MOSFET Power Dissipation Calculator
Calculate MOSFET conduction loss, switching loss, total power dissipation, junction temperature, and efficiency for power electronics design
I²·R_DS(on) conduction · linear-ramp switching loss · θJA modelAPI slug mosfet-power-dissipation
Operating point
Loss modelα 0 %/°CQg 20 nCVg 10 VCoss 0 pF
0 to 2 %/°C
0 nC to 10 µC
1 to 20 V
0 pF to 100 nF
Parts and stress
| Part | Value | Must handle |
|---|---|---|
| Q1Switch | 10mΩ | VDS ≥ 20 VID ≥ 5 ATj ≥ 39.8 °C at TA 25 °CRθJA ≤ 40 °C/W |
Where the 390 mW goes
Biggest levers: lower-R_DS(on) FETs and faster switching edges.
Waveforms at the nominal operating point
Ideal switching, 2 periods- Conduction LossPcond
- 250mW
- Switching LossPsw
- 120mW
- MOSFET Efficiency (1 − P_loss/P_in)η
- 99.4%PASS
- Junction Temperature, Coolest CornerTj,min
- 39.8°C
Formula
International Rectifier AN-1070 (J. Cerezo), P_CONDUCTION, P_SWITCHING and P_GATE; Infineon, D. Graovac, M. Pürschel and A. Kiep, MOSFET Power Losses Calculation Using the Data-Sheet Parameters (2006); M. Keagy, Calculate Dissipation for MOSFETs in High-Power Supplies (Electronic Design, 2002)
| Symbol | Quantity | Your value |
|---|---|---|
| ID | Drain current (A) | 5 A |
| RDS(on) | On-resistance (Ω) | 10 mΩ |
| VDS | Drain-source voltage (V) | 12 V |
| fsw | Switching frequency (Hz) | 100 kHz |
| tr | Rise time (s) | 20 ns |
| tf | Fall time (s) | 20 ns |
Assumptions
6 items
- 01Conduction I_D²·R_DS(on) and switching ½·V_DS·I_D·(t_r + t_f)·f_sw, with linear edges and no ringing (IR AN-1070; Infineon, Graovac et al. 2006), and the junction at T_A + R_θJA·P through one fixed thermal resistance, without the transient impedance a pulsed load sees.
- 02One drain current serves both losses, the RMS current for conduction and the switched current for the edges, which holds only for a flat current at a duty cycle near 100 %; the efficiency is taken against V_DS·I_D, and as the calculator takes no duty cycle its waveforms are drawn at 50 % duty.
- 03With a temperature coefficient α, R_DS(on) is the 25 °C value and rises as R_DS(on)·(1 + α·(T_j − 25 °C)) (M. Keagy, 2002); the junction temperature is that loop’s steady state, solved in closed form, and where α·R_θJA·R_DS(on)·I_D² ≥ 1 there is none (thermal runaway) and those results are absent.
- 04The output capacitance’s ½·C_oss·V_DS²·f_sw is lost in the channel at each hard turn-on and heats the junction; the gate drive Q_g·V_g·f_sw is dissipated in the driver, so it is in the loss model’s total and efficiency but not in the junction temperature; reverse recovery and body-diode conduction are not modelled.
- 05The drain current and the ambient are ranges: the junction temperature, the loss and the efficiency are evaluated at every pair of their corners and each worst case is reported with its corner; a minimum ambient may be below 0 °C, and a bound of exactly 0 reads as the nominal.
- 06The drain-source rating is the voltage sold at or above 1.5 times V_DS, and the current rating is the highest drain current.
Worked example
48 V motor bridge FET, 10–20 A
A FET in a 48 V motor bridge carries 15 A (10 to 20 A) and switches at 20 kHz with 50 ns edges; it has 4 mΩ at 25 °C, rising 0.5 %/°C, 1.2 nF of output capacitance and 90 nC of gate charge at 12 V, on a heatsink giving 15 °C/W, in −40 to 70 °C ambient (40 °C typical). The motor runs near full duty, so one current serves both losses. How hot does the junction get, and at which corner?
| Step | Expression | Result |
|---|---|---|
| Switching loss at 15 A | ½ × 48 V × 15 A × (50 ns + 50 ns) × 20 kHz | 720 mW |
| Output-capacitance loss | ½ × 1.2 nF × (48 V)² × 20 kHz | 27.65 mW |
| Conduction loss at 25 °C | (15 A)² × 4 mΩ | 900 mW |
| Self-heating feedback | 0.5 %/°C × 15 °C/W × 900 mW | 0.0675 |
| Junction temperature at 15 A and 40 °C | 40 °C + 15 °C/W × (720 mW + 27.65 mW + 900 mW × (1 + 0.5 %/°C × (40 °C − 25 °C))) / (1 − 0.0675) | 67.59 °C |
| Conduction loss at that junction | (15 A)² × 4 mΩ × (1 + 0.5 %/°C × (67.59 °C − 25 °C)) | 1 092 mW |
| Junction temperature, worst at 20 A and 70 °C | 70 °C + 15 °C/W × 3.35 W | 120.2 °C |
| Voltage the rating must cover, with margin | 1.5 × 48 V | 72 V |
| Drain-source rating, the next one sold | ≥ 72 V | 80 V |
| Gate-drive loss | 90 nC × 12 V × 20 kHz | 21.6 mW |
| Efficiency with gate drive, at 15 A | 1 − (1.839 W + 21.6 mW) / (48 V × 15 A) | 99.7 % |
How it works
The MOSFET power dissipation calculator determines conduction losses, switching losses, and thermal requirements for power transistor applications — essential for motor drives, DC-DC converters, and high-current switching circuits. Power electronics engineers, inverter designers, and thermal management specialists use this tool to prevent device failure and optimize efficiency. According to Erickson & Maksimovic's 'Fundamentals of Power Electronics', total MOSFET power loss comprises conduction loss Pcond = Irms² × Rds(on) and switching loss Psw = ½ × Vin × Iout × (tr + tf) × fsw. For silicon MOSFETs, Rds(on) increases 40-100% from 25°C to 125°C junction temperature per Infineon application note AN-2014-02 — always use hot Rds(on) for thermal calculations. Gate charge Qg determines driver power and switching speed: Pgate = Qg × Vgs × fsw dissipates in the driver circuit. Modern GaN FETs achieve 50% lower switching losses than silicon at 500 kHz due to 5× faster switching (10 ns vs 50 ns), enabling >99% efficiency in server power supplies per Efficient Power Conversion (EPC) design guides.
