• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Subscribers 64 subscribers
  • Views 838 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.

UCC28951: 2KW OBC design - How to fix resonance issue during debugging

Prem Kumar G G
Prodigy 110 points
Part Number: UCC28951
Other Parts Discussed in Thread: UCC21222,

Hello,

The Onboard charger requirement is 200V to 400V input, 48V output - 36A load.

Some basic details about main parts:

Full bridge transformer:

Primary inductance - 1025uH, Leakage inductance - 2.5uH

Full bridge MOSFET: IPW65R190CFDFKSA1, Coss - 86pF

Shim Inductor - 2uH

MOSFET Driver: UCC21222

Operating frequency: 230kHz

While the converter operating at above 300V DC input, at light load (3A) we observed failures in primary MOSFET due to over temperature.

We assume it could be because of the converter is not operating in ZVS mode due to resonance not met.

So now we are operating at 150V DC input and output voltage is set to 24V. We want to ensure ZVS before proceeding with the higher input voltage.

I have attached the test observation for your checking. The waveform includes voltage across primary and primary current.

In order to achieve resonance, we also tested with 330pF external capacitor connected across Drain and Source of Full bridge MOSFETs.

But we have not found better result.

We are continuing to tune the circuit to meet the resonance.

Please check the test observation and give your opinion about the waveform.

Also suggest us how to meet the resonance at light load conditions.

Please let us know if you require more details about the design or any other information.

It will be very helpful for us if you support.

PSFB_150vin_24vout_min_load_testing.xlsx

  • 0
    TI__Expert 7080 points

    Hi Prem,

    An expert will get back to you soon !

  • 0 in reply to Jaden Ning
    TI__Genius 9140 points

    Hi Prem,

    I have below suggestions to you.

    • if frequency upto 230kHz is a must, I would suggest you can use UCC28951 light load burst mode function to decrease the mosfet switching loss.
    • increase slim inductor value to help achieve ZVS at load that is lower than before.
    • check whether dead time is enough to discharge the Coss of mosfet,if not, please increase the deadtime.

  • 0 in reply to David Huang33
    TI__Genius 9140 points

    Hi Prem,

    are you able to evaluate the suggestion I offer? And do you have any progress on this issue?

  • 0 in reply to David Huang33
    Prodigy 110 points

    Hi David,

    Thanks for your suggestion.

    There is lot of improvement in the testing now.

    Below are the major modification we have done.

    1) Adaptive delay implementation - ADEL is tied with CS now.

    2) Dead time resistor changed to 15kohm

    3) Rsum resistor changed to 18kohm - Previously we were using 90kohm which caused some disturbance when increasing the load. Then we arrived at required slope compensation of 300mV/us. But from magnetization part the slope is only 32mV/us. So we reduced Rsum resistor for the additional slope compensation.

    Now we are able to increase the Input voltage to 380V without major heat dissipation is MOSFET and no failure in device.

    I have attached the recent result at 8amps load condition.

    Now we are in resonance tuning stage.

    Now our concern is A and B MOSFETs (same leg) are having more heat dissipation cause above 80°C.

    But C and D MOSFETs are having very less heat dissipation.

    We are using thermal gun to observe the temperature.

    Please check the waveform and tell us what could be the reason for heat dissipation only in A and B MOSFETs.

    PSFB_Load_testing_05_11_2020.xlsx

  • 0 in reply to Prem Kumar G G
    TI__Genius 9140 points

    Hi Prem,

    For PSFB topology, one leg is easy to realize ZVS, the other leg is not so easy to realize ZVS.

    at your test condition, I guess Mosfet A and B still can not realize ZVS. Please double confirm with VGS and VDS of Mosfet A or B to see whether or not it achieve ZVS.

    if this is just the ZVS issue, I suggest increase slim inductor value to help here. You can increase the inductor to lower the output power that can let all mosfet realize ZVS.

  • 0 in reply to David Huang33
    Prodigy 110 points

    Hi David,

    We will check VDS and VGS of MOSFET A and B.

    Kindly check and reply for all below points:

    1)      Connecting External capacitor across Drain and Source to achieve resonance is good or bad – Please give your opinion on connecting external capacitor

    2)      We are not using MOSFET for secondary rectification. Currently we are using only diode for secondary rectification. Will this impact ZVS?

    3)      Please send a sample waveform of VGS and VDS with achieving ZVS – This help us a lot for cross verifying our waveform

    4)      We tried increasing Shim inductor – It reduced the heat dissipation in A and B MOSFET – But the temperature of Shim inductor is going above 90°C – We are using toroidal powder iron core – What could be the reason for more heat dissipation in shim inductor

    5)      When using the Shim inductor value as per the calculation in the datasheet, we are not meeting the ZVS. What could be the reason for this?

  • 0 in reply to Prem Kumar G G
    TI__Genius 9140 points

    1) Adding the capacitor on DS of mosfet will need more inductor and time to discharge the voltage on VDS from high (for example 400V) to low(for example zero), so it is bad for ZVS achieving.

    2) There is no imfluence on ZVS when using Diode.

    3) if we can achieve, VDS will follow to zero before VGS raising.

    4)The increasing loss on Slim inductor help to achieve ZVS of the mosfet.

    5)you need to consider the leakage of transformer as part of the Slim inductor.

  • 0 in reply to David Huang33
    Prodigy 110 points

    Hi David,

    Thanks for the answers.

    We had good improvement after following your suggestion.

    We have also realized ZVS at significant load condition.

    Now the heat dissipation on MOSFET is very minimum while we increasing the load.

    But now our concern is the heat dissipation on shim inductor.

    Shim inductor Part No: 2301-V-RC - 10uH 20A toroidal inductor

    We have reduced number of turns to make it 2.7uH at which we are able to achieve ZVS.

    In this condition the temperature of shim inductor is increasing above 75°C when 10Amps loaded in the output.

    So we are not able to increase the load further because it may cause failure in shim inductor.

    Please guide us how to design or select the shim inductor. Please suggest any suitable part in your view.

    Please send if there is a design document for shim inductor designing.