Ceramic Package Selection White Paper

Ceramic Package Selection

A Component Engineering White Paper — choosing the right ceramic package by thermal path, power class and reliability target

Queendom LEDs · Component Engineering Group

1. Why Ceramic Package Selection Is a System-Level Decision

The package is the component’s thermal interface, its mechanical anchor and its protection against the environment. In high-power and high-reliability designs the package choice constrains the achievable drive current, the junction-to-case thermal resistance, the soldering process window and the service life of the fixture. Because those constraints are fixed at specification time, a package chosen on price alone typically reappears later as thermal derating, colour shift or field returns. This paper sets out how to choose a ceramic package by three measurable axes — power class, thermal path and reliability target — and how to avoid the recurring selection errors in this product family.

It complements two application notes in our resource center: the Ceramic Package Selection Guide, which covers the series-by-series mapping, and the Ceramic LED Absolute Maximum Ratings note, which covers derating and current limits in detail.

2. What a Ceramic Package Changes

Compared with a plastic leadframe package, a ceramic substrate such as AlN (aluminium nitride) or Al₂O₃ (alumina) changes four properties at once. It raises thermal conductivity, it lowers the coefficient of thermal expansion mismatch against the die, it removes the organic encapsulant from the primary thermal path, and it permits higher reflow and higher operating temperatures. Each of these matters, and each has a cost consequence.

It is worth separating the two dominant ceramic materials before going further. AlN is chosen when the design is thermally limited: it conducts heat several times faster than alumina and is the default substrate for 3 W and above, for COB assemblies and for automotive exterior emitters. Al₂O₃ is chosen when the design is electrically or mechanically limited but not thermally extreme: it is cheaper, still far better than plastic, and adequate for mid-power industrial parts where the board carries most of the heat. Choosing between them is a thermal calculation, not a preference.

A second distinction that decides more programs than it should is the encapsulation. A ceramic substrate can still be fitted with an organic lens or silicone encapsulant, and those organics reintroduce the very failure modes the substrate was chosen to avoid: yellowing under UV, outgassing at temperature, and delamination under cycling. Ceramic substrate plus organic encapsulation is a valid industrial part. Ceramic substrate plus inorganic encapsulation is what UV and automotive programs generally require. Confusing the two is one of the most common specification errors in this family.

3. Choosing by Power Class

The first axis is the electrical power per emitter. Power class determines the minimum thermal path you can get away with, and therefore whether ceramic is indicated at all.

4. Choosing by Thermal Path and Board

The second axis is the thermal path from junction to ambient. A ceramic package only pays off if the path downstream of the package can actually remove the heat. Specifying ceramic onto a poor board is a common and expensive error.

4.1 Path continuity

Trace the path continuously: junction → die attach → substrate → solder → thermal pad → vias → heatsink. Every interface adds resistance, and the worst interface sets the ceiling. A ceramic package bonded with an inadequate solder void is thermally worse than a plastic package bonded perfectly.

4.2 Board and via design

For ceramic packages above roughly 1 W, the board should have a dedicated thermal pad with a via farm directly beneath the package footprint, and the pad should be connected to an internal or bottom copper plane. The via count, drill diameter and copper thickness together determine whether the ceramic package’s advantage is realised or wasted.

5. Choosing by Reliability Target

The third axis is the qualification and life target. This is where ceramic packages are least substitutable: reliability targets translate directly into package and interconnect requirements.

Reliability targets should be written as measurable statements rather than as adjectives. “High reliability” is not a specification. “Lumen maintenance above 70 % of initial value after 50,000 hours at a 105 °C board temperature, with no delamination observable at 1000 thermal cycles between −40 and +125 °C” is a specification, and it points immediately at a package family. The exercise of writing the target this way usually resolves the package question before any supplier conversation begins.

6. The Cost Question: When Ceramic Pays for Itself

Ceramic packages carry a higher unit price, and the temptation is to evaluate that price against the price of a plastic alternative. That comparison is almost always wrong, because the plastic alternative is not free: it carries a cost that appears later and elsewhere. Three lines of cost are worth quantifying before deciding.

7. Common Mistakes and How to Avoid Them

7. How to Write the Specification

A ceramic package requirement should be written so that two engineers reading it arrive at the same part. The following fields are the minimum that make a requirement unambiguous, and each maps directly onto a package decision made above.

8. Worked Example: Industrial High-Bay Fixture

Consider a high-bay fixture intended for a foundry aisle: 150 W total, twelve emitters, 45 °C ambient in summer, continuous operation, five-year service interval. The package decision proceeds through the three axes in order.

The same fixture specified with 24 emitters at 6.25 W each would still generally call for ceramic, but with a lower substrate requirement and a larger board area. The point of the exercise is that the package decision follows from four numbers, not from a preference, and the numbers can be written down before any supplier is contacted.

10. A Selection Checklist

12. Conclusion

Ceramic package selection is a three-axis decision: power class sets the minimum thermal path, the board must be able to complete that path, and the reliability target determines whether an organic package can ever qualify. Selecting ceramic by default is as wrong as never selecting it. The reliable method is to derive the requirement from the environment and the life target, then pick the least expensive package that meets it.

Download the full white paper

This white paper is published as part of the Queendom LED technical library. For datasheets, test reports and application notes referenced above, see the LED Components Support and Resource Center sections. Engineering enquiries: sales@queendomlamp.com.