Wolfspeed Power Applications Forum

Wolfspeed Technical Community:

Technical Questions About the CAB450M12XM3 SPICE Model Equations and Physical Interpretation

always
always Contributor Level 1
edited November 18 in Module Products

Dear Wolfspeed Technical Support Team,

I am a master's student in electrical engineering, currently conducting research on SiC MOSFET behavioral modeling and inverter loss analysis.

I am studying the official SPICE model of the Wolfspeed CAB450M12XM3 power module. I would greatly appreciate your clarification regarding the physical meaning and implementation of several model equations and parameters.

1. Physical meaning of VthFwd(Tj)

In the forward conduction model, the effective gate-source voltage is expressed as:

Vgs_eff = Vgs - Vths/(1+max(Vgs-10,0)^3) + VthFwd(Tj)

What is the physical meaning of VthFwd(Tj)?

2. Temperature lookup table extending to 1975°C

The VthFwd(Tj) lookup table extends to 1975°C, whereas the maximum operating junction temperature specified in the datasheet is 175°C.

Why does the lookup table extend to such a high temperature? Is this intended for numerical convergence, parameter extrapolation, or another modeling purpose?

3. Model validity below 25°C

The VthFwd(Tj) and RTJ(Tj) lookup tables both begin at 25°C, while the datasheet specifies an operating junction temperature range of −40°C to 175°C.

Has the SPICE model been validated for junction temperatures below 25°C? If so, how are the device characteristics represented in this temperature range?

4. Threshold-voltage adjustment expression

The effective gate-source voltage includes the following adjustment term:

Vths/(1+max(Vgs-10,0)^3)

What is the physical or modeling purpose of this expression? Why does the adjustment decrease when Vgs exceeds 10 V, and why is a cubic function used?

5. Reverse channel conduction

The forward channel current model includes the following term:

b(Vgs)*a(Vgs)*min(Vds,0)

This term allows the channel current model to represent conduction when Vds is negative.

Why is reverse channel conduction included in the MOSFET model? What is its physical significance, and how does it interact with the separate third-quadrant conduction model, including the body diode and reverse-recovery behavior?

6. Determination of MOSFET operating regions

The forward channel current is described by an arctangent-based continuous function:

IdFwd = a(Vgs)*atan(b(Vgs)*Vds) + c(Vgs)*Vds

Unlike conventional MOSFET models, this equation does not explicitly contain a threshold voltage or define separate cutoff, linear (ohmic), and saturation regions.

How are the cutoff and saturation regions identified in this model? Is there an explicit criterion or boundary for distinguishing the cutoff, linear, and saturation operating regions, or are these transitions represented implicitly by the fitted functions?

If possible, could you also recommend any technical papers, application notes, or documentation explaining the derivation and physical interpretation of these SPICE model equations?

These questions are important for accurately describing the model principles and applicability in my master's thesis.

Thank you very much for your time and assistance.

Best regards

Tagged:

Comments

  • Forum_Moderator
    Forum_Moderator Wolfspeed Employee - Contributor Level 5

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

  • ChadF
    ChadF Wolfspeed Admin - Contributor Level 2

    Hello always,

    Thank you for such a detailed question, and overall interest in our SPICE Models. These questions go beyond the scope of what Wolfspeed can publicly share about our SPICE models. Please reach out if you have any other questions, and we wish you the best of luck with your master's thesis.

    Best regards,

    Chad F.

  • ChadF
    ChadF Wolfspeed Admin - Contributor Level 2

    Hello always,

    I am going to close this post. If you have any other questions please submit a new post.

    Best wishes,

    Chad F.

This discussion has been closed.