Post Go back to editing

Link Down Issue for EVAL-ADIN1100 using 1-pair cable

Category: Hardware
Product Number: ADIN1100

Hi All,

My name is SK Lee. I'm in charge of DFAE in WTK. WTK is one of distributor in South Korea.

 

I have a question for 10BASE-T1L PHY Chip. (Part # : ADIN1100)

 

(Background)

  1. As we know, ADI have some experience for communication distance depending on the cable spec.
  • Cable Spec : BELDEN / 74040NH / 1-Pair / Solid Core / 18AWG / Impedance 100ohm / 1,500m
  • Cable Spec : BELDEN / 3076F / 1-Pair / Stranded / 18AWG 7/26 / Impedance 100ohm / 535m
  1. We are using 2 EA of the EVAL-ADIN1100 boards as below and have established a 600 meters using ethernet cable using CAT.5 AWG 24 reliable cable.

      

  • We are using the GUI tool supported in ADI website. (Ethernet PHY Evaluation Version : 1.8.100.5320)
  • Test Result : Link Up and Packet Test OK
  • So far, OK

 

  1. Now We are trying to establish a link using traditional 1-Pair Cable of customer, But we couldn’t make a link establishment. The cable spec we are using as below picture.

                 

     

  • Style : UL 2464
  • AWG Size : 18AWG (34/33AWG) Stranded
  • 1-Pair (without twisted)
  • Isolation resistance : 15Mohm
  • Impedance : unknown
  • Test Result : Link Down (Link Failed)
  • EVAL-ADIN1100 : 2.4V peak to peak mode

  

(Question)

  1. How can we make a link establishment using above 1-pair cable ?
  2. Are there any modification parameter in GUI tool as below display ?
  3. Are there any cable specification we should check ? Impedance ?

              

 

Parents
  • Hi SK Lee,

    1. Considering you have the EVAL-ADIN1100EBZ boards powered up and the cable connected correctly between P101 - 10Base-T1L connectors, you should be able to bring up a link. Unless, the Insertion loss and/or the Return loss of the cable are considerably off from the 10BASE-T1L standard specifications. 

    2. No modification of any parameters is needed in the GUI. 

    3. Yes, insertion loss and return loss are specified in the 10BASE-T1L standard. Our 10BASE-T1L devices have a built in headroom to accommodate cables that are compliant and non-compliant with the IEEE standard which may vary from cable to cable. As per the IEEE802.3cg 10BASE-T1L specifications;

    Insertion loss for the 2.4Vpp transmission level:

    Return loss for any transmission level:

    What I am suspecting, is that either the insertion loss of the cable is too high across the frequency range of interest, or the impedance of the cable is not only deviated from the 100 ohm nominal impedance, but also varying across the construction of the cable. An non-ideal impedance is okay but if this varies along the cable too much, it may be harder for the transceiver to work with it.

    If you measure the insertion loss and return loss of the cable, we can figure out which is the dominating failing factor.

    Kind Regards,

    Shazia

  • Hi, SMohammadi,

    Thank you for your answer.

    I will check the impedance and insertion loss of customer's cable.

  • Hi SK Lee, 

    Happy to assist! If there are no further issues, could you please close the thread? Please feel free to reopen or start a new thread anytime if you have any additional questions or need further assistance.

    Regards,

    Shazia 

Reply Children
No Data