NFC WLC with ST25R3916 and ST25DV-i2c

[OSilv.1]

My project is to charge and receive data from a device that has on board energy storage (70F supercap). 300mW of received power is plenty for my application. I thought NFC WLC with iso15693 would be perfect for the job.

From ST's NFC WLC webpage:

https://www.st.com/content/st_com/en/support/learning/st25-education/nfc-for-wireless-charging.html#stm32trust-overview

I see NFC can charge up to 1 watt with an output power on the ST25R3916 of 1.8W. The webinar on the website demonstrates received power of 0.5W.

However, my X-Nucleo-NFC04A1 is only receiving 0.013W on its EH pin with the X-Nucleo-NFC06A1 as the reader. This was measured via a 50 ohm resistor between V_EH and ground. The rest of the board is supplied via 3.3 input, so I assume all of the received power is dissipated in the resistor.

My first thought is that the reader isn't outputting 1.8W. I have tried configuring the RF regulator at 0x2C and TX driver at 0x28 to output max power, which should cause the rf regulator to limit current, correct? I am not seeing the i_lim flag turn on however. Is it possible to ensure that the ST25r is outputting 1.8W? And, if it already is, is the efficiency really only 0.013W/1.8W = 0.7%?

Any help is much appreciated!

[Ulysses HERNIOSUS]

Hi OSilv.1,

the X-NUCLEO-NFC06A1 is not able to output 1.8W with its given antenna and matching. The output power is mostly defined by the matching/antenna impedance. Please see AN5276 for more details.

BR, Ulysses

[OSilv.1]

Hi Ulysses,

The X-NUCLEO_NFC06A1 "Getting Started" data sheet claims up to 1.7W of output power on page 2. Is this correct?

Thank you,

Oliver

[OSilv.1]

If max power is not 1.7W, what is the maximum output power for the X-NUCLEO-NFC06A1?

Thank you!

[JL. Lebon]

Hello OSilv,

Please note that EH (Energy Harvesting) feature of the ST25DV is not NFC WLC (WireLess Charging).

The two things are different.

Intent of EH is to power small devices that consume very much power (diodes, up to a small MCU), not to charge a battery. The tag takes its power from the RF field, and if there is more power that needed, the rest is output on the V_EH pin. But it will not deliver more than a few mW (~10mW in best conditions).

NFC WLC on the other end is a protocol defined in the NFC standard, the specifies a communication protocol (on top of NFC) and a power transfer method. When used in WLC, the ST25DV is there to manage mainly the NFC communication protocol. The power transfer is not done through the ST25DV EH pin, but is made directly on the antenna using a bridge rectifier. This helps to "harvest" more power.

Best regards.

[OSilv.1]

Thank you JL. Lebon, that cleared things up for me. I am wondering if adding a rectifier and PMIC on the receiver antenna will change the matching circuit on the reader? Also, have you seen if the added circuitry interferes with communications? I would love to see a reference schematic or application note on the subject. If ST offered a course for designing NFC WLC (for example, the process used to design the devices that were used in the webinar) I would happily pay to enroll.

I tried adding a rectifier made of four diodes (https://www.onsemi.com/pdf/datasheet/1n4001-d.pdf) to the receiver, but saw only microwatts on the output. I assume this is due to the large forward voltage required. I've ordered a rectifier IC with low forward voltage: https://www.mouser.com/ProductDetail/750-CDBHD140L-G. Hopefully this will solve the issue.

Any guidance would be very appreciated. Thanks again for your and Ulysses' quick responses!

[OSilv.1]

Hello @JL. Lebon_O​ and @Ulysses HERNIOSUS_O​,

I've replaced two serial and two parallel capacitors on the X-nucleo-nfc06A1 to bring the matching impedance down to 3.5 - 1j ohms. I've set the driver output resistance to the lowest setting. I've placed an antenna (4.6 uH) and rectifier bridge (4X MMBD330T1G onsemi | Discrete Semiconductor Products | DigiKey) at an optimal distance above the reader. This receiver circuit is also resonant at 13.56MHz. This provides 2.4V at the rectifier output, however adding a one-ohm resistor (a place holder for a battery) between rectifier positive and negative drops the rectified voltage to nearly zero.

I've attempted other resistances such as 10, 100, and 1000 ohms, the best power supplied to them has been only 2mW (less than the 10mW I've measured from the V_EH pin of the st25dv using a reader of a higher matching impedance. I feel I am missing something fundamental. Any help would be greatly appreciated.

Thank you,

Oliver

[Marcel Rossignol]

Hello Oliver,

there are several points which you should take into account for NFC WLC:

• WLC needs a reader antenna matching which take already the mutual coupling with the tag antenna into account. This maximize the power transfer.
  • A high power VNA measurement would be needed to determine the mutual coupling impedance between reader and tag antenna (10dBm VNA output power is to less to activated the bridge rectifier on the tag side to "see" the tag load!). This impedance is depending on the antenna distance, the VNA output power, the tag tuning and tag load.
  • High antenna Q-factor is very welcome on the reader side (removing the serial resistors in the matching network)
  • EMC filter cut off- frequency is on the Nucleo board around 11.7MHz. This is to low for WLC! Suggestion: cut off - frequency of around 20.3MHz (110nH/560pF)
  • Take care on the EMC coil choice! Very low DCR (high Q) and high rated current is needed (e.g.: Coilcraft 0805LS-111)
  • Reader and tag antenna resonance frequency should have NOT the same value.
    • Reader: 13.56MHz
    • Tag: 15 - 18MHz (this is also tag load depending)
  • The AAT of the reader could help to find a better matching during power transfer. This is an interesting option if no high power VNA measurement is available and the "trial and error" method is needed :)

I hope this brings you a little bit further.

