How Heat Sinks Are Tested: Thermocouples, Wind Tunnels and Spec Sheets
Heat sink performance is measured, not assumed: a known power is fed into a heated block or real device on the sink, temperatures are read with thermocouples at defined points, and thermal resistance is the temperature difference divided by the power. The number is only as honest as the test — thermocouple attachment, airflow measurement, and ambient control decide whether the spec sheet number survives contact with your product.
Every heat sink datasheet carries a thermal resistance curve, and every one of those curves came out of a test someone ran. The gap between that test and your application is where most thermal surprises are born: the vendor tested with a copper heat source bolted to a perfect flat base in unobstructed air, and your product has a small die, a stamped clip, and a crowded enclosure. This guide explains how the tests actually run so you can read a spec sheet — or run a verification yourself — without being fooled.
The Basic Measurement: Power In, Temperature Out
Thermal resistance from case to ambient is defined as (Tcase − Tambient) / P. The test therefore needs three controlled things: a known dissipated power, a correct temperature reading at the case, and a stable, known ambient. Power usually comes from a heater element — a resistor, a thermal test die, or the real component driven at a set current. The sink is mounted exactly as in production, the whole assembly is allowed to reach equilibrium (often 30–60 minutes), and the temperatures are logged.
| Test element | What it does | Practical notes |
|---|---|---|
| Heated resistor / test die | Provides known power | Calibrate the resistance-versus-temperature curve |
| Real component | Tests the actual interface | Power varies with temperature — measure it live |
| Power supply + meter | Sets and records power | Record P at equilibrium, not at switch-on |
| Thermocouple logger | Records temperatures | Log until drift < ~0.1–0.2 K over several minutes |
| Ambient control | Fixes the test condition | Still-air tests need a draft-free box or chamber |
Takeaway: equilibrium is the word most lab tests cheat. A sink pulled from a 10-minute run reports a lower resistance than the same sink at 60 minutes because the aluminum is still warming and the delta-T looks flattering. Log the trend and wait for flatness.
Thermocouple Placement: Where the Error Comes From
A thermocouple measures the temperature of its own junction, and that junction is easily fooled. Taped to the surface of a heat sink base, a thermocouple reads the tape and the air around it as much as the metal beneath. The accepted hierarchy: drilled-and-staked or soldered junctions read true; epoxied junctions read close; taped junctions read low by several degrees; and a junction left loose under a device reads whatever it wants.
| Attachment method | Typical error | Use |
|---|---|---|
| Soldered or brazed into a drilled hole | ±0.5 K class | Reference-grade case temperature |
| Staked into a drilled hole with epoxy | ±1–2 K | Production verification |
| Epoxied flat to the surface | ±2–5 K | Quick checks only |
| Kapton or aluminum tape | ±3–8 K (reads low) | Never for pass/fail decisions |
| Loose under the device | Unpredictable | Never |
Takeaway: on a verification test the difference between soldered and taped thermocouples can exceed the entire margin you designed for. Drill a 1–1.5 mm hole into the base or the device mounting tab, stake the junction in it, and read the number that is actually true. The heat sink mounting methods guide shows where those measurement points sit on real assemblies.
Wind Tunnels and Airflow: The Forced-Air Test
Forced-air ratings need controlled airflow, and "controlled" is doing real work in that sentence. A proper setup is an open-loop or closed-loop wind tunnel with a flow straightener so the air reaching the sink is uniform, plus a way to measure velocity — an anemometer probe in the duct or a calibrated pressure drop across a nozzle. The result is a curve of thermal resistance against face velocity or volumetric flow, usually reported in LFM (linear feet per minute) or m³/h.
| Airflow term | Typical value | Why it matters |
|---|---|---|
| Face velocity | 1–5 m/s (200–1000 LFM) | What the datasheet curve is plotted against |
| Volumetric flow | Fan rating in m³/h or CFM | Not the same as velocity at the fins |
| Pressure drop | Pa across the sink | Weak fans stall on dense fin arrays |
| Free-air vs ducted | Vendor's choice | Unshrouded tests flatter the sink |
Takeaway: match the test condition to your product. If the sink sits behind a fan blowing across it in open space, the free-air curve applies; if it is inside a ducted channel, the vendor's pressure-drop data matters more than the velocity curve. A fan's free-air flow rating is meaningless once the sink resists it — the operating point is where the fan curve and the sink pressure curve cross.
