Thermal Shock Test Methods for Assemblies

Thermal shock testing moves an assembly between two temperature extremes as quickly as the equipment allows. It is not the same test as thermal cycling, even though both use temperature, and the difference determines which one is appropriate for a given question. Thermal shock is fast, severe and useful for screening, while thermal cycling is slower and closer to the conditions a product actually experiences.

Thermal Shock Compared With Thermal Cycling

Thermal cycling applies a defined rate of change, usually a few degrees per minute, and holds the assembly at each extreme for long enough that the whole mass reaches temperature. Thermal shock transfers the assembly between two baths, so the surface changes temperature almost instantly while the interior follows more slowly.

The consequence is a steep temperature gradient through the assembly. That gradient produces stresses that a slow cycle does not, which makes the test efficient at revealing weaknesses but less representative of a real application. A product that passes thermal shock may still fail in a slower cycle, and the reverse is also true.

Liquid to Liquid and Air to Air Methods

Liquid to liquid transfer is the harsher method. The assembly is moved between two baths of fluid held at the extremes, and the heat transfer is efficient because the fluid contacts the whole surface. The transfer time is short and the ramp is limited mainly by the thermal mass of the assembly.

Air to air transfer uses two chambers and a mechanism that moves the assembly between them. The ramp rate is lower because air transfers heat less efficiently, and the test is closer to a fast thermal cycle. The choice between the methods should be stated with the test, because the same temperature extremes produce different results in the two.

Sample Preparation and Monitoring

The samples have to represent the product. They should come from production, be fully assembled, include the finishes and the coatings that the product has, and have the same joints that are of interest. A coupon with a simplified construction does not reproduce the stress distribution of the real assembly.

Monitoring during the test is what turns it into a measurement rather than a survival exercise. Continuously monitoring the resistance of a daisy chained net detects a crack when it opens or when its resistance rises, which gives the number of cycles to failure rather than only the final state. Intermittent monitoring can miss an event that closes again, so continuous monitoring is preferred.

Temperature Extremes and Dwell Time

The extremes should be chosen from the product requirement rather than from a default. The hot extreme is often limited by the components or by the material, and the cold extreme by the mechanical behaviour of the solder. A test that exceeds the product limits can produce a failure that would never occur in service, which misleads the evaluation.

The dwell time has to be long enough for the whole assembly to reach the extreme. Where the dwell is too short, the interior never reaches the temperature and the test is milder than intended, which makes a marginal design look better than it is. Measuring the assembly temperature during the first cycles confirms that the dwell is adequate.

<img src="https://www.gopcba.com/wp-content/uploads/2026/06/Thermal-Management-1.jpg" alt="Cross section of a solder joint after thermal shock testing” />

Failure Criteria and Failure Analysis

The failure criterion should be defined before the test. An electrical open is an unambiguous criterion, while a resistance rise requires a threshold. Where a physical criterion such as a crack length is used, the sample has to be sectioned, which means the test is destructive and the sample size has to be planned accordingly.

The failure has to be analysed rather than only counted. A crack at the interface, a crack through the bulk alloy, a delamination of the laminate and a pad crater all indicate different root causes, and the analysis is what makes the test useful. The examination methods and the interpretation are the same as those used for solder joint assessment.

Applications and Limitations

Thermal shock is well suited to screening a process, to comparing two materials or two designs quickly and to revealing a gross weakness. It is not well suited to predicting a service life, because the gradient it imposes is not the one the product sees. Where a life prediction is required, a defined thermal cycling test is the appropriate method.

The test method should be recorded with its parameters: the extremes, the dwell, the transfer time, the monitoring method and the failure criterion. Without those, two results cannot be compared even if both are described as thermal shock tests.

Interpreting the Result

The result is a comparison, not an absolute. Two designs tested under the same conditions can be ranked, and a design can be compared against a known good reference. Extrapolating the result to a different condition requires a model with assumptions that should be stated rather than implied.

The gopcb engineering team uses thermal shock as a screening tool for process changes and reserves the slower thermal cycling method for life evaluation. Where a product is specified with a cycle requirement, the requirement is tested with the method that matches it, and the screening test is used only to detect a change before the longer test is run.

Checks Before Release

The acceptance criteria should be written before the work starts, so that the decision is made by the specification rather than by the person inspecting. The environment around the process, including temperature, humidity and cleanliness, sets limits on what the process can hold.

A result that cannot be reproduced is not a result, and reproducibility should be demonstrated rather than assumed. The tooling, the material and the profile form one system, and a change to any of them should be assessed against the other two before it is released.

Related reading: our fabrication notes, board quality and design release notes cover the same ground.

Verification and Records

Sampling is a compromise between cost and confidence, and the sample size should follow from the failure rate that has to be detected.

FAQ

Is thermal shock a substitute for thermal cycling? No. It applies a steeper gradient and a different stress distribution, so it answers a different question.

Why monitor resistance continuously? Because a crack can open and close, and intermittent monitoring can miss the event or record a later time than the true failure.

Can liquid to liquid testing damage good assemblies? It can produce failures that the product would never see, so the extremes and the dwell should be set from the product requirement.