnetic in the impedance analyzer, you are measuring the real Rm and the imaginary Xm of the impedance because it’s a complex number.
Usually, the equipment provides you the L, dividing Xm by w. At low frequencies, you see the inductance, and at higher frequencies, the inductance becomes negative because it starts to dominate the capacitance.
One of the most influential researchers in the world in Magnetics, Charles Sullivan, published a methodology for measuring this Rac. In the following picture, he shows the equivalent circuit of an inductor.
Here, you can see that what we call Rac, here is called Rw and is in series with the resistance of the core. Therefore, if you want to know this value precisely, you need to measure the whole impedance, separate the effects of the parallel capacitance, remove the complex impedance of the inductance and them, and separate the core's resistance from the resistance of the windings at each frequency.
Most engineers have been measuring the Rm, thinking they have the Rac. That’s a typical mistake.
In his article, Charles proposes a method [1] to measure the resistance of the core, and then, get the Rac. His method consists of adding an extra winding with a very thin wire and doing some measurements and operations. It’s not a straightforward method.
Last year, Frenetic worked with the University of Oviedo to study this, and we got a new method. We are writing a conference article about it. Our method is not straightforward either, just a different approach. The most exciting part of the investigation was the verification of the measurements. Maybe in the future, we will use this method for building an AI-based model for Rac.
Reference
[1] C. R. Sullivan, "Computationally efficient winding loss calculation with multiple windings, arbitrary waveforms, and two-dimensional or three-dimensional field geometry," in IEEE Transactions on Power Electronics, vol. 16, no. 1, pp. 142-150, Jan. 2001, doi: 10.1109/63.903999.
Frenetic Updates
These are some of the most important updates of FO.
The main waveform page includes a breakdown of losses, magnetic field, and four new temperatures.
Also, a new feature allows you to work collaboratively in a design with your team. You can see who is editing the design at each moment.
In the Core Optimizer, you can filter by maximum losses, Bmax or both, and all see the inductance and gap in each design.
New Consultancy Service
I have decided to experiment with a Consultancy Service for one month three weeks, having a maximum of 2 consultancy sessions per week.
Therefore, in total, I’m offering
eight sixconsultancy sessions.
Before the meeting, I will ask you to email me the questions or topics you want to discuss. If I can´t help you, I will be honest and refund you the money.
The Consultancy topics I will cover are:
- Magnetic design
- Integration of Magnetics in your Power Converter
- Frenetic Online Demo Design
Follow the link to book a consultancy with me.