Wolfspeed Power Applications Forum

Wolfspeed Technical Community:

CAB760M12HM3 Simulation of Current Imbalance under Vths Mismatch in Synchronous Buck Topology

Options
JGJH
JGJH Contributor Level 1

Dear Wolfspeed Technical Support,

I am currently simulating the CAB760M12HM3 module in a synchronous buck converter topology, focusing on the current imbalance behavior caused by V<sub>th</sub> mismatch between two parallel high-side switches, using LTspice.

According to Infineon's application note "Paralleling Power MOSFETs in High Current Applications" (e.g., Figure 5), it is clearly observed that the turn-on timing differences caused by V<sub>GS(th)</sub> variation lead to noticeable saturation delays in the drain current of each MOSFET. Based on this, I configured the simulation with the following setup:

  • Two half-bridge modules in parallel, differing only in V<sub>th</sub>
  • The lower-side switch (body diode/FWD) is biased at -4V
  • Both high-side MOSFETs receive the same gate signal simultaneously

However, in the synchronous buck configuration, the drain currents of the parallel devices converge too quickly, and I am unable to observe the saturation delay difference as clearly as shown in Figure 5.

I would like to ask the following:

  1. Are there any testbench examples or parameter configurations available that would help emphasize the current saturation delay between parallel MOSFETs in a synchronous buck converter using the CAB760M12HM3?
  2. When simulating with the same module in a Double Pulse Test (DPT) structure, I am able to observe a much more distinct current saturation delay between devices due to V<sub>th</sub> mismatch.
    Could the lack of visible delay in the synchronous buck converter be due to the influence of the large output inductor in the converter structure?
  3. Would it be possible to obtain a reference simulation example using CAB760M12HM3 that replicates the structure and behavior shown in Figure 5?

Although the application note attributes the effect to parasitic inductances (e.g., 1 nH, 3 nH), I have implemented similar parasitics in my simulation, but no clear delay or imbalance is observed.

Thank you very much for your time and support. Any guidance or reference material you could provide would be greatly appreciated, as it would be a significant help in validating my experiments and understanding the dynamic behavior of the module.

Sincerely,
Junhyeok

Tagged:

Comments

Sign In or Register to comment.