MOSFET Switching Loss Calculation

Hello! In our last lesson, we explored the concept of hard switching loss, establishing that it's caused by the overlap of voltage and current during the switching transitions. We saw that this loss is proportional to the switching frequency, creating a major design conflict.

Today, we're moving from the "what" and "why" to the "how much." Your goal is to learn how to calculate the switching losses that occur during the turn-on and turn-off of a MOSFET. This is a critical skill for evaluating the thermal performance of a power converter and for appreciating the efficiency gains offered by Zero Voltage Switching (ZVS).

We will cover two methods: a simplified triangular approximation and a more precise method using energy values directly from a component's datasheet.

The Triangular Approximation of Switching Loss

As we saw previously, the instantaneous power loss during a switching event peaks when both voltage and current are significant.

MOSFET Switching and Conduction Losses — This graph illustrates the relationship between the drain-source voltage ($v_{DS}$), drain current ($i_{DS}$), and the resulting power loss. The triangular power spikes during the turn-on and turn-off intervals represent the energy lost in each transition.

If we approximate these power spikes as triangles, we can calculate the energy lost in each transition. The energy ( $E$ ) is the area of the triangle, which is $\frac{1}{2} \times base \times height$ .

Height: The peak power, which is approximately the bus voltage ( $V_{b u s}$ ) times the load current ( $I_{l o a d}$ ).
Base: The duration of the transition. These are the rise time ( $t_{r}$ ) and fall time ( $t_{f}$ ) specified in the MOSFET's datasheet.

The energy lost during one full switching cycle (one turn-on and one turn-off) is:
$E_{S W} = E_{o n} + E_{o f f} \approx \frac{1}{2} V_{b u s} I_{l o a d} t_{r} + \frac{1}{2} V_{b u s} I_{l o a d} t_{f}$

To get the average power loss ( $P_{S W}$ ), we multiply this energy per cycle by the number of cycles per second, which is the switching frequency ( $f_{s w}$ ). This gives us the most common formula for estimating hard-switching overlap loss.

Overlap Switching Loss (P_sw)

The following resource, 'MOSFET Power Loss Calculator' from DMC, clearly defines this formula and its components.

Please read the short section titled 'Overlap Switching Loss (P_sw)'. Focus on the formula provided and the explanation of how performance is affected. This will solidify the relationship between voltage, current, frequency, and transition times.

The formula you just saw is:
$P_{S W} = \frac{1}{2} V_{b u s} I_{l o a d} (t_{r} + t_{f}) f_{s w}$

This equation is your primary tool for a first-order estimation of switching losses. It clearly shows that loss increases linearly with switching frequency, bus voltage, and load current.

A Worked Example

Now, let's watch a video that walks through a concrete numerical example. The video calculates the energy loss per transition first and then multiplies by the frequency, which is an equivalent way of applying the same principle.

Power Electronics - MOSFET Power Losses

The video 'Power Electronics - MOSFET Power Losses' provides an excellent, step-by-step calculation of the different losses in a MOSFET.

Please watch from 2:43 to 8:31. The video first explains the calculation for turn-on (2:43 - 3:59) and turn-off (3:59 - 5:09) losses separately. Then, it presents a full numerical example (5:09 - 8:31). Pay close attention to how the values for voltage, current, time, and frequency are plugged into the formulas to find the power loss in Watts.

As you saw in the video, the total power loss is the sum of conduction loss and switching loss. The switching loss itself was broken down:

Turn-on Energy ( $E_{o n}$ ): $\frac{1}{2} V_{D D} I_{D, o n} t_{o n} = \frac{1}{2} (200 V) (10 A) (18 n s) = 18 μ J$
Turn-on Power ( $P_{o n}$ ): $E_{o n} \times f_{s w} = 18 μ J \times 100 k H z = 1.8 W$
Turn-off Power ( $P_{o f f}$ ): Calculated similarly to be $1.8 W$
Total Switching Power ( $P_{S W}$ ): $P_{o n} + P_{o f f} = 1.8 W + 1.8 W = 3.6 W$

Note that the video's definition of transition time is slightly more detailed, combining current rise time and voltage fall time. For datasheet calculations, using the overall rise ( $t_{r}$ ) and fall ( $t_{f}$ ) times is a common and effective simplification.

Test your understanding!

