Enter values and press Calculate.
Heat flows from the silicon junction → device case → heat sink → ambient air. Each step has a thermal resistance in °C/W, and they add up in series like electrical resistors. The temperature rise above ambient equals the power dissipated times the total thermal resistance.
| Symbol | Meaning |
|---|---|
| θJC | Junction-to-case (from the device datasheet) |
| θCS | Case-to-sink (thermal pad / paste, ~0.2–1 °C/W) |
| θSA | Sink-to-ambient (the heat sink's rating) |
| TJ | TA + P × (θJC+θCS+θSA) |
A lower θSA means a bigger/better heat sink. If the required θSA comes out negative, no heat sink can keep the part cool — you must reduce power, lower ambient, or pick a device with lower θJC.
Yes for anything above a watt or two. Paste (or a pad) fills microscopic air gaps between the case and sink, dropping θCS from ~1 °C/W dry to ~0.2–0.5 °C/W.
Still-air heat sinks are quiet and reliable. Adding a fan (forced convection) can cut the effective θSA by 2–4×, letting a smaller sink handle the same power — useful when space is tight.
More surface area and fins mean a lower θSA (better cooling). Datasheets rate sinks in °C/W — pick one at or below the value this calculator gives.
Yes, but their heat adds up. Sum the power dissipations for the θSA calculation, and remember each device still has its own θJC and pad in series with the shared sink.
Compute the junction temp with no sink (θSA very large). If TJ stays below the maximum with margin, you don't. Otherwise size one with the first tab.
Dry metal-to-metal ≈ 0.5 °C/W; with thermal paste ≈ 0.2–0.4 °C/W; with an insulating mica/silicone pad ≈ 0.5–1.2 °C/W.
Keep TJ at least 20–30 °C below the datasheet maximum for reliability and to allow for hot days and aging.
Rearrange Tj = Ta + P(θJC+θCS+θSA). The heatsink must provide θSA ≤ (Tj(max) − Ta)/P − θJC − θCS, and designing to about 80% of Tj(max) leaves sensible margin.
The case-to-sink thermal resistance, set by the mounting interface. A dry joint can add several °C/W; thermal compound or a gap pad typically brings it to a few tenths. On low-power parts it is negligible, but on a device dissipating tens of watts it can dominate.
Only for a rough estimate. θJA assumes a specific test board in still air and rarely matches a real installation. For heatsink design, use θJC and add the interface and heatsink resistances yourself.
A great deal — natural convection heatsink ratings often improve by a factor of two to four with modest airflow. The figure quoted on a heatsink datasheet is usually for natural convection unless airflow is stated.
The air immediately around the heatsink, not room temperature. Inside a sealed enclosure that is commonly 15–30 °C above the room, and using room temperature instead is a frequent cause of thermal failure.
Yes, for natural convection. Fins must run vertically so warm air can rise between them. A heatsink mounted with horizontal fins can lose a substantial fraction of its rated performance.