I am designing a bioimplantable circuit for a biomedical implant. The goal is to have an extremely tiny device in a flexible substrate (Dupont Pyralux) that must be powered wirelessly.

My problem comes here. I do not have much knowledge about antennas. I have been told to design an antenna that must work at 13.56 MHz. I am following the design of John Rogers (page 5, figure 2,b), Implantable, wireless device platforms for neuroscience research.

Then I have some doubts about the design that I would like to receive some feedback about it.

  • How would you design a PCB antenna with that operating frequency (13.56 MHz)?
  • How would you address this problem? I need a primary and secondary coils working in that frequency.

I use Eagle but I don't know how to design, for example, a loop antenna in that software. If you recommend to me any other software for designing the antenna, simulating it and attaching it to my circuit, it would be great.

  • \$\begingroup\$ I am in an antenna design course right now, but am not an antenna designer. Common antennas such as patches, dipoles, and loops have been analyzed in papers such as Microstrip Antenna Technology by Carver. In the course we start with equations from such papers, and then model the antenna in a simulator such as Ansys Electronics Desktop. The simulation results are typically close to the equations, but usually we have to adjust the dimensions to meet the design requirements. \$\endgroup\$ – DavidG25 May 6 '19 at 18:24
  • \$\begingroup\$ Also, if you use a resonant antenna, it is going to be pretty big for 13.56 MHz. Quarter wavelength will be something like 2.5m. \$\endgroup\$ – DavidG25 May 6 '19 at 18:28
  • \$\begingroup\$ Yes, we are going to quarter wavelength. That's the approximate size of the primary coil? The antenna must be design in the PCB so, we count only with a single layer or even double layer design. Where would you start designing that? \$\endgroup\$ – Mufasa May 6 '19 at 18:31
  • \$\begingroup\$ Again, I’m not an expert on this, but I’d read a paper that describes analysis and design of the antenna you want to make and then I would simulate it. \$\endgroup\$ – DavidG25 May 6 '19 at 18:35
  • \$\begingroup\$ I already did it. I will try to simulate it then. I need to figure out how to link the antenna and the circuit. Cheers. \$\endgroup\$ – Mufasa May 6 '19 at 18:45

Microchip has technical information about antenna design at 13.56 MHz. Start with AN710 "Antenna Circuit Design for RFID Applications".

  • \$\begingroup\$ We want to build pour own antenna, we cannot order any commercial antenna cause that would be unaccurate for our design. \$\endgroup\$ – Mufasa May 6 '19 at 22:02
  • \$\begingroup\$ @mufasa AN710 is a 50-page book on how to design planar loop antennas for 13.56 MHz, with example designs, matching networks and equations. Why not start by reading it carefully? \$\endgroup\$ – tomnexus May 7 '19 at 5:01

For a 13.56 MHz system, you need two coil antennas, because inductive coupling is the best way to transfer energy and data for that given frequency.

The term "RFID" (Radio Frequency Identification) is your friend. Narrow your search down to "RFID antenna design" and you will find a lot of information on the internet on how to design such antennas.

  • \$\begingroup\$ Ok. Do you recommend me any software to design the antenna and simulate the performance? \$\endgroup\$ – Mufasa May 6 '19 at 22:02
  • \$\begingroup\$ @mufasa Ansys FEKO is a good option. It's capable of this sort of design and more affordable (perhaps 1/10 of the price) of the heavyweights like CST. If you are a student/researcher at an educational institution you may be offered a discounted or free version. \$\endgroup\$ – tomnexus May 7 '19 at 5:06
  • \$\begingroup\$ I can recommend FEMM, a free open-source tool. I did some RFID antenna design with that. (femm.info/wiki/HomePage) \$\endgroup\$ – Stefan Wyss May 7 '19 at 5:08
  • \$\begingroup\$ Cheers to both of you! \$\endgroup\$ – Mufasa May 7 '19 at 8:01

Implantable loop antenna have a loading effect due to the Dk =80 of water so the capacitance of fluid has 80 times more than air and this reduces the self resonant frequency of any tiny coil inductance by sqrt (80) or almost f/9.

So one approach is find a RFID or charger coil that is more than 9x your desired frequency and tune lower with a cap as required. Since saline is conductive it has some loss or load resistance, so get an RLC calculator or Nomograph to choose your Q and BW and resonant frequency vs impedance.

Check first what materials are allowed and consider Teflon, polyamide, and other inert elastomer dielectrics for compatibility. Since these are low Dk e.g. 4 compared to 80 of water, the net effect is the the capacitance in series.

  • \$\begingroup\$ We are already using polyamide (PDMS) for the substrate. We ordered the pyralux dupont substrate that is pretty good for this purpose. I guess that given an inductance (obtained from the geometric shape of the coil) we can calculate the capacitance and then we can just design the coil in a CAD software. Is it right? \$\endgroup\$ – Mufasa May 6 '19 at 22:00

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.