System-on-package (SOP) architectures take advantage of compact, high-performance designs to place the maximum amount of functionality on a subsystem that can then be mounted on a lower-cost, lower density interconnect board. Embedding passive components is a key technology in achieving these goals since this enables smaller SOP substrate footprints or, equivalently, higher functional density, along with better power distribution, increased design flexibility and improved reliability. The resulting footprint areas of integrating capacitors will have more of an effect on the layer count of SOP assemblies than will integrating resistors due to the rather low specific capacitances of most embeddable dielectrics, but the situation is improving steadily. It may be necessary to use two different dielectric materials to cover the entire required range. The inherently lower parasitic inductance of embedded capacitors makes them much more useful in decoupling than surface mount capacitors, enabling more robust power distribution and decreased power/ground noise. The key to this performance enhancement in large boards is the use of a thin dielectric to decrease the inductance but, for the smaller SOP substrates, the dielectric constant must also be high to provide sufficient decoupling capacitance in the reduced area.
No takes yet. Share an insight, caveat, or question.
R. Ulrich (2004) studied this question.
Synapse has enriched 5 related papers. Consider them for comparative context: