• State
  • Locked Locked
  • Replies 5 replies
  • Subscribers 38 subscribers
  • Views 397 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.

PMP11302: PMP11302 Re-Design Support

Amal Babu
Prodigy 90 points
Part Number: PMP11302
Other Parts Discussed in Thread: UCC28881

Hi,

I am currently designing a power supply for my own application. My input constraints are bit different from the specifications of the PMP11302 design.

Input spec:

Input : 1200V

Output :12V

Load current:100mA (max).

So as to meet the input spec i did the following changes:

  • Changed the voltage rating of the Capacitors at the input and the value.
  • Changed the MOSFET ratings to a higher one.
  • Changed the Freewheeling diode and the diode across inductor (D10 & D1) to two diodes in series at both places to handle the high voltage.

Following are my concerns regarding the redesign of this EVM based on my new spec:

  1. Do i need to change the value of balance resistor connected across the Capacitor as i changed the value of the capacitor?
  2. Do i need to be worried about the external MOSFET Specifications like Input Capacitance (Ciss) which determines how fast the FET turns OFF and ON? Especially the value of R13?
  3. How do i calculate the exact voltage across the FET when a high voltage of 1200V is applied at the input? I understand that the zener regulates based on the zener current and can vary for different input voltages.
  4. Do i need to change the snubber circuit connected in parallel to the FET upon changing the part?
  5. How much voltage can be expected across the diodes D10 and D1 during the off times of the device? Shall i consider replacing the two diodes in series with a single diode with a higher spec like 1500V with reverse recovery time<75nS? How much would be the max power dissipation or the WCCA Current flowing through these diodes?
  6. Can i expect a stable output from PMP11302 at no load condition? Is it mandatory to have a zener or load across the output of UCC28881 as per the datasheet recommendations to get a stable output?
  7. During the testing of the PMP11302 design , i happen to short the feedback resistor R8 and now i am observing high voltages at the output. I am seeing a no load voltage of around 60V at the output after this. Need your expert advise in identifying where the design could have gone wrong as i shorted the feedback resistor? I can also see the device is no more supporting any loads as per its spec. Please throw some light on this scenario? 

Thanks in advance for the support.

  • 0
    TI__Mastermind 24375 points

    Hi Amal,

    Thank you for reaching out.

    1. The bleeder resistor might not need change. It will depend on if there is any imbalance seen.

    2. External MOSFET needs to hav sufficient charge (gate capacitance) to keep on during ON state of the internal MOSFET. The gate resistor should be small as possible to reduce the turn on and turn off delays. It is better to start with a 10ohm value and reduce appropriately.

    3. I am attaching an app note for you to check this criteria.

    https://www.ti.com/lit/ta/ssztbz6/ssztbz6.pdf?ts=1714551803671&ref_url=https%253A%252F%252Fwww.google.com%252F

    Also you can verify this with the design reports, where the excess voltage above the zener clamp voltage will have to be handled by the external FET

    4. This has to be checked according to waveforms.

    5. 1500V with a recovery time suggested should be safe.

    6. Yes, pre load resistor needed

    7. I am not sure why would you short R8. the FN pin needs 1V to maintain regulation with on off control.

    Regards,

    Harish

  • 0 in reply to Harish Ramakrishnan5
    Prodigy 90 points

    Thankyou for the reply.

    " I am not sure why would you short R8. the FN pin needs 1V to maintain regulation with on off control."

    We were trying to Analyze the worst case scenario by shorting the feedback resistor value to see the effects. Usually before implementing it in any design we do both constructive and destructive analysis to see the performance of the device.

    Since the device operates on high voltage its important to analyze its performance. 

    As described earlier we did short the feedback and observed a peak output voltage of 60V.The output of the SMPS IC was directly connected to the LDO without a Load which resulted in damaging the LDO.(TLV70433). We had even removed the Zener as we had changed the feedback resistor to get a 12V output from PMP11302. So in short the output of SMPS IC was tied directly to the input of LDO.

  • 0 in reply to Amal Babu
    TI__Mastermind 24375 points

    Hi Amal,

    Thank you for clarifying. When FB is always zero, it could be possible that UCC28881 is always pulsing at 62khz causing increase of output voltage. In that case it may necessary to include output OVP to cover the worst case scenario.

    Regards,

    Harish

  • 0 in reply to Harish Ramakrishnan5
    Prodigy 90 points

    Dear harish,

    As described above i changed the input caps from three to 4 10uF caps in series to withstand high input voltage and changed the FET Q1 to IPD90R1K2C3.

    i successfully tested the circuit for a phase to phase voltage of 415VAC.

    When i tried to load the circuit above 415VAC , the output is suddenly drooping down to 9V to 10V from 12.8V.

    Please note it was working fine when i tested above 415VAC phase to phase with the original design with the FET from STM and three caps in series. What could have gone wrong in this?

    Please help with this.

  • 0 in reply to Amal Babu
    TI__Mastermind 24375 points

    Hi Amal,

    You have to find what potentially could cause this issue. One thing I suspect is probably the on time current limit could be hit causing the voltage to drop. Also you might want to check if inductor current runaway protection where Ton < Ton_to could be triggered and voltage could drop.

    Thank you

    Regards,

    Harish