Wireless Energy Transfer simulation (inductive coupling power transfer)

HELP!!! Wireless Energy Transfer simulation (inductive coupling power transfer)

Hi guys,

Well i am working on a project to transfer wireless energy using two resonating coils, and i would like to simulate different loop types and estimate the EM power transfer, but the problem is that i'm a newbie with simulation software, and acctually i'm confused about which software to use (Ansoft Maxwell, Matlab, Comsol Multiphysics, Solidworks, ... )???

Can anyone please help me with some suggestions and instructions on how to design the model?

Thank you in advance

If you understand transformers with loose coupling, and know how to measure or calculate the coupling coefficient between the inductors, you can get your complete simulation in every spice based simulator. Many people use LTspice (that is free and fully functional). Of course when you know the coupling coefficient (or M), you can put your circuit in whatever circuit simulator you prefer (or already have).

If you want to simulate the coupling coefficient, you need an EM simulation package or at least a magnetics simulation package.

You can simulate your complete "problem" in an EM simulator, but this is very computer time consuming and may be limited to linear circuits. So in my opinion best route is to characterize the inductors (that is Lprim, Lsec, k and the loss resistance). These 4 parameters you can enter in a circuit simulator where you can add capacitors, rectifiers, etc.

You may search for some literature on coupled LC circuits (simple 2 resonator band pass filters). Literature on MF and RF RFID may also help you with simple approximate formulas for guessing the magnetic field strength.

Thank you WimRFP for your reply

The thing is, i want to simulate a standard model, to see how it works, after that i will modify the circuit as i want.

And for the EM simulation package, if i wanna simulate the whole model in it, which one can you suggest? I don't care if it's paid or free, just need to know which one is better. As for the circuit simulator, i use Multisim, is that ok?

I read a thread yesterday, someone who wanted to simulate the same problem as mine, the suggestion was

I did similar simulations. The coils are actually antennas, why not?! So, you can really normally simulate this in fullwave EM simulation tool and perform coils optimisation if needed.
Here method of moments (unbounded domian) would be preffered, because I expect later on you will move coils much further away than 1.5 feet.

If you could share problem specs (coils structure, topology, orientation, placement over ground, operating frequency) I could try to simulate that.

Click to expand...

So, can i simulate the coils as antennas?

Thansk again

If you can get Lprim Lsec and k from measurements, you can get the values from full EM simulation too. Simulation saves you soldering. I expect most inaccuracy in the loss resistance of the coupled inductors.

You can use an antenna simulator (I used IE3D and very old NEC2D) for determining k in coupled inductors). Your coils will very likely be very small w.r.t. the free space wavelength belonging the operating frequency. To avoid convergence/accuracy problems, you may simulate on a somewhat higher frequency.

Once you have the transformer parameters from the EM simulation, you can run your challenge in Multisim including all non-lineair components (like rectifiers, power switches, etc).

:-( :?: :?: :?: :?: I'm so sorry, but i'm not following you! :-( I don't know what i have to do first! Can you please guide me step by step?
Should i simulate the coils in the full EM simulator first? For now, i'm willing to try arbitrary numbers, just to create the model! Do you have a ready example that i can use to see how it works?

All i want to do is simulating a model to send ~1W to the receiver at a distance of ~1.5 feet. So what i have to do?
I have built a real circuit but it's a bit bigger and i want to optimize the coils to make them smaller, and reduce the heating effect!

Also, i have Maxwell, HFSS and Comsol Multiphysics.

As i told you before, i'm a newbie, and maybe this is my first time to simulate something! :-(

Sorry again for bothering you!

Some steps to take (in my opinion, based on my own limited experience):

Understand coupled inductors (transformer with loose coupling) and inductively coupled 2 resonator circuits and field around inductors. When your coils contain more then one turn, you may need to know something about proximity effect as this may very likely introduce more loss then the skin effect.

When ferrite or other magnetic material is required, master yourself ferrite materials in power applications. For your case, I don't think your coils will contain ferrite, so you may limit yourself to air coils.

