2
\$\begingroup\$

I would like to use 1 RFID reader (13.56Mhz, protocol iso14443a) with 2 antennas in a way that I can recognize the location of the tags- if they stand on antenna #1 or #2.

For that, I tried to use 2 RF switches that can work from 10Mhz to 3Ghz (or any other option that includes 13.56Mhz in it). I used the RF switches on the TX lines- TX1 and TX2 of the differential output of the reader, and with the combination of the RF switches I tried to control one antenna at a time.

The idea is to switch every X ms between the antennas and by that know where the tag is. the control is with 2 GPIOs (1 & 0 to activate one antenna and 0&1 to activate the 2nd antenna).

From the material I found on the internet, i saw that it can also be done using high-speed RF relays, but I also saw that they are more expansive.

This is the RF switch I checked: http://www.rfmd.com/store/downloads/dl/file/id/30282/rf3023_product_data_sheet.pdf

And this is my reference design:

2 RF switches for to controol 2 antennas

After I tested in the lab this RF switch I notice that i keep getting recognition from both antennas in all modes. It seems like the component keep shortened the RFC(5) to both RF1(1) and RF2(3), that it doesn't really disconnects between RFC and RF1 (or RF2).

  1. Please advise why do you think it happens. howcome that this component doesn't fully disconnect the other net?
  2. Assuming I need a very low cost solution, do you have other alternatives?

Thank you!

Dudi

\$\endgroup\$
  • \$\begingroup\$ iI've sucesfully used ordinary relays at this frequency. 13Mhz has more than 20M wavelength centimetre scale devices are practically invisible. \$\endgroup\$ – Jasen Nov 22 '15 at 10:26
  • \$\begingroup\$ The part is rated up to 4GHz, it is no wonder it is expensive and seems overkill for an application at 14MHz. Isolation is spec'd for 34dB at best so the question rises: How much signal attenuation do you measure in your application? \$\endgroup\$ – jippie Nov 22 '15 at 11:06
  • \$\begingroup\$ measure the voltages on all the RF pins with an oscilloscope and check that they are in range. \$\endgroup\$ – Jasen Nov 22 '15 at 11:09
  • \$\begingroup\$ Are you building your RF circuit on one of those proto boards that has many holes for poke in wires? If so be advised that these boards have capacitance from every connection point to the next connection point. That in addition to the flying wires used to connect up such type of circuit leads to coupling all over the place. \$\endgroup\$ – Michael Karas Nov 22 '15 at 11:12
  • \$\begingroup\$ How are both antennas connected to their transmit circuits? \$\endgroup\$ – Andy aka Nov 22 '15 at 20:28
1
\$\begingroup\$

schematic

simulate this circuit – Schematic created using CircuitLab

  1. It sounds like the isolation between the antennas isn't great. Maybe the antennas just aren't physically far enough apart, or their wires are too close together, resulting in crosstalk.
  2. Switching every few mS probably rules relays out. Here's a diode switcher you could use that only uses common components. The buffer amplifiers should provide approximately 50 ohm output impedance to reduce attenuation. It only needs one GPIO. You could also use jfets for switching but they're less common.
\$\endgroup\$
  • \$\begingroup\$ Thanks! I succeeded doing it by better soldering and keep on distance between the 2 antennas. I'll try the switching with my architecture, and if I'll see any problems I might try your suggestion. Good job! \$\endgroup\$ – Dudi Nov 26 '15 at 9:12

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.