On-Wafer Measurement Errors Due to Unwanted Radiations on High-Q Inductors

Abstract

This paper investigates the disagreements that may occur between on-wafer measurements and electromagnetic (EM) simulations of high-Q inductive devices. Such disagreements are highlighted on a planar spiral inductor and a 3-D solenoid which exhibit measured maximum Q-factors of 27 and 35 respectively, while 42 and 45 were expected from EM simulations. Both devices are fabricated on high-resistivity substrates. A radiative interaction is identified between RF probe and inductive device under test. By using EM simulations, extra-losses associated with this parasitic effect are fully modeled through the calculation of radiation and dissipation related Q-factors. Adjustments of onwafer probe setup are proposed to reduce this parasitic effect. Finally, the 3-D solenoid inductor is characterized using a new experimental fixture, and the maximum Q of 45 predicted by EM simulation is retrieved.

High $Q$ Coil Measurement for Inductive Power Transfer

“…When connecting an IPT coil directly to a coaxial measurement system an unintentional conductor, caused by the electrically thick cable shield, is connected to the coil. This causes additional radiation losses as the coil approaches its self-resonant frequency [23] and also causes the resonant frequency of the coil to shift.…”

High-Performance Compact 3-D Solenoids for RF Applications

“…Parasitic interactions between the on-wafer setup and the inductor may have very visible effects when the Q-factor is high. These effects are minimized by tuning the internal dimensions and width of the metal ring surrounding the device under test (DUT) and connected to the ground tips of GSG probe [10]. Finally, inductance and Q-factor are derived from the measured Γ without de-embedding access wires and pads in order to keep the best measurement accuracy and to avoid additional errors.…”

“…As a consequence, Q and SRF value are underestimated, up to 18 % and 5 % respectively, while inductance values are slightly overestimated (+0.25 nH). However, the comparison with electromagnetic (EM) simulations carried out under the same conditions is more accurate [10].…”