• State Resolved
  • Locked Locked
  • Replies 11 replies
  • Subscribers 64 subscribers
  • Views 2847 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.

TI chip tps23756 POE issue

Ronak Shah
Prodigy 225 points
Other Parts Discussed in Thread: TPS23756, TPS23756EVM

Hi,

We have successfully made IEEE 802.3at enabled POE device as per TI tps23756 as described in below PDF.
Issue We are facing is:
As IEEE 802.3at enabled PSE device needs proper start-up state mechanism as per IEEE standard. I think we are not able to go into the classification stage of the initialization.
But when we connect oscilloscope's probe(earthed with ground) to VDD, device goes into classification and normal operation. And PD successfully get 54 voltage from PSE.

Any body as face same issue with this device can guide me for board bring-up!!! And help will be appreciable!!!

Thanks,
RNK.

slvu329a.pdf
  • 0
    TI__Mastermind 42795 points

    Please diagram the POE system including power supplies and grounds. This issue appears to be system ground related.

    Please also provide oscilloscope waveforms of the power interface between PSE and PD so it can be established where/when the handshake failure occurs.

  • 0 in reply to Eric Wright
    Prodigy 225 points

    Hi Eric,

       I have attached POE block diagram and waveform observed. Internal circuit is designed like TPS23756EVM. With Passive POE, PD is working fine.

    As shown in block diagram when I connect ground of DSO, the PD starts up.

    Attached wave (observed on PD RJ45) repeats in each 2.78 seconds. And it's Peak is 7.8V.

    Waveform scales- X-axis:2.00V/Div and Y-axis: 100mS/Div.

    I am also unable to understand waveform. Pl. explain me a wave.

    Thanks,

    Rnk.

  • 0 in reply to Ronak Shah
    Prodigy 225 points

     

    pl. find Waveform.

    Thanks,

    RNK.

  • 0 in reply to Ronak Shah
    TI__Mastermind 42795 points

    From the waveform, it appears that  the PD detection phase failed (not classification). The voltage levels are below 10V (detection range). Refer to figure 23 in the TPS23756 datasheet (lower waveform which is 10V/div) for the phase voltages during PD detection, classification, and ramp up.

    The best way to view the power interface (ethernet cable wire pairs) voltages is with a differential probing method. Use a common ground (chassis or earth) for two scope probes and look ad VDD and VSS with two probes. Then use the oscilloscope to view the difference (A-B) waveform.

    Also, when you connect the scope ground to VDD are you sure your scope is not  just completing the 54V circuit directily between the PSE and PD (IE, are you sure that the proper phase sequence occurred; detection, classification, ramp up vs. just direct application of 54V to the PD)?

    Please also disclose the PSE model # and manufacturer.

  • 0 in reply to Eric Wright
    Prodigy 225 points

     

    Hi Eric,

              Yes, I am sure that DSO probe is not completing 54V circuit. I am using MIT-09-56D PSE. I have captured waveform,when the proper phase sequence occurred.(on connecting probe.)

    Pl. observe the wave.

    X-axis: 20V/Div,    Y-axis: 100ms/Div

    Peak = 59V

    Regards,

    RNK.

  • 0 in reply to Ronak Shah
    Prodigy 225 points

    Hi Eric,

                I have also observed 25 ripples (range: 10.8V to 20.0V) before PD gets 59V. Are those ripples class and mark? And before those 25 ripples, there is some signal ranging from 7.60V to 10.00V (which is attached in next post).

    Pl. observe the other wave.

    X-axis: 20V/Div,    Y-axis: 20ms/Div

    Thanks,

    RNK.

  • 0 in reply to Ronak Shah
    Prodigy 225 points

    Hi Eric,

               pl. find Waveform.

    X-axis: 5V/Div,    Y-axis: 40ms/Div

    Thanks,

    RNK.

  • 0 in reply to Ronak Shah
    TI__Mastermind 42795 points
    The fuzzy and strange waveforms are an artifact of this PSE. Can you try another vendor PSE? Waveform general state voltages are shown in figure 23 of the TI datasheet.
  • 0 in reply to Eric Wright
    Prodigy 225 points

     

    Hi Eric,

                I have changed PSE, start using IEEE 802.3af compliant ( MIT-07D). I got some strange waveforms diff. than IEEE 802.3at compliant PSE (MIT-09-56D).

    Pl. find attached file for result observed.

    X-axis: 20V/div  Y-axis: 40mS/div.

     

    Thanks,

    RNK.

  • 0
    Prodigy 15 points

    Hello Ronak,

    please kindly get it back to Alan.chen@shinecitech.com once you receive this.

    we're looking for a chance to support PoE Transformer to you!

    Best Regards

    Alan

  • 0 in reply to alan chen74832
    Prodigy 225 points

    Hi Eric,

               As I have done some bring-up with my POE device. I found that "It doesn't have earthing." So we I touch earth's track or connect digital GND of system to DSO, It gets detection and starts. And by changing Rden resistor to 25.9K Ω, It gets detection and starts (for both 802.3at & 802.3af PSE).

              So Eric, Pl. suggest me the solution accordingly. And why do I need to change Rden?

    Thanks,

    RNK.