This case study shows that neither the junction-to-ambient thermal resistance nor the junction-to-case thermal resistance will predict the die temperature of a package in a set top box application with reasonable accuracy. Instead junction-to-board thermal resistance, θJB, or junction-to-board thermal parameter, ΨJB, is proposed as the proper metric to be used to estimate die temperature under steady state conditions. Packages used in this work were 352 PBGA and 272 PBGA mounted on a 2-layer PCB which was designed to simulate a generic application. Nearby components with significant power dissipation were simulated using heater strips with specified power inputs. In the study, the die size varied from 7.1 mm × 7.1 mm to 11 mm × 11 mm. Measurements were taken in a generic set top box at various power inputs to the packages via thermal die and to the PCB via heaters.

This content is only available via PDF.
You do not currently have access to this content.