Design considerations for bifilar coils for Wireless Power Transfer

Question

I found many different DIY videos and posts demonstrating the concept of wireless power transfer using inductive coupling, however, they have limited explanation on theoretical calculations and design aspects. Thus, I hope to find some guidance on this forum. I focused my attention on the pancake and bifilar coil geometries since they seem to be compatible with the design constraints of the receiving coil. Since the receiving coil has design limitations, all bets are on the transmitter coil, so it must be more powerful to compensate for the low receiving efficiency. The goal is to be able to receive at least 100mW of power.

The Receiving Coil

The receiving coil will be a simple pancake coil PCB printed with the thickness of traces being 300um. The width and the spacing of the "flat wires" would be equal and can be >4mm, and the outer coil radius 100mm. These parameters would dictate the number of turns in the coil. At this point I am not concerned about the N1/N2 winding ratios of the transmitting and receiving coils since I am planning on using a voltage regulator at the receiver to maintain the required voltage.

The Transmitting Coil

The transmitting coil would be the Tesla's bifilar pancake coil - I would just pick a random speaker wire and make as many turns as needed to approximate the outer diameter of the receiving coil. It is not clear, though, how to wind the double-stranded wire: two strands side-by-side on the same plane (left), or one strand on top of the other forming a stack of two parallel coils (right)?

I have also seen some speculations on how to connect the two coils of the bifilar coil, and it is not clear which one is suitable for my application: the one that connects internal end of one coil to the external end of another coil (left), or the one that connects two internal ends (right)?

The questions:

1. What are the design recommendations for the Receiving coil (given above constraints)?
2. What are the effects of the two winding approaches, and which one is more suitable for my application?
3. How the coil interconnection approaches affect the overall power transmission efficiency and which one is more suitable for my application?
4. Should I have thicker wire with less turns or a thinner wire with more turns for the transmitter coil?
5. In both connection scenarios I end up with three points (x,y,z) to which I could connect the driver circuit. What are the connection approaches and driving options that are valid? I found several examples that use a single transistor and a battery to drive the transmitter coil, but I was hoping for a slightly more complex and effective suggestion.
6. Is the spiral receiving coil (shown above) compatible with a bifilar type transmitting coil?

Answer 1

1. What are the design recommendations for the Receiving coil (given above constraints)?

Designing the coils for wireless power is a rather complicated topic. You can find heaps of research papers on the topic. It depends on many parameters such as your working frequency, sizes of the coils, transfer distance, power level, target efficiency, etc. If you are after the high-efficiency, then you should maximize coil quality factors (\$Q_{\rm{TX}}\$ and \$Q_{\rm{RX}}\$) and the coupling coefficient (\$k\$). The figure of merit for high efficiency is \$k^2 Q_{\rm{TX}} Q_{\rm{RX}}\$. You should minimize your coil resistance to maximize the coil quality factor and increase the coupling between them, which will again be a complex optimization problem. There are some options for you to choose: The use of Litz wire to reduce skin effect losses, and optimize the number of turns and the gap between tuns to optimize \$kQ\$ product. For instance, you can find such numerical optimization in this paper.

2. What are the effects of the two winding approaches, and which one is more suitable for my application?

For the winding of two turns, one strand on top of the other is better because you can use the available space effectively. you can have more turns for a given outer diameter of the coil to have higher inductance (and mutual inductance), and/or you can introduce a small gap between turns to reduce proximity effect resistance.

3. How the coil interconnection approaches affect the overall power transmission efficiency and which one is more suitable for my application?

The logic is pretty simple. You should make sure current in the winding flows in the same direction so the effective field from both windings added up. Otherwise, the field created by two windings will be canceled each other.

If you use the left one, you should connect your source between terminals 'x' and 'y' so that two windings are in series - the effective total number of turns will be 2N.

If you use the right one, then you should connect the source between terminals 'y' and 'x-z connected together', so that two windings are in parallel-the effective coil resistance will be half.

4. Should I have thicker wire with less turns or a thinner wire with more turns for the transmitter coil?

This is again a complex optimization problem for maximum \$Q=\omega L /R\$. A higher number of turns will increase the coil inductance \$L\$ (having a positive effect on \$Q\$), but it will also increase coil resistance \$R\$ (having a negative effect on \$Q\$). If your working frequency is low (say, below \$50~\rm{kHz}\$), then you can assume the proximity effect resistance is negligible. Then you may approximate \$L\propto N^2\$ and \$R\propto N\$, which makes a higher number of turns a better choice.

5. In both connection scenarios, I end up with three points (x,y,z) to which I could connect the driver circuit. What are the connection approaches and driving options that are valid?

refer response to Q3

I found several examples that use a single transistor and a battery to drive the transmitter coil, but I was hoping for a slightly more complex and effective suggestion.

Now you are referring to the configuration of the driving circuit where you have many options. I suggest the use of a current fed push-pull oscillator, for example, the circuit configuration given in this application note can be a good choice.

6. Is the spiral receiving coil (shown above) compatible with a bifilar type transmitting coil?

Yes, provided that you configure bifilar coil correctly.