ST25R3911B: Lower the output power

[luigiantuono]

Hello,

I am working with an NFC reader with an ST25R3911B chip. The problem I am experiencing is the output power. The power emitted by the antenna is too high; therefore, my task is to reduce the transmitted power.
To do this I have relied on reading your application notes. From this extensive reading I realized that there are many possible solutions to reduce the output power, including:
- reducing the supply voltage of the chip;
- using a single-ended antenna instead of differential antenna configuration;
- designing an antenna with higher impedance;
- change the antenna driver output resistance;
- change the chip internal regulated voltage.
Unfortunately, I am working with an already defined differential architecture that excludes the first three options (because would be optimal to not change the current HW structure of the PCB). This allows me to hope for only the last two.
First of all I am assuming that these last two options do not require a change in the PCB hardware, if this is true could you give me a hand in figuring out how to implement these last two options at the firmware level (register addresses to overwrite and how)?
Thank you for your time.

Sincerely,
Luigi

[Ulysses HERNIOSUS]

Hi Luigi,

• driver resistance can be influenced by setting droff* bits - please be aware that when you are using protocols using AM modulation it you may also need to adapt dram* settings or use automatic adjustment
• regulator based power reduction: Use reg_s=1 and change rege_* bits.

BR, Ulysses

[luigiantuono]

Can you please confirm that the registers you are referring to are:

- for driver resistance : 0x27

- for regulator register : 0x2A

if so, I will try both of the approaches, and I will be back in case I need more help. Thank you. 

Kindly,

Luigi

[luigiantuono]

Hello again Ulysses,

unfortunately, I cannot see any reduction in the power output, and also, using a spectrum analyzer, I saw a weird peak around 27.12MHz (which is both the second harmonic and the frequency of the XTAL we are using).

I am afraid that I am doing something wrong in the other registers. When re-flashing the chip I started fresh with the configuration suggested in the AN5592 for the differential antenna as shown in the image:

It does not add up to me, however, that by changing the parameters you suggested in the previous response, from the Spectrum Analyzer, I see no significant difference. Do you have any suggestions on what to do? 

If, by any chance, I am doing something extremely ***, I attach here the register's content of the chip to complete the description of the problem.

Thank you for your time.

Sincerely,

Luigi

[Ulysses HERNIOSUS]

Hi Luigi,

not sure how you are performing any of the tests. For seeing an effect of course the chip has to be put to ready mode (en=1) and field needs to be enabled (tx_en=1).

Then you should see immediate effect on your spectrum analyzer when switching register 27h e.g. from 00h to ffh. 

If still not working as expected, then please share the step-by-step details of what you are doing!

BR, Ulysses

[luigiantuono]

Hi Ulysses,

I acquired some pictures of the Spectrum Analyzer so that you are able to see what's going on. I also saved the registers' content while coupling between tag and reader was happening so there should be no unexpected values in the registers. 

Just to give you a bit of context, I am using a passive NFC tag with a ISO/IEC 14443-3 Type A  based interface.

This is what I can see from the Spectrum Analyzer when 0x2A = 0xC0 , 0x27 = 0x7F (scenario1):

While this is what I saw on the Spectrum Analyzer when 0x2A = 0x00 , 0x27 = 0x00 (scenario 2)

 I will attach both or registers' content marking them with scenario #. 

Thank you for your time Ulysses, I appreciate it. 

Hope you have a good day,

Luigi

[Ulysses HERNIOSUS]

Hi Luigi,

I am also a bit surprised but as I am more into software - would expect a much higher 13.56MHz. I am not too familiar with spectrum analyzers. Maybe @Travis Palmer  can have a look and has some idea.

What was the tag position in the reader field? Which reader antenna? Where do you measure?

For the register settings: I would play with one setting at a time to begin with. Setting reg_s=0 as in your second scenario causes using the values from register 2Bh (result of Regulator adjustment). So unless you read that register  you will not know the value.

Setting register 27h to ffh should essentially disable the field (drivers all high-Z). I would start with that then move to feh, ... to see effects.

BR, Ulysses

[Travis Palmer]

Hello Luigi,

I don't know why your 2nd harmonic (27.12MHz) is so high.

You can see a nice reduction on 13.56MHz from -61dBm to -71dBm. 

How do you connect the SA to the PCB? If you are probing at the RFO directly, it is quite clear to me. The chip is outputting a rectangular signal, which is converted by the EMC filter into a sinus. 

The easiest way of confirming the output power and output power reduction is to measure the antenna voltage via a scope. Additionally i would recommend to verify the RFI voltage to be max 3V (idealy around 2.5V to 2.7V).

Please let me know, if this helped.

BR Travis

[luigiantuono]

Thanks for your answers,

I am going to answer back in chronological order. So starting from @Ulysses HERNIOSUS :

- The tag is in the reader field during the dump of register's content while the reader antenna is a custom printed antenna. I am sure of the correct NFC pairing between the passive tag and the reader cause I set a blinking led feedback on the custom printed PCB that host the ST25R3911B.

- I am measuring the spectrum using a SA as in the setup shown in the attached image:

The setup is installed inside a metal box.

I am quite sure that this partially answers to @Travis Palmer 's questions anyway I would like to give to you some more details regarding the HW configuration we used for the ST25R3911B. This is only becauise now my main issue is to reduce the power associated to the second harmonic of the signal. After that I weill go on with the lowering the output power of the chip itself.

Assuming that we are using an antenna (for the NFC) that has this characteristics:

Using then a tool for the antenna matching I found out that the suggested matching circuit is this:

Which is not that different (at least for the EMC filter) with respect to the one used on the custom PCB I am referring to, which is the following:

So, assuming that the coupling circuit is well designed (I hope that makes sense to you as well), my final guess to try to solve the problem involves AM modulation.

In both scenario1 and scenario2 files that I attached to my previous message, I saw that the contents of register 0x09 has bit 5 set to zero. This implies OOK modulation which in the case of an ISO/IEC 14443-3 Type A tag is not working properly.
I was wondering, the Analog preset command, from the information obtained from the configuration register (0x03) should automatically set the correct modulation type for the specific tag but this does not happen. Assuming this is the problem (I'm not sure because I don't know for sure how the chip behaves and whether it changes the settings and then resets them with each coupling cycle), could you help me understand this properly?

Also, if you find any obvious issue with the HW matching circuit please let me know because it seems fine to me and I don't want to go in the wrong direction keeping you busy with my questions.

Hope you're having a good day.

Kindly,

Luigi

[Travis Palmer · Accepted solution]

Hello Luigi,

Getting one step back, I try to summarize:

Initially you wanted to reduce the output power of the ST25R3911B. 

Measuring the field strength with a SA you have seen that the higher harmonics are quite high.

The explanation for this is, that NFC is based on magnetic coupling. Two loop antennas (some people are quite strict and claim that this are coils and no antennas) are communicating via magnetic coupling. The SA uses a normal E-Field antenna, which is most probably "only" sensing the electric near field of the PCB. 

In other words, I have doubt that this setup is appropriate to evaluate the magnetic field strength of NFC systems. I would recommend to measure the field strength according to ISO10373-6 chapter 5.3 or 7.1.1.

Do you have any issues regarding the 2nd harmonic - 27.12MHz and any emission test / emission standard? 

Please let me know, if this helps you.

br Travis