• State TI Thinks Resolved
  • Locked Locked
  • Replies 22 replies
  • Answers 3 answers
  • Subscribers 18 subscribers
  • Views 2797 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

How many tags can TRF7960A read?

user4616934
Intellectual 675 points
Other Parts Discussed in Thread: TRF7960A, TRF7970A, TRF7960

TRF7960A works on 5V full power output and connected external PA(4W) and external big antenna, RFID software is based on new firmware, How many tags can TRF7960A read?

Thanks.

  • 0
    TI__Guru*** 128125 points
    Hi Brian,

    It's not easy to quantify that without testing ourselves, I don't think we have any of the equipment needed to replicate your situation though. I can ask Eddie is he has any experience with something relatable to your application.

    Sorry for the lack of replies on the other thread, but today is a US work holiday so we are all out of office today as well :)
  • 0 in reply to Ralph Jacobi
    Intellectual 675 points
    Hi Ralph,
    Ok, thank you for your attention to my question of multiple tags reading.
    For new code porting new your new firmware, today I found that sometimes it stopped to read tag when I placed some tags on atenna plane. I do not know if there is any problem on porting your new firmware or other issues.
    Sorry for disturbing your holiday.
    Have a nice holiday.
    Thanks.
    Best Regards
    Brian
  • 0 in reply to user4616934
    TI__Guru*** 128125 points

    Hi Brian,

    No worries. I was provided a couple slides which may help you understand the range limitations and the impact that multiple tags can have in the RF field: HF Tag Read Ranges.pdf

    Regarding the issue you found, I don't think I quite understand the orientation you described. Can you send a picture of the tag location and antenna location so we can try and figure out if this is an RF issue or firmware issue?

  • 0 in reply to Ralph Jacobi
    Intellectual 675 points

    Hi Ralph,

    Thanks for sharing these relation curve figures of read range with me, it is very helpful to analyze my application.

    I flatted these tags at random on the antenna plane like the following picture, two kinds of problems can occur:

    1. Reader stopped to detect tags; (can not see SPI/IRQ signals output on scope)

    2. Reader is detecting tags, but can not get any tags UID or data. (can see SPI/IRQ signals output on scope)

    Thanks.

    Best Regards,

    Brian

  • 0 in reply to user4616934
    TI__Guru* 90906 points

    better check your antenna tuning and Q first  - you should have no trouble reading those (if they are full size cards) how big is the antenna? what is the tuning and Q of it? 

  • 0 in reply to Josh Wyatt
    TI__Guru*** 128125 points
    Hi Brian,

    In addition to Josh's suggestion of checking your antenna tuning values which is important for reliable tag detection - especially with so many cards - couple suggestions related to firmware as well.

    When you say you can't see any SPI communication, do you mean from the TRF7970A replying to SPI commands being issued, or none at all? If none at all, that means your host processor locked up and that behavior needs to be debugged to see what in the firmware caused the issue and then handle that case properly.

    When you do get SPI and IRQ communication, what IRQ results do you get? Can you check which functions you are in and/or which commands you have sent out which result in those errors?

    Another thing I wonder about, have you checked how big your stack is and if there is any chance for it to overflow? The Anticollision routine is recursive and will suck up some stack each time it is called, so you need to have enough stack space to support as many iterations as you may see in your application of the function call. Stack overflow definitely could cause a lot of problems, so I would definitely investigate your stack usage. If you are using CCS, I can provide some tips on how to determine how much stack you are using right now.
  • 0 in reply to Ralph Jacobi
    Intellectual 675 points
    Hi Josh,
    There are only these parameters(work frequency:13.56MHz/Impedance: 50ohm/SWR:<or =1.2/) in the document that antenna vendors provided, I have asked them for Q and tuning value and have not gotten their reply. It's 25cm*25cm copper tubing antenna.
    Thanks
    Best Regards
    Brian
  • 0 in reply to user4616934
    Intellectual 675 points
    Hi Ralph,
    I do not have network analyzer to measure Q value and tuning value, hope to get these parameter values from the vendor.
    For the first problem, I will check it next Monday;
    For the second problem, TRF7960X went into NO_RESPONSE_RECEIVED status, it did not detect any tag of these tags on antenna like no tag present.
    I need some time to understand how to use stack in anticollision routine, how much stack space do these iterations need? could you explain more about this? I am using ARM development system, not using CCS.
    Thanks.
    Best Regards
    Brian
  • 0 in reply to user4616934
    Intellectual 675 points
    I am using two kinds of antenna (PCB shown in above figure and copper tubing). The two problems above also can be reproduced on copper tubing antenna as well.
  • 0 in reply to user4616934
    TI__Guru*** 128125 points
    Hi Brian,

    I am not sure if there are any differences with IAR vs CCS on this - I don't think there are - so I would expect each call to use 16 bytes of stack space.

    Are you using the No Response Interrupts for ISO15693? Or just your own timeout? How long is the timeout if your own?

    How much time is there between when you turn the RF field on for the TRF7960A and when you send out the anticollision command?
  • 0 in reply to Ralph Jacobi
    Intellectual 675 points

    Hi Ralph,

    The antenna's Q value is about 50, sorry, they did not reply me it's work frequency deviation. 

