Reader loses access to NT3H2211 in pass-through mode

[leon_wu]

I'm using pass-through mode to transfer file via NFC. It works most of time. But from time to time the RF side can't access the NFC (not even able to read session register). I2C side still works as expected. The value of NC_REG is 0x7d (PTHRU_ON_OFF = 1, FD_OFF=11, FD_ON=11, TRANSFER_DIR=1,NFC_TO_I2C). NS_REG is 0x21 (RF_LOCKED=1, RF_FIELD_PRESENT=1). It can recover after detach and re-attach the reader. Is there anyway to recover it from NFC or I2C side. switching VCC off and on can't fix it.  Please help.

Thanks.

[Kan_Li]

Hello,

Would you please help to check if there is RF reset event during the transfer? Thanks for your patience!

Have a great day,
Kan

-------------------------------------------------------------------------------
Note:
- If this post answers your question, please click the "Mark Correct" button. Thank you!

- We are following threads for 7 weeks after the last post, later replies are ignored
Please open a new thread and refer to the closed one, if you have a related question at a later point in time.
-------------------------------------------------------------------------------

[leon_wu]

Hi Kan,

Thanks for your reply. Can you please let me know how to check the RF reset event? Thanks.

Best Regards,

Leon

[Kan_Li]

Hi Leon,

You may set FD pin for simple field detection and probe this pin during the transfer.

Have a great day,
Kan

-------------------------------------------------------------------------------
Note:
- If this post answers your question, please click the "Mark Correct" button. Thank you!

- We are following threads for 7 weeks after the last post, later replies are ignored
Please open a new thread and refer to the closed one, if you have a related question at a later point in time.
-------------------------------------------------------------------------------

[leon_wu]

Hi Kan,

The FD pin is low with the trigger setup in your reply. I have a debug pc in a lab and sometimes I remote desktop (windows) this debug pc while the tag is attached to a reader. It seems irreverent but I haven't observed this behaviour while I'm directly using the debug PC. Thanks.

Best Regards,

Leon

[Kan_Li]

Hello Leon,

FD pin is open drain type, so you have to add a pull-up on this pin as below:

did you implement like that? Please kindly clarify.

Thanks for your patience!

Have a great day,
Kan

-------------------------------------------------------------------------------
Note:
- If this post answers your question, please click the "Mark Correct" button. Thank you!

- We are following threads for 7 weeks after the last post, later replies are ignored
Please open a new thread and refer to the closed one, if you have a related question at a later point in time.
-------------------------------------------------------------------------------

[leon_wu]

Hi Kan,

Yes, the FD pin is pulled up.

I have another question which is not relevant to the previous one, as following:

For my application I need to exchange messages between PC and my Firmware. Once the Firmware received a message from PC side(FD pin active), it writes a reply message to the SRAM by disabling pass-through mode, changing the direction to I2C_to_RF and then enabling pass-through mode. I noticed that the FD pin stays inactive after I changed the direction,  but still I can write the SRAM from my firmware side (I2C). Is this looks right to you. I was assuming the FD pin should be active in this case. Thanks.

Best Regards,

Leon

[Kan_Li]

Hello Leon,

Referring to the spec, pass-through mode is not disabled during transfer, please refer to the following for details.

and per the data sheet,

Maybe that is why NFC side lost the communication.

You may refer to https://www.nxp.com/docs/en/application-note/AN11579.pdf  for more details.

Hope that helps,

Have a great day,
Kan

-------------------------------------------------------------------------------
Note:
- If this post answers your question, please click the "Mark Correct" button. Thank you!

- We are following threads for 7 weeks after the last post, later replies are ignored
Please open a new thread and refer to the closed one, if you have a related question at a later point in time.
-------------------------------------------------------------------------------

[leon_wu]

Hi Kan,

I understand that the pass-through mode needs to be enabled during the transfer.  The pass-through mode is only disabled when I change the direction of transfer and it is enabled immediately afterwards. The section 4.2 of the application note has the following description:

The I²C side either can poll on the bit SRAM_RF_READY in the registers or listen on the FD pin to get low. Afterwards it can either write new data in the SRAM or switch the data flow direction via turning the pass-through mode off first, switch the direction and then turn it on again.

Best Regards,

Leon

[Kan_Li]

Hi Leon,

Actually you need not switch the direction manually, when the last page in the SRAM was read/written, the NTAG I²C plus arbiter automatically switches the communication side. but did you just update part of the SRAM and then switch the direction? Is that your case? Please kindly clarify.

Thanks for your patience!

Have a great day,
Kan

-------------------------------------------------------------------------------
Note:
- If this post answers your question, please click the "Mark Correct" button. Thank you!

- We are following threads for 7 weeks after the last post, later replies are ignored
Please open a new thread and refer to the closed one, if you have a related question at a later point in time.
-------------------------------------------------------------------------------

[leon_wu]

Hi Kan,

Please check the following case:

RF side writes a message of 64 bytes into SRAM. I2c side detects the change of fd pin and reads the 64 bytes (after the last page is written by  RF side, i2c can read).  Because I2c side read the last page of SRAM, the RF side again has the ownership. If I want to write a reply message from the I2C side, I believe I need to change direction. Thanks

Regards

Leon

[Kan_Li]

Yes, direction should be changed in that case, but you have to make sure RF side knows how to handle this, you know, the NFC reader may have NAKs during the direction switching, and if the reader thinks it is an error and quit the session, the connection will be lost as well.

Have a great day,
Kan

-------------------------------------------------------------------------------
Note:
- If this post answers your question, please click the "Mark Correct" button. Thank you!

- We are following threads for 7 weeks after the last post, later replies are ignored
Please open a new thread and refer to the closed one, if you have a related question at a later point in time.
-------------------------------------------------------------------------------