The Impact of KGD and Module Repair on Multichip Module Cost

The impact of known good die (KGD) probability on multichip module (MCM) yield and cost has been modeled and systematically analyzed. The current work extends the researchers’ previous MCM modeling effort involving single chip populations (a single KGD probability) to modules containing complex, multiple chip populations. Most of the analysis is performed on modules with dual populations (that is, two chip types). In order to develop a total cost picture for an MCM versus the respective KGD probabilities of the underlying chip populations, it was necessary to develop new algorithms to calculate the number of modules necessary to ensure one working MCM, the chip cost as a function of KGD, and chip repair cost as it relates to the complexity of module testing and the actual removal and replacement operations. The combination of these models and algorithms produces cost surfaces in dual (KDG) probability space that contain optimum or minimal points. Associated with the cost minimums are specific KGD probabilities for each chip type in the population. Thus, one only pays for improved KGD probability up to the values that minimize the overall module cost. Repair has a direct impact on the overall module yield and cost. The authors have shown that the first repair makes the most significant improvement in yield and cost reduction. Higher numbers of repairs also have positive impact in large modules. In the dual-population case, there is an optimum number of repairs to minimize cost, and as repair costs increase; this optimum number appears to decrease.