• State TI Thinks Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 2 answers
  • Subscribers 63 subscribers
  • Views 110 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.

ZVS not happening!

Sivadharshan Kamaraj
Prodigy 60 points
Part Number: LM5034

Tool/software:

Dear team,

                 I have used LM5034 in my latest design for +28V to +50V conversion with a power of 250W.

The attachment contains the waveform of the VDS of Primary MOSFET and VGS of the same.  The clamping capacitor value has been chosen based on the following formula.

DMIN - 0.279

Lmag - 41.4 uH

F - 500 kHz

Based on the data above, the calculated clamp capacitor value is 12.7 nF.  The actual value used in 20 nF.

Also, the programmed dead time is 50ns.

What can I do, to achieve zero voltage switching in active clamp MOSFET and primary MOSFET?

CH1: OUT1 (Gate pulse of primary MOSFET), CH2: VDS of primary MOSFET.  CH3: Input current.

Kindly support in this regard.

  • 0
    TI__Guru 74115 points

    Hi,

    The ZVS needs the help to get energy from the transformer primary side leakage inductance along with proper dead time. In your input voltage range, ZVS is agreed not to be sought from past two decades practice of active clamp forward since the leakage energy demanded is so large to achieve ZVS that efficiency benefit is not obtained from resulted conduction losses increase and duty cycle reduction.

    Practically, you can play with the dead time to get so called Vds valley switching instead of Vds Zero Voltage Switching.

  • 0 in reply to Hong Huang
    Prodigy 60 points

    Hi,

    The reference for this design has been taken from the attachment (slua303.pdf).  They have discussed about ZVS there. 

    If  ZVS will not happen, what is the required leakage energy and how to calculate the same to achieve ZVS?

    How much efficiency can be improved after achieving ZVS?

    The observed efficiency is 83%.

    Also, throw some light on Vds valley switching.

  • 0 in reply to Sivadharshan Kamaraj
    TI__Guru 74115 points

    Hi,

    In applications of 48V input or nearby and frequency like below 600kHz switching losses are not meaningful with much later investigation than the application note you cited.

    Pursuing ZVS makes efficiency lower. So only need to pay attention of Lm and a leakage inductance 1 to 2% is ok. You can make estimation how much power losses from ZVS saved. Based on we know very insignificant in the applications like yours.

    It looks the dead time is too small then causing VI crossing power losses. You can increase the dead time to see if any help. Increase dead time you can see Vds is at a value between Vin and zero when MOSFET turns on which is called valley switching.

    You need to measure and find where the power losses are and how much then deal with each. 

    - dead time to cause MOSFET losses

    - transformer core and winding

    - SR MOSFET shoot through

    - MOSFET Rdson

    - MOSFET gate charge

    others

    ZVS can only have effect together with the dead time on the first. All others on the list not related to ZVS but what we know they are usually the reasons to cause efficiency low.