Practical tips
- Per Texas Instruments' 'GaN FET Design Guide', replace silicon MOSFETs with GaN at fsw > 200 kHz — GaN's 10× lower Qg and zero Qrr reduce total losses by 40-60%, enabling 1 MHz+ operation without heatsinks
- Use thermal interface material (TIM) with θTIM < 0.5°C/W for surface-mount packages — Bergquist Gap Pad 5000S35 achieves 0.3°C/W, reducing Tj by 15-20°C compared to bare PCB mounting
- Implement adaptive dead-time control to minimize body diode conduction — TI UCC21520 isolated driver adjusts dead-time from 20-100 ns based on load current, reducing dead-time losses by 30%
Common mistakes
- Using 25°C Rds(on) for thermal calculations — silicon MOSFET Rds(on) increases 1.5-2× at operating temperature; a 10 mΩ device at 25°C may exhibit 20 mΩ at 150°C, doubling conduction losses
- Neglecting switching losses at high frequency — at 500 kHz, switching losses often exceed conduction losses; a 10 A/48 V MOSFET with 30 ns total switching time dissipates 3.6 W in switching alone
- Ignoring reverse recovery losses in body diode — synchronous buck dead-time causes body diode conduction; silicon diode Qrr = 100-500 nC causes additional 0.5-2 W loss at 200 kHz
Frequently asked questions
5 questions
How do I calculate MOSFET power dissipation?
Per Infineon application note AN-2014-02: Ptotal = Pcond + Psw + Pgate. Pcond = Irms² × Rds(on)_hot. Psw = ½ × Vds × Id × (tr + tf) × fsw. Pgate = Qg × Vgs × fsw (dissipated in driver). For synchronous rectifiers, add body diode losses: Pdiode = Vf × Id × tdead × fsw × 2. Total accuracy typically ±15-20% due to switching waveform non-ideality.
What factors influence MOSFET power dissipation?
Primary factors: (1) Load current (Pcond ∝ I²), (2) Switching frequency (Psw ∝ fsw), (3) Operating voltage (Psw ∝ Vds), (4) Temperature (Rds(on) ∝ T^1.5 for silicon), (5) Gate drive voltage (lower Vgs increases Rds(on)). Secondary factors: gate resistance, Miller plateau charge, reverse recovery charge. GaN and SiC devices exhibit lower temperature coefficient (1.2-1.4× from 25°C to 125°C vs 1.6-2× for silicon).
Why is power dissipation important?
Excessive dissipation causes thermal runaway and device failure. Per MIL-HDBK-217F, MOSFET failure rate doubles for every 10-12°C increase above 100°C junction temperature. At Tj = 175°C (typical silicon maximum), failure rate is 16× higher than at 125°C. Thermal cycling (on/off) causes additional mechanical stress — solder joint fatigue limits automotive MOSFETs to 10,000-100,000 thermal cycles per Infineon reliability data.
How do I select a MOSFET for minimum power loss?
Per TI power design guide: (1) Low-frequency applications (<100 kHz): minimize Rds(on), ignore Qg (choose large die for low conduction loss), (2) High-frequency applications (>500 kHz): optimize Rds(on) × Qg product (figure of merit), (3) Calculate optimal die size where Pcond = Psw (balanced losses minimize total). GaN achieves 10× better Rds(on) × Qg FOM than silicon, dominating high-frequency designs.
What is the relationship between Rds(on) and temperature?
Silicon MOSFETs exhibit positive temperature coefficient: Rds(on)(T) = Rds(on)(25°C) × (T/298)^α, where α = 1.5-2.5 depending on voltage rating. Per Infineon datasheets: 40 V devices α ≈ 1.5, 100 V devices α ≈ 2.0, 600 V devices α ≈ 2.3. This means Rds(on) at 125°C is 1.5-2.0× higher than 25°C value. SiC MOSFETs have lower coefficient (α ≈ 1.0), maintaining lower losses at high temperature. This calculator takes the linear form Rds(on)(25°C) × (1 + α × (Tj − 25°C)) with α in %/°C, typically 0.35-0.5 %/°C (M. Keagy): at 0.4 %/°C, Rds(on) at 125°C is 1.4× its 25°C value.
Shop Components
As an Amazon Associate we earn from qualifying purchases.
- DC-DC Buck Converter ModulesAdjustable step-down converter modules for bench and prototype useAmazon →
- LDO Voltage Regulator KitAssorted low-dropout linear regulators for prototypingAmazon →
- Electrolytic Capacitor KitAluminum electrolytic capacitor kit for power supply filteringAmazon →
- Power Inductor KitAssorted shielded power inductors for switching supply designsAmazon →