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.

TPS23753A: PMP7355 design issue, 20hz saw tooth output on no load

Part Number: TPS23753A
Other Parts Discussed in Thread: PMP7355, , , TPS23753

Hi,

I have implemented a PMP7355 design on a custom board, which connects to a TPS65217C PMIC chip, and the LED connected to the SYS output pin flashes for about 50ms once.

Scoping the output of the 5V PoE shows a saw tooth of 4.7V to 7.5V at 20Hz. This is preventing the PMIC from starting up.

  

Our custom PMP7355 design has the following parts substituted:

1. R18 is 0.22 ohms, so that extra current allowed for 12V AUX input operation, delivering 10W output.

2. T2 is a POE13F-50L, which is different in inductance and turns (still 5V output) from the original POE13P-50L flyback used in PMP7355

3. A current sense resistor (10 mOhms at 1W) inserted between VDD and VDD1 pins for current measurements.

Does anyone have any ideas on how to fix this?

In parallel to this I will obtain a TPS23753AEVM-004 board and swap out the components for 5V operation, and test with a TPS65217C evaluation board.

The requirement of our design is 5V at 10W output, and the PoE transformer needs to be lower than 10mm height to fit in the enclosure, hence the use of POE13F-50L or HA3802-BL from Coilcraft.

Thanks in advance,

Kyran McGlasson
SAGE Automation - Australia

  • Hi Kyran,

    When you see this behavior:
    1. What is occurring at the input? Does this occur with the 12V adapter? With PoE input?
    2. How much load is present on the output?
    3. Additionally, could you look at Vc?

    Thanks,
    Tom Amlee
    PoE PD Apps Engineer
  • Thanks Tom,
    I will check this and get back to you.
    Regards,
    Kyran
  • Another question that I forgot to ask, in PMP7355, D4 is connected to the low side (GND) of the secondary winding, I have seen in the datasheet for TPS23753a that it is connected in series with the output, the high side of secondary winding, why is that? Could we move it to the high side without any issues?
    And C27 connected to the high side of the primary winding to GND, this is missing in the datasheet. What does this do? Can we leave it out?
    Thanks, Kyran.
  • Hi Kyran,

    There will be no difference in functionality if you move D4 to the high side of the secondary, but EMI results may change.

    C27 is an isolation capacitor that should be kept for transient protection. Additionally, it helps with common mode noise.

    Regards,
    Tom Amlee
    PoE PD Apps Engineer
  • Hi Kyran,

    The change from POE13P-50L to POE13F-50L is a significant change in inductance & turns ratio. In addition to the question I posed previously, could you post your schematic?

    Thanks,
    Tom Amlee
    PoE PD Apps Engineer
  • Thats exactly what I thought. We rushed head first into this design, using the "F" version of the flyback transformer as we have height restrictions preventing us using the "P" version, and didnt look into the detail of the inductance and turns ratio.

    The problem I have is that I dont have a working evaluation board to compare the scope measurements with.

    We have now ordered TPS23753AEVM-004, and will change the components required for 5V operation.

    I think we will change our design to match this board as it also has a flat flyback transformer that will fit in our enclosure. It should be arriving sometime today.

    Thanks again Tom. Schematic pdf attached.

    Regards,
    Kyran

    Power_PoE.pdf

  • Hi Kyran,

    I've checked your schematic & find no issues. Keep me in the loop if there are any updates & I can provide assistance.

    Also, do you have responses to the questions I posed, initially?

    Thanks,
    Tom Amlee
    PoE PD Apps Engineer
  • Hi Tom,

    I have validated the changes to TPS23753AEVM-004 for 5V operation and can report that it performs as expected. As mentioned earlier the original design used a POE13F-50LD that has a primary inductance of 40uH. When calculating the minimum inductance required to achieve continuous current conduction mode using slva305c (Designing with the TPS23753 section 2.6.5) it appears the POE13F-50LD does not meet the requirement.

    Switching out the flyback transformer with any of these options: C1174-AL or HA3802-BL (meeting the height requirement); will provide significant inductance (150uH) to maintain continuous current conduction, and prevent the excessive noise that I am experiencing.

    So we are now designing the PoE circuit following TPS23753AEVM-004 circuit, using 0.39 ohms for the Rcs current sense resistor, to achieve 10W when using 12V adaptor power. The user guide states efficiency of 87% at 5V 1.5A.

    Interesting to note that this is similar in some ways to TPS23753AEVM-235, and the efficiency is 92% at 5V 1.5A. I wonder if this could be achieved with an HA3802-BL flyback transformer.

    So for now I consider this ticket resolved, it may be closed.

    Thanks again for your assistance.

    Kind Regards,

    Kyran McGlasson
    SAGE Automation