Problem background：
The normal usage environment is for the wristband to swipe the card irregularly. After a period of use, such as 100 units, dozens of units cannot swipe the card, and there is no response when swiping the card. It does not mean that data writing fails, but rather that NFC cannot receive a single message, resulting in complete failure.
After the wristband is placed out of battery, the power on function returns to normal.
Analysis already done：
Analysis already done：
1. Use the t4t format NFC virtual card, attached are the init function and handle function. The handle does not handle any communication or complex algorithms, only setting some variables.
2. The operation of the card reader on NFC cards is:
1. Reset the antenna (equivalent to removing the card once)
2. Card search (equivalent to activating a card and obtaining UID)
3. Send RATS command (equivalent to switching the card to APDU mode)
4. Send multiple APDU instructions to read files:
Select DF file
00 A4 04 00 07 D2 76 00 00 85 01 00
Select MF
00 A4 00 0C 02 E1 04
Write file (write 2 bytes)
00 D6 00 00 02 00 01
3. The conclusion drawn from the testing conducted by the card reader manufacturer is that if the wristband is pulled back in the magnetic field and continues to send instructions, there is a very small probability that it will freeze and fail, with hundreds of repetitions occurring once.
4. Our testing method here is to continuously swipe the card: up to 500+swipes show no response, down to 41 swipes show no response, a total of thousands of swipes, with a median of around 200 swipes showing no response once. When not responding, all messages will not be received.
5. I added a piece of code that executes every 10 minutes
After an exception occurs, the NFC will not trigger any events, including NFC error callbacks, while the chip is still working properly.
Could you please help me check if I know the cause of the problem, or if there are better avoidance and analysis solutions?
nfc_t4t_emulation_stop(); nrf_delay_ms(20); nfc_t4t_emulation_start();
When this code exists, any unresponsiveness can be restored.
But currently, the problem has not been identified.
But currently, the problem has not been identified.
One additional phenomenon: The card reader manufacturer said that when the card reader approaches the NFC card, there will be two exit messages: one is the initial reset antenna, and the other is the final disappearance of the magnetic field. But when I tested it, if my card reader kept sticking to the wristband, the nfc_callback on the 840 side would continuously receive NFC_T4T_EVENT_FIELD_ON，NFC_T4T_EVENT_FIELD_OFF without stopping. I left it for 10 minutes without getting stuck, so I didn't continue studying. I don't know if this phenomenon is helpful for analyzing the problem.
After an exception occurs, the NFC will not trigger any events, including NFC error callbacks, while the chip is still working properly.
Could you please help me check if I know the cause of the problem, or if there are better avoidance and analysis solutions?