You are reviewing a buck converter design. The MOSFET datasheet lists the following typical values:

Rise Time ( $t_{r}$ ): 30 ns
Fall Time ( $t_{f}$ ): 20 ns

The converter operates under these conditions:

Bus Voltage ( $V_{b u s}$ ): 48 V
Load Current ( $I_{l o a d}$ ): 10 A
Switching Frequency ( $f_{s w}$ ): 500 kHz

Using the triangular approximation formula, calculate the total switching loss ( $P_{S W}$ ).

Show answer

The total switching loss is 6.0 W.

Here is the calculation:

Sum the transition times: $t_{r} + t_{f} = 30 ns + 20 ns = 50 ns$
Apply the formula:
$P_{S W} = \frac{1}{2} V_{b u s} I_{l o a d} (t_{r} + t_{f}) f_{s w}$
$P_{S W} = 0.5 \times 48 V \times 10 A \times (50 \times 1 0^{- 9} s) \times (500 \times 1 0^{3} Hz)$
$P_{S W} = 0.5 \times 48 \times 10 \times 50 \times 1 0^{- 9} \times 500 \times 1 0^{3}$
$P_{S W} = 240 \times (25000 \times 1 0^{- 6}) = 240 \times 0.025 = 6.0 W$

Beyond Approximation: Using Datasheet Switching Energy

The triangular approximation is a great starting point. However, it simplifies a complex process. In real-world inductive circuits, effects like the reverse recovery of the freewheeling diode add to the turn-on losses, making the power spike non-triangular.

For this reason, many MOSFET datasheets provide characterized switching energy values, typically listed as $E_{o n}$ and $E_{o f f}$ , measured in Joules (often microjoules, $μ J$ ). These values are measured in a standardized test circuit that mimics a typical converter, so they provide a more realistic basis for loss estimation.

When you have $E_{o n}$ and $E_{o f f}$ from a datasheet, the calculation for total switching power becomes very direct:
$P_{S W} = (E_{o n} + E_{o f f}) \times f_{s w}$

The key challenge is that these datasheet values are only valid for the specific test conditions listed (e.g., a certain voltage, current, and gate resistance). If your application's conditions differ, you must scale these values. This is a very practical skill for design review.

The following resource provides a detailed example of this exact process.

Power MOSFET Tutorial - Datasheet Extrapolation Example

This application note, 'Power MOSFET Tutorial' from Microchip, explains how to perform a realistic loss calculation. The example shows how to take switching energy values from datasheet graphs and scale them to a specific application's operating voltage and gate drive resistance.

Please read the section titled 'Datasheet Extrapolation Example' (page 11). Don't worry about the initial conduction loss part. Focus on the steps starting from 'For turn-on switching loss...'. Observe how the author: Finds initial E_on and E_off values from a graph (Figure 18). Scales these energies by the ratio of the application voltage to the test voltage. Applies a second scaling factor based on the application's gate resistance (Figure 19). Finally, calculates the total switching power using the scaled energy values.

This method is more involved, but it yields a much more accurate estimate of the real-world switching loss. For your goal of reviewing PCBs and component choices, understanding how to interpret and scale these datasheet graphs is invaluable. It allows you to assess if a chosen MOSFET is truly a good fit for the converter's operating conditions.

Conclusion

You now have two methods to quantify the power dissipated during hard switching.

Key Takeaways:

Triangular Approximation: A quick and useful method for estimating switching loss using the formula $P_{S W} = \frac{1}{2} V_{b u s} I_{l o a d} (t_{r} + t_{f}) f_{s w}$ . The required rise ( $t_{r}$ ) and fall ( $t_{f}$ ) times are found in the MOSFET datasheet.
Datasheet Energy Method: A more accurate method that uses the manufacturer-characterized switching energies ( $E_{o n}$ and $E_{o f f}$ ). The calculation is $P_{S W} = (E_{o n} + E_{o f f}) \times f_{s w}$ . This method often requires scaling the datasheet values to your specific operating conditions.
Regardless of the method, switching loss is a major contributor to total power loss, especially at high frequencies and high voltages.

We have now established what hard switching loss is and how to calculate it. In the next lesson, we will examine the final piece of the "hard switching problem": the generation of electromagnetic interference (EMI). Understanding these three aspects—V-I overlap, calculable power loss, and EMI—will give you a complete picture of why ZVS is so crucial for modern, high-performance power converters.

Recommended resources

The Triangular Approximation of Switching Loss

A Worked Example

Beyond Approximation: Using Datasheet Switching Energy

Conclusion