Problem with NFC Tag detection using ST25R95

[Mohammed Ali]

We have designed a a custom PCB with a custom antenna using ST25R95 NFC Tag reader, also the antenna tuning has been done. We found an issue during our testing, at our production site the tags are not getting detected but the same PCBs are working fine in our office.

We have observed that the AC mains supply is stable at both locations but the mains frequency isn’t stable at the production site. We are using a standard power adapter with input rating 110-240V 50/60Hz Output 5V 4A. Can the mains frequency variation lead to NFC chip not reading tags or it has to do something with the environment (EMI)? 

Also the DC voltage on the ST_R0 Pin is around 7.8V in both the environments (office as well as production site) indicating that the RF field is on if I'm not mistaken.

[Brian TIDAL]

Hi,

Can you elaborate on " the tags are not getting detected"?

• which technology is being used: NFC-A, NFC-B, NFC-F and/or NFC-V?
• which error code is being received: 0x87 (timeout)?
• Is it a random error (i.e error on some sendReceive commands) or a systematic error (i.e. error on all sendReceive commands)

Can you share a logic analyzer trace (SPI CLK/MOSI/MISO/CS + IRQ_IN + IRQ_OUT) so that I can have a look to the errors?

Also, can you probe the RF field with a scope in both cases (connect the GND of the probe to the tip of the probe and put the probe in the field and you should be able to capture the field)? Can you probe the 5V output voltage from the power adapter?

Can you try to swap the power adapter between you office and your production site?

Can you share the schematics of your custom board so that I can review them?

Rgds

BT

[Mohammed Ali]

Hi Brian,

We are using ISO/IEC14443A. I cannot tell you exactly what is the error code that we are receiving since the debug print is disabled in the firmware.

Also we are using UART interface for communication and the Board is enclosed in the final product hence it is difficult to get the trace of UART.

We tried probing the RF field and were able to pickup the 13.56MHz carrier and at both the sites the we observe no difference. Also dc level at ST_R0 is same that is 7.8V at both the sites. We also probed the 5V output from the adapter and found the voltage is stable and ripple is around 30mVpp.

1. Field pickup at the office site
   
2. Field pickup at the Production site
   

• So the problem is we have 10 units at the Production site with 10 dedicated power adapters and none of them are able to detect any tag (we are using a buzzer indication once a tag is detected).

• When the same set of 10 units with same 10 power adapters are tested at the office site, the units are working seamlessly.

• Also we observed that if we connect the power adapter from UPS line to Raw main power line, the units at the production site starts detecting the tags.

Sharing the schematics for your reference

 

[Brian TIDAL]

Hi,

the UPS is likely creating EMI issues. A second order LC low pass filter may be needed.

Rgds

BT

[Mohammed Ali]

Hi Brian,

Can you please share reference schematics for the same?

Where exactly should we add the 2nd order LPF?

FYI we are into production now and we will not be able to change the design. Can you suggest any other alternatives to minimize this issue?

Can we try by adding ferrite beads/chokes on the power line to minimize this issue?

Attaching a rough block diagram for your reference, Please suggest accordingly

Thanks in advance,

[Ulysses HERNIOSUS · Accepted solution]

Hi,

good old debugging rule from the 9 debugging rules of David J Agans: "Quit thinking and look":

Measure the power supply output close to the NFC reader using oscilloscope on your production plant while it is failing to read NFC tags. See if it is really bad. Then apply ferrite beads, etc. to see the improved supply.

BR, Ulysses