• State TI Thinks Resolved
  • Locked Locked
  • Replies 10 replies
  • Answers 4 answers
  • Subscribers 20 subscribers
  • Views 3436 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

4 W HF RFID reader with copper tube antenna read range

Tomas Petlach
Prodigy 110 points

Other Parts Discussed in Thread: TRF7960A

Hello Team,

I would need to clarify a few things about the read range of ISO15693 tags of the long range reader designed by the TI 4W HF RFID power amplifier application note with Cu tube loop antenna.

We designed longe range reader with TRF7960A and power amplifier according to app note (sloc132). The output of the amplifier is tuned with spectrum analyzer for maximum output power 4 W on the maximum PA supply voltage of 12 V. In the next step the we tune the output phase shift to 45 degrees with 50 ohm load at the output connector.

1. Is it appropriate to change the phase shift by the change of the output capacitance (C112, C113) or by the change of the inductance of coil L105 (we'd rather change the inductance)?

Next we constructed the copper tube loop antenna according to the TI HF antenna cookbook and design notes with slightly less size of 480x480 mm and with gamma matching. With vector network analyzer we tuned the antenna to 50 ohm @13,56 MHz and Q of 17. The return loss of the antenna has very similar characteristics as the antennas in app note (sloc132), it has -10 dB near the 13,15 and 13,80 MHz frequencies and -14 dB at 13,56 MHz.

One thing that I need to clarify is the Q and the bandwidth of the slightly mistuned antenna in app note. As I understand, the transmitting bandwidth is very narrow due to the very slow data rate and amplitude modulation, but the receiving bandwidth is 484 kHz on both sides from 13,56 MHz due to subcarrier frequencies.

2. Should't be the Q of the mistuned antenna lower than 20 due to the offset of the band center fequency from the 13,56 MHz?
3. How does it help to mistune the antenna to prevent the transmitter power from saturating the reiver inputs? I assumed, that with the better antenna matching (lowest return loss) the power from transmitter is more effectively transmitted by the antenna and less power returns beck to the reciver. What is the difference in this case?

We tested a read range of couple TI tags on 1W output power with following results:

circular inlay D 24 mm RI-I16-112A-03: read range of 16 cm
circular inlay D 32 mm RI-I17-112A-03: read range of 26 cm
rectangular inlay 75x45 mm RI-I02-112A-03: read range of 46 cm

I found on this forum a TI document with read ranges tested with 300 x 300 mm antenna for different tag types (7318.TI HF-I Read Ranges.ppt). When I focus on tag -CB1A CARD (large rectangle)(assume that corresponds to RI-I02-112A we used) the read ranges with 300 x 300 mm antenna TI used are even slightly better than we achieved with our 480 x 480 antenna. We tried mistune the antenna to the lower and to the higher frequencies, but with no better read ranges than mentioned earlier. Another thing is, that the increasing of the transmitter output power from 1 W to 4 W increased the read range only by 4 cm to 50 cm.

4. Can you suggest what we should to focus on to check that everything is set as it should be?
5. Is there a way to find out directly if the receiver is saturated by the transmitter signal? Is it measurable at some point of signal path?

We also tried to design "simpler" wire 1 turn loop square antennas.
The 1st has dimensions of 450 x 450 mm and is wound from 1 mm^2 stranded wire with inductance 2,66 uH.
The 2nd is 650 x 650 mm from 2,5 mm^2 stranded wire with inductance 4,37 uH.
We used both the transformer matching and capacitive matching with the same results.
With the transmitt power of 1 W and RI-I02-112A tag we achieved read range of 40 cm with the 1st smaller antenna and the read range of 45 cm with the larger 2nd antenna. We tried to mistune these antennas both to the higher and to the lower frequencies, but those were the best results.

6. Is there some problem with wire loop antennas, that would cause worse read ranges? I understand that using the wire instead the tube results in higher inductance, but with proper matching to the same impedance and Q, should be the results very similar?

Thanks and regards,

Tomas

  • 0
    TI__Guru* 91296 points

    Tomas -

    Sorry for the delay - we had short week last week - so i simulated this sloc132 design today and i found that in the simulation, either side of the L105 were not very good looking signals. I changed C111 and C112 value from 68pF to 100pF, then both to 220pF and got noticeably better waveforms on both sides and phase shift appears correct.

