Thermal Shock Testing of Plated Holes and Vias on PCBs

Thermal shock testing exposes plated holes to repeated changes in temperature and then inspects them for cracks, and it is the standard way to demonstrate that a barrel can survive the assembly process. The test is simple to describe and difficult to interpret, because the result depends as much on the coupon, the cycle and the evaluation method as it does on the plating itself.

What Thermal Shock Does to a Plated Hole

The laminate expands far more than the copper when the board is heated, so the barrel is stretched along its axis on every hot excursion and compressed on every cold one. A ductile deposit deforms and recovers, while a brittle one accumulates damage until a crack forms where the stress is highest.

The damage is progressive rather than instantaneous. A hole that survives ten cycles may fail at fifty, which is why the test is defined with a cycle count and an evaluation point rather than as a single pass or fail event.

Test Methods Compared

Air to air thermal cycling moves the sample between two chambers at defined temperatures, and it is the closest to the conditions a product sees in service. Liquid to liquid testing moves the sample between two baths, which transfers heat far faster and produces a more severe result in fewer cycles.

The two methods are not interchangeable, and a result from one cannot be quoted against a limit written for the other. The faster method is used to screen processes quickly, while the slower one is used where the result has to represent service conditions. Our plating thickness notes describe the deposit being tested.

Air to Air and Liquid to Liquid

In an air chamber the sample reaches temperature slowly, and the ramp rate depends on the thermal mass of the coupon and the airflow around it. In a liquid bath the transfer is almost immediate, so the sample experiences a sharp shock that is difficult to reproduce exactly the same way twice.

The liquid method is popular because it is quick and inexpensive, and it is often used as a process control check rather than a qualification test. Where a specification demands a particular method, the same method has to be used for the qualification and for the routine checks, otherwise the results cannot be compared.

Sample Selection and Coupon Design

The sample has to represent the product, which means the coupon should carry the same hole sizes, the same layer count and the same copper thickness as the panel it represents. A coupon that only carries large holes will pass a test that the smallest hole on the product would fail.

The coupon should also include the highest aspect ratio hole on the design, because that is the hole with the most strain in the middle of the barrel. Our test coupon notes describe how the sample is organised on the panel.

<img src="https://www.gopcba.com/wp-content/uploads/2026/09/271-1.jpg" alt="Microsection of a plated hole after thermal cycling” />

Cycle Count, Temperature Range and Dwell

The temperature range sets the amount of strain per cycle, and the dwell time determines whether the sample actually reaches the extremes. A short dwell on a thick coupon produces a test that looks severe on paper and is mild in reality, because the centre of the board never reaches the set temperature.

The cycle count then determines the severity of the whole test. A hundred cycles is a common screening figure, while a product qualification may run several hundred, and the choice should be made with the intended service environment rather than by habit.

Evaluation by Microsection

Evaluation is destructive, so the sample is sectioned after the test and examined under magnification. The section shows cracks through the barrel wall, cracks that follow the interface between the copper and the laminate, and the plating thickness that produced them.

The section should be taken through the same holes that were measured before the test, so that the before and after condition of one barrel can be compared. That comparison is what makes the result useful as process feedback rather than as a certificate.

Acceptance Criteria and Interpretation

The acceptance criteria usually describe the maximum crack length as a percentage of the wall thickness, or state that no crack may extend through the full thickness of the copper. The criteria should also say how many holes are examined and what proportion must pass.

Interpretation matters because a single cracked hole in a sample of thirty is a different finding from one in a sample of three. The sample size, the location of the holes and the reading of the section all have to be recorded for the result to mean anything. Our hole copper notes describe the copper that carries the current.

Using the Result to Improve the Process

A failure points at a specific part of the process, and the useful work is to identify which. Cracks with a thin deposit at the fracture suggest a plating distribution problem, while cracks with full thickness copper suggest a ductility problem in the chemistry.

Repeating the test after a change is what confirms the diagnosis. Without a before and after result, a process adjustment is an opinion, and the same failure tends to reappear on the next lot. Our aspect ratio guide describes where the geometric limits lie.

Specifying the Test in the Order

The purchase specification should name the test method, the temperature range, the dwell, the number of cycles, the sample selection and the evaluation criteria. Where the customer already has a standard, quoting it by number and revision is clearer than paraphrasing it.

The specification should also state what accompanies the boards, because a test report without the section photographs is difficult to interpret later. Attaching the coupon layout to the order removes most of the ambiguity about what was actually tested.

Process Control and Verification

On a design of this kind, plated hole is the item that decides how the rest of the board is arranged. The process window is set by the narrowest step in the flow, so an improvement anywhere else shows up as margin rather than as yield until that step is addressed. A short note on the drawing about handling, storage or packaging is often worth more than an extra decimal place on a tolerance.

Process Control and Verification

On a design of this kind, plated hole is the item that decides how the rest of the board is arranged. The process window is set by the narrowest step in the flow, so an improvement anywhere else shows up as margin rather than as yield until that step is addressed. A short note on the drawing about handling, storage or packaging is often worth more than an extra decimal place on a tolerance.

FAQ

How many cycles should a plated hole survive? It depends on the specification, but a common screening requirement is that no barrel cracks after a defined number of cycles within a stated temperature range. Qualification for a demanding product may double or triple that figure.

Can the test be run on the product board instead of a coupon? It can, but the board is destroyed and the result is not comparable with previous lots. A coupon from the same panel gives the same information while leaving the product intact.

How does gopcb run thermal shock testing? We test coupons from the production panel, use the method and cycle count agreed with the customer, section the holes afterwards, and report the crack findings with the plating thickness so the result can be tied back to the process.