ST25R500 Unexpected Reading Distance Results

[Chen Gu]

Help Needed: ST25R500 Antenna Matching and Unexpected Reading Distance Results

Hello everyone,

Thank you for taking the time to read my post.

I am currently using the ST25R500 and designing a reader with the goal of achieving the maximum possible reading distance. I have been working on the antenna design, impedance matching, and the internal transmitter impedance settings of the ST25R500, but I have encountered some confusing results and would really appreciate your advice.

My antenna is a 2-turn enamelled-wire coil. For the impedance matching, I am using the impedance matching software officially recommended by ST.

Based on the recommendations in the datasheet, I set my target impedance to:

10 Ω − j4 Ω

As far as I understand, this target impedance should be within the recommended range.

I entered my antenna parameters into the official matching software and used its built-in simulator. I then adjusted the values of the matching components, as shown in Figures 1 and 2.

However, after actually building the circuit and measuring the antenna feed point with a NanoVNA, I obtained a significantly different result from the simulation.

The measured impedance was approximately:

Z ≈ 23.55 Ω − jX

The corresponding reactance is approximately equivalent to:

C ≈ 3.11 nF @ 13.56 MHz

My NanoVNA measurement setup is shown in Figure 3.

(This is the schematic diagram of the wiring I did during the test. On the actual PCB, it is not powered on. ST25R500 should default to the high-impedance state.)

This leads to my first question:

Why is there such a large difference between the impedance obtained from the official ST matching software simulation and the impedance measured in the actual hardware using the NanoVNA?

There is another issue that is even more confusing to me.

Based on the actual antenna impedance measured by the NanoVNA, I further adjusted the internal transmitter impedance setting of the ST25R500.

My initial understanding was that if the actual antenna/feed impedance was around 20 Ω, then setting the ST25R500 internal transmitter impedance to approximately 20 Ω should provide relatively good impedance matching.

I am using a 5 V supply, and I also applied the 1.75 factor specified in the datasheet when converting the internal impedance setting. Therefore, I set the internal impedance to approximately 20 Ω.

However, the actual test results showed a very counterintuitive phenomenon.

When the ST25R500 internal transmitter impedance was set to approximately 20 Ω, the reading distance was not optimal.

When I further reduced the internal impedance, and eventually set it to approximately:

2 Ω

the actual reading distance became significantly longer than when the impedance was supposedly matched.

This is very confusing to me.

According to the conventional understanding of impedance matching, the output impedance of the transmitter and the load impedance should be reasonably matched in order to achieve better power transfer. Therefore, I originally assumed that:

Transmitter output impedance ≈ Antenna/feed impedance

should result in a better reading distance.

However, my actual test results seem to show the opposite:

Internal impedance ≈ 20 Ω → shorter reading distance

Internal impedance ≈ 2 Ω → longer reading distance

Therefore, I would like to ask the following questions:

1. How should the target impedance in the official ST matching software be interpreted?
   Is the target impedance the actual load impedance that should be presented to the ST25R500 RFO1/RFO2 outputs?

2. Why is there such a significant difference between the simulation result from the official ST software and the impedance measured with the NanoVNA?
   In this situation, should I use the simulated result as the reference, or should I primarily rely on the actual antenna/feed impedance measured at 13.56 MHz?

3. What is the actual relationship between the ST25R500 internal transmitter impedance setting and the external antenna/load impedance?
   Is it incorrect to simply assume that the best condition is achieved when the chip output impedance is equal to the antenna impedance?

4. Why does reducing the ST25R500 internal impedance from approximately 20 Ω to approximately 2 Ω result in a longer reading distance?
   Could this be related to the characteristics of the internal power amplifier/output stage, output current, load impedance, harmonics, matching network, or the actual voltage/current at the antenna?

5. If my primary goal is to maximize the reading distance rather than minimize power consumption, which parameter should I focus on optimizing for the ST25R500?
   Should I primarily optimize the transmitter output power, antenna current, antenna voltage, magnetic-field strength, or the power-transfer efficiency at the RFO output?

My actual measurements seem to indicate that simply achieving the best conventional impedance match does not necessarily result in the maximum reading distance.

Therefore, I suspect that there may be more involved here than conventional conjugate impedance matching, possibly related to the actual operating characteristics of the ST25R500 output stage.