Know how to measure relevant parameters, think of, H field, inductance, k or M, loss (heat loss due to resistance and dielectric materials), impedance (phase and amplitude), etc

If you are going to develop your own power source, you need to understand power conversion topologies for DC to AC conversion. Some applications can work very well with self-oscillating circuits such as the Royer oscillator. Be prepared that self-oscillating circuits may become unstable in case of coupled resonators, this can be very frustrating if you don't now what happens. Strange behavior in oscillators is due to the phase behavior of the input impedance of coupled resonators.

Based on a small signal analysis (frequency sweep in pspice) you can assess where the losses are (assuming you modelled the loss resistors in your inductors and capacitors.

Study what others did, this may inspire you, and may give you some guidance. Disadvantage is that you may close your eyes for new concepts.

Select your favorite simulator where you can simulate coupled inductors and non-linear components to run time domain simulations with actual components. Note that spice based simulators allow direct entry of equations and have high level blocks such as integrators, limiters, multipliers, etc. By using these, you can save simulation time. Any spice-based simulator can be used.

If you want to simulate coil behavior and coupling, you need an EM simulator, HFSS can be used. Though not necessary, understanding the Smith chart and S-parameters maybe helpful as several EM simulators have their focus on RF component and antenna design. using a full EM simulator enables you to find self-resonant issues that may hinder you, or may result in EMC problems. If you are familiar with HFSS, I would go for HFSS.

I would always check simulations with measurements as a simulation is an approximation of reality and you have to make a trade-off between simulation accuracy, simulation time.

In your case, I would start with the inductors as you have to find a compromise between size of coils and efficiency. Small coils will result in low efficiency. This is because of k*Qloaded is of importance. small inductors have low k (at certain distance), so Qloaded must by high. With low k, the unloaded Q of the resonators will no longer be large w.r.t. loaded Q, hence reducing efficiency.

Thank you so much for your help, i really appreciate it

I will start to design my coils first, and i'll do as you said.

Thank you again mate

One more thing! Lprim and Lsec must have the same inductance value? If not, why we want to change their value if we can design the coil as we need?
On what basis the value of the inductance is selected?
What's the difference between the series/parallel capacitors in the circuit (for resonance), or when do we use the series/parallel and for what (pros and cons)?
If you can point me to anything that i can get some knowledge from about this field (ebook, book, ect...)!

Thank you again

Lprim and Lsec can have whatever value, it depends on what you want to make or simulate. They can also have different values (for example small diameter primary coil and large diameter secondary coil).

Looking to "some steps to take" in posting #5, what is your current knowledge on circuit analysis and power electronics?

My knowledge is good, not that great, but i can deal with it
But for now, i wanna understand all the basics of inductive coupling, and things that affect it (increasing or decreasing the frequency, size of the coils, series and parallel resonance, ...).

They can also have different values (for example small diameter primary coil and large diameter secondary coil).

Click to expand...

Yes, but they also can have the same inductance value and not the same size/shape! I was wondering if that could affect the efficiency of the transfer?

I have an example, I wanna design this circuit:

Sec-coil size: 10 mm Ø;
Prim-coil size: 80 mm Ø;
Power to transfer: 1 W (Example:5v , 200ma);
Distance between coils: 10 cm;

So i need to design the oscillator!

Now, what frequency should i choose and on what basis?

After choosing the frequency comes the tuning of the coils, should i use series or parallel resonance? I found this in some paper "Literature review shows that with series resonance power transfer to the external coil from the source unit can be maximized whereas with parallel resonance the output voltage of the internal unit can be maximized." So they chose to use the series resonance in the primary coil and the parallel in the secondary.

What are the other calculations that i need to do?

Please, don't get me wrong, i have some knowledge, but i'm not trying to use it, cos i wanna understand the concept from someone who already has experience with this matter, that's it.

PS: I wanna make this circuit in reality, not simulate it.

Thanks in advance