• State
  • Locked Locked
  • Replies 5 replies
  • Subscribers 33 subscribers
  • Views 174 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

DP83TC817S-Q1: static Loopback Test for MII mode

Jan Balzer
Prodigy 90 points
Part Number: DP83TC817S-Q1
Other Parts Discussed in Thread: DP83TC811

Tool/software:

Dear Suppor,

We are currently trying to analyze a problem with the DP83TC817-Q1. We are developing boundary scan tests.

A loopback has probably already been implemented here.

We are currently trying a static loopback test via the MII loopback function of this module.

We set this loopback mode via the MDIO interface register 0x00 to the value 0x6100.

Then we try to drive static levels at TX0 and TX1 and then try to measure these levels back at RX0 and RX1.
The TX_EN signal is driven high during this process.

But unfortunately we cannot measure any change in the RX signals. We can measure and describe the register values. That all seems to work.

Now the question is whether this can work statically and what other conditions may have to be met to realize such a static loopback test. (Without using functional protocols)

It would be nice if you could support us here.

Thanks Jan Balzer

  • 0
    TI__Genius 9209 points

    Hi Jan,

    Would be happy to help.

    Can you tell us a tittle more about the problem or testing you are doing? The DP83TC811 supports JTAG testing, but the DP83TC817 does not have specific pins for it. Is the boundary scan you are doing related to JTAG testing? 

    Does a ping go through the PHY?

    Regards,

    Nick

  • 0 in reply to Nick Peabody
    Prodigy 90 points

    Hello Nick,

    we have another device with boundary scan functionality. So we can with the other device drive and measure static level on the nets to the PHY.

    So we can control with the other boundary scan device the MDIO interface. Now we want to activate the MII loopback mode via the MDIO interface of the PHY. So we think it should be possible. We can read and write the rigister and so we can check it.

    So we set this loopback mode via the MDIO interface register 0x00 to the value 0x6100.

    The next idea was to drive with the other boundary scan device and high and low level to the TX0 and TX1 pins of the PHY.

    Now we think that we can expected these high and low level on the RX0 and RX1 pins. So we could measure these driven levels on the TX pins on th RX pins with the other boundary scan device on this board.

    But unfortunately we can not measure the static dirvel levels on the TX pins on the RX pins. 

    No is the question: Are other conditions necessary for this test strategy?(set other registers) Is this even possible?

    So we could check the connection between the boundary scan device and the ETH PHY.

    Thanks

    Jan

  • 0 in reply to Jan Balzer
    TI__Genius 9209 points

    Hi Jan,

    In loopback modes, we commonly supply a packet generator in order for packets to go through the loopback. We do not mention using static voltage as a test.

    Jan Balzer said:
    So we could check the connection between the boundary scan device and the ETH PHY.

    Can the boundary scan device be used to generate ethernet packets for this test?

    Best regards,

    Nick

  • 0 in reply to Nick Peabody
    Prodigy 90 points

    Hello,

    no the boundary scan device can not send ethernet packets.

    We thought of a simple connection test that we could realize with a loopback test.

    We would only check the connection between the boundary scan device and the ETH PHY on the digital interface. 

    See you here a possibility to realize this somehow with the possibilities of the ETH PHY? To drive simple static levels and measure them back again?

    We see different loopback modes? 

    But if I understand correctly the PHY answers only in the RX pins if the PHY resceive a complete ETH packet? right?

  • 0 in reply to Jan Balzer
    TI__Genius 9209 points

    Hi Jan,

    Yes, that is the case.

    The loopback mode you are doing is the shallowest through the PHY. It's typically not done with static voltage levels.

    Best regards,

    Nick