If anyone has experience designing NFC/RFID readers using the ST25R500, ST25R3916, or other ST25R-series devices, I would greatly appreciate your help in analyzing this behavior.

In particular, I would like to understand the entire signal/power chain:

RFO output stage → matching network → antenna → actual antenna voltage/current → magnetic-field strength → tag coupling → reading distance

Any advice or technical explanation would be greatly appreciated.

Thank you very much!

[Travis Palmer]

Hi Chen Gu,

First many thanks for the very detailed documentation! It helps a lot to reproduce your steps.
Regarding your Question: 
Why is there such a large difference between the impedance obtained from the official ST matching software simulation and the impedance measured in the actual hardware using the NanoVNA?
The matching impedance of Z=23 -jx is not concerning to me at all. There are parasitic which cannot be estimated in the simulation since they are related to PCB layout and PCB stack-up. For example, the dimension of the parallel capacitance track width and its capacitance to the below metal layer. It creates an additional parallel capacitance of typically ~3pF to 10pF which is shifting your resonance point lower in frequency. Using the strong capacitive matching (as you did) such additional parallel capacitance is increasing the matching impedance. Please take a closer look at chapter 8.4 Matching network behavior of AN6092 for finetuning the matching impedance. 
Regarding your Question about internal transmitter impedance:
In the frequency range of 13.56MHz the impedance matching is not so important. For our application it would be even an issue. If we have the same source and load impedance, it would mean that the same power is dissipated in the transmitter as in the load. Building an IC which outputs around 2W, it would also dissipate 2W. At a thermal resistance of 50°C/W the device would heat up to 100°C. The efficiency is then around 50% (a bit less because of the internal LDO and logic dissipating additional power). 
The ST25R family is designed to have quite low driver resistance to minimize the losses in the driver. It allows to achieve an efficiency up to 83%. 
 

In your case (source impedance 20Ohm, Load impedance 20Ohm) the output power would be around 200mW at ~100mA driver current. A very reduced read range is expected. 

Regarding your other questions:
How should the target impedance in the official ST matching software be interpreted?
The target matching impedance is the most important criteria to adjust the output power according to your power budget. Not everyone wants to achieve 2W of output power. Choosing the lowest driver resistance at a higher matching impedance (15 to 20Ohm) might give you satisfying read range results at a very efficient matching. 
Why is there such a significant difference between the simulation result from the official ST software and the impedance measured with the NanoVNA?
Has this question been sufficiently answered above?
What is the actual relationship between the ST25R500 internal transmitter impedance setting and the external antenna/load impedance?
Has this question been sufficiently answered above?
Why does reducing the ST25R500 internal impedance from approximately 20 Ω to approximately 2 Ω result in a longer reading distance?
Please have a look at chapter 4 Dissipated power calculation of AN6092. It explains the relationship between dissipated power, ouput power and efficiency. You can put the formulas in an excel file to compare different scenarios fitting to your application. For example:
 

To correctly do the load balancing you would need to go the other way around. Instead of increasing the transmitter impedance, you would need to lower the matching impedance.
Zsource  ~ 2*0,63, ZLoad = 1,26Ohm
But the result would violate the Abs max ratings specified in the datasheet: 
 

And the device would stop working.
If my primary goal is to maximize the reading distance rather than minimize power consumption, which parameter should I focus on optimizing for the ST25R500?
The relation between output power and reading distance is not linear. Which means 2W does not offer twice the read range compared to 1W output power. To achieve highest output power you should:
•    Select highest supply voltage
•    Lowest driver resistance
•    A suitable LDO drop (300 or 350mV)
•    High quality EMC inductors
•    A suitable system Q-factor 
The target matching should then be chosen according to the power you want to spent during Tx and your expected read range result. 
Please let me know if this answers your questions.

BR Travis

[Chen Gu]

Thank you, Travis Palmer.

I will carefully consider your suggestions, as well as the relevant technical documentation provided by ST.

At the moment, I am achieving a reading distance of approximately 40 cm. However, I am still hoping to further improve the reading distance if possible.

I will continue optimizing the antenna and matching network based on your suggestions and the ST documentation. Thank you again for your help and guidance.