What Spec Sheet Numbers Actually Promise
A published resistance of, say, 1.2 K/W at 300 LFM means: under that specific airflow, with that specific heat source size and mounting, the vendor measured 1.2 K/W. Change any variable and the number moves. The most abused variables are source size (a large copper block spreads heat into the base far better than a small die) and ambient (a "25 °C" test with the test rig's own waste heat nearby is not 25 °C at the sink).
| Spec sheet condition | Typical real-world difference | Effect on resistance |
|---|---|---|
| Large copper test block | Small real die | +10–30% (spreading loss) |
| Bare flat base, fresh grease | Clips, pads, or dry joints | +20–100% |
| Open bench airflow | Enclosure blockage | +10–25% |
| 25 °C lab ambient | 45–60 °C internal ambient | Same K/W, higher absolute temps |
| New fin array | Dust, recirculation | Creeps up over life |
Takeaway: treat datasheet resistance as the best case, then add the deltas your product actually imposes. The derating table in our heat sink sizing calculation article quantifies each one — source size, orientation, altitude, blockage — so the spec-sheet number becomes a starting point rather than a promise.
Running Your Own Verification — and What the Factory Should Send
You do not need a wind tunnel to verify a passive sink: a power resistor, a thermocouple staked into the base, and a draft-free box are enough to check natural-convection resistance within a few percent. For forced-air designs, replicate the product's fan and duct rather than borrowing a lab tunnel. Whichever route you take, the pass criteria should be written before the test: maximum case temperature at the rated power after equilibrium at the worst ambient, with the production interface material.
When the sink is custom-made, ask the factory for the test record — the dimensional inspection report is standard, and a thermal test on the actual sample closes the loop between the drawing and the physics. At BQUQ, machined and extruded sinks ship with the usual ISO9001 batch documentation, and the CNC-machined heat sinks and extruded heat sinks lines will quote from your thermal spec within 12 working hours — include your test method in the inquiry and the sample can be validated against the same conditions your product will see.
Email sc@bquq.com or WhatsApp +86 137 1315 7787 with your PDF/DXF/STEP file. An engineer reviews it and replies with price, lead time and DFM notes on working days.
Frequently Asked Questions
Q: How is heat sink thermal resistance measured?
A: A known power is dissipated through a heater or real device mounted on the sink; temperatures at the case and in the ambient are logged until equilibrium; resistance equals the temperature difference divided by the power. Airflow, if any, is measured and reported with the result.
Q: How accurate are heat sink datasheet thermal resistance figures?
A: Accurate for the exact condition tested — usually a large copper heat source on a flat base in open airflow. Real products add 10–30% or more from smaller sources, interface variations, and enclosure blockage, so treat the datasheet number as a best case and derate it.
Q: Where should I mount a thermocouple to measure case temperature?
A: In a small drilled hole (1–1.5 mm) in the device tab or sink base, with the junction staked or epoxied in place. Surface-taped thermocouples read several degrees low and can fail a good design or pass a bad one.
Q: What is the difference between LFM and CFM in heat sink specs?
A: LFM is the air velocity at the fins (linear feet per minute); CFM is the volumetric flow the fan moves. A high-CFM fan can still deliver low velocity at the fins if the flow is blocked or bypasses the sink, which is why resistance is plotted against velocity, not fan rating.
Q: How long should a heat sink thermal test run?
A: Until the temperature trend flattens — typically 30–60 minutes for metal sinks and longer for large passive assemblies. Short tests return flattering numbers because the sink is still warming toward equilibrium.
Related Articles
- heat-sink-sizing-calculation — More from the BQUQ Thermal Management engineering series.
- heat-sink-selection-workflow — More from the BQUQ Thermal Management engineering series.
- heat-sink-mounting-methods-guide — More from the BQUQ Thermal Management engineering series.
Data Sources and Verification
Tolerances, cycle times and price ranges in this guide come from BQUQ production records at our Dongguan plant, where CNC machining (±0.005 mm), stamping, custom springs and heat sinks run under one roof. BQUQ is an ISO 9001:2015 certified factory; the certificate and batch inspection reports are available on request with every quotation.
Related Resources
- About BQUQ: an ISO9001-certified source factory in Dongguan running four production lines under one roof.
- Heat sinks and thermal parts: extruded, CNC-machined and stamped options from the thermal line — extruded heat sinks, CNC-machined heat sinks, stamped heat sinks.
- Industry trends: manufacturing, material market, and sourcing analysis for buyers.
- Technical articles: engineering guides and process comparisons — more where this article came from.
- FAQ hub: quick answers on CNC, stamping, springs, and heat sinks.
- Case studies: real parts and real numbers from projects we engineered and delivered.
- Contact us: send your drawing and get a quote within 12 working hours.
Authored by the BQUQ Engineering Team. BQUQ is an ISO9001-certified source factory in Dongguan, China, running CNC machining, metal stamping, custom springs, heat sink and collet lines under one roof. Send drawings to sc@bquq.com or WhatsApp +86 13713157787 for a quote within 12 working hours. www.bquq.com
