The test profile has four parameters. Most engineers specify two of them correctly. The ramp rate and the dwell time — the ones that actually determine what failure mode the test finds — are the ones most often guessed. A profile with the correct temperature range and cycle count but insufficient dwell time is not running the test that was specified. It is running a different test with the same name on the documentation.
The four parameters and what each one does
Temperature extremes (T-low and T-high). The range the product is cycled through. A wider range produces larger thermal excursions and faster fatigue accumulation per cycle. The range is set by the applicable standard or by the deployment environment analysis. For JEDEC JESD22-A104 Condition B: -55°C to +125°C. For automotive electronics to ISO 16750-4 Class VI: -40°C to +150°C. For consumer electronics: -40°C to +85°C. These ranges are not interchangeable. A component qualified at -40/+85 has not been qualified for automotive underhood use at -40/+150.
Ramp rate. How fast the chamber transitions between T-low and T-high, in °C per minute. Faster ramps produce higher thermal gradients across the DUT during transition, which increases stress per cycle and changes the acceleration factor. Most standards specify ramp rates between 3°C/min and 20°C/min. The achievable ramp rate under your specific DUT load — which is almost always lower than the empty-chamber spec sheet figure — must be verified before writing the profile. The calculation: required cooling power (watts) = DUT thermal mass (J/K) × ramp rate (K/s). If required power exceeds the chamber's rated capacity at the relevant temperature, the specified ramp rate cannot be achieved with that load. The decision this drives is at Benchtop or Floor-Standing Environmental Chamber? The Decision Comes Down to One Number.
Dwell time. How long the product remains at each temperature extreme before the ramp begins. The critical requirement: dwell must be sufficient for the DUT — not just the chamber air — to reach thermal equilibrium at the setpoint. This is the parameter most often under-specified. A 10-minute dwell is sufficient for a small PCB. The same 10-minute dwell applied to a 5 kg metal assembly may leave the core of the assembly 20°C away from setpoint when the ramp begins. The DUT never reaches the specified extreme. The test is less severe than intended. The only way to verify thermal equilibrium is a thermocouple attached to the DUT thermal mass — not confirmation that the chamber air has reached setpoint.
Number of cycles. How many complete hot-cold-hot transitions the product undergoes. The Coffin-Manson relationship links plastic strain per cycle to cycles to failure. A wider temperature range at the same ramp rate requires fewer cycles to represent the same accumulated fatigue as a narrower range at more cycles. JEDEC JESD22-A104 specifies 1,000 cycles for high-reliability qualification. The physics of what those cycles accumulate is covered in Temperature Cycling Testing: You're Probably Testing the Wrong Failure Mode.
Dwell time: the calculation
Thermal equilibrium time for a DUT is estimated from its thermal time constant: tau = m × Cp / (h × A), where m is mass in kg, Cp is specific heat capacity in J/kg·K, h is the convective heat transfer coefficient in W/m²·K (typically 15–50 for forced air in a chamber), and A is surface area in m². Dwell time should be at least 3 × tau for 95% equilibrium, or 5 × tau for 99%. For a 500 g aluminium assembly (Cp approximately 900 J/kg·K, surface area 0.05 m², h approximately 25 W/m²·K): tau is approximately 360 seconds. Minimum dwell for 95% equilibrium: 18 minutes. A profile specifying 10-minute dwell for this assembly is under-specifying by almost a factor of two.
This calculation is an estimate. The definitive verification is a thermocouple measurement on the DUT during a trial run. The thermocouple reading confirms when the DUT — not the chamber air — has reached setpoint temperature.
Writing the complete profile specification
A complete temperature cycling test profile specification contains: T-low and T-high in °C; ramp rate in °C/min (with statement of whether this is the required chamber air rate or the required DUT rate); dwell time at each extreme in minutes, with the equilibrium criterion stated; number of cycles; starting condition (hot or cold); and monitoring requirements during the test (powered or unpowered DUT, continuous electrical monitoring or end-of-test only). A profile missing any of these elements is ambiguous. The standard it references — IEC 60068-2-14, JEDEC JESD22-A104, MIL-STD-883 Method 1010 — fills in defaults for unspecified parameters, but those defaults may not match the deployment environment. Specifying all parameters explicitly eliminates ambiguity between the test engineer, the chamber operator, and the quality record.
Profile verification before the first formal test
Before placing product in a chamber for a formal qualification test, run the profile with the actual DUT and fixture loaded, with thermocouples attached to the DUT at multiple locations, and verify: that the chamber achieves the specified ramp rate under load; that the DUT reaches T-low and T-high within the dwell period; that the temperature gradient across the DUT during transition is within acceptable limits; and that the chamber holds setpoint within the specified tolerance during dwell. A formal test that fails because the profile was never verified against the actual load is an expensive way to discover that the dwell time was insufficient. How the chamber achieves these conditions mechanically is covered in Inside the Box: How an Environmental Test Chamber Actually Works.
