Junction-to-case thermal resistance-still a myth?

After a brief review of MIL-STD-883C: Method 102 and SEMI Std. Doc. Hash 1321 for junction-to-case thermal resistance ( theta /sub JC/) measurements, an experiment to measure theta /sub JC/ of different K packages while altering the extrinsic variables is described. It is concluded that junction-to-case thermal resistance is not a physical constant: measurement conditions have a significant influence on this data. Thus, in order to make use of theta /sub JC/ data, it should be accompanied by measurement conditions, such as die size, power, and environment (type and intensity). These measurement conditions, all within the Standard's guideline, are discussed and recommendations for proper presentation and use of the information are emphasized.<<ETX>>