Wolfspeed Power Applications Forum

Wolfspeed Technical Community:

ZVS in DAB

Options
Khaledsailan
Khaledsailan Contributor Level 1
edited November 26 in General

Hi can any one tell me how to achieve ZVS in the DAB from the Hardware side and from the software side. and how to implement it?

Comments

  • Forum_Moderator
    Forum_Moderator Wolfspeed Employee - Contributor Level 5
    Options

    Thank you for your post, it has been approved and we will respond as soon as possible.

  • ChrisN
    ChrisN Wolfspeed Employee - Contributor Level 4
    Options

    Hello Khaledsailan,

    In the simplest control scheme (single phase shift), the diagonal switch pairs in the primary and secondary of a dual active bridge (DAB) operate at a 50% duty cycle. Under the right operating conditions with proper design, when diagonal switch pairs turn off, the current through the leakage inductor starts to flow through the body diode of the complementary switch positions during the converter deadtime. This means that these devices are not blocking voltage when they turn on, thus achieving zero voltage switching (ZVS). The nuance of a DAB is designing the converter to achieve ZVS in the most operating conditions possible. Our partner, Texas Instruments, has a more detailed guide with diagrams found here showing how the circuit achieves ZVS, and Wolfspeed has more information in the whitepaper found here and in the CRD60DD12N-GMB design files.

    For a software implementation, many controllers with a power-electronics focus such as the Texas Instruments C2000 line of controllers include features to introduce a phase shift between one or more of the pulse width modulation (PWM) outputs. The C2000 version of this feature is called EPWM_setPhaseShift. I recommend using these integrated features for the software.

    Thanks,
    Chris N.

  • Khaledsailan
    Khaledsailan Contributor Level 1
    Options

    Thank you chris,

    The problem that we have it is at low load because even with TPS it is not possible to achieve the ZVS, for that either from the Hardware or from the software we have to react. from Hardware snubber can be used or redesign the inductor , and from the software some method like adaptive dead time can be used. The question is which direction is easier Hardware or software? if software is anybody knows how the adaptive dead time can be implemented?

  • ChrisN
    ChrisN Wolfspeed Employee - Contributor Level 4
    Options

    Hello Khaledsailan,

    Typically, designers prefer to make changes in software rather than hardware since the updates can be made rapidly without requiring assembly/fabrication time. Most power electronics focused controllers have an integrated, easy-to-change deadtime module built in. For example, the Texas Instruments C2000 series of controllers includes deadtime control by varying the rising and fall edge delay counts (EPWM_setRisingEdgeDelayCount and EPWM_setFallingEdgeDelayCount). You could update these values to achieve variable dead time.

    I recommend using a software tool such as PLECS or LTspice to simulate your converter before trying hardware or software changes. This will enable you to test whether the proposed changes will fix your problem. Wolfspeed provides models for both of these simulation tools for download here.

    Thanks,

    Chris N.

  • ChrisN
    ChrisN Wolfspeed Employee - Contributor Level 4
    Options

    Hello Khaledsailan,

    I hope that answered your question. I will close this discussion for now but if you have a follow up question, please "Start a New Discussion" and we would be glad to support you further. 

    Thanks,

    Chris N.

This discussion has been closed.