    Q value needed for the RFID / NFC reader antennas is generally set by the bandwidth of the tag response. if using ISO15693 double subcarrier tag response, then BW should be around 848kHz, which would then put the Q value (when using 13.56MHz as fundamental) at 15.99 or 16. Staying under 20 is generally recommended as a result. 

    regarding the higher inductance values you were using - this is generally not recommended because then you will have a tough time tuning the antenna in this band to 50 ohms. (the heavier the coil, less shunt cap is needed....so then in the range above about 3uH the shunt cap becomes almost nothing, which means you have less control available to you though changing that cap value and its almost impossible to tune it correctly.)

    can you check the waveforms and performance after making that simple change on the front end and let us know how that worked out for you/

  • 0 in reply to Josh Wyatt
    TI__Guru* 91296 points

    Tomas -

    spent a little more time on this - see below - in the simulation, i made the parts have the same designations as what is in sloc132 so you could easily see which ones to change (and these are now also standard values) - simulation results looked alot better - see what you think. also, notice that diodes are gone and only one RX channel is being used. this may be enough for you to be successful. 

  • 0 in reply to Josh Wyatt
    Intellectual 260 points
    Hi Josh,

    I'm using the old value of parts. If I change as above (remove envelop detector, and using CX, CY), is it right?
  • 0 in reply to chia se
    TI__Guru* 91296 points
    maybe so - you would be wise to check that RX voltage level before connecting to the RX pin...the buffer resistor there might need to change value if the level is too high.
  • 0 in reply to Josh Wyatt
    Prodigy 110 points
    Hello Josh,

    thank you for the suggestions.

    We have short week this week :-) so I will test the modifications tommorow and let you know the results.

    Before that I have a few questions:
    You have changed also the inductances L101 through L104 slightly from 538 nH (in sloc132) to 560 nH, is that critical or it could be let on previous values?
    Without the second receiver input used, I thing there is no need for setting the phase shift on L105, right? Is there a higher risk of receiving holes in receiving space with receiving only one phase of signal?
    How do you judge simulation results? What parameters do you watch in result?

    Thank you.

    Tomas
  • 0 in reply to Tomas Petlach
    TI__Guru* 91296 points
    the waveforms themselves is what i was looking at - with the other values I saw misshapen signals where there should be sine waves and ringing, overshoots, etc.

    the second receive channel could be used if needed, but you might not need it at all if you have slow moving application.
  • 0 in reply to Josh Wyatt
    Intellectual 260 points

    Hi Josh,


    Thanks so much! I try the change and inform the result! I hope it works well.

  • 0 in reply to Josh Wyatt
    Prodigy 110 points
    Hi Josh,

    a few results from yesterday measurements. I did modifications according to your 1st post. Increasing both capacitors C111 and C112 to 220p changed the phase shift to 15,6 ns. With this phase shift the read range lowered from 46 cm to 38 cm. Then I tuned the phase shift back to 9,6 ns by lowering the C112 to almost the same value as earlier. The read range increased, but only to 42 cm.
    The 2nd modifications you suggested we could't make yet, because 560 nH air inductor propably isn't a standard value, because both of our suppliers (Farnell and Digikey) doesn't have greater than 538 nH inductors. Also the Coilcraft and Abracon coil manufacturers have the maximum value of 538 nH of air inductors for these currents. So we will have to find the other disrtibutor or try to make the coils by wounding our own. As soon as I have the results, I'll post it, but it won't be earlier than in 2 weeks because of our company holiday.

    Thank you for now.

    BR

    Tomas
  • 0 in reply to Tomas Petlach
    Prodigy 65 points

    Hi Tomas!

    Will you be using this circuit it ( See the attached files7711.appnote_trf796x_pwramp_4w.pdf)? I designed the circuit according to the attached files ,But  read range is less than 10cm,  why? Could you please offer  your details design of circuit and antenna ? 

    Please suggest.

    Thanks

    Yuan

  • 0 in reply to user4397992
    Prodigy 110 points
    Hi Yuan,

    we made the reader and antenna exactly according to the TI app notes and it works on the distance of about 40 cm for 75x45 mm cards. But this project is suspended for some time due to another projects, so we ended with this results for this time.