This article is trying to make sense out of confusing information regarding the behavior of a MOSFET during switching sequences, in numerous technical articles.

We are not attempting to explain the physics behind a MOSFET structure. For those interested to find more about a MOSFET structure, we recommend the SGS-Thomson technical articles mentioned in the references. The purpose of the article is to present a power supply design engineer with facts that will help design a MOSFET driving circuit, calculate the estimated losses for critical events, predict the efficiency of a power supply, estimate the junction temperature for critical components and various stresses, and ultimately, helping make decision to optimize a design.

The MOSFET switching events are analyzed for an inductive load, diode clamping circuit, the only one that applies to a switching power supply. The datasheet information or technical articles regarding resistive loads have little or no relevance to switching a MOSFET in a switch mode power supply. Also the article is considering only 500V/600V MOSFETs, most relevant and for switch mode power supplies.

Capacitors reference designators are the same as in SGS-Thomson articles. Check the references if a more detailed explanation of their significance is necessary.

And now the comments:

• Gate voltage has a sharp rise at t1 due mainly to the MOSFET source inductance.
• Internal MOSFET channel will carry also the Coss discharge current.
• At t3 the drain voltage will reach Vx, usually around 25V for 500V MOSFETS. After this time Vds*Idrain loss should be considered part of the conduction loss.
• Q3, gate charge associated with drain voltage reaching Vx, is much smaller then Q3+Q4, commonly specified in a MOSFET datasheet.

Common errors and misconceptions:

Error: Drain voltage will decreased linearly to zero during t2 - t4 period, when gate voltage is constant (gate plateau voltage).

Reality: Drain voltage will decrease much faster reaching Vx voltage, during t1 - t2 period. Calculating the switching loss associated with this period of time, considering that the drain voltage will decrease linearly for the entire "plateau" period, will results in huge errors.

Error: Drain voltage will reach zero at t2b (the end of the diode reverse recovery period).

Reality: Drain voltage will decrease much faster reaching Vx voltage, during t1 - t2 period. Calculating the switching loss associated with this period of time, considering that the drain voltage will decrease linearly for the entire "plateau" period, will results in huge errors.

Error: MOSFET capacitances cannot be used to determine switching behavior, you need gate charges values.

Reality: MOSFET capacitances, if fully characterized, can fully explain (together with other parameters) the switching behavior, without the need for gate charges.

Error: MOSFET datasheets give enough information to characterize the switching behavior of a MOSFET in your application.

Reality: Most MOSFET datasheets, for reasons not discussed in this article, are not giving enough information to be able determine the switching behavior in a typical application for which the part was intended to be used. Time characteristics (turn-on delay time, rise time, turn-off delay time, fall time) are measured in conditions that bear no resemblance with a typical application of the device, so, the best for a design engineer would be to disregard them altogether. An attempt to use them to "guestimate" the real useful data, it is time consuming and usually not done by engineers.

A notable exception: Motorola datasheets, similar with the one mention in the references for MTW20N50E, are providing the information that really matters in determining the switching behavior. To the best of our knowledge, current at the time this article was last updated, Motorola datasheets were the only ones providing all following critically needed characteristics: Vx voltage, gate charge needed for drain voltage to drop from Vdd to Vx, high voltage capacitance variation in a readable chart, low voltage capacitance variation in a readable chart, Ciss and Crss variation at low voltage, zero Vgs and Vds, zero Vds and Vgs zero to 10V, internal source inductance.

Other Considerations:

• SMPS Power Supplies is using the above described correct theory regarding MOSFET switching to accurately calculate the switching losses in PFC hard switching and soft switching topologies. Combined with our accurate models for diodes (with voltage drop, reverse recovery time and reverse recovery current being functions of operating temperature, forward current, dI/dt), the design spreadsheets (ADH2450Des__.xls, ADH8100Des__.xls) are the most accurate design tools for designing and predicting the performances of a switching power supply.

References:

• Study of a Model for Power MOSFET Gate-Charge, SGS-Thomson Microelectronics, Power MOS Devices Data book, 1st Edition - June 1988.
• Gate Charge Characteristics Lead to Easy Drive Design for Power MOSFET Circuits, by M. Melito and F. Portuese, SGS-Thomson Microelectronics, Designers' Guide to Power Products, Application Manual, 2nd Edition - June 1993.
• MTW20N50E, Designer's Data Sheet, Motorola - REV 4, 1996.

SMPS Intellectual Property

This article contains information for which SMPS Power Supplies and its partners may claim Copyright and/or Trademark rights and may be subject of a Patent application. Also SMPS Power Supplies and its partners may claim the status of "First to be published", relative to ideas published in this article. Any third parties may quote reasonable parts of this article without contacting us, assuming that the source is clearly identified and a link to the full article is included. If you wish to incorporate information from this article within a commercial product, you should contact us for permission.

• First LCD Consulting internal document: 19 Oct 1991
• Major update, SMPS Power Supplies internal document: 10 Feb 1998
• Web first published: 3 Aug 2002
• Last Revision: 3 Sep 2005