Best regards,

Marcel

[OSilv.1]

Hi Marcel,

Thank you very much for your insights! This is what I need to progress my design. Unfortunately, I do not have an expensive VNA, just a NanoVNA.

I have one more question:

To do both wireless charging and communications (iso15693) at the same time, the quality factor should not be too high. I have read to keep the Q factor below 35 for this reason. Is 35 reasonable, or could I go higher?

Thank you again!

Best,

Oliver

[Marcel Rossignol]

Hi Oliver,

well, the NanoVNA is better than nothing :)

But you should keep in mind that your impedance measurement do not reflect the true picture!

However it is important to measure the antenna parameter for the matching and the matching impedance with the tag antenna on the reader antenna.

Just a short definition of Q factor: I would separate between reader Q, tag Q and system Q (Q if the tag is close to the reader antenna)

Your statement regarding the reader Q is true for the usual NFC matchings. But for WLC is different, because you always operate reader and tag in very close distance with a relatively high load. This reduces the system Q tremendous (<5!!). Basically the reader Q can be as high as possible in an NFC WLC system, the tag load decreases the system Q in a range where communication is not an issue.

Best regards,

Marcel

[OSilv.1]

Hi Marcel,

Thank you for your timely and informative replies! I am hoping to operate up to 3cm of distance with at least 100mW of power transfer. I also require communications even after the battery (super capacitor) is fully charged. I would assume that the system becomes a usual NFC system at that point (in terms of load and thus lower quality factor of the reader), correct?

On the topic of impedance matching, I am not seeing much change in impedance (measured from the ST25r3916 outputs) unless the tag antenna is brought within 2mm of the reader antenna. I am thinking that this is due to the capacitive effects between the two antennas. For the operating distance I require, there is little to no change in impedance. Does this indicate an issue in the coupling?

Best,

Oliver

[Marcel Rossignol]

Hi Oliver,

thanks for the information regarding your operation distance requirements!

This changes a lot: My assumption was that you want to do WLC in some millimeter distance :)

If I talk about NFC WLC, I talk about operation distances of few millimeters (NFC forum defines two operation volumes. The bigger one is a cylinder with a height of 5mm and a diameter of 10mm).

So 3cm distance with at least 100mW is most likely not reachable without PA. I have to say "most likely", because I have never tested your requirements.

But e.g.: 200mW@4cm distance are for sure not possible without PA.

Basically the coupling should be fine, but at 3cm distance the coupling factor is already very small which leads on one hand to very low mutual coupling and less impedance change and on the other hand to high power transfer losses.

You could increase the coupling factor a bit by increasing the reader antenna. A rule of thumb (for the 3cm operating distance) is that the reader antenna diameter should be ~10mm bigger than the tag antenna diameter. But I would not expect a dramatic increase of the power transfer value.

Sorry for the bad news, but I missed the point with your 3cm operation distance!

Best regards,

Marcel

[OSilv.1]

Hi Marcel,

Thank you for your insight. We can move our distance requirement to 1.5cm if you believe we can achieve at least 100mW at that distance.

If we do need a power amplifier, this could be placed at the outputs of the ST25r3916, correct? Would the RF inputs need larger capacitor voltage dividers to accommodate the larger voltages?

Your help has been very useful. If STMicro offered a class on the subject of antenna design (with and without power amplifiers and other augmentations) I would happily pay to enroll. Could you suggest a reference book I could study from?

Thank you,

Oliver

[Marcel Rossignol]

Hi Oliver,

to give you a reliable answer regarding the 100mW@1.5cm, I have to do some tests in the lab (Next week I´m on vacation. So in two weeks I could have a look on this requirement). But this requirement is more relaxed and could work.

Yes, that´s correct. A Class - E PA could do the job here and yes, the input voltage divider capacitors have to be larger.

Currently we do not offer such a class.

I could recommend the following book: RFID Handbook (Klaus Finkenzeller)

Best regards,

Marcel

[OSilv.1]

Hi Marcel,

That is good to hear. Thank you for offering to do some tests in the lab, I am really looking forward to what you find. I am still in the process of assembling a matching network with the lower inductance (and lower DCR) EMC inductors (https://www.coilcraft.com/en-us/products/rf/ferrite-core-chip-inductors/0603-(1608)/0603af/0603af-111/).

I will look through the handbook you suggested. Thank you for the recommendation.

Best,

Oliver

[Marcel Rossignol]

Hi Oliver,

the testing will be delayed. But I have not forgotten :)

Best regards,

Marcel

[Schuyler]

Hi,@JL. Lebon_O​ 

I noticed that the purpose of EH is to power small devices (diodes, up to small MCUS) that consume a lot of power.Therefore, I would like to ask how to power stm32L476rg without power through EH in X-NUCLEO-nfc07A1, because L4 is a low-power MCU.

Best regards,