Temperature cycling vs. thermal shock: the profile distinction
Temperature cycling profiles specify a ramp rate — the chamber transitions between extremes at a controlled rate in a single zone. Thermal shock profiles specify a transfer time — the DUT moves physically between two pre-conditioned zones in under 30 seconds. They are not interchangeable. A temperature cycling chamber cannot run a thermal shock test regardless of how fast the ramp rate is set. The failure mechanisms differ, the equipment differs, and the standards differ. The full distinction is at Thermal Shock Testing: Why Slow Ramps Miss the Failures That Matter. The standards governing both — IEC 60068-2-14 Test Nb for cycling and Test Na for shock — are decoded in IEC 60068 Decoded: The Global Environmental Testing Standard Behind Most Product Qualifications.
Writing the dwell time specification correctly
The most common dwell time error is specifying the chamber air temperature as the equilibrium criterion rather than the DUT temperature. A specification that says "hold at -40°C for 15 minutes" is ambiguous: does "hold at -40°C" mean the chamber air is at -40°C, or that the DUT has reached -40°C? For a small PCB, the difference is negligible. For a 3 kg aluminium assembly, the difference can be 25°C or more — the chamber air reaches -40°C while the core of the assembly is still at -15°C. The test is less severe than specified. The solder joints are cycling through a smaller temperature range than the profile claims.
Write it correctly: "Hold until thermocouple on the DUT reads within 2°C of setpoint for a minimum of 5 minutes, or hold for [X] minutes, whichever is longer." That specification is unambiguous and verifiable. Run a trial cycle with the actual DUT and fixture loaded, with a calibrated thermocouple attached to the thermal mass of the assembly, before the first formal test run. Record the time from chamber air reaching setpoint to DUT reaching setpoint. That time, plus a 5-minute margin, is your dwell specification. The calculation is in the Loaded Ramp Rate Calculator.
The ramp rate that matters: DUT rate, not chamber rate
The ramp rate in a test profile is sometimes specified as the chamber air ramp rate and sometimes as the required DUT temperature change rate. They are different numbers. The chamber air can ramp at 5°C/min while the DUT — if its thermal mass is high — ramps at 2.5°C/min. The failure mode that temperature cycling targets (CTE mismatch fatigue) is driven by the temperature change rate at the solder joint, not the temperature change rate of the chamber air. A profile that achieves 5°C/min chamber air ramp but only 2.5°C/min DUT ramp is applying half the intended mechanical stress per unit time. The test is less severe than the profile number implies.
If the applicable standard specifies a ramp rate without clarifying whether it means air or DUT, check the test method document. IEC 60068-2-14 specifies the ramp rate as the rate of change of the test chamber air temperature. JEDEC JESD22-A104 specifies that the temperature change rate should be measured at the component. When in doubt, measure both and report both.
Choosing the right test condition from the standard
JEDEC JESD22-A104 defines ten test conditions (A through J) ranging from 0/+70°C for standard commercial plastic packages to -55/+125°C for high-reliability applications. Selecting the wrong condition is more common than it should be — it usually happens when engineers default to condition B (-55/+125°C) for a product that will operate in a controlled indoor environment that never sees below -10°C. The test is more severe than required, which costs time and money but doesn't invalidate the result. The reverse error — selecting a condition that is less severe than the deployment environment — produces a qualified product that fails in the field. Select the test condition by starting with the measured or specified deployment environment, not by picking a "standard" condition from the table. The automotive context — which conditions ISO 16750 specifies for each vehicle installation location — is at Automotive Environmental Testing: The Standards Stack That Governs Every Component You Ship.
The acceptance criteria that the profile doesn't specify
A temperature cycling test profile specifies the environmental conditions. It does not specify what constitutes a pass or fail. The acceptance criteria must be defined separately — and they must be defined before the test begins, not after the results are in. Typical acceptance criteria: no electrical failures during continuous monitoring (if the DUT is powered throughout); no physical damage visible under X magnification post-test; resistance measurements within Y% of pre-test values; no change in functional performance parameters. The most rigorous programmes add destructive physical analysis on a defined sample — cross-sectioning of solder joints, X-ray of BGAs, metallographic examination of PTH barrels. A product that passes 1,000 cycles electrically but shows 60% crack propagation in the solder joints has given you important information about its remaining life margin. That information only exists if you looked for it. The standards that govern acceptance criteria — JEDEC, IEC, MIL-STD — define test conditions and sample sizes but generally leave acceptance criteria to the programme engineer or the customer specification.