4
\$\begingroup\$

I'm designing a small board that integrate a PN512 IC along with other components (MCU, etc). I need to build a proof-of-concept device for a presentation with the suit people and everything is ready but the PN512 antenna/matching circuit design. My guidelines for a PCB antenna are:

  1. It must be single layer. (I've seen single layer designs with a 0 ohm resistor as a jumper over the traces)
  2. It can't be more than 10x10cm or 4x4in.
  3. I need it to successfully transmit data between the PN512 and an external NFC device, 5cm (2 in) or less range will be more than fine.
  4. I own a 200MHz scope to play with (Sadly, no network analyzer).
  5. My knowledge of the "black magic" behind this is very limited, but I'm learning as i go.

So my question is:

It is possible to build an antenna with the above requirements? Do you know of any (open) design available to get as starting point?

I've found many design guides for 2.4GHz antennas, but nothing helpful for the 13.56MHz range. Any help will be highly appreciated!

\$\endgroup\$
1
  • 2
    \$\begingroup\$ Datasheet for the suit people would be amusing to see :-) \$\endgroup\$ Dec 8, 2012 at 21:40

3 Answers 3

6
\$\begingroup\$

How about this document about antenna design for the PN512? The rectangular antenna on page 47 looks like it will meet your needs.

Texas instruments also makes some fairly decent HF antenna design guides.

I've fabricated some of these rectangular antennas for 13.56MHz NFC before (the ones specified by TI), they're fairly simple and work well as can be expected.

\$\endgroup\$
1
  • \$\begingroup\$ +1 for app note. Cool! Nice find. \$\endgroup\$ Dec 9, 2012 at 23:56
2
\$\begingroup\$

Context

I could give you all the math and physics, but I think you just want the "do this" answer, so here goes:

The NFC part of your design problem relaxes a number of the usual antenna design requirements since it is "near-field" (the NF in NFC). From an end-user performance metric (can it read/write the tag), it isn't super sensitive to the impedances and the antenna impedances don't change a lot given the small-scale and close proximity. Basically, it's all bad, so a little more or less bad doesn't matter a lot. NFC was designed to discourage long-range reading (snooping).

Solution

Follow the Philips (NXP) reference design for the front-end and then draft an antenna something like:

  • 2" x 2" loop
  • with a few turns (say 4)
  • with as filamentary a conductor as possible (your PCB fabricator's minimum track width)
  • with minimum gap between turns (your PCB fabricator's minimum gap width)

Implementation

  • To minimize cross-overs, spiral inwards.
  • Obviously, ensure that there is no metal near the loop and specifically inside of it.
\$\endgroup\$
1
\$\begingroup\$

I think that you can find all your answers in Panasonic's design tool, which also offers a simulation of the distance that you can achieve. https://b2bsol.panasonic.biz/semi-spt/apl/en/tool/nfcdesignnavigator/

Additional resources can be found from NXP, Nordic Semiconductors and Texas Instruments. Search for Nordic's appnote called "nwp_026 NFC Tuning", that will give you an insight of what equipment you might need. I have recently got to work an NFC tag, just using the math calculations from that app note, even without the equipment, and it matched the distance simulated in Panasonic's tool.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

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