What Is Thermal Power in Electronics? Junction Math and Heatsink Sizing

Thermal power in electronics is the rate at which a component converts electrical energy into waste heat, measured in Watts (W). If a power MOSFET drops 2 volts across its drain-source channel while passing 4 amps, its thermal power dissipation (P_D) is 8 watts. That 8W does not vanish; it must be physically moved from the silicon junction to the surrounding air. If your thermal management path cannot evacuate that 8W fast enough, the junction temperature rises until the silicon fails.

The Thermal Path: Junction-to-Ambient Math

To manage thermal power, you must understand thermal resistance, denoted as R_theta (R_θ) and measured in °C/W. This metric tells you how many degrees Celsius a specific interface will rise for every watt of heat pushed through it. The goal is to keep the junction temperature (T_J) below the manufacturer's absolute maximum rating.

The fundamental equation for thermal equilibrium is:

T_J = T_A + (P_D × R_θJA)

Where T_A is the ambient air temperature, P_D is the dissipated thermal power, and R_θJA is the total thermal resistance from the junction to the ambient air. However, R_θJA is not a single physical barrier; it is a series of resistances, much like resistors in series in an electrical circuit:

If you are relying purely on the PCB copper pour without a discrete heatsink, R_θSA becomes the thermal resistance of the PCB itself. According to Texas Instruments application note SNVA489, a standard 1-square-inch pad of 2 oz copper on an FR4 board yields roughly 40 °C/W to 50 °C/W in still air. This is why high-power surface-mount parts require extensive thermal via arrays to stitch heat to inner and bottom copper layers.

Heatsink Selection: A Real-World Sizing Example

Let us apply this math to a concrete scenario. You are driving a 12V stepper motor using an IRLZ44N logic-level MOSFET in a TO-220 package. Based on your duty cycle and R_DS(on) at your specific gate drive voltage, you have calculated the worst-case thermal power dissipation to be 6 Watts.

The Design Parameters:

• P_D: 6 W
• T_J (Target Max): 125°C
• T_A (Worst-case enclosure ambient): 45°C (electronics inside a sealed enclosure easily run 20°C above room temperature).
• R_θJC: 1.5 °C/W (from the IRLZ44N datasheet).
• R_θCS: 0.8 °C/W (using a Bergquist GAP PAD T-550 thermal pad, avoiding messy thermal grease in a production build).

Step 1: Calculate Required Total Thermal Resistance (R_θJA)
R_θJA = (T_J - T_A) / P_D
R_θJA = (125 - 45) / 6 = 13.33 °C/W

Step 2: Calculate Required Heatsink Resistance (R_θSA)
R_θSA = R_θJA - R_θJC - R_θCS
R_θSA = 13.33 - 1.5 - 0.8 = 11.03 °C/W

You need a heatsink with a thermal resistance of 11.03 °C/W or lower in natural convection. Browsing DigiKey's extruded aluminum heatsinks, the Aavid Thermalloy 593202B00000G is a standard TO-220 clip-on heatsink rated at approximately 10.5 °C/W in still air. It costs roughly $1.85 in low quantities. Because 10.5 is less than 11.03, this part will keep your MOSFET safely under 125°C. If you were to skip the heatsink and rely on the bare TO-220 tab (which has an R_θJA of ~62 °C/W in still air), the junction temperature would hit a catastrophic 417°C, instantly destroying the silicon.

Reading the Derating Curve and Failure Signatures

How hot is too hot for this part? The datasheet's absolute maximum rating for T_J is typically 150°C to 175°C. However, you must consult the Power Derating Curve. This graph plots allowable thermal power dissipation against ambient or case temperature. It starts at 100% power at 25°C and slopes linearly downward, hitting 0W at the maximum junction temperature. If your enclosure ambient hits 80°C, the derating curve might dictate that your MOSFET can only safely dissipate 50% of its nominal wattage. Ignoring this curve is the number one cause of late-stage prototype failures.

When thermal limits are breached, components do not always fail with a dramatic pop. Chronic thermal stress leaves specific forensic signatures:

• Solder Joint Creep: Repeated thermal cycling causes the coefficient of thermal expansion (CTE) mismatch between the silicon die, the leadframe, and the PCB to physically tear the solder joints apart. You will see micro-cracks under a microscope, leading to high-resistance connections.
• Electromigration: At sustained temperatures above 125°C, the high current density causes metal atoms in the silicon's internal interconnects to physically migrate. This eventually creates open circuits or short circuits inside the IC.
• FR4 Delamination: If the PCB itself gets too hot, you risk exceeding the glass transition temperature (Tg) of the fiberglass. Standard FR4 has a Tg of 130°C; high-Tg boards are rated for 170°C. Exceeding this causes the resin to soften, leading to via barrel separation and board warping.
• Thermal Runaway (BJTs): Unlike MOSFETs, which have a positive temperature coefficient that naturally limits current as they heat up, Bipolar Junction Transistors (BJTs) have a negative temperature coefficient. As a BJT heats up, its V_BE drop decreases, causing it to draw more current, which creates more thermal power, leading to an uncontrollable feedback loop until it melts.

Airflow and Enclosure Tweaks: What Actually Buys You Degrees

If your R_θSA calculation demands a heatsink so large it will not fit inside your enclosure, you must alter the environment. What airflow and enclosure changes actually buy you thermal headroom?

1. Forced Convection (The Biggest Win)
Adding a small fan drastically alters the boundary layer of stagnant air clinging to the heatsink fins. A standard 40mm x 10mm Sunon MagLev fan pushing just 11 CFM across a TO-220 heatsink can drop its effective R_θSA by 50% to 70%. The Aavid 593202B00000G mentioned earlier drops from 10.5 °C/W in still air to roughly 4.0 °C/W at 200 LFM (linear feet per minute) of airflow. This allows you to use a physically smaller, cheaper heatsink.

2. The Chimney Effect (Passive Enclosure Design)
If you cannot use a fan, you must optimize natural convection. Heat rises. Place intake vents at the lowest physical point of your enclosure and exhaust vents at the highest point. Ensure the heatsink fins are oriented vertically so the heated air can flow straight up through the channels. Horizontal fin orientation in a sealed box traps the heat and effectively increases R_θSA by up to 30%.

3. PCB Copper Weight and Thermal Vias
For surface-mount parts like the D2PAK or QFN, the PCB is your heatsink. Upgrading from 1 oz copper to 2 oz copper on the outer layers reduces the lateral thermal spreading resistance. Furthermore, placing an array of 0.3mm thermal vias directly under the exposed thermal pad, filled with epoxy and capped (via-in-pad), transfers heat to the bottom layer, effectively doubling your radiating surface area.