• State Resolved
  • Locked Locked
  • Replies 3 replies
  • Subscribers 63 subscribers
  • Views 262 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.

TPS2372: Tstart Fault or Class/Detect Error

Jake Green
Prodigy 10 points
Part Number: TPS2372
Other Parts Discussed in Thread: BOOST-PSEMTHR8-097, TPS2373-4EVM-758, , TPS23881B, TPS23881, BOOST-PSEMTHR-007

Tool/software:

Hi I am a long time user of TI products and first time posting a question to the forum.

Let me list out my setup:
I have BOOST-PSEMTHR8-097 w/ TPS23881B1EVM-24, the jumpers on the board are setup for 2P-30W and I have the TPS2388x GUI running and I have setup the ports as auto mode and have applied 56V to the EVM. I also have a TPS2373-4EVM-758 as a "known good" board that I am using to confirm that the evaluation board is working correctly. I also have my own custom PCB that I am trying to test and get working with POE.

1. I have fully tested that the onboard non-isolated DCDC is working perfectly, I cut the trace after the TPS2372 to confirm that the DCDC is fully working and the problem is not within the power stage of the circuits that I have designed.

2. The board seems to be fully getting through the detection, class, and mark POE detection, but for some reason the chip fails to "connect" the VSS to the Return and connect the path to my DCDC converter.

Troubleshooting steps:

  • Removed all ancillary components in the POE path so that it is basically just the full bridge rectifiers and the components required for the TPS2372 + a 0.1uF 100V cap
    • Components removed are a shunt resistor - replaced with a 0 ohm jumper (10mohm), ferrite beads - replaced with a jumper on VDD/VSS and several other filter capacitors. The placement of those components don't seem to cause any problems.
  • Verified DEN = 24.9k, CLSA = 63.4ohm, CLSB = 90.4ohm, Autoclass connected to VSS and Open (makes no difference, but I did confirm that the TPS23881B does "see" that it is requesting autoclass), I am leaving IRSHDL_EN, MPS_DUTY, AMPS_CTL all open and not connected to anything
  • Verified that REF is 49.9k and I have hooked scope up to pin and it does generate 1.5V and does turn on
  • the TPS23881 is detecting that the channel is connected, R is valid, it sees that the board is requesting class 8, but reports the following error via the GUI:
  • This seems very confusing to me as it seems to be fully detecting the POE, but failing to actually do the last step and apply the load to the port.
  • Here is my scope image:

CH2 = voltage 10v/div, CH4 = current using 25Mhz AC/DC probe in 50ma/div

To my untrained eye this looks good or like the first part of the communication is working perfectly, but for some reason the voltage does not go up to the ~50V instead it goes up to like 33V. Which my understanding is basically telling the POE PD to disconnect rather than connect.

Any help is extremely appreciated!!!

- Thanks Jake

  • 0
    TI__Mastermind 37101 points

    Hi Jake,

    Thanks for reaching out!

    The reply may be delayed due to the limited resource and Lunar New Year. We will get back to you in two weeks. Thanks for your patience! 

    Best regards,

    Diang

  • 0 in reply to Diang Xing
    TI__Mastermind 37101 points

    Hi Jake, 

    1. BOOST-PSEMTHR8-097 has 8 x 2-pairs port. This can only go up to Class-4, not Class-8. BOOST-PSEMTHR8-097 at Class-8 may run into power demotion mode (request Class-8, assign Class-4)

    Class-8 needs a 4-pairs port, and you may need a BOOST-PSEMTHR-007 with 4-paris port.

    2. If you are running open load test, make sure you have a 1.5kOhm, 1W rated resistor between AMPS_CTL to VSS for maintaining power signature.

    3. The VDD-VSS voltage is strange to me. It should be closed to the PSE board's input voltage. Can you measure both VDD-VSS and RTN-VSS at the same screen?

    Best regards,

    Diang

  • 0 in reply to Diang Xing
    Prodigy 10 points

    Hi Diang,

    So your comment got me sparked on a whole bunch of diagnostics. The comment about the voltage looking strange was what I locked on to. I also hooked up the scope to the VDD-VSS of the TPS2373-4EVM-758 and totally agreed with your comment about it looking strange... well the problem was not with any of the TI parts :). It was my "optimization" of the layout of the OnSemi FDMQ8205A full bridges that I was using. They are VERY serious about the gate pins being routed in the strange way where the INPUT 2 = G1 & G2 and the INPUT 1 = G3 & G4. I had flipped them so that I did not need to use vias on my PCB. Which could have explained the weird voltages. I cut up a few traces and installed a few bodge wires and it works now. Thanks so much for the comment about the voltages and that was the key that unstuck by brain!

    Issue solved! Posting this for others to learn from my mistakes!

    Thanks,

    Jake