August 12, 2015: I discovered something interesting on the original antenna design today. I had copied the antenna design from a TI reference design without really thinking about it. When I would use the prototype with this antenna design, the range was much worse than other antennas. I calculated the antenna inductance using http://www-smirc.stanford.edu/spiralCalc.html

TI's patch reference had 3 turns of 550 um spacing with 380 um turn width and an OD of 30850 um, leading to something like 600 nH, where the RF430FRL152H datasheet expects 2.66 uH. This is off by a factor of 4.4. Once I realized this, I added a larger capacitor to shift the frequency to where it needed to be. I put a 150 pF cap in parallel with the antenna and got much more range. However, I think that capacitor is effectively shunting a lot of current from the IC itself, reducing power transfer and range. I'll need to redesign the antenna for the next version. Based on the calculator, 6 turns, 200 um space and trace, 33000 um OD will get me closer to 2 uH.