The design, build and verification of the Forcesix measurement setup

Abstract

The goal of this study is to develop a measurement setup that can accommodate the water-cooled modules present in the innovative e-beam lithography tool developed by Mapper Lithography and to perform flow verfication measurements on them. The overall objective is to verify the stage-stability requirements set for the Matrix-machine with regards to FIV by observing the induced cooling forces. This requires measuring in 6-DOF, over a wide frequency range (10-300 Hz) and at a very low noise level (e-11 N^2/Hz). Most challenging for this design is to be able to observe the FIV while in the presence of a variety of dominant environmental disturbances. The final design consists of a mass-optimized triple mass-spring-damper (MSD) system, weighting 828 kg. It uses six low-noise piezoelectric force transducers to observe the 6-DOF reaction-forces exerted by the modules. These modules under testing are supported by the piezos through custom designed stiff-flexible struts with a high axial/radial stiffness ratio. This improves measured signal and protects the piezos from damaging bending moments. This sensitive part of the measurement setup is isolated from floor vibrations by a double MSD Vibration Isolation (VI) platform (granite stones on airmounts). Acoustic shielding has been achieved by a custom designed enclosure that disconnects at the bottom granite stone. Flow vibrations in the supply tubing are discharged at various stages. Water flow is provided under constant pressure and flow rate by a hydrostatic pressure vessel. This prevents measuring distinct resonances from asynchronous motor characteristics inherent to a centrifugal pump. Verification measurements have been performed showing a noise floor characteristic at the level of the theoretically predicted effect of all disturbances combined (2.5 e-11 N^2/Hz). The main findings of this study are: - when aiming to measure very low-level dynamic reaction forces (0.35 microN-rms) in the presence of dominant disturbances that transmit through parasitic stiffnesses, quartz piezoelectric sensors proof to be a better solution when compared to (seismic) accelerometers. - flow vibrations induced in supply tubing can have a significant impact on the measured signal, if the stiffness train that connects the sensor with the measurement setup is relatively low. An effective method to minimize this disturbance is to discharge the bulk of the input to different stages of the vibration isolation platform, if present. - of all disturbances, environmental acoustics have shown to be most difficult to shield. The most effective means to reduce its effect is to fully enclose the sensitive part of the measurement setup and to rigidly connect this casing to a heavy mass with an attractive transfer path to the sensor e.g. the bottom stage of a two MSD VI platform. - when measuring direct forces using sensitive piezoelectric sensors that cannot withstand transverse loading / bending moments, stiff-flexible support struts with a high axial/radial stiffness ratio (roughly > 500) are found to be a solution. Concluding, the design process detailed in this thesis describes a method on how to effectively design a low-level reaction force measurement system, while in the presence of a variety of disturbances. In the report, generic design guidelines are listed that can serve as a reference to others when aiming to measure low-level FIV.