3
\$\begingroup\$

I recently stumbled upon a two-in-one NFC key designed by N-O-D-E and was astonished.

Although its original purpose is for security, I love the idea of using tact switch to decide which NFC chip is responding to the reader. So I wanted to build myself a mifare 1k version of it.

But as stated in the description:

The antenna itself was designed using this useful online generator, and it has enough inductance to run one chip at a time. It's also intentionally tuned just outside of the 13.56MHz NFC standard, for security, so you have to basically place the tag directly on top of the reader for it to work.

This part just gave me headache. I wanted to make myself one of this is for the convenience of carrying multiple NFC chip in a small package, so I'd like to make it respond to the reader from a reasonable distance.

I have definitely no experience in antenna designing so I did a little research.
Most Mifare tags use thin wire coil as its antenna. Those antennas have large hollow in the middle and as far as I know, if the turns of the coil are same, larger antenna hollow (area) can make the tag respond from greater distance.

But, in N-O-D-E's design, he uses much wider wire and filled the whole antenna. enter image description here



which makes me really curious.
Most NFC antennas printed on a PCB are wide wire with few turns, why don't they make thinner wire with more turns and larger hollow?

At fist I was wondering if filling antenna area would have negative affect on inductance, but from ST edesign suite I know that filling the middle area may even increase inductance. enter image description here

The design of on board antenna and coil antenna seems conflicting to me.
I wonder which kind of design is best suited for pcb printing and would like to know the difference between using wide and thin wires for antennas.

\$\endgroup\$
  • 2
    \$\begingroup\$ These NFC systems don't really use what antenna and RF people call "antennas", as they all work in the near-field. I suspect that it is important here to get a very good Q-factor, which requires the ESR of the coil to be low. Skin-effect probably is not too much of an issue yet at these frequencies. \$\endgroup\$ – Joren Vaes Jun 2 '18 at 5:13
  • \$\begingroup\$ Thanks for your advises! I did read the technical document AN1445 and now I know the inductance should be between 300nH and 3µH. And the more turns may have more coupling problem. So now my question becomes: If whenever I use my tag, there will be no other NFC tags near it, Is it better for me to increase turns to aim for inductance that's closer to 3µH? (e.g. Does greater inductance means greater response distance?) \$\endgroup\$ – Jimmy.D Jun 2 '18 at 6:37
  • \$\begingroup\$ And because I'm not an engineer of any kind, I'm sorry that I can't really talk the math part of antenna designing... I can only try to draw an antenna in Eagle using parameters given by ST edesign suite. It should work if I draw it correctly, right? \$\endgroup\$ – Jimmy.D Jun 2 '18 at 6:52
  • \$\begingroup\$ @Nipo Oh I see! I've never seen this document before. It's what I've been looking for so long! (Using none-technician word to explain technical things.) Thanks a bunch! \$\endgroup\$ – Jimmy.D Jun 2 '18 at 7:00
2
\$\begingroup\$

Interesting reference documents:

They basically say that:

  • reading distance will be related to antenna diameter,
  • inductance should be kept in the 0.3-3µH range.

Here you are facing "passive" (in spec terminology) NFC communication. This is done through load modulation. Communication medium is a transformer[1].

Being a transformer, voltage at secondary coil poles is a factor of the number of turns. So the more turns, the more voltage (and the less current). Because we use digital ICs with clamping diodes on secondary side, having more voltage there makes little interest: it will flow through clamping diodes and will basically be useless.

Another important design consideration for a PCB with both antenna and components is that you should not block the magnetic field with the chips and traces. Keep antenna trace at the outer region of board, put components in the middle.

1: Then talking about antenna is somewhat wrong, but this is the usual terminology, so we'll keep it. "Active" communication, on the other hand, is much more like using antennas -- that's why NFC antenna requirements are different depending on whether you intend to use passive only or both communication methods.

\$\endgroup\$

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.