Reading PCB Tg Data: Test Methods and Thermal Qualification

Glass transition temperature, or Tg, describes a change in the mechanical behavior of a PCB resin system. It helps explain expansion and stiffness during heating, but its value is meaningful only alongside the measurement method and specimen condition. A material comparison that mixes DSC, TMA and DMA results can rank laminates incorrectly before the design work even begins.

For a production board, the useful question is whether the complete material and construction can survive assembly and service. Answering that requires Tg, expansion data, thermal endurance information and evidence from the intended manufacturing process.

What Changes at the Glass Transition?

Below its transition region, the cured resin is relatively stiff. As temperature rises through that region, molecular motion increases and the resin becomes more compliant. The glass reinforcement continues to influence the composite, so a glass-reinforced laminate does not behave like an unsupported block of resin.

Tg is neither a melting point nor a universal damage threshold. A thermoset PCB substrate does not simply melt when its Tg is crossed. Nevertheless, lower stiffness and changed expansion behavior can impose additional movement on copper features and bonded interfaces. Cooling may reverse the resin transition while leaving any cracks or interfacial damage accumulated during the excursion behind.

This distinction explains why a board can pass briefly above Tg during a qualified assembly process yet still require a different assessment for sustained operation. Temperature, time, moisture, mechanical constraint and repeated exposure all matter. A single Tg number contains none of that history.

Compare Test Methods Before Comparing Numbers

The common methods observe different physical responses. Differential scanning calorimetry, or DSC, detects a heat-capacity change. Thermomechanical analysis, or TMA, follows dimensional behavior. Dynamic mechanical analysis, or DMA, examines mechanical response under an oscillating load. Reporting conventions within a method also matter.

Method	What the measurement follows	What to check on the report
DSC	Change in thermal response associated with heat capacity	Heating program, reported transition criterion and scan history
TMA	Specimen movement as temperature changes	Measurement axis, load, expansion slopes and specimen preparation
DMA	Temperature-dependent stiffness and damping	Frequency, deformation mode and the criterion used to assign Tg

Table 1: Questions to ask when comparing Tg measurements; methods are not interchangeable.

Do not apply a fixed numerical correction to translate one method into another. A useful example appears in Panasonic's low-Dk material data: the listed R-A555(W)/R-A550(W) system has a Tg of 160°C by TMA and 200°C by DMA. Those are method-specific entries for the same listed system, not two interchangeable material grades or a universal 40°C offset.

Sample conditioning deserves similar attention. Moisture and cure state can affect the response. If an incoming-material result differs from a finished-board result, first check whether the samples received equivalent preparation and thermal history. A discrepancy warrants investigation; it does not immediately prove that the supplier sent the wrong laminate.

*Figure 1: Three measurement routes to Tg. The illustration is conceptual and contains no measured curves or acceptance limits.*

Connect Tg to Expansion and Via Loading

The most consequential dimensional change is often through the board thickness. Glass fibers restrain much of the in-plane expansion, whereas the resin-rich thickness direction can expand more strongly. Copper lining a plated hole must accommodate movement imposed by the surrounding laminate.

Look for both expansion coefficients below and above the transition, as well as total Z-axis expansion over a stated temperature interval. Two materials with similar Tg can have different expansion slopes. Their accumulated thickness changes during the same reflow cycle can therefore differ.

For a first-order comparison, divide the temperature excursion into intervals and add the expansion contribution from each interval. With a constant coefficient within an interval, relative expansion is approximately the coefficient multiplied by the temperature change. If the transition is crossed, using only the room-temperature coefficient understates the movement.

This calculation is a screening tool. It does not predict via fatigue by itself. Barrel length, copper distribution, plating quality, hole geometry and the thermal cycle influence the strain that the interconnect actually experiences. A higher Tg is useful when it addresses the identified failure mechanism; it is not a substitute for checking total expansion.

Keep Tg, Td and Time-to-Delamination Separate

Thermal decomposition temperature, Td, describes a chemical-degradation measurement, commonly associated with a specified mass loss. Time-to-delamination measurements such as T260 or T288 describe exposure at a fixed temperature under a particular test procedure. Long-term thermal ratings address another question again.

These properties are related to the material system but cannot be inferred from one another. A board selected for repeated assembly exposure may need good delamination resistance even when its service temperature is modest. A board intended for prolonged elevated-temperature operation needs evidence relevant to that duration, not just a successful short reflow trial.

When reading a comparison, retain the measurement conditions beside the values. A time-to-delamination figure measured with copper removed should not silently be treated as equivalent to a result from a copper-clad specimen. Equally, a high Td does not authorize operation close to that value. A destructive test endpoint is not a recommended service condition.

Translate the Application into a Qualification Plan

Start by documenting the board's temperature history. Include fabrication exposures where relevant, the planned number of assembly passes, any allowed rework and the expected service cycle. Distinguish oven settings from measured temperatures on the actual assembly.

Next, identify which features carry the greatest mechanical risk. A thick board with long plated holes presents a different problem from a thin board carrying a large area-array package. In the first case, through-thickness movement may dominate the investigation. In the second, package-to-board displacement and local bending can deserve more attention.

Then compare candidate materials against the requirement that actually matters. Request the exact laminate and prepreg designations, method-specific Tg data and the corresponding expansion and thermal-exposure information. Do not use broad labels such as standard, mid-Tg or high-Tg as if they were complete specifications.

A useful release question is: if the board house proposes another grade with the same headline Tg, what additional evidence would make it acceptable? The answer may include compatible dielectric properties, equivalent construction availability, expansion behavior, solder-process limits and test results on representative coupons. Defining that evidence before fabrication avoids an uncontrolled material substitution later.

Manufacturing Determines Whether the Material Is Used Correctly

A qualified material still needs a compatible fabrication process. Lamination must follow a validated resin-system profile so that flow and cure are appropriate for the construction. Drilling and hole preparation must leave surfaces suitable for reliable metallization. Storage and moisture handling must follow the relevant material and assembly guidance.

Avoid blanket instructions to increase cure temperature, lengthen chemical treatment or bake every board. A change intended to help one resin system can harm another construction or surface finish. Process settings belong to a validated manufacturing recipe, not a generic high-Tg rule.

For a discrepancy investigation, keep three records together: incoming material identity, the process history and the test report. Comparing Tg values without those records can hide a conditioning difference or a process deviation. Conversely, those records can show that a measured difference is expected rather than evidence of a faulty batch.

Inspection should also address the relevant failure path. A material-property test and an electrical test answer different questions. Neither replaces an appropriate assessment of plated interconnects after the intended thermal exposure.

Conclusion

Read Tg as a method-dependent material property within a larger thermal specification. The strongest qualification links the reported value to expansion, time at temperature, board construction and controlled manufacturing. That makes the material decision explainable and makes later substitutions easier to evaluate.

Frequently Asked Questions

Q1: Can two datasheets report different Tg values for the same resin system?

Yes. Different measurement methods and reporting criteria can produce different values. Check method, conditioning and specimen history before interpreting the difference as a change in grade or quality.

Q2: Does assembly above Tg mean that the PCB has failed?

No. Passing above Tg changes the resin response but does not automatically establish damage. Acceptance depends on the qualified assembly exposure and evidence that the construction and interconnects remain suitable afterward.