    For stack, I did not see software crash on uVison, it seems that it can not be stack overflow issue, it is pending on SPI writing because of SPI controller has been busy all the time.

    Both (NO_RESPONSE_RECEIVED_15693 and NO_RESPONSE_RECEIVED) are used, it can went into the two status of TRF796X when running separately. there is 6ms time between them. 

    Thanks.

    Best Regards

    Brian

  • 0 in reply to user4616934
    TI__Guru*** 128125 points
    Hi Brian,

    That Q factor doesn't make any sense. How did you measure or determine it?

    I don't entirely follow what you mean by it went into the two status with 6ms time between them... if you get the NO_RESPONSE_RECEIVED_15693, then it should send out a slot marker and try for the next slot, so if it is not doing that then there is an error in firmware porting.
  • 0 in reply to Ralph Jacobi
    Intellectual 675 points

    Hi Ralph,

    I have not considered Q value and tuning of antenna yet, and I have not instrument to measure them now, Could you tell me how can I determine Q and tuning? then I can customize right antennas from the provider.

    If bad Q and tuning is used, can firmware detect this case to handle it?

    For what you asked, there is 6ms time between turning the RF field on for the TRF7960A and sending out the anticollision command?"

    NO_RESPONSE_RECEIVED_15693 interrupt is used in anticollision handling in code. what is the difference between NO_RESPONSE_RECEIVED and NO_RESPONSE_RECEIVED_15693?

    Thanks

    Best Regards

    Brian

  • 0 in reply to user4616934
    TI__Guru*** 128125 points
    Hi Brian,

    The firmware will not be able to detect or handle hardware issues. If the data received by the transceiver is garbage due to bad RF, then nothing can be done in the firmware to recover that data. The best you can do is reset the part and try again, which won't result in any changes until the hardware side is resolved. Hence why we have our concerns with that.

    The NO_RESPONSE_RECEIVED_15693 is specifically for the ISO15693 anticollision routine to represent the No Response Interrupt that is received when using ISO15693 Slot Markers as per Register 0x0D settings at the timing defined by Register 0x07. This is used to optimize the ISO15693 anticollision procedure by using IRQ interrupts instead of CPU timers. This is done because when going through the 16 slots there are timeouts that do not necessarily mean no tag is present, but just that the tag doesn't respond to that slot.

    The NO_RESPONSE_RECEIVED is the generic case where the MCU determined Timeout occurs and it is treated as the tag not replying. This should prompt the function to return a fail (if there is a return statement) and kick out to re-start detection of tags.
  • 0 in reply to Ralph Jacobi
    Intellectual 675 points

    Hi Ralph,

    For ISO15693，BW is 846kHz when one subcarrier is used and Q is about 16, Is it right if I select this Q value?

    Thank you for your clear description for two TRF796X status.

    But sometimes there is some problem on SPI writing, it seems that SPI work is not stable, CMU SPI are connected to two devices(TRF7960 and EEPROM), Can it be SPI hardware design issue?

    Thanks

    Best Regards,

    Brian

        

  • 0 in reply to user4616934
    TI__Guru*** 128125 points
    Hi Brian,

    One thing you could be running into is that the TRF79xxA SPI lines are not tri-stated even when SS is used and therefore when using multiple devices on the SPI lines with the TRF79xxA devices, it is recommended to drive EN low to turn the chip off which will allow the SPI lines to act as high impedance then.

    If you need to use the EEPROM at the same time as the TRF79xxA, then I recommend placing tri-state buffers on the TRF79xxA SPI lines to avoid any issues with the EEPROM communications.

    You can find some details about this on the document here on page 5: www.ti.com/.../sloa205.pdf
  • 0 in reply to Ralph Jacobi
    Intellectual 675 points
    Hi Ralph,
    Thanks so much for letting me know this SPI issue when two devices work at the same time, but now I only debug TRF7960A and do not make EEPROM to work by setting /CS signal to high level, it seems that instable SPI writing is not caused by this SPI issue you talked. Maybe too many HW modification on current board can cause this problem, now I only can wait to look if there is this same problem on new board.
    Thanks
    Best Regards
    Brian
  • 0 in reply to user4616934
    Intellectual 675 points

    Hi Ralph,Who can I ask some problem about TI HF power amplifier reference board? 

    Thanks.

    Best Regards,

    Brian

  • 0 in reply to user4616934
    TI__Guru*** 128125 points
    Hi Brian,

    I think Eddie might know a bit about it? If you want to post your questions about that here, I can ask him to check the thread?
  • 0 in reply to Ralph Jacobi
    Intellectual 675 points

    Hi Ralph,

    Ok, can I use your design totally and only remove auto bias? 

    Thanks.

    Best Regards

    Brian

  • 0 in reply to user4616934
    TI__Mastermind 38590 points

    Hi Brian,

    I recommend using the updated amplifier design attached. TRF7970 Exernal Power Amplifier for general applications_V53(1).pdf 

  • 0 in reply to Eddie LaCost
    Intellectual 675 points

    Hi Eddie,

    Thank you so much for sharing this with me, it is very helpful for continuing my design. I will take some time to understand this and come back to you.

    Thanks

    Best Regards,

    Brian