• State Resolved
  • Locked Locked
  • Replies 17 replies
  • Answers 1 answer
  • Subscribers 33 subscribers
  • Views 1855 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.

DP83867E: Gigabit Ethernet Compliance failing on Peak Output Voltage and Transmitted Distortion Tests

jmbats
Prodigy 70 points
Part Number: DP83867E

Hi,

We are using multiple DP83867E PHY on our board. I am currently doing Gigabit Ethernet Compliance Testing on each PHY of the board. I have done the tests on 4 PHYs so far. I just observed that the PHYs are mostly failing on Voltage Point A, and Voltage Point B for Peak Output Voltage (Peak Differential Output Voltage and Level Accuracy Test) for some differential pairs and some would fail on either Voltage difference between Point D and Point A and B and Voltage difference between Point C and Point A and B on top of the other measurements. Also, it is failing on Percentage of eye opening within limit for Transmitter Distortion for some differential pairs. All other tests for gigabit ethernet compliance are passing.

I just want to understand and pinpoint what will be the issue here on the board and how we can improve it or make it pass in the next revision as we plan to redesign the board. Is this a board issue or something on the ethernet compliance test setup that I have to carefully check and perform? I am using R&S test suite for this setup.

BTW. I have performed the gigabit ethernet compliance test as well on the DP83867ERGZ-R-EVM evaluation board with the same setup and those tests are passing except for the pair B transmitted distortion test with and without disturbing signal.

Your input and information would be highly appreciated. Thank you.

  • 0
    TI__Mastermind 26752 points

    Hi,

    Please send me the compliance report as well as your board schematic.

    Thanks,

    David

  • 0 in reply to David Creger
    Prodigy 70 points

    I am afraid that I cannot give you the board schematic as it is a confidential one. I can ask permission about it. 

    For the compliance report, can I directly send it to you? Please give me your email or anything that I can send you privately. Thank you.

  • 0 in reply to jmbats
    TI__Mastermind 26752 points

    Hi,

    Please send me a private message here on E2E.

    Thanks,

    David

  • 0 in reply to David Creger
    Prodigy 70 points

    kindly check your inbox. thanks.

  • 0 in reply to jmbats
    TI__Mastermind 26752 points

    Received, thanks. Let me go through the report and get back to you.

    Thanks,

    David

  • 0 in reply to David Creger
    TI__Mastermind 26752 points

    Hi,

    Failing these compliance tests generally points to a schematic or layout issue. I will not be able to offer much support if you cannot share these documents. However, I can point you to several resources that will be useful. Please have a look at the following:

    How to Configure DP838xx for Ethernet Compliance Testing

    Ethernet PHY PCB Design Layout Checklist

    Take a special look at section 9.2.1.1 and section 11 of the datasheet

    Looking at your report, it seems the failures are marginal and may not impact throughput. Getting a pass will likely require a component/layout/software iteration.

    Thanks,

    David

  • 0 in reply to David Creger
    Prodigy 70 points

    Hi,

    Thanks for the reply.

    I know that some are marginal for the output peak voltage. However, for the transmitter distortion test, it is not marginal and it is failing on percentage of eye opening within limit that bad. Can you check on this and how this failing measurement can affect the performance of board?

    Thanks.

  • 0 in reply to jmbats
    TI__Mastermind 26752 points

    Hi,

    Failing these compliance tests may affect the data transmission and reception, but I cannot say by how much. Unfortunately I cannot be of much assistance without the schematic and layout files, but I recommend you to go through the resources I have listed above.

    Thanks,

    David

  • 0 in reply to David Creger
    Prodigy 70 points

    Thanks. 

    But probably for now we can refer to the DP83867ERGZ-R-EVM evaluation board as it failed on pair B transmitted distortion test with and without disturbing signal. Is this a normal behaviour of the evaluation board?

  • 0 in reply to David Creger
    Prodigy 70 points

    I retested again the evaluation board DP83867ERGZ-R-EVM to check the repeatability of the tests (run 5 times each test on each pair) and here is the result:

    Peak Output Voltage Without Disturber
    PAIR A - Run 3 Marginal Fail on the Voltage difference between Point D and Point A and B
    PAIR B - PASSED
    PAIR C - PASSED
    PAIR D - PASSED

    Transmitter Distortion Without Disturber
    PAIR A - PASSED
    PAIR B - FAILED (all runs)  on Percentage of eye opening within limit
    PAIR C - Marginal Pass on Percentage of eye opening within limit
    PAIR D - PASSED

    Peak Output Voltage With Disturber
    PAIR A - (Marginal) fail on Voltage difference between Point D and Point A and B on all runs
    PAIR B - Run 3 (Marginal) fail on Voltage difference between Point D and Point A and B
    PAIR C - Run 1-4 (Marginal) fail on Voltage difference between Point D and Point A and B
    PAIR D - PASSED

    Transmitter Distortion With Disturber
    PAIR A - Marginally PASSED on Percentage of eye opening within limit
    PAIR B - FAILED (all runs) on Percentage of eye opening within limit
    PAIR C - Marginally PASSED on Percentage of eye opening within limit
    PAIR D - Marginally PASSED on Percentage of eye opening within limit

    Please check your message for the test report. Thanks. ScopeSuite software version 3.6.0 installed on RTO 1044 Rohde & Schwarz oscilloscope and firmware version 3.70.1.0 (oscilloscope and diff probe have been calibrated)

    May I ask as well the results you have for this evaluation board? I am willing to go thru the process or sign the NDA if required. Thanks.

     

  • 0 in reply to jmbats
    TI__Mastermind 26752 points

    We can share the compliance results of the EVM under NDA, yes. I will have someone contact you to start this process.

    Earlier you shared a report of our EVM passing every test except for TX distortion on Pair B. Has something changed since then?

    Our compliance report was done using a Tektronix setup. In the past we have seen varying results between oscilloscope vendors. Do you have access to a Tektronix setup you can try this on?

    Thanks,

    David

  • 0 in reply to David Creger
    Prodigy 70 points

    The previous one and the current one have most likely the same result but I have run the current one 5 times producing inconsistency in some tests (passed status is not repeatable). The consistent part is the failing test on Pair B transmitter distortion.

    Can you give me the tektronix model you are using? Or probably you can include it later together with the result so we will try to replicate your setup here. Thanks.

  • 0 in reply to jmbats
    Prodigy 70 points

    Also, can I request to perhaps run the failing tests 5 times as well for the EVM (for Peak Output Voltage and transmitter distortion tests only both with and without disturber for all pairs). I just want to know the results.

    Also, is there a way to at least boost a little bit the output voltage like probably writing to some register? Thank you so much

  • 0 in reply to jmbats
    TI__Mastermind 26752 points

    Hi,

    The compliance report we are working to share will include the oscilloscope model. Someone should be reaching out to you shortly. 

    Thanks,

    David

  • 0 in reply to David Creger
    Prodigy 70 points

    Thank you so much. I will be waiting.

  • 0 in reply to jmbats
    TI__Mastermind 26752 points

    Hi,

    Please send me a private message with your email address.

    Thanks,

    David

  • +1 in reply to David Creger
    Prodigy 70 points

    Hello,

    For the sake of the future readers of this thread, the problem we encountered was due to the fact that we were using an old software and oscilloscope from R&S. We currently have in possession the new software and oscilloscope and so far we are getting a good measurement. We can already close this thread but I will still be waiting for the test report from